INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

SKELETAL MUSCLE HAS EXTRAORDINARY abilities to adapt to changing conditions, including the ability to atrophy and hypertrophy in response to altered loading. Hypertrophy of skeletal muscle fibers in response to an increased mechanical load is a complex event marked by large-scale remodeling of the fiber architecture. The process involves the activation and subsequent fusion of satellite cells to the muscle fibers (29, 31) and the increased synthesis of structural and contractile proteins. Decreased protein degradation and increased protein synthesis cause the accumulation of contractile and structural proteins, resulting in the addition of myofibrils and an accompanying increase in fiber cross-sectional area (CSA).

Muscle fiber type is also plastic and believed to be regulated primarily by contractile activity via calcium-mediated signaling pathways. The calcium/calmodulin-dependent phosphatase calcineurin has recently been shown to play an important role in fiber-type transitions (9, 11) and may also be involved in compensatory fiber hypertrophy in rodent muscle (11). However, skeletal muscle fiber-type transition is often dissociated from hypertrophy (e.g., electrical stimulation or endurance training effects). Therefore, it seems likely that contractile activity-related pathways interact with a mechanosensitive signaling pathway to determine phenotypic effects.

Nitric oxide (NO) synthase (NOS) is a candidate for the role of mechanosensor in skeletal muscle fibers. Both the endothelial (eNOS) and neuronal (nNOS) isoforms of NOS are expressed in skeletal muscle (23, 32) and are upregulated in skeletal muscle after exercise training (4) or electrical stimulation (27). Furthermore, changes in NOS expression and activity are directly related to loading patterns, with unloading causing downregulation and reloading, resulting in a normalization of expression (28, 35). These studies implicate NO in the responses of skeletal muscle to activity. Finally, pivotal work by Tidball et al. (36) has demonstrated that NOS activity is necessary for load-induced upregulation of structural cytoskeletal proteins in vivo and for stretch-induced increases in gene expression in cultured myocytes. These data suggest that NOS may be a key transducer of mechanical strain on the cytoskeleton into chemical signals affecting gene expression. Furthermore, because eNOS and nNOS activities are calcium/calmodulin dependent, NOS activity may be an important site of modulation of the calcium-signaling pathway in skeletal muscle. Therefore, we tested the hypothesis that in vivo pharmacological blockade of NOS activity in the rat would prevent skeletal muscle fiber-type transition and fiber hypertrophy in the plantaris during chronic overload. Our data support the conclusion that endogenous NO production is an important component of the signaling pathway that is responsible for adult skeletal muscle adaptations to mechanical overload.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Animals. Thirty-eight female, Sasco-line Sprague-Dawley rats (4 mo old, ~250 g) were purchased from Charles River Laboratories (Wilmington, MA). Rats were individually housed in polycarbonate cages in the Texas Woman's University Animal Facility on a 12:12-h light-dark cycle (light 0700-1900). Before the experiment was begun, the animals were randomly assigned to one of four groups: control (Con)-nonoverloaded (Non-OL) (Con-Non-OL, n = 9), Con-overloaded (OL) (Con-OL, n = 9), NOS blocked-Non-OL [N^G-nitro-L-arginine methyl ester (L-NAME)-Non-OL, n = 10], and NOS blocked-OL (L-NAME-OL, n = 10). The protocol for this study was approved by the Institutional Animal Care and Use Committee at Texas Woman's University.

Systemic inhibition of NOS activity. The pharmacological inhibition of NO synthesis was achieved in the NOS-blocked groups (L-NAME-OL and L-NAME-Non-OL) by administration of the competitive NOS inhibitor L-NAME (Cyclopss Biochemical, Salt Lake City, UT) in the rats' drinking water at a concentration of 0.75 mg/ml. Pilot experiments indicated that this concentration produced a dose of ~100 mg · kg body mass¹ · day¹. This dose has been shown to effectively inhibit NO synthesis (33) and skeletal muscle NOS activity (3) in the rat without apparent adverse effects.

Synergist ablation surgery. Chronic overload of the plantaris was induced by surgical, bilateral removal of the synergist muscles to the plantaris. The rats were anesthetized with an intraperitoneal injection of a cocktail containing ketamine (75 mg/kg body mass) and xylazine (10 mg/kg body mass). With the use of aseptic technique, a midline incision was made in the skin of the hindlimbs from the popliteal fossa to the Achilles tendon region. A second longitudinal incision was made through the hamstrings, exposing the proximal heads of the gastrocnemius and the soleus muscle. The proximal origins of the two heads of the gastrocnemius were carefully isolated from other muscles, nerves, and vessels and then sectioned. Next, the distal tendon was transected to remove the gastrocnemius without disturbing the plantaris. Finally, the soleus muscle of the rats was isolated in the same manner, sectioned, and removed. The hamstring incision was closed with 4-0 silk suture. The overlying skin was closed with sterilized metal wound clips and treated with a topical antibiotic cream to avoid infection.

Experimental protocol. The day before the animals underwent surgery, L-NAME was added to the drinking water of the two appropriate groups (L-NAME-OL and L-NAME-Non-OL). Twenty-four hours after the L-NAME treatment was begun, two groups received bilateral synergistic ablation surgeries (Con-OL and L-NAME-OL), and the other two groups had sham surgeries performed (Con-Non-OL and L-NAME-Non-OL). After surgery, the rats were housed individually and allowed to ambulate freely in the cage for 14 days. Drinking water was replaced, body mass was recorded every other day throughout the experimental period, and the dose of L-NAME, per kilogram of body mass, was calculated for each L-NAME rat. In addition, the rats were examined to ensure proper healing of the wounds. On day 15, the rats were anesthetized with an intraperitoneal injection of a cocktail containing ketamine (75 mg/kg body mass) and xylazine (10 mg/kg body mass). After blood was collected by direct cardiac puncture, the rats were killed via cervical dislocation. Immediately after the rats were killed, the plantaris muscles were removed, trimmed of connective tissue and fat, weighed on an analytic balance, frozen in liquid nitrogen at resting length, and then stored at 80°C for subsequent histochemical and biochemical analyses.

NOS assay. Tissues were homogenized in 20 volumes of homogenizaton buffer containing 25 mM Tris · HCl (pH 7.4), 10 mM EDTA, and 100 µg/ml phenylmethylsulfonyl flouride. The homogenate was centrifuged at 20,000 g for 15 min at 4°C, and the pellet (membrane fraction) was resuspended in one-half of the original volume of homogenization buffer. Protein concentration of the membrane fraction was determined by a dye-binding technique (Bio-Rad, DC Protein Assay) with the use of BSA as the standard. NOS activity was quantified by the conversion of [³H]arginine to [³H]citrulline by using a modification of the procedure described by Rouet-Benzineb et al. (30). Aliquots from the resuspended pellets (100 µl) were incubated in 50 mM HEPES (pH 7.4) with 100 mM [³H]arginine (50 Ci/mmol), 120 µM NADPH, 60 mM L-valine, 12 mM L-citrulline, 1.2 mM MgCl₂, 0.2 mM CaCl₂, 10 µg/ml calmodulin, 3 µM tetrahydrobiopterin, 1 µM FAD, and 1 µM flavin mononucleotide. The reaction was carried out for 30 min at 30°C with and without 2 mM L-NAME and terminated by adding 2 ml of 20 mM HEPES (pH 5.5) containing 2 mM EDTA. Calcium-independent activity was determined by the addition of 1 mM EGTA to the reaction mixture. After incubation, samples were mixed with 700 µl of equilibrated Dowex AG50W-X8 resin (sodium form) and 4 ml of termination buffer and then centrifuged briefly to sediment resin. [³H]citrulline was quantified in the supernatant by liquid scintillation. Specific NOS activity was obtained by subtracting [³H]citrulline formed in the presence of L-NAME from the total [³H]citrulline recovered. Calcium-independent activity was determined by evaluating [³H]citrulline formation in the absence of calcium (the presence of EGTA). All samples were run in triplicate for total and calcium-independent NOS activity.

Total protein assay. Portions (~50 mg) of the plantaris samples were weighed and then homogenized in deionized water. Next, 6 N sodium hydroxide were added to the homogenates in a sufficient quantity to achieve a final normality of 1.0. The samples were then allowed to digest at 37°C overnight. The samples were neutralized with 6 N HCl before a dye-binding protein assay (Bio-Rad, Hercules, CA) was performed.

Myosin heavy chain composition. The SDS-PAGE protocol of Talmadge and Roy (34), as modified by Bamman et al. (5), was used to separate all four isoforms (I, IIa, IIx, and IIb) of myosin heavy chain (MHC) found in adult rat skeletal muscle. Plantaris muscle sections (50 mg) were homogenized in 1 ml of homogenization buffer. The buffer consisted of 250 mM sucrose, 100 mM KCl, 5 mM EDTA, and 20 mM Tris, pH 6.8. The homogenates were centrifuged for 15 min at 2,500 g in the cold (5°C). The supernatant was removed and discarded without disturbing the myofibril pellet. The pellet was resuspended in 500 µl of wash buffer (175 mM KCl, 2 mM EDTA, 0.5% Triton X-100, and 20 mM Tris, pH 6.8). The same centrifugation was repeated, and the supernatant was once again discarded. The wash buffer process was repeated, and the myofibril pellet was resuspended in 500 µl of resuspension buffer (150 mM KCl and 20 mM Tris, pH 7.0). Myofibrillar protein samples were diluted to a constant protein concentration (0.2 µg/µl) by sample buffer [1.0% -mercaptoethanol, 4.0% SDS, 16.0% 1.0 M Tris (pH 6.8), 20% glycerol, and 0.2% bromophenol blue], boiled for 5 min at 100°C, and packed in wet ice for 5 min. Samples were then stored at 80°C until the gel was run.

Histochemistry. A series of frozen 12-µm-thick cross sections were cut from the midportion of the plantaris with a cryostat (Leica Instruments, Nussioch, Germany) that was set at 20°C. The sections were mounted on slides, air dried, stained for myofibillar ATPase (7), and mounted with Crystal Mount (Fisher Scientific) aqueous mounting medium. Individual fibers were sampled from each section, as described by Armstrong and Phelps (2), for microphotometric analysis of CSAs and fiber-type proportions. Briefly, each muscle section was subjectively divided into three regions (superficial, middle, and deep). One field of view from each region was randomly selected and digitized for analysis with an image analysis system that was calibrated by a stage micrometer. The system consisted of a Leed Olympus BX60 microscope (Leed Instruments, Irving, TX), along with a CCD-72 series camera (DAGE-MTI, Michigan City, IN). Every fiber in the each field of view was counted (~300 fibers/section) and classified as type I, IIA, or IIB/X, based on ATPase staining intensity in serial sections after preincubation in alkaline (pH 10.4) or acid (pH 4.2 and 4.6) conditions. "Edge effects" were controlled by examining only those fibers completely visable within the field of view. The Scion Image program was used to determine the percentage of the total fiber number comprised by each fiber type, the percentage of the total muscle CSA comprised by each fiber type, and the average CSA for each fiber type. Fiber-type percent by CSA was calculated as the sum of CSAs for a given fiber type divided by the sum of CSAs for all fiber types times 100. Fiber-type percent by number was calculated as the total number of fibers of a given type divided by the total number of fibers analyzed for that muscle times 100. The fiber-type percent by number gives an assessment of the proportion of fibers expressing a given fiber-type phenotype. On the other hand, the fiber-type percent by CSA assesses the proportion of the muscle area composed of a given fiber type, independent of differences in fiber size among the different types.

Statistical analyses. Each of the dependent variables was analyzed by using 2 × 2 (Con/L-NAME × Non-OL/OL) factorial ANOVA to determine main effects and interactions. Tukey's post hoc test was applied where significant interactions were found to determine individual group differences. Significance was established as P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

For the L-NAME groups, the dose of the NOS inhibitor received by each rat was calculated post hoc, based on the concentration of L-NAME in the drinking water and the volume of water consumed each day. Mean (±SE) L-NAME dose for the two L-NAME groups did not differ (L-NAME-Non-OL = 114.2 ± 7.4 mg · kg¹ · day¹; L-NAME-OL = 121.6 ± 10.6 mg · kg¹ · day¹; P > 0.05). Furthermore, the mean (±SE) volume of water consumed did not differ among any of the experimental groups (Con-Non-OL = 42.9 ± 1.8 ml/day; Con-OL = 42.5 ± 1.7 ml/day; L-NAME-Non-OL = 39.9 ± 1.3 ml/day; L-NAME-OL = 39.4 ± 1.3 ml/day; P > 0.05).

Morphometric characteristics. Initial body mass did not differ among groups. Mean body mass was significantly reduced in both OL groups at day 3 (2 days after ablation surgery) but returned to values similar to those of the Non-OL groups by day 5. On the last day of the treatment period (day 15), mean body mass (±SE) was as follows: Con-Non-OL = 249.7 ± 5.9 g, Con-OL = 244.8 ± 4.3 g, L-NAME-Non-OL = 243.6 ± 4.2 g, L-NAME-OL = 230.7 ± 6.7 g. The final mean body mass of the L-NAME-OL group was 7.6% lower (P < 0.05) than that of the Con-Non-OL group. Body mass did not differ among groups at any other time points during the treatment period.

NOS activity. NOS activities in the particulate fraction of plantaris homogenates are presented in Table 2. L-NAME treatment reduced total and calcium-dependent NOS activity, compared with the Con groups. However, no significant effects of overload on total calcium-dependent or calcium-independent NOS activity were observed within the Con or L-NAME groups. Calcium-dependent NOS activity was essentially undetectable in the L-NAME groups, whereas total NOS activity was reduced by an average of 53%. L-NAME treatment did not significantly affect calcium-independent NOS activity in the plantaris.

MHC isoforms. Figure 1 reports the mean (±SE) MHC isoform profile for the plantaris as a percentage of the total MHC pool. The proportion of type I MHC was significantly higher (+44%) in the Con-OL group, whereas the percentage of type IIb MHC was significantly lower (32%), compared with the Con-Non-OL group. No significant differences in the relative proportions of the MHC isoforms were observed in the L-NAME groups compared with the Con-Non-OL group. These data indicate a significant interaction between overload and L-NAME treatment on MHC isoform expression, such that L-NAME treatment prevents the overload effect on myosin isoform transition.

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Myosin heavy chain (MHC) composition of plantaris determined by SDS-PAGE on myofibrillar protein. L-NAME, NG-nitro-L-arginine methyl ester; Non-OL, nonoverloaded. * Significantly different from control-Non-OL group, P < 0.05.

Histological measurements. Histochemical analyses of the plantaris, including mean (±SE) fiber CSAs and histochemically determined fiber-type profiles, are presented in Table 3 and Fig. 2.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Cross-sectional area (CSA) of histochemically determined fiber types in nondehydrated 12-µm cryostat sections. OL, overloaded. Significantly different from * control-Non-OL group and  L-NAME-Non-OL group: P < 0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

These data support our hypothesis that endogenously produced NO plays an important role in the mediation of skeletal muscle hypertrophy and fiber-type transitions during chronic mechanical overload. We report that blockade of NOS activity with L-NAME significantly blunts the overload-induced fiber hypertrophy and accumulation of muscle mass in the plantaris. Furthermore, L-NAME treatment moderately inhibited the effect of plantaris overload on the expression pattern of MHC isoforms in this muscle. To our knowledge, these data are the first to document an effect of NOS activity on muscle mass and fiber type in a chronic skeletal muscle hypertrophy model. These observations extend the work of Tidball et al. (36), who first identified the NOS enzyme as a mechanotransducer that exerts control over cytoskeletal genes in muscle, and we confirm a functional significance for NOS signaling during hypertrophy.

Our study employed in vivo models of muscle hypertrophy (surgical synergist ablation) and NOS inhibition (systemic L-NAME treatment) to identify the meaningfulness of NOS signaling in skeletal muscle. However, it should be noted that systemic blockade of NOS activity has physiological consequences throughout the body. Most notably, L-NAME treatment has significant effects on hemodynamics (i.e., increased total peripheral resistance and mean arterial pressure; Ref. 10). It is certainly conceivable that these side effects could influence the response of the plantaris to chronic overload.

Effects of chronic L-NAME treatment on skeletal muscle blood flow. The importance of NO in the regulation of skeletal muscle blood flow remains controversial. Recent reports failed to show an acute effect of L-NAME on resting skeletal muscle blood flow in the rat (10) or exercise-induced hyperemia in humans (13). Conversely, other studies have identified NOS activity as an important mediator of skeletal muscle hyperemia during exercise in the rat (18, 24) and horse (19). Nevertheless, we believe that the observed effects in the present study are more likely due to blockade of intramuscular NOS activity for the following reasons. First, skeletal muscle blood flow is not necessarily elevated during the hypertrophy process. Armstrong et al. (1) showed no difference in blood flow in the remaining plantaris and soleus muscles 30 days after the synergist gastrocnemius was removed. Additionally, Egginton et al. (12) found that capillary-to-fiber ratio was not associated with the changes in mRNA or protein distribution of several growth factors in the stretch-overloaded extensor digitorum longus muscle. Measurements of muscle blood flow showed no change after 2 and 8 wk of overload (12). Second, chronic L-NAME treatment in our Non-OL group did not alter plantaris mass. Furthermore, 3 wk of L-NAME treatment and treadmill exercise reduced arteriogenesis but did not change hindlimb muscle mass-to-body mass ratio (22), suggesting that, if L-NAME does cause resting or exercise-induced muscle ischemia in the rat, it does not directly affect skeletal muscle mass.

Central effects of chronic L-NAME treatment. NO is also produced endogenously in the central nervous system (CNS) and influences a myriad of physiological systems. Therefore, systemic NOS blockade via L-NAME has central side effects. One prominent role of NO in the CNS is the regulation of autonomic neural activity. Although controversy exists, most studies indicate that endogenously produced NO reduces overall sympathetic outflow (reviewed in Ref. 39). Therefore, L-NAME administration would be expected to increase sympathetic activity. This and/or other centrally mediated effects could influence the skeletal muscle hypertrophy response through multiple pathways. For example, central effects of NOS blockade may affect autonomic control of regional blood flow and influence the production of growth factors, including growth hormone, insulin-like growth factor I, and insulin. These potential central effects of NOS activity on skeletal muscle hypertrophy warrant further study.

Effects of L-NAME treatment on insulin responses. L-NAME administration results in a decrease in glucose tolerance (3). This effect was attributed to a reduced insulin response to a glucose challenge, rather than to direct effects on skeletal muscle. Because insulin is known to stimulate protein accretion, this L-NAME-induced reduction in insulin release could interfere with skeletal muscle hypertrophy. After exercise, insulin participates in muscle hypertrophy by inhibiting the increase in protein degradation (37). Our observation that plantaris mass did not differ between Non-OL Con and L-NAME groups argues that L-NAME does not significantly alter the normal balance of protein synthesis and degradation during the first 2 wk of L-NAME treatment. It is possible, however, that L-NAME may have blunted an overload-induced increase in growth factor production. Because we did not measure levels of circulating insulin or other growth factors in the present study, we cannot determine the contribution of this potential mechanism to the reported L-NAME effect.

NO signaling in skeletal muscle. NO is produced endogenously in skeletal muscle by both nNOS and eNOS (26) and is known to be an important signaling molecule in skeletal muscle. Mechanical activity influences NO production in skeletal muscle by regulating NOS activity in the short term (28) and protein expression of nNOS and eNOS in the longer term (4, 35). In a series of elegant studies, Tidball and colleagues have demonstrated that NO production in skeletal muscle responds to loading (35), influences gene regulation (36) and sarcomere addition (21) in response to various models of altered use, and controls calpain-mediated proteolysis (20). Although skeletal muscle expresses both calcium-dependent isoforms of the NOS enzyme, it is believed that nNOS, a subsarcolemmal, costameric protein associated with dystrophin and -syntrophin (17), is the most responsive to altered use and loading patterns in muscle (8, 28, 35, 36). Koh and Tidball (21) used a selective nNOS inhibitor to show that stretch-induced sarcomere addition is dependent primarily on nNOS activity. Therefore, future studies should focus on the role of skeletal muscle nNOS during overload-induced hypertrophy.

NOS activity and skeletal muscle hypertrophy. Given our hypothesis that endogenous NOS activity is important for hypertrophy, we expected NOS activity to be elevated in the hypertrophied plantaris. However, neither calcium-dependent nor total NOS activity was altered in the plantaris of the Con-OL group compared with the Con-Non-OL group (Table 2). The citrulline assay used in this study assesses maximal in vitro enzyme activity, which is directly related to enzyme content. Because the activity of the constitutive isoforms of NOS is regulated by calcium-calmodulin binding, our overload treatment may have increased endogenous production of NO without changing NOS protein content or maximal activity.

Control of skeletal muscle phenotype. These data are the first to report a role for NO in controlling skeletal muscle fiber type. Recent work has identified calcineurin as an important regulator of fiber-type-specific genes in skeletal muscle (9). Dunn et al. (11) showed that inhibition of calcineurin phosphatase activity with cyclosporin A prevents fiber-type transition and hypertrophy in mouse skeletal muscle after 30 days of overload. Contrary to these reports, Bodine et al. (6) failed to support a crucial role for calcineurin in the load-induced slow myosin expression and hypertrophy in skeletal muscle. Instead, these authors implicated the phosphatidylinositol 3-kinase/Akt pathway as a key mediator of skeletal muscle hypertrophy and fiber-type changes. Future studies should examine how the activation of muscle NOS may interact with these pathways.