It has been demonstrated that endurance exercise and chronic acceleration, i.e., hypergravity, produce comparable adaptations in a variety of physiological systems, including decreased adiposity, increased energy metabolism, and altered intermediary metabolism. Similar adaptations have not been demonstrated for skeletal muscle per se. To further differentiate between these general responses with respect to gravity and exercise, this study tested the hypothesis that chronic exercise (voluntary wheel running) and chronic acceleration (2 G via centrifugation) will induce similar changes in muscle myosin heavy chain (MHC) isoform expression in rat plantaris, a fast extensor, and in rat soleus, a slow “antigravity” extensor. The experimental design involved four groups of mature male rats (n = 8/group): 1 G and 2 G with running wheels, and 1 G and 2 G controls without running wheels. The primary observations from the study were as follows: 1) 8 wk of 2 G are an adequate stimulus for MHC compositional changes in rat plantaris and soleus muscle; 2) both exercise and +G caused an increase in the slow MHC1 isoform in soleus muscle, suggesting that loading is a primary stimulus for this shift; and 3) 2 G and exercise appeared to have differential effects on the plantaris muscle MHC isoforms, with 2 G causing an increase in MHC2b, and exercise causing a decrease in MHC2b with a concomitant increase in MHC1, suggesting that factors other than enhanced loading, possibly locomotor activity levels, are the primary stimulus for this shift.
- wheel running
endurance exercise and chronic acceleration produce comparable adaptations in a variety of physiological systems (for reviews, see Refs. 4 and 5). Chronic exposure to force environment greater than Earth’s gravity (+G) and endurance training leads to an increased usage of fatty acids as metabolic substrates, a decrease in percent adipose mass, with no change or a slight increase in lean body mass, an increase in maximum O2 uptake, and an increased resting metabolic rate (4, 10, 17, 22). Animals adapted to chronic +G exposure also exhibit dramatic increases in exercise thresholds, e.g., increased maximum O2 uptake, relative to nonadapted controls (5, 6).
Because muscle is a major metabolic organ and provides the mechanical work requirements for posture and locomotion (which are presumably increased during both +G and endurance exercise), we hypothesized that muscle would also exhibit similar adaptations to exercise and +G, as assessed by myosin heavy chain (MHC) analysis. The functional properties of individual myofibers are largely dictated by the specific MHC isoform(s) they express (20). As such, the ratios of different MHC isoforms determine the phenotype of the muscle fiber, e.g., slow oxidative, fast glycolytic, etc. (7). Assaying for changes in the MHC isoform profile of myofibers following stimulation, e.g., exercise or +G, thus provides significant and specific information regarding the extent and type of muscle adaptation.
Although anatomical changes in the muscles of animals exposed to +G have been reported (6), only one study has evaluated MHC isoform changes in the muscles of mature animals exposed to chronic +G (18). This study (18) failed to demonstrate any changes in the MHC profile of the soleus and medial gastrocnemius, possibly because the study's 2-wk duration was too short to elicit MHC isoform responses. In contrast to +G, skeletal muscle adaptations to endurance exercise have been evaluated in many studies (8, 9, 11, 15, 24), including the effects on muscle MHC isoforms (8, 15). The myosin adaptation also appears to have a duration-response relationship (8) and is muscle specific. In general, endurance exercise, in contrast to resistance-training paradigms, does not promote an increase in fiber size, e.g., hypertrophy; rather, the muscle adaptations are confined to mitochondrial changes and fast-slow MHC isoform shifts (11, 12).
To test the hypothesis that +G and endurance exercise will produce comparable adaptations in muscle, this study examined the MHC composition of select muscles from rats after 8 wk of exposure to 1) chronic exercise (wheel running); 2) chronic acceleration (2 G via centrifugation); 3) chronic exercise and chronic acceleration (wheel running at 2 G); and 4) nonexercise/nonacceleration sedentary (1 G) controls. Both a slow-twitch, e.g., soleus, and fast-twitch agonist, e.g., plantaris, muscle were evaluated. The primary observations from the study were as follows: 1) 8 wk of 2 G are an adequate stimulus for MHC compositional changes in rat plantaris and soleus muscle; 2) both exercise and +G caused an increase in the slow MHC1 isoform in soleus muscle, suggesting that loading is a primary stimulus for this shift; and 3) 2 G and exercise appeared to have differential effects on the plantaris muscle MHC isoforms, with 2 G causing an increase in MHC2b and exercise causing a decrease in MHC2b, with a concomitant increase in MHC1, suggesting that factors other than enhanced loading, possibly locomotor activity (ACT) levels, are the primary stimulus for this shift.
MATERIALS AND METHODS
Care of the rats in the experiment met the standards set forth by the National Institutes of Health in their Guidelines for the Care and Use of Experimental Animals, and the protocol was approved by the UC Davis Institutional Animal Care and Use Committee. Male Sprague-Dawley rats (∼250 g), purchased from Simonson Laboratories, were used for the experiment. Preliminary analysis confirmed that all rats were willing to run on the wheel (W), including the rats in the control (C) group. Four groups (n = 8/group) of rats were studied and included: 1-G and 2-G groups, with (1GW and 2GW) and without (1GC and 2GC) running wheels.
Thirty-two rats were implanted intraperitoneally with biotelemetry transmitters (VM-FH disc; Minimitter, Sunriver, OR) to record body temperature (Tb) and locomotor activity (ACT). A surgical plane of anesthesia was initiated and maintained by using 3% isoflurane in pure medical-grade oxygen, administered by an adjustable isoflurane vaporizer (Viking Medical Products, Medford Lakes, NJ). Using aseptic techniques, we performed a midline celiotomy and inserted a sterilized transmitter into the peritoneal cavity. All incisions were sutured and treated with lidocaine and a topical antibiotic. The rats recovered on a heating pad with Tb constantly monitored via a colonic probe. The online Tb telemetry data allowed daily health monitoring of each animal without stopping the centrifuge.
Housing and centrifugation.
Following 14 days of postoperative recovery, the experimental recording began (e.g., day 1 of 8 wk). The 2-G (wheel and no wheel) groups were placed on a 4.5-m-diameter centrifuge. The 1-G (wheel and no wheel) groups were housed in an adjacent room with identical housing and ambient conditions. The animals were individually housed in standard vivarium cages with food (Lab Diet) and water ad libitum. Each cage was placed on top of a telemetry receiver interfaced to a microcomputer data acquisition system (Data Sciences). Tb values were recorded at 5-min intervals, and ACT data were collected in 5-min bins. The wheel was also interfaced to the data acquisition system, and the animals had ad libitum access (no wheel locks).
The 2-G animal cages were housed inside centrifuge modules, which provided ventilation, a 24-h light-dark cycle (light-dark 12:12 h), an ambient temperature of 25 ± 1°C, and visual isolation. Modules containing the animal cages were mounted with one degree of freedom, thereby ensuring that the net G field was always ortho-normal to the cage floor. Centrifugation was interrupted twice weekly for ∼15- to 20-min periods for animal husbandry.
General animal health was monitored daily in all animals via recorded Tb telemetry data. Body mass was measured twice weekly, and animal cages were changed every week.
At the end of the 2-G exposure, the rats were removed from the centrifuge and euthanized by decapitation, and the muscles were excised, weighed, then snap-frozen in liquid nitrogen, and stored until analyzed. The analyzed muscles included the soleus (anti-gravity, slow twitch) and plantaris (locomotion, fast twitch). Total muscle protein content (mg/g) was determined for each animal/muscle in all groups (Bio-Rad DC Protein Assay). For the MHC analysis, the myofibril MHCs were separated by SDS-PAGE (26). The SDS-PAGE gels were stained, photographed, and scanned for the quantification of MHC isoforms.
Citrate synthase assay.
Citrate synthase (CS) enzyme activity for the soleus and plantaris from all groups was determined by using frozen muscle homogenates. The assay has been previously described (23).
Locomotor and wheel ACT.
Locomotor ACT levels for all animals were recorded via telemetry. Briefly, small movements of the transmitter (i.e., animal) result in changes in signal strength at the receiving antenna. These changes in signal strength allow monitoring and quantification of gross motor ACT. In addition, wheel activity (revolutions) was recorded with a magnetic switch and the Vital View Data Acquisition System (Minimitter, Sunriver, OR).
Group differences were determined by using a two-way ANOVA. Specific mean comparisons were made by using Tukey's honestly significant difference post hoc test (SPSS). An α < 0.05 was considered statistically significant. All data are presented as means ± SD.
Figure 1 shows the mean body mass responses of all four groups throughout the study. The four groups did not differ in body mass during the preexperimental 1-G period (305 ± 6 g). At 2-G onset, both 2-G groups exhibited an anticipated decrease in body mass. After the first 4 days of 2 G, the 2-G groups began to increase body mass at a growth rate similar to pre-2 G; however, both groups maintained significantly lower (P < 0.001) body masses from the 1GC group for the duration of the study. After the first week of the experiment, the 1GW group, which was introduced to the wheel at the time of 2-G onset, began to diverge from their prewheel growth curve. Similar to the 2-G groups, the 1GW group maintained a significantly lower (P < 0.001) body mass from the 1GC group during the experimental period. The body masses of the 1GW group were not significantly different from those of the 2-G groups from the 4th wk of 2-G exposure on. At the end of the experimental period, the 1GC group (427 ± 4.5 g) was significantly heavier than the 1GW (380 ± 5.5 g), 2GC (388 ± 6.7 g), and 2GW (383 ± 9.7 g) groups.
As shown in Fig. 2, after 2 G, the absolute soleus weights of the 2GC (171 ± 10 mg) and 2GW (168 ± 14 mg) groups were not significantly smaller than 1GC (176 ± 5 mg) or significantly larger than the 1GW (164 ± 10 mg) group. Furthermore, the body mass-adjusted wet weight of the soleus was not significantly different (P > 0.05) in the 2GC (0.44 ± 0.01 mg/g) or 2GW (0.44 ± 0.01 mg/g) groups compared with the 1GC (0.41 ± 0.01 mg/g) group; the 1GW (0.43 ± 0.02 mg/g) group was similar to the 1GC. In contrast, the absolute plantaris muscle weight was significantly smaller (P < 0.01) in the 2GC (398 ± 17 mg) and 2GW (395 ± 23 mg) compared with the 1GC (465 ± 17 mg) and 1GW (430 ± 20 mg) groups. Similar to the soleus, the mass-adjusted plantaris mass did not differ between the 1GC (1.21 ± 0.08 mg/g), 1GW (1.19 ± 0.06 mg/g), 2GC (1.11 ± 0.08 mg/g), and 2GW (1.10 ± 0.06 mg/g) groups. Thus the changes in the plantaris weight appear proportional.
Total muscle protein.
As shown in Fig. 3, the protein content of the soleus muscles in the 2GC (254 ± 14 mg/g) and 2GW (247 ± 10 mg/g) groups was significantly higher (P < 0.05) than in the 1GC (220 ± 17 mg/g) group; the 1GW (237 ± 9 mg/g) group was not significantly different from either the 1GC or the 2G groups. The plantaris of the 2GC (218 ± 12 mg/g) and 2GW (212 ± 18 mg/g) groups had significantly lower total protein concentrations than either the 1GW (252 ± 12 mg/g) or 1GC (247 ± 18 mg/g) groups (P < 0.05). There were no differences in plantaris total protein content between the 1GC and 1GW groups.
MHC isoform distribution.
All muscle homogenates were subjected to SDS-PAGE and separated with good resolution. The overall order of migration by MHC types was (bottom to top of gel) MHC 1 > 2b > 2x > 2a. This migration order was identical to that from other laboratories using the same protocol (26). Typical MHC isoform separations of rat soleus and plantaris are shown in Fig. 4.
Figure 5 is a histogram detailing the mean (±SD) MHC isoform content of the soleus and plantaris for each group. The 1GC group soleus (Fig. 5A) contained MHC1 and MHC2a, with a preponderance of type MHC1 (∼90%). In contrast, the 1GW, 2GC, and 2GW groups demonstrated a significant decrease in MHC2a, with a concomitant significant (P < 0.01) increase in MHC1. Furthermore, in the soleus, only 25% of the 2-G rats (2GW and 2GC) demonstrated the presence of any MHC2a; that is, there was an apparent and significant (P < 0.01) shift to 100% slow-type MHC1 in these groups. A similar shift from MHC2a to MHC1 was seen in the 1GW group.
For the plantaris (Fig. 5B), the 1GC group contained primarily MHC2x and MHC2b, with some MHC2a and MHC1. In comparison, the 1GW group showed a significant (P < 0.01) increase in slow MHC1 and a significant decrease (P < 0.01) in the fast-type MHC2b. In contrast to the similar soleus isoform patterns between the 1GW and 2-G groups, the 2-G groups did not show an increase in either MHC1 or MHC2a or a decrease in fast-type MHC2b in the plantaris. Rather, as seen in Fig. 5, both 2G groups demonstrated a significant decrease in MHC2x and a significant increase in fast MHC2b compared with the 1GC and 1GW groups (all P < 0.01).
Compared with the 1GC group (28.5 ± 1.3 μmol·g−1·min−1), soleus CS activity (expressed as μmol·g muscle wt−1·min−1) was significantly increased (P < 0.05) in the 1GW group (34.1 ± 1.1 μmol·g−1·min−1). Soleus CS activity was not, however, significantly different in either the 2GC (27.9 ± 1.5 μmol/g min) or 2GW (29.1 ± 1.5 μmol/g min) compared with the 1GC group (Fig. 6A). In contrast, plantaris CS activity (Fig. 6B) was significantly higher (P < 0.01) in the 1GW (28.9 ± 2.9 μmol·g−1·min−1), 2GC (27.3 ± 2.1 μmol·g−1·min−1), and 2GW (26.8 ± 2.2 μmol·g−1·min−1) compared with the 1GC group (17.3 ± 1.8 μmol·g−1·min−1).
The 1-G baseline mean ACT (44 ± 4 counts/5-min bin) was not different between any of the groups (1GC, 2GC, 1GW, 2GW). On the first day of the experiment, the 2GC and 2GW exhibited a significant decrease from 1 G (P < 0.001) in ACT counts. During early 2 G (days 2–6), daily mean ACT was highly depressed in both groups (∼6 counts/bin). Also, during the first ∼5 days of the experiment, i.e., 2 G, the 2GW rats did not attempt to run on the wheel. During late 2 G (days 40–56), mean ACT had recovered, e.g., it was steady state to ∼40% of 1 G in both the 2GC and 2GW rats (∼20 and 17 counts/bin, respectively; Fig. 7A). With respect to wheel activity, the 1GW rats averaged 3.5 miles of wheel running per day. In contrast, during late 2 G, the 2GW rats averaged ∼0.37 miles of wheel running per day; ∼10% of the running of their 1GW counterparts (Fig. 7B).
This study demonstrated for the first time the effects of long-term chronic acceleration on the MHC composition of mature rat muscle. The primary findings of this study were as follows: 1) 8 wk of 2 G are an adequate stimulus for MHC compositional changes in rat plantaris and soleus muscles; 2) both exercise and +G caused an increase in the slow MHC1 in soleus muscle, suggesting that loading is a primary stimulus for this shift; 3) 2 G and exercise appeared to have differential effects on the plantaris muscle, with 2 G causing an increase in MHC2b, i.e., faster muscle, and exercise causing a decrease in MHC2b with a concomitant increase in MHC1, i.e., slower muscle, suggesting that another factor, possibly locomotor ACT levels, is a primary stimulus for this shift; and 4) the lack of a statistical interaction of wheel running and gravity on any measured variable strongly suggests that the use of a voluntary wheel paradigm may need to be reconsidered in future studies, evaluating the additive effects of gravity and exercise. In the end, the findings of this study for the soleus support our hypothesis, whereas our hypothesis is rejected for the plantaris.
Response of body and muscle masses.
Body mass was decreased in 2 G and with exercise relative to controls after the 8-wk experimental period. The maintenance of a lower body mass appears to be a regulated phenomenon at 2 G and, presumably, during chronic exercise. For example, chronically accelerated animals, when fasted briefly, recover lost body mass upon refeeding just as rapidly as with earth gravity controls (21). This finding suggests that the decreased body size in the 2-G environment is not a result of an inability to acquire food or an incapacity for synthesis (22). This is an important point, as nutritional deficiency has been linked with muscle atrophy.
The lack of change in absolute soleus mass, the “anti-gravity” extensor, was surprising, as the expectation was a hypertrophic response consequent of the presumed increased loading by hypergravity exposure. As such, this observation challenges the assumption that 2 G effectively increases the load on the soleus muscle. One explanation for the lack of a hypertrophic response may be the orientation of the gravitational stimulus with respect to the animal, e.g., the force is directed through the dorsoventral axis of the animal (−Gx). That is, to achieve true “loading” of the soleus in a rat, or any quadruped for that matter, it may be necessary to orient the rat vertically so that the resultant force vector is directed through the rostrocaudal axis of the animal (+Gz), as it would be in a standing biped. Nevertheless, it is clear that, without electromyogram recordings, we cannot definitively state that 2 G is indeed increasing the load at the level of the soleus. Finally, when normalized to body mass, the soleus mass was not significantly different from that of controls. In contrast to the soleus, after 8 wk, the absolute mass of the plantaris was significantly lower in both 2-G groups. However, neither 2-G group had significantly smaller plantaris masses when normalized to body mass. Thus, similar to the soleus muscle, the plantaris changes are proportional to the changes in body mass.
The results of the present study demonstrate that exposure for 8 wk to either 2 G or voluntary exercise can alter the MHC isoform content of the soleus and plantaris muscles. This study only evaluated the end-point MHC changes, however, and thus the actual dose/duration necessary to induce these changes cannot be determined from the available data. Interestingly, in general, the dose-response relationship between exercise duration and muscle MHC phenotype has been evaluated in very few studies (8). For the present study, a 60-day duration was chosen, in part, because other studies examining physiological changes resulting from either +G or exercise have demonstrated that the physiological system(s) under investigation largely reaches steady state by 60 days (13, 14, 27). Thus, although one could not make the a priori assumption for muscle that 60 days is “steady state,” any changes that have or ultimately will take place will likely have occurred by the 60-day point.
At 2 G, the mechanical work requirements of the soleus are presumably increased (work = potential energy = mass·gravity·height) and thus could bias the soleus toward a slower, more fatigue-resistant phenotype. Moreover, during short-term submaximal exercise, blood flow to the soleus increases, presumably reflecting increased motor unit recruitment and metabolic demands (1). Indeed, after either the 2 G or exercise exposure, the soleus muscle of both the 2-G and exercise groups became even slower; that is, a significant transition of MHC2a to MHC1 occurred. In 75% of the 2-G animals, this transition was to 100% MHC1. Although the transition to a slower phenotype is consistent with increased mechanical loading of the soleus, the transition could also be due to a number of other factors, such as changes in posture and/or neuromotor activity, e.g., increases in MHC2a recruitment patterns. It should also be pointed out that, despite the similar MHC transition in the 2-G and exercise groups, the motor units, temporal aspects of recruitment, and metabolic demands may not have been identical between the exercise and 2-G stimuli.
In contrast to the soleus, the plantaris did not demonstrate a similar MHC shift between the 1-G exercise and 2-G groups. Specifically, the plantaris of the 2-G animals became “faster,” as is evidenced by an increase in the MHC2b relative content. In contrast, the 1-G exercise group responded with a decrease in MHC2b with a concomitant increase in MHC1. This “slower” MHC phenotype in the 1-G exercise group and faster constitution in the 2-G groups suggests that the plantaris may be more responsive to stimuli other than the presumed loading by hypergavity per se. For example, when locomotor ACT levels are compared between the 2-G and 1-G groups, it is immediately apparent that the 2-G groups were significantly less active than either 1-G group. It is unclear why the animals reduced their ACT at 2 G, although this ACT response is identical to that documented in previous studies (13, 14). Nevertheless, it is perhaps not surprising that the effect of this lower ACT would manifest in a muscle such as the plantaris, whose motor units are relatively quiescent during resting (nonlocomotor) conditions. That the lower ACT levels can explain the differential MHC response of the plantaris in the 2-G groups seems to be supported by the findings of other laboratories that have shown the plantaris to be exquisitely sensitive to inactivity paradigms (25).
The CS assay has been routinely used to assess the “oxidative” capacity of a muscle (3). Increases in CS activity are thus used as an indicator of a training effect. In this study, the 1GW group evidenced a significant increase in CS activity in both the soleus and plantaris muscle compared with control values. The increase in CS in the 1GW group is consistent with a training effect and parallels the shift to a slower MHC phenotype in both muscles for this group. Despite the similar shift to a slower MHC phenotype in the 2-G groups, neither 2-G group evidenced a change in CS activity of the soleus muscle. Rather paradoxically, however, we documented an increase in CS activity in the plantaris of the 2-G group. Here, it is difficult to reconcile an increase in CS activity in muscles that have shifted to a faster, and hence more “glycolytic,” MHC phenotype. The available evidence suggests that, as the plantaris adapted to 2 G, it became faster and increased its oxidative capacity. This relatively rare fast-oxidative phenotype may reflect the significant increase in fatty acid utilization known to occur following adaptation to +G (13).
Locomotor and wheel ACT.
ACT levels are an important variable when considering the effects of an experimental paradigm on MHC compositional changes. This is because muscles, particularly those with locomotive or postural, i.e., antigravity function, are very sensitive to increases or decreases in ACT levels. For example, an unloaded soleus muscle, i.e., reduced weight-bearing ACT, exhibits a marked downregulation of the MHC1 isoform and concomitant expression of the fast MHC2x (2). Furthermore, lower ACT paradigms have been shown to increase the amount of MHC2b in the plantaris muscle, effectively making a fast muscle an even faster muscle phenotype (25). Thus the interpretation of the findings herein concerning MHC expression must take into consideration the impact of the reduced or increased ACT levels seen in the 2-G groups and 1GW group, respectively. Unfortunately, without electromyogram measurements, an accurate quantification of muscle ACT is not possible.
In conclusion, this study demonstrated that muscle MHC composition can be altered by endurance exercise and chronic acceleration. The MHC responses to these two paradigms are not the same and appear to be muscle specific. Moreover, as the 2-G groups' ACT levels were less than those of the 1-G groups, the presumed loading component of hypergravity may actually be more complex than originally assumed. To this end, this study has generated some novel observations, primary among these, the plantaris' response to long-term acceleration. That the plantaris became faster in response to chronic acceleration and that this occurred in the face of highly attenuated locomotor ACT levels suggest that the muscle requires more than gravitational loading to maintain a “normal” phenotype. Thus the suggestion that the skeletal muscle deconditioning seen in astronauts can be selectively eliminated by intermittent or chronic loading via centrifugation may only be viable, in part, for muscle with a primarily motor function, e.g., plantaris. Our data would suggest that an equally important aspect of such a countermeasure may be locomotor ACT while loaded. If true, this is consistent with the demonstrated corollary that the loading state imposed during locomotion is critical in influencing the pattern and magnitude of MHC changes in fast-twitch muscle (24). Last, it is unclear why 2 G promotes the development of a fast-oxidative phenotype for the plantaris, although changes in substrate availability may, in part, underlie this phenomenon.
This study was supported in part by National Aeronautics and Space Administration Grant NNA-04CC82G (C. A. Fuller), the Achievement Rewards for College Scientist Scholar Foundation Award (P. M. Fuller), and National Institute of Arthritis and Musculoskeletal and Skin Diseases Grant AR-30346 (K. M. Baldwin).
We thank Fadia Haddad and Ming Zeng for assistance with the SDS-PAGE protocol for the MHC isoform separation.
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 © 2006 the American Physiological Society