Adaptations to repeated bouts of injury-inducing lengthening contractions were studied in mouse anterior crural muscles. Five bouts of 150 lengthening contractions were performed in vivo, with each bout separated by 2 wk of rest. Three primary observations were made. First, there was little, if any, attenuation in the immediate isometric torque losses after lengthening contractions at “physiological” stimulation frequencies (i.e., <125 Hz), although there was a pronounced decrease in torque loss at higher frequencies between the first and second bouts. Second, the immediate losses in strength that occurred after all five lengthening contraction bouts could be explained in part by excitation-contraction uncoupling. Third, the most important adaptation was a significant enhancement in the rate of recovery of strength after the lengthening contractions. It is probable that the accelerated rate of strength recovery resulted from the more rapid loss and subsequent recovery of myofibrillar protein observed after the fifth bout.
skeletal muscle fibers are injured when the intensity of muscular exertion exceeds that to which an individual is accustomed. It is widely accepted that the injury primarily occurs when the muscle is actively lengthened. In humans, lengthening contraction-induced muscle injury is characterized by swelling of the muscle, infiltration of inflammatory cells, disruptions in sarcomeric architecture, catabolism of force-bearing proteins, release of muscle proteins into the blood, and delayed-onset muscle soreness (e.g., Ref. 1). In mice, we have shown that lengthening contraction-induced muscle injury is also characterized by swelling of the muscle, infiltration of inflammatory cells, damage to myofibers, catabolism of force-bearing proteins, and release of muscle proteins into the blood (11, 13, 19, 31, 33, 34, 36). However, the most important functional consequence of this type of soft tissue injury is the substantial (e.g., ∼50%) and prolonged decrement (e.g., 2–6 wk) in muscular strength, which has been demonstrated in both human (4, 10) and animal models (11, 13, 19, 34).
The mechanisms responsible for the strength loss after initiation of lengthening contractions are not completely understood. The attenuation in force appears to be due in part to localized damage to muscle fibers, which Morgan (22) attributed to nonuniform stretching of sarcomeres in series during the lengthening contractions. The high stresses and strains in the muscles cause disruption of specific contractile (e.g., Ref. 11) or cytoskeletal proteins (e.g., Ref. 18) that presumably contribute to the strength loss. Our laboratory (13, 31, 36) and others (2) have also shown that excitation-contraction (E-C) uncoupling is important in the strength deficit early after injury induction (≤5 days) in mouse skeletal muscle (33). However, loss of force-bearing proteins appears to be responsible for the prolonged strength deficit (11). We have estimated that E-C uncoupling accounts for most (∼60–75%) of the strength deficits out to 5 days after initiation of the injury in our mouse model, whereas loss of force-bearing elements within the muscle is responsible for the remaining force deficits from 14 to 28 days (11, 33).
Once contraction-induced muscle injury occurs, there is no known medical intervention that hastens the recovery of muscle strength. In fact, the only recognized method of attenuating future lengthening contraction-induced muscle injury is conditioning with lengthening contractions, i.e., the repeated-bout effect (e.g., Refs. 4, 5, 21, 23). Studies on humans and rodents have generally shown that a single bout of lengthening contractions can almost completely protect the muscle from injury when the release of muscle protein (e.g., creatine kinase) is used as the injury marker (4, 30). However, solely on the basis of human studies, it appears that the loss in maximum strength immediately after the injury induction is somewhat resistant to adaptation (3–5, 23, 24). The most striking functional adaptation to repeated bouts of lengthening contractions is a faster recovery in the strength deficit (3, 5, 23, 25). In contrast to the number of reports on human subjects, few studies using laboratory animals have investigated this phenomenon, and fewer could be found in which it was attempted to determine the underlying mechanisms. Therefore, the first objective of this study was to characterize the functional adaptation of mouse anterior crural muscles to repeated bouts of lengthening contractions. The primary force-producing anterior crural muscles in the mouse are the tibialis anterior (TA) and extensor digitorum longus (EDL), both of which are almost exclusively fast-twitch (>95%) muscles (27). The second objective was to investigate potential mechanisms responsible for the persistent loss, but accelerated recovery, of muscle strength with repeated bouts of lengthening contractions. It was hypothesized that the continuing initial loss of muscle strength is due to E-C uncoupling, whereas more rapid recovery in strength with repeated bouts is related to an attenuation of loss, and faster recovery rate, of myofibrillar proteins.
Three studies were performed. In the first, functional adaptations of mouse anterior crural muscles to five bouts of 150 maximal lengthening contractions separated by 2 wk of recovery were determined. We have shown that this lengthening contraction protocol causes a significant loss of strength (∼50%) requiring a prolonged period (≤4 wk) of recovery time (11, 13, 19, 34). Isometric torques of this muscle group as a function of stimulation frequency (20–300 Hz) were measured in anesthetized adult female ICR mice (n = 10) before and immediately after each of the lengthening contraction bouts. Peak torques were recorded during the 150 maximal lengthening contractions that were performed in each bout. The recovery of anterior crural muscle strength after injury was also evaluated in vivo by measuring isometric torque as a function of stimulation frequency (20–300 Hz) at 5 days after the fifth injury bout (only 6 of the 10 mice were studied). The selection of 5 days as a measure of recovery of isometric torque was based on the observation that human skeletal muscle shows accelerated recovery in strength in 5 days after a third injury bout compared with the initial injury bout (23).
It is possible that differences in strength deficits between the first and fifth injury bouts could stem from changes in the maturation of the mice over the 8 wk. Therefore, the second study investigated the possible contribution of animal maturation to the adaptation of anterior crural muscle strength in five female ICR mice. The anterior crural muscles performed five bouts of isometric contractions (i.e., torque-frequency assessment) separated by 2 wk of recovery. After the isometric contractions in the fifth bout, anterior crural muscles performed the 150 lengthening contraction injury protocol, and torque as a function of stimulation frequency was measured immediately after and 3 days after injury induction.
In the third study, the mechanisms responsible for the loss and recovery of isometric torque of the anterior crural muscles after the fifth lengthening contraction bout were examined in 16 adult female ICR mice. As in the first study, isometric torques as a function of stimulation frequency (20–300 Hz) were measured in vivo before and immediately after each of the five lengthening contraction bouts. In the first two studies, we found that isometric strength of the anterior crural muscles was nearly recovered 5 days after the fifth injury bout and that this rapid recovery was not explained by physical maturation. Therefore, the mechanisms of strength loss were investigated, both in the anterior crural muscle group in vivo and in individual anterior crural (i.e., EDL) muscles in vitro, immediately after (n = 6) and 3 (n = 6) and 5 (n = 4) days after the fifth injury bout.
Immediately after the last in vivo assessment of isometric torque at the respective sampling times after the fifth injury bout, mice were anesthetized with pentobarbital sodium, and the left and right TA muscles were dissected free, weighed, frozen in liquid nitrogen, and stored at −80°C for later analyses of total, myofibrillar, and soluble protein contents. In addition, the EDL muscles were dissected free and peak isometric twitch (Pt), maximal isometric tetanus (Po), and caffeine-induced forces were determined in vitro. EDL muscles were subsequently weighed, frozen in liquid nitrogen, and stored at −80°C for later analyses of total, myofibrillar, and soluble protein contents.
E-C uncoupling in the muscles was assessed by two methods. First, we directly evaluated E-C uncoupling in the EDL muscle in vitro with caffeine as previously described (13, 36). Caffeine bypasses the normal E-C coupling pathway by directly activating sarcoplasmic reticulum (SR) Ca2+ release channels (9). Similar relative reductions in Po and in caffeine-elicited force are interpreted as reflecting alterations in myofibrillar Ca2+ sensitivity or disruption of force-generating and/or -transmitting elements. On the other hand, greater percent reductions in Po compared with caffeine-induced forces (i.e., increases in the caffeine-induced force-to-Po ratio) after injury induction reflect a failure in the E-C coupling process. Second, E-C uncoupling in the anterior crural muscles was indirectly assessed by comparing low- and high-frequency strength losses. Greater reductions in low-frequency isometric torque compared with high-frequency isometric torque have traditionally been used as an indicator of E-C uncoupling (6, 14).
Female ICR mice (n = 31), 8–12 wk old, were used in the three studies. The mice were housed in groups of five to six animals per cage, supplied with food and water ad libitum, and maintained in a room at 20–22°C with a 12-h photoperiod. For the in vivo injury induction and torque measurements, mice were anesthetized with 0.33 mg/kg fentanyl citrate, 16.7 mg/kg droperidol, and 5.0 mg/kg diazepam (12). Mice were euthanized with an overdose of pentobarbital sodium. All animal care and use procedures were approved by the institutional animal care and use committee and met the guidelines set by the American Physiological Society.
In vivo muscle strength testing and injury induction.
The general protocols for measuring anterior crural muscle strength and inducing injury have been described previously (19, 34). Briefly, with the knee fixed at 90°, the left foot of the anesthetized mouse was attached to a machined aluminum shoe, which in turn was attached to the shaft of a servomotor system (Aurora Scientific, 300B dual-mode). Contraction of the anterior crural muscles was initiated by a pair of platinum needle electrodes (25 gauge) inserted through aseptically prepared skin for stimulation of the left common peroneal nerve. Peak torque was then optimized by varying stimulation voltage in a series of 5–10 isometric contractions (200-ms trains of 75-μs biphasic pulses at 300 Hz); the contractions were separated by 45–60 s. Next, torque as a function of stimulation frequency (20, 40, 60, 80, 100, 125, 150, 200, 250, 300 Hz) was measured during 10 isometric contractions (200-ms trains of 75-μs biphasic pulses), with 45 s between contractions. The left anterior crural muscles were then injured by performing 150 lengthening contractions, in which the muscles were stretched while being stimulated from 20° of ankle dorsiflexion to 20° of ankle plantarflexion at an angular velocity of 2,000°/s. This movement was preceded by a 100-ms isometric stimulation, so the contraction lasted a total of 120 ms (11, 13, 19, 34). The entire injury protocol took 28.5 min (10–12 s between contractions). Beginning 3 min after the last lengthening contraction, the measurement of torque as a function of stimulation frequency was repeated. Muscle torques as a function of stimulation frequency were also measured 5 days after the fifth injury bout in the first study, at 3 days after the injury bout in the second study, and at 3 and 5 days after the fifth injury bout in the third study.
In vitro twitch, tetanus, and caffeine-induced contracture experiments.
In the third study, after the fifth in vivo injury protocol, EDL muscles were dissected free at 0, 3, or 5 days postinjury and studied at 37°C by using an in vitro preparation as previously described (11, 13, 19). Briefly, muscles were mounted in a chamber containing a Krebs-Ringer bicarbonate buffer (pH 7.4) with (in mM) 144 Na+, 126.5 Cl−, 6 K+, 1 Mg2+, 1 SO42−, 1 PO43−, 25 HCO3−, 1.25 Ca2+, 0.17 leucine, 0.10 isoleucine, 0.20 valine, 10 glucose, and 10 μg/ml gentamicin sulfate and 0.10 U/ml insulin (the buffer was equilibrated with 95% O2-5% CO2 gas). The distal tendon was attached by silk suture and cyanoacrylate adhesive to a fixed support, and the proximal tendon was attached to the lever arm of a servomotor system (see above). Muscle length in the chamber was set at its anatomic resting length (Lo), a length halfway between the muscle's minimum and maximum in vivo lengths. Anatomic Lo was set by adjusting resting tension to 4.4 mN, a force previously determined to correspond to the anatomic Lo in mouse EDL muscle (31). Resting muscle length was set to the anatomic Lo instead of physiological Lo (i.e., the muscle length for which tetanic force is maximized) because the mouse EDL muscle anatomic Lo is reliably set by our procedure, and the process for determining physiological Lo for the fast-twitch EDL muscle at 37°C induces fatigue because of the repeated tetani. Isometric twitch (Pt; 0.2 ms pulse at 150 V) and tetanic (Po; 200 ms trains of 0.2 ms pulses at 300 Hz) contractions were initiated at 7 and 8 min into the incubation, respectively. After the measurement of Po, the EDL muscles were exposed to Krebs-Ringer solution containing 50 mM caffeine for 6 min, and peak contracture force was recorded (13).
Protein fractionation and measurement.
In the third study, TA and EDL muscle protein contents were separated into myofibrillar and soluble protein fractions by a modification of the methods described previously (11). All buffers, homogenates, and suspensions were maintained on ice. Muscle samples were homogenized in 400 μl of (in mM) 250 sucrose, 100 KCl, 20 MOPS, and 5 EDTA (pH 6.8). After we removed an aliquot for total protein measurement, the homogenate was centrifuged for 15 min at 5,000 g (4°C). The supernatant (termed “soluble protein fraction”) was removed, and the pellet (termed “myofibrillar protein fraction”) was resuspended in 400 μl of (in mM) 150 KCl and 20 MOPS (pH 7.0). Aliquots from the total protein pool and myofibrillar and soluble protein fractions were measured for protein content by use of the Bradford assay (BioRad).
Body weight was analyzed with a group (injured and control) by bout (1st, 2nd, 3rd, 4th, 5th) ANOVA with repeated measures on the bout factor. Data from in vivo injury measurements (e.g., lengthening torque, isometric torque-frequency) were combined from the first and third studies. Changes in total work done on the anterior crural muscles during the lengthening contraction protocol were analyzed by single-factor (bout, i.e., 1st, 2nd, 3rd, 4th, 5th) repeated-measures ANOVAs. Peak lengthening contraction torques were analyzed by a contraction (1st, 150th) by bout (1st, 2nd, 3rd, 4th, 5th) ANOVA with repeated measures on the bout and contraction factors. Isometric torque as a function of stimulation frequency was assessed with a bout (1st, 2nd, 3rd, 4th, 5th) by condition (preinjury, postinjury) by frequency (20, 40,… , 300 Hz) ANOVA with repeated measures on all factors. Changes in absolute and relative torques from pre- to postinjury were analyzed by using a bout (1st, 2nd, 3rd, 4th, 5th) by frequency (20, 40,… , 300 Hz) ANOVA with repeated measures on the bout and frequency factors. Recovery of isometric torque after the fifth injury bout was analyzed by using separate condition (preinjury, time postinjury) by frequency (20, 40,… , 300 Hz) ANOVAs with repeated measures on the condition and frequency factors for each time point. Only data from the third study were used to analyze in vitro contractile mechanics and protein contents. Muscle wet weight, total protein, myofibrillar protein, soluble protein, Pt, Po, and caffeine-induced force were evaluated with time (0, 3, 5 days) by leg (injured, contralateral control) ANOVAs with repeated measures on the leg factor. When significant main effects or interactions were found, differences in group means were tested with Student-Newman-Keuls post hoc tests. An α level of 0.05 was used for all analyses. All values are presented as means ± SE.
As described above, in vivo torque data were combined from the first and third studies, whereas only the muscles from the third study were used to analyze in vitro contractile mechanics and protein concentrations. Unless otherwise stated, all changes reported below were statistically significant (P < 0.05).
The mean initial body weight of all mice was 29.3 ± 0.3 g. There were no differences in body weights among mice from the three studies at any of the first three respective bouts. However, mice from the maturation study (i.e., study 2) were slightly larger (≤10%) than mice from the initial adaptation study (i.e., study 1) at bouts 4 and 5. Overall, mice were 5.1 ± 1.1, 8.1 ± 1.1, 13.1 ± 1.3, and 15.4 ± 1.7% heavier at the time of the second, third, fourth, and fifth injury bouts, respectively. To control for this maturation, all torques were normalized to body weight.
In vivo lengthening torque measurements during the injury protocols.
Changes in peak torques produced by the anterior crural muscles over the 150 lengthening contractions during the five bouts are shown in Fig. 1. The peak torque of the first lengthening contraction in the first injury bout was 182.9 ± 2.9 N·mm·kg−1, or 194% of the torque during the preinjury maximal isometric tetanic contraction. Compared with the first lengthening contraction of the first injury bout, the peak torque during the first lengthening contraction increased 7.9% during the fifth injury protocol. The relative decreases in peak lengthening torques from the first to the last lengthening contractions were 46.9 ± 1.3, 37.7 ± 2.0, 34.0 ± 0.7, 31.4 ± 0.7, and 30.4 ± 0.9% for the first, second, third, fourth, and fifth injury bouts, respectively. The relative decreases in peak lengthening torques for bouts 1, 2, and 3 were different from each other. However, bouts 3, 4, and 5 were not different from each other. The work done on the anterior crural muscles during the lengthening contraction bouts progressively increased until the fifth bout (Fig. 1, inset).
In vivo isometric torque-frequency measurements before and after the injury protocols.
Peak isometric torques produced by the anterior crural muscles as a function of stimulation frequency before and after the five injury bouts are shown in Fig. 2. Few changes occurred across bouts in peak isometric torque measured before the lengthening contractions (i.e., preinjury torque). Specifically, there were no differences in isometric torques among all bouts at stimulation frequencies of 60 and 80 Hz and no differences among the first four bouts for all the stimulation frequencies. However, preinjury peak isometric torque before the fifth injury bout was greater than that of the first bout at 20, 200, 250, and 300 Hz.
Compared with the preinjury torques, peak isometric torques measured after each injury bout were significantly reduced at all stimulation frequencies (Fig. 2). The primary adaptation in postinjury isometric torque occurred between injury bouts 1 and 2. Compared with the first injury bout, the postinjury isometric torques after the second bout were improved at all stimulation frequencies (Fig. 2). However, the degree of improvement in postinjury isometric torque from injury bout 2 through injury bout 5 was stimulation frequency dependent. Postinjury isometric torques did not change from bouts 2 through 5 at stimulation frequencies <80 Hz (Fig. 2). Also, there were no significant differences in relative decreases in isometric torque from bouts 2 through 5 at stimulation frequencies <80 Hz (Fig. 3), nor were there any differences in absolute torque deficits among the five injury bouts for stimulation frequencies of 40, 60, and 80 Hz (Fig. 4). Thus the ability of the anterior crural muscles to produce isometric torque at relatively low stimulation frequencies after lengthening contraction-induced injury improved very little after the second injury bout.
Although most of the improvement in postinjury isometric torque occurred between bouts 1 and 2 (Figs. 2–4), anterior crural muscle strength generally improved with subsequent injury bouts at stimulation frequencies >80 Hz (Figs. 2 and 3). For example, postinjury isometric torques increased from the preceding bout after the second, fourth, and fifth injury bouts at stimulation frequencies of 200, 250, and 300 Hz (Fig. 2). However, it is interesting to note that the primary reason for the attenuation in relative strength loss at the higher frequencies stems from increases in preinjury isometric torque values and not from decreases in absolute strength loss (Fig. 4). There were no differences in absolute torque loss among the last four injury bouts at stimulation frequencies >80 Hz (Fig. 4).
In addition to the apparent lack of adaptation of low-frequency torque production to repeated injury bouts, the low-frequency torques of the injured muscles were reduced to a greater extent relative to the preexercise condition than the high-frequency torques (Fig. 3). Isometric torques were reduced 77–92% after the five injury bouts at stimulation frequencies <125 Hz. In contrast, isometric torques were only reduced between 34 and 58% after the five injury bouts at stimulation frequencies >150 Hz.
In vivo isometric torque measurements during recovery after the fifth injury bout.
The recovery of isometric torques produced by the anterior crural muscles after the fifth lengthening contraction bout is shown in Fig. 4. Immediately after the fifth injury bout, isometric torques were reduced between 34 and 82% (Figs. 3 and 5A). Isometric torques produced 3 and 5 days after the fifth injury bout were not significantly different from those produced before the fifth injury bout (Fig. 5, B and C).
In vivo isometric and lengthening torque measurements in age-matched control mice.
Isometric torques produced in the initial exercise bout were not significantly different between age-matched and “adapted” mice, except for minor (≤10%) differences at stimulation frequencies of 150 and 200 Hz. Moreover, there were few differences (i.e., 60 and 80 Hz) between age-matched and adapted mice in preinjury isometric torques at the fifth exercise bout. There were no differences in peak torques between the first and fifth isometric contraction bouts in age-matched control mice at any stimulation frequency (Fig. 6). The peak torque of the first lengthening contraction from control mice during the fifth bout was 228.3 ± 8.2 N·mm·kg−1 or 211% of the torque during the preinjury maximal isometric tetanic contraction. The relative decrease in peak lengthening torque from the first to the last lengthening contraction was 42%. This relative decrease in lengthening torque in control mice was not different from that (47%) in the mice performing the first injury bout in the first study. However, these relative lengthening torque decreases were greater than the decrease (30%) in the mice that performed the fifth injury bout in the first study.
Peak isometric torques in the age-matched control mice were significantly reduced (−40 to −86%) immediately after injury at all stimulation frequencies (Fig. 6). There were few (i.e., 60, 125, and 150 Hz) differences between age-matched and adapted mice in postinjury isometric torques associated with the fifth exercise bout. Injured age-matched mice produced isometric torques lower than those produced in adapted mice from the fifth bout but the same as those in adapted mice from the first injury bout at stimulation frequencies of 60, 125, and 150 Hz. In contrast to mice adapted to lengthening contractions in the first study, age-matched control mice demonstrated little recovery of anterior crural muscle strength 3 days after injury induction (Fig. 6). Decreases in isometric torques 3 days after injury ranged from 39 to 70%. Isometric torques in age-matched mice were significantly lower than those in adapted mice 3 days after injury at nearly all stimulation frequencies (i.e., 20–40, 100–300 Hz).
Muscle wet weight and protein contents during recovery after the fifth injury bout.
TA and EDL muscle wet weights and total, myofibrillar, and soluble protein contents immediately after and at 3 and 5 days after the fifth injury bout are shown in Tables 1 and 2, respectively. On average, the wet weight of the injured TA muscle (65.0 ± 1.2 mg) was greater than that of the contralateral control TA muscle (59.5 ± 1.2 mg) across the three time points after the fifth injury bout. However, there were no differences between injured and contralateral control TA muscles in total, myofibrillar, or soluble protein contents (Table 1). Therefore, the increases in wet weight in the injured TA muscles were attributable to water.
The wet weight of the injured EDL muscle (12.7 ± 0.3 mg) was also greater than that of the contralateral control EDL muscle (12.1 ± 0.3 mg) across the three time points after the fifth injury bout (Table 2). In contrast, the mean total protein content was reduced 8.1% in the injured EDL muscle compared with the contralateral control muscle across the three time points after the fifth injury bout. However, myofibrillar protein content in the injured EDL muscle was only significantly reduced (18%) at 3 days after the fifth injury bout when compared with the contralateral control muscle. As in TA muscle, the elevated wet weights were caused by increased water.
In vitro EDL muscle force after the fifth injury bout.
The relative changes in Po and caffeine contracture force after the fifth injury bout are shown in Fig. 7. For the sake of comparison, data from a previously published study (13) showing recovery of EDL muscle Po after one bout of the same lengthening contraction protocol are also shown in Fig. 7. There were no differences in Po in the contralateral control muscles immediately after and at 3 and 5 days after the fifth injury bout. The specific Po of the contralateral control EDL muscles was 25.9 ± 0.4 N/cm2. However, EDL muscle Po was reduced 33% immediately after the fifth injury bout. EDL muscle Po started to recover soon after the fifth injury bout, with Po only being reduced 17 and 7% at 3 and 5 days, respectively. Despite the apparent accelerated recovery in the strength deficit, the absolute Po values in the injured EDL muscle were still different from values in the contralateral control muscle at these times.
There were no differences in Pt (54.9 ± 1.0 mN) of the contralateral control muscles immediately after and 3 and 5 days after the fifth injury bout (Pt data are not shown). However, EDL muscle Pt was reduced 49% immediately after the fifth injury bout. EDL muscle Pt started to recover soon after the fifth injury bout, with Pt only being reduced 17 and 13% at 3 and 5 days, respectively.
EDL muscle caffeine-induced contracture force did not display the same relative changes as Po or Pt between injured and contralateral control muscle across the three time points after the fifth injury bout. The mean caffeine-induced force for all the contralateral control muscles was 96.7 ± 1.8 mN, which was 23% of the Po. Although the relative changes in caffeine-induced force between injured and contralateral control muscles were −7, 12, and 10% at 0, 3, and 5 days after injury induction, respectively, none of these differences was significant (Fig. 7).
Three primary observations were made in this study. First, a repeated-bout effect was not readily apparent for the immediate isometric strength losses after lengthening contractions at “physiological” stimulation frequencies (i.e., ≤125 Hz), although there was a pronounced repeated-bout effect at higher frequencies between the first and second bouts. Second, the immediate losses in strength that occurred after all five lengthening contraction bouts may be explained in part by E-C uncoupling. Third, the most important repeated-bout effect was a significant enhancement in the rate of recovery of strength after the lengthening contractions. Each of these observations will be considered in turn.
Persistent immediate losses in strength.
In general, there was little attenuation in the loss of torque produced by the mouse anterior crural muscles immediately after the lengthening contraction protocol over the five bouts at the lower stimulation frequencies, whereas peak torques immediately postinjury were generally increased with each successive bout at frequencies over 150 Hz. Voluntary recruitment of motor units normally occurs with stimulation frequencies that elicit incompletely fused tetanic contractions. For example, it was recently reported that the mean firing rates of “fast” motor units ranged from ∼80 to 110 Hz in TA muscle during locomotion in conscious rats (8). If mouse anterior crural muscles display similar stimulation frequencies during motor unit recruitment, then stimulation frequencies below 125 Hz should represent more physiological frequencies. Thus there was a beneficial repeated-bout effect in torque production at high “nonphysiological” stimulation frequencies but not at lower more physiological frequencies. This corresponds with data from human studies employing maximal voluntary isometric contractions (3, 5, 23, 25). For example, Newham and coworkers (23) have shown that three bouts of maximal voluntary lengthening contractions of the forearm flexor muscles in humans result in a relatively constant initial peak strength deficit (∼50%).
Several observations support the notion that, in our in vivo injury model, persistent strength losses immediately after the five bouts of lengthening contractions do not stem from metabolic fatigue. First, our injury protocol (150 lengthening contractions with 120-ms train duration and 10–12 s between contractions) is less metabolically stressful than most other injury protocols (e.g., 450 lengthening contractions with 600-ms train duration with 2 s between contractions; Ref. 17). Second, we have shown that the performance of 150 concentric contractions, which imposes a greater metabolic stress on the muscle than lengthening contractions, does not result in any maximum strength deficits in mouse anterior crural muscles after the protocol (34). The modest reductions in strength at low stimulation frequencies after the concentric contractions recover within 24 h (34). Third, the maximum isometric strength deficit immediately after injury in mouse EDL muscle does not significantly recover at 3, 6, 12, or 24 h when our in vivo injury model is used (11, 19), as would be expected if metabolic fatigue was compounding injury-induced strength deficits. Fourth, long-lasting (i.e., ≥3 days) impairment of SR Ca2+ release during isometric contractions after injury (13) is not consistent with observations of rapid recovery of force production and/or SR Ca2+ release associated with fatigue (26).
On the basis of data from the present study, it appears that the persistent strength deficit observed in the EDL muscle immediately after the fifth injury bout also is caused by E-C uncoupling. This conclusion is based primarily on the observation that the relative reductions (i.e., injured vs. contralateral control) in Po (−33%) far exceed the relative changes in peak caffeine contracture force (−7%) immediately after injury induction (Fig. 7). If the decrease in Po resulted only from disruption or loss of force-producing or force-bearing elements within the muscle, then the relative decreases in caffeine contracture force should have been similar to those for Po. By directly activating the SR Ca2+ release channel (9), caffeine presumably bypassed the site of E-C coupling dysfunction and resulted in near-normal caffeine-induced force production.
The present results do not appear to correspond as closely with previous animal studies from several other laboratories. Although we recently reported that isometric torque deficits persist immediately after a second bout of 50 lengthening contractions in mouse anterior crural muscles (32), in three other rodent studies (17, 20, 29) the investigators concluded that there is a rapid adaptation in the muscle strength deficit after only one bout of lengthening contractions. These investigators based their conclusions on a single time point (i.e., 2 or 3 days) after the second injury bout (17, 20, 29), so it is possible that strength deficits occurred immediately after injury in these studies as well. Presumably, these investigators did not want to confound the measurement of strength deficit due to injury with that of metabolic fatigue because the nature of their “injury” protocols may have predisposed the muscle to fatigue.
It is also important to emphasize when the most meaningful adaptation occurred in the present study. The most significant change in the ability of the anterior crural muscles to withstand lengthening contraction-induced injury occurred between bouts 1 and 2. Compared with the first injury bout, the postinjury isometric torques increased at all stimulation frequencies after the second injury bout, albeit very little at the low frequencies. Thus the largest percent and absolute isometric torque changes occurred between bouts 1 and 2, with much smaller changes occurring between bouts 2 and 5 (Figs. 3 and 4). For example, relative changes in postinjury peak torque during isometric twitch contractions (i.e., 20 Hz) improved 10% from bout 1 to bout 2 and then deteriorated 3% from bout 2 to bout 5. Although relative changes in maximal isometric tetanic torque (i.e., at 300 Hz) improved during the five injury bouts, the largest improvement (13%) came after the second bout, whereas smaller improvements (7%) occurred after the next three bouts. Furthermore, absolute isometric torque deficits did not improve after the second injury bout at stimulation frequencies >80 Hz (Fig. 4). One possible explanation for these observations is based on the hypothesis proposed by Schwane and Armstrong (30). They suggested that prior lengthening contractions remove “stress-susceptible” muscle fibers, thereby resulting in decreased muscle damage and release of myocellular proteins into the circulation with subsequent exercise bouts. Because we do not observe decreases in the total number of muscle fibers with our in vivo injury model (11), it is conceivable that some fibers have stress-susceptible sarcomeres that are prone to injury that are removed after the first lengthening contraction bout. This would be consistent with Morgan's “popping sarcomere” hypothesis (22). Hence, damage and loss of myofibrillar proteins could contribute to the strength loss after the first bout if the protein loss resulted from segmental myofiber necrosis, but this contribution to the strength loss would predictably diminish after subsequent bouts. In support of this idea, we observed no significant reduction in myofibrillar protein content of the TA muscle after the fifth injury bout (Table 1), whereas myofibrillar protein content in this muscle is significantly reduced by 21% 5 days after one injury bout (unpublished observation). In addition, LaPointe and coworkers (17) recently reported that the repeated-bout effect was associated with significant reductions in inflammatory markers in rat skeletal muscle and thus presumably less damage to muscle after a second bout of lengthening contractions.
Thus our results and those from human studies (3, 5, 23–25) indicate that the immediate loss in isometric strength after lengthening contractions is not easily overcome with repeated bouts, unlike the rapid adaptations that occur in other markers of muscle injury, e.g., plasma levels of muscle cell proteins (4, 30) and delayed-onset soreness (4).
Role of E-C uncoupling in the immediate strength losses.
It is widely accepted that E-C uncoupling is a primary mechanism contributing to strength loss associated with low-frequency fatigue. In recent years, E-C uncoupling has also been shown to contribute to the immediate force deficit associated with lengthening contraction-induced injury in single muscle fibers (2) and in whole slow-twitch (36) and fast-twitch (13) muscles. Furthermore, using confocal laser scanning microscopy and Ca2+ minielectrode techniques, our group has (13) demonstrated that the excitation process responsible for initiating Ca2+ release from the SR is significantly impaired out to 3 days after injury of mouse EDL muscle.
On the basis of data from the present study, it appears that the persistent strength deficit observed in the EDL muscle immediately after the fifth injury bout is caused by E-C uncoupling. This conclusion is based primarily on the observation that the relative reductions (i.e., injured vs. contralateral control) in Po (−33%) far exceed the relative changes in peak caffeine contracture force (−7%) immediately after injury induction (Fig. 7). If the decrease in Po resulted only from disruption or loss of force-producing or force-bearing elements within the muscle, then the relative decreases in caffeine contracture force should have been similar to those for Po. By directly activating the SR Ca2+ release channel (9), caffeine presumably bypassed the site of E-C coupling dysfunction and resulted in near-normal caffeine-induced force production. This conclusion is based on the presumption that the caffeine and Ca2+ sensitivity of the myofilaments and the Ca2+ handling by the SR were not significantly affected by either maturation of the muscle (i.e., 8-wk study duration) or the five bouts of training with lengthening contractions. It appears unlikely that maturation of the skeletal muscles would enhance either SR or myofilament sensitivity to either Ca2+ or caffeine. For example, the sensitivity of the SR Ca2+ release channel to caffeine and the relative caffeine-induced force both decrease in mouse EDL muscle from birth to 60 days (28). Mouse EDL muscle exhibits some subtle changes in myosin phenotype from the age of 4 to 12 wk, with increases (∼20%) in fast myosin heavy chain isoforms and a decrease (∼20%) in the slow myosin heavy chain isoform (27). However, this type of change in the myofilament phenotype would be expected to result in a decrease in sensitivity to Ca2+ and caffeine (7). Whether five bouts of lengthening contractions induced alterations in the sensitivities of the myofilaments and/or the SR to Ca2+ and caffeine is unknown. However, the observation that myofibrillar protein is reduced 3 days after injury whereas caffeine-induced force is unchanged suggests that EDL muscles adapted to lengthening contraction-induced injury may display some type of altered SR Ca2+ handling, e.g., greater SR Ca2+ content (16). Therefore, comparing relative reductions in Po and caffeine-induced force may overestimate the contribution of E-C uncoupling to the isometric strength loss. Nonetheless, additional indirect evidence supporting E-C uncoupling as the primary factor for the strength deficit after the fifth injury bout comes from the observation that the relative reductions in EDL muscle twitch force (−49%) were greater than those in tetanic force (−33%). Greater relative reductions in twitch force compared with tetanic force have long been interpreted as an indicator of E-C uncoupling (6, 14), although this change could also indicate increased muscle compliance (15).
There is no direct evidence for E-C uncoupling being the primary mechanism for the strength deficit in the anterior crural group as a whole (i.e., TA muscle) after performance of lengthening contractions. However, indirect evidence suggests that this is the case. First, EMG data indicate that excitation failure in the TA muscle occurs downstream from the neuromuscular junction and plasmalemma after either a single bout (34) or multiple bouts (32) of injury. Second, there are disproportionate reductions in isometric torque produced at low stimulation frequencies compared with those at high stimulation frequencies after a single bout (34) and after multiple bouts (Fig. 3). Theoretically, low-frequency force production is affected to a greater extent than high-frequency force output when E-C coupling failure occurs (6, 14). Third, the proportion of the injured muscles showing histologically identified damage represents only a fraction of the relative strength deficit in most injury models (35). Fourth, previous studies indicate that the degree of muscle damage is reduced with lengthening contraction training (e.g., 30) even though there is little attenuation of the force loss (e.g., Ref. 23). Fifth, there are no significant reductions in total, myofibrillar, or soluble protein contents at any time in the TA muscle after the fifth injury bout (Table 1). If myofibrillar protein loss is minimal, then it is likely that some other mechanism(s) (e.g., E-C uncoupling) is responsible for the torque deficits observed in the anterior crural muscle group immediately after the fifth injury bout.
The persistence of the E-C uncoupling through repeated bouts of lengthening contractions lends credence to the hypothesis that the disruption of excitation is a protective mechanism that prevents more serious damage to the muscle structure when it is exposed to high-stress contractions. It would be of considerable interest to know how far into a lengthening contraction conditioning program the E-C uncoupling would continue. If in fact it does serve as a biological strategy to preserve the muscle, it could be hypothesized that this particular component of the force attenuation would never show significant adaptation.
Faster recovery of strength with repeated bouts.
The most striking adaptation of the mouse anterior crural muscles to repeated bouts of lengthening contractions was in the enhanced recovery rate of strength after the injury. This muscle group showed complete recovery of isometric torque by 3 days after the fifth injury bout (Fig. 5). The faster recovery of anterior crural muscle strength after the fifth injury bout cannot be explained simply by maturation. Age-matched control mice that performed a single injury bout at the same stage of maturation as the experimental mice when they performed the fifth injury bout exhibited little recovery of peak isometric strength by 3 days. The range in isometric strength loss at different stimulation frequencies (20–300 Hz) 3 days after this single injury bout was 39–70%, whereas the range in strength loss immediately after injury induction in the first study was 40–86%. Previously, our laboratory has shown that the range in isometric strength loss of the same muscle group at varying stimulation frequencies (20–400 Hz) was 35–72% 3 days after a single bout of lengthening contractions (34). This accelerated recovery in anterior crural muscle strength that we observed after five bouts of exercise with lengthening contractions in the present study is particularly impressive considering the increases in work done on the muscles over the five bouts. Work done on the muscles normalized to body weight during the fifth bout was ∼28% greater than that done during the first bout (Fig. 1, inset).
In this study, muscle strength was measured both in the anterior crural muscle group in vivo and in one of the muscles in the group (EDL) in vitro. In general, the adaptation in strength in EDL muscle was similar to that of the anterior crural muscle group as a whole after the five bouts of lengthening contractions. Because the TA muscle produces ∼89% of the isometric tetanic torque of the total anterior crural muscle group (19), it is conceivable that EDL muscle strength deficits could have been different from those of the group as a whole. However, the reduction (33%) in EDL muscle Po immediately after the fifth injury bout was similar to the reduction in anterior crural muscle peak tetanic torque (34%). Although the recovery in absolute strength of the EDL muscle lagged behind that of the muscle group as a whole, the recovery in EDL muscle Po was also accelerated compared with recovery from a single injury bout. EDL muscle Po was reduced only 17 and 7% at 3 and 5 days, respectively, after the fifth injury bout. In contrast, EDL muscle Po was reduced ∼50 and 35–49% at 3 and 5 days, respectively, after a single bout using the same injury model (11, 13).
The mechanisms underlying this enhanced rate of recovery with repeated bouts are unknown. However, it appears that the EDL muscles exhibited an accelerated loss and recovery of myofibrillar protein after the fifth injury bout. Compared with the contralateral control muscles, myofibrillar protein content of the injured EDL muscles decreased 15% by 3 days after the fifth injury bout and had recovered to control values by 5 days (Table 2). In contrast, we previously observed (11) that myofibrillar protein content of unconditioned EDL muscle was unchanged at 3 days (−4%) and significantly reduced at 5 (12%) and 14 (18%) days after a single injury bout. Collectively, these results suggest that, in the EDL muscle, E-C uncoupling and the myofibrillar protein disruption recover faster after the fifth injury bout compared with the first bout.
Adaptation of skeletal muscle to repeated bouts of lengthening contractions has been thought to involve a neural component (i.e., altered motor unit recruitment) as well as an intrinsic myocellular component (23, 32). For example, we recently concluded that changes in motor unit recruitment in human skeletal muscle occur with repeated bouts of lengthening contractions (32). However, given the number of studies investigating the adaptation of skeletal muscle to repeated bouts of lengthening contraction-induced muscle injury, it is surprising that few studies could be found in the literature that attempted to elucidate the mechanisms responsible for the adaptation of the muscle fibers per se. This remains a question that warrants further investigation.
This research was supported by the Omar Smith Chair and a seed grant from the College of Education and Human Development at Texas A&M University and by the Office of Research and Sponsored Programs and the College of Education at Georgia State University.
We thank the following Georgia State University graduate students for assistance in performing the protein assays: Stephanie Hinz and Eric Arnold.
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 © 2004 the American Physiological Society