- Email: [email protected]
Effects of immobilizing a single muscle on the morphology and the activation of its muscle fibers
Experimental Neurology 194 (2005) 495 – 505 www.elsevier.com/locate/yexnr
Effects of immobilizing a single muscle on the morphology and the activation of its muscle fibers Marie-Agne`s Giroux-Metgesa, Jean-Pierre Penneca, Julien Petitb, Julie Morela, He´le`ne Talarmina, Mickae¨l Drogueta, Germaine Dorangea, Maxime Giouxa,* a
Laboratoire de Physiologie, Faculte´ de me´decine, 22 Avenue Camille Desmoulins, CS 93837, 29238 BREST Cedex 3, France b Faculte´ des sciences du sport et de l’e´ducation physique, Avenue Camille Jullian, 33607 Pessac Cedex, France Received 10 August 2004; revised 10 January 2005; accepted 22 March 2005 Available online 25 April 2005
Abstract A single muscle of Wistar female rats, either soleus or peroneus longus, was immobilized by fixing its cut distal tendon to the bone during 8 weeks. We observed a transitory weight loss in both muscles; the mean fiber cross-sectional area (CSA) showed a reduction at day 30, followed by an increase at day 60. The time course of the activation of the immobilized muscle was evaluated by recording the chronic electromyographic (EMG) activity during short periods (1 min every other day) of treadmill locomotion. During immobilization, the integrated EMG amplitude of the soleus increased, reaching a maximum at 4 weeks, but remained close to control values during 8 weeks for the peroneus. The median frequency (MF) of the power density spectrum of the soleus EMG was not statistically different between immobilized and control muscles, while MF of the immobilized peroneus EMG was permanently higher than that of control muscles. This suggests two different modes of adaptation in motor unit command, depending on the muscle profile, which could be concomitant with the restoration of muscle fibers CSA after 8 weeks. D 2005 Elsevier Inc. All rights reserved. Keywords: Immobilization; Soleus muscle; Peroneus muscle; Electromyographic activity
Introduction Muscle immobilization could lead to several changes such as atrophy, or modifications in its motor drive. During extensive immobilization performed by plaster cast or joint pinning, muscle atrophy is presumably the consequence of the reduction in the motor drive. Previous studies have demonstrated that muscular atrophy was probably consequent to a reduction in the daily use of the muscle concerned, due to a decrease in its voluntary motor activity. In fact, muscular weight loss has been shown to be correlated with a decrease in electromyographic (EMG) activity. As regards the soleus fixed in a neutral position, Fournier et al. (1983) have demonstrated that integrated
* Corresponding author. Fax: +33 2 98016313. E-mail address: [email protected] (M. Gioux). 0014-4886/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2005.03.008
EMG amplitude was reduced to 50% of the control value after 4 weeks of immobilization. The data also showed that muscle weight/body weight ratio decreased to approximately 50% of the control value. Hnik et al. (1985) have also reported a great loss in muscle weight to about 67% of the control, with a simultaneous decrease in EMG activity (cumulative activity, number of counts per minute) to 10% of the control value on soleus immobilized in the shortest position. These different techniques often involve several joints within one single limb and sometimes within both limbs in the same animal. These models then lead to a quite complete disuse of the hindlimb. However, other models exist in which immobilization is limited to a single muscle. In the model described by Petit and Gioux (1993), immobilization is performed by the fixation of the cut distal muscular tendon of one single muscle to the adjacent bone. Since there is no joint immobilization or fixation, this does not
496
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
lead to major alterations in the behavior of the animals which are allowed to move freely and, moreover, this does not lead to limb disuse. However, it has been shown that this model leads to muscle atrophy that mainly affects type I fibers even if a partial recovery toward control values is observed in motor unit contractile properties at the 8th week of immobilization of the peroneus longus muscle. Thus, the question is: are these morphological modifications of muscle fibers related to some changes in their neural activation? The evolution of mean CSA of muscle fibers and EMG activity was analyzed after single muscle immobilization. Because most studies have demonstrated that some changes are, partly, dependent on the muscular profile, this experiment was conducted successively on two different muscles: a slow-twitch muscle, the Soleus, and a fast-twitch one, the Peroneus Longus. Muscle fiber cross-sectional area (CSA) after ATPase staining was measured at the end of the 4th and 8th weeks of immobilization, in order to estimate the amount of atrophy in the 3 main fiber types. Repeated recordings of EMG activity of the single immobilized muscle was performed in order to evaluate the modifications in muscle activation which could be the consequence of adaptations in the motor command to the lack of shortening of the immobilized muscle whereas global use of the hindlimb and joint mobility are retained. Because EMG characteristics are related to the mechanical characteristics of the contraction produced, variations in EMG of immobilized muscles could only be evaluated with a standardized motor task. Thus, we recorded the EMG activity in animals submitted to short periods of treadmill running, in both controls and animals with a single fixed muscle.
Methods Animal groups The experiments were performed on two groups of female Wistar rats weighing 389 T 36 g (Centre d’e´levage De´pre´, St. Doulchard, France) and 2 months old at the beginning of experiments. All rats had previously been selected according to their ability to walk on a motor-driven treadmill and conditioned to walk at constant speed (30.5 cm s 1), which is more easily obtained with female rats. This velocity was chosen because it corresponded to the spontaneous velocity range observed by authors who have studied normal locomotion in the rat (Clarke and Parker, 1986). Another session of experiments was conducted to study the EMG variations. Concerning EMG, data were recorded from 10 rats for Soleus (3 controls, 7 immobilized muscles) and from 7 rats for Peroneus Longus (3 controls, 4 immobilized muscles). Fiber cross-sectional areas were measured in another group of animals (for Peroneus Longus: 5 control muscles, 5 muscles immobilized during 4 weeks and 6 muscles immobilized during 8 weeks; for
Soleus: 5 control muscles, 4 muscles immobilized during 4 weeks and 5 muscles immobilized during 8 weeks). All procedures were performed according to our ethical regional committee recommendations. The experiments were authorized by a departmental agreement no. A29019-3 and performed according to the recommendations of the European Community directive no. 86/609. Muscle fixation For each surgical procedure, animals were anesthetized with an intraperitoneal injection of a mixture of Xylazine (10 mg kg 1) (Rompun 2%, Bayer) and Ketamine (100 mg kg 1) (Imalge`ne, Merial) which provided a 2-h sedation. Surgery was performed under aseptic conditions. Great care was taken not to damage blood vessels and nerve branches. For both series of experiments, the muscle under study was immobilized by securing its distal tendon to the bone. Firstly, the left leg was placed in an external device in order to maintain the ankle joint at 90-. Thus, the muscle was set to this resting length. Secondly, the distal tendon was dissected and, then, firmly attached to the bone with a nylon thread (gauge 6.0) which was passed through a hole (made with a drill) in the distal end of the fibula, before being cut from its distal insertion and just below the fixation. Soleus and Peroneus Longus are monoarticular muscles and act as ankle extensors. After this immobilization, both proximal and distal tendons were fixed to the same bone. Due to this fixation procedure, the immobilized muscles could not be passively stretched. Furthermore, the anatomical layout of neighboring muscle insertions could not lead to any stretching of the immobilized muscles. Neighboring muscles could only induce a very slight shortening of the immobilized muscles (Greene, 1955). The motor activation of immobilized muscles could lead to an isometric or a slightly concentric contraction because of the in-series elasticity of their tendons (Fig. 1). EMG electrode implantation Electrodes used for chronical recording were made from two Teflon-insulated multistranded stainless steel wires (75 Am diameter). The surgical procedure and anesthesia were the same as for muscle fixation. Firstly, the proximal ends of the electrodes were soldered to a connector secured to the skull by screws and dental cement. Secondly, the distal ends were inserted subcutaneously up to the muscle. The recording surfaces were standardized by removing a 1.5mm length of insulation and the ends were bent back to form a hook and, then, gently pushed between the muscle fibers; the inter-electrode distance was set at 5 mm. The wires were stitched to the fascia using a 7.0 Prolene suture. The same type of suture was used to close the fascia. At the end of each experiment, the location of the implanted devices was verified and we checked to ensure that the immobilized muscles remained firmly fixed. When the
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
497
analyses were performed for each sequence and the duration of the bursts was measured. Firstly, the amplitude of every burst in the sequence was obtained by dividing the area under the rectified and filtered EMG signal by the duration of the burst and then the mean burst amplitude for the sequence was calculated. Secondly, a power spectral analysis was performed using a fast Fourier transform of the raw EMG signal. For each burst in the sequence, the power density spectrum (PDS) and the median frequency (MF, which divides the PDS in 2 parts of equal energy content) were calculated. Then, the mean median frequency for the sequence was calculated (Spike 2 software, Cambridge Electronics Design). In order to clarify the graphs and because the variations in the mean parameters were found to be low, the values obtained during 3 consecutive recording sessions were averaged. At terminal surgery inspection, animals were excluded from the analysis when the electrodes were out of place and/ or the tendon fixation had slipped. Fig. 1. Surgical procedure for left muscle fixation. (1) Fibula. (2) Soleus or peroneus longus muscle. (3) Hole of 0.3 mm diameter made with a drill. (4) Calcaneum. After setting the muscle to its neutral length, the distal tendon was firmly attached to the distal end of the fibula with a strong suture (Prole`ne 6.0). One single muscle was, then, fixed at its two ends on the fibula in the left hindlimb.
electrodes were out of place, the animals were excluded from the analysis (3 rats from a total amount of 20 animals). A single muscle was recorded in each rat. Electrode implantation was performed in either the soleus or the peroneus longus of the left hindlimb with the muscle either fixed or used as control. Thus, we proceeded to collect EMG recordings in the 4 animal groups (control soleus, immobilized soleus, control peroneus, and fixed peroneus muscles) from the sixth day after implantation. Electromyogram recording and analysis Recordings began 6 days after surgery when the hindlimb had completely healed up and when the animals were able to walk in a normal way without any impairment. EMG was periodically recorded during a single period of 1-min regular walk, once a day, every other day, for up to 8 weeks. This task is different from training because of its short duration (1 min), low intensity (walking at 30.5 cm s 1) even though it is repetitive, and steady. The signal was amplified (Medelec MS6, band width: 8 Hz –8 kHz) and stored on a digital tape recorder (DTR 1803, Biologic France, sampling frequency: 48 kHz) for off-line analyses. For every recording, raw EMG was digitally rectified and low-pass filtered (50 Hz) using a Biopac Labpro software (MP 100 system, sampling frequency 5 kHz). The threshold used to define the start and the end of a burst was arbitrarily set to twice the background noise amplitude of the raw signal. For each recording session, a sequence which contained 15 T 3 successive bursts was selected. Two
Muscle trophicity Histological control Fiber cross-sectional area measurement. Because only the CSA of the 3 main fiber types was evaluated, as in the precedent study on the same model (Petit and Gioux, 1993), the ‘‘ATPase method’’ (Brooke and Kaiser, 1970) was used. Prior to this final experiment, each animal was killed with an overdose of pentobarbital (intraperitoneal injection). CSA was measured in 14 soleus muscles and 16 peroneus muscles. For each point of analysis and for each fiber type, the number of cells/number of muscles are shown in parentheses (Table 1). The muscles were quickly removed and immediately frozen at constant length in isopentane precooled with liquid nitrogen ( 150-C), then stored at 80-C until histochemical processing. Transverse sections of 8 Am thickness were cut, stained for myofibrillar ATPase activity after preincubation at pH 9.4, 4.63, or 4.35, and with Hematoxylin– Eosin. HE slides were checked by a histologist in order to verify the absence of necrotic structural alterations. Muscle fiber cross-sectional areas were measured on a microscope Olympus IX 70, with a camera using Labview software. The gross morphology of immobilized muscle fibers was similar to the corresponding control muscle fibers. The fibers were classified following their myofibrillar ATPase activity according to Brooke and Kaiser’s criteria (Brooke and Kaiser, 1970). Representative fascicles with fibers cut perpendicular to their long axes were then chosen for fiber area measurement. The mean cross-sectional area was calculated for each fiber type. Connective tissue. In order to evaluate the proportion of fibrosis, the ratio of connective tissue per muscle was calculated from the estimated surface of connective tissue
498
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
Table 1 Evolution of mean cross-sectional area in the immobilized muscles Fiber type
Soleus
Peroneus longus 2
Mean CSA (Am ) T SD (no. of cells/no. of muscles)
% Total number/group
Mean CSA (Am2) T SD (no. of cells/no. of muscles)
% Total number/group
I Controls 4 weeks 8 weeks
6775.8 T 2875.3 (698/5) 3738.6 T 1892.3 (907/4)** 5368.9 T 2790 (1028/5)**
92.8 97.3 96
2462.7 T 1066.3 (547/5) 2034.6 T 974.9 (367/5)** 2972.5 T 1229.5 (255/6)**
31.4 32.2 24.3
IIA Controls 4 weeks 8 weeks
2647.4 T 1822.2 (40/5) 2252 T 996.4 (16/4) 4510.5 T 2668.6 (33/5)**
5.3 1.7 3
2927.3 T 1420.3 (580/5) 2793.2 T 1198.6 (396/5) 4045.6 T 1748.5 (346/6)**
33.3 34.7 33
IIB Controls 4 weeks 8 weeks
3375.5 T 1018.47 (14/5) 3002.4 T 1476.4 (9/4) 5775.6 T 2207.7 (10/5)**
1.9 1 1
4590.2 T 3101.4 (616/5) 4161.9 T 1458.75 (378/5) 5417.5 T 2304.2 (448/6)**
35.3 33.1 42.7
Data are expressed as mean T SD. Numbers in parentheses represent no. of cells analyzed/no. of muscles. Fiber-type proportions are expressed in percentage of total cell number per group. ** Statistical difference (P < 0.01) between fiber CSA in immobilized muscles and corresponding control value.
obtained by subtracting the cumulated surfaces of fiber CSA from the whole surface of the slide examined. Muscle weight/body weight ratio was measured in 10 controls, 11 animals for immobilized soleus (6 after 4 weeks and 5 after 8 weeks immobilization), and 11 animals for fixed peroneus longus (6 after 4 weeks and 5 after 8 weeks immobilization). Protein content and immunoblotting of myosin heavy chains. Other samples of the muscles which were frozen for CSA measurement were homogenized in homogenization buffer (0.3 M sucrose, 20 mM HEPES, 1 mM azide sodium, pH 7.4). The protein content was measured with Bradford method (Bradford, 1976). The amount of protein per muscle was measured after extraction in control and immobilized muscles in order to evaluate the changes in protein synthesis at the same time points (4 and 8 weeks of immobilization). Proteins were diluted 1:1 with 2 Laemmli sample buffer (S3401, SIGMA), boiled and subjected to SDS-PAGE. The semi-quantitative analysis of myosin heavy chain (MHC) content was performed according to the protocol of Talmadge and Roy (1993). After separation by one-dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on 5% gels, proteins were transferred onto nitrocellulose membranes (Millipore). Subsequently, fast MHC were detected with monoclonal antibodies antifast skeletal myosin (M 4276, Sigma) and goat anti-mouse IgG-1 phosphatase alkaline conjugate (M32108, Tebu) used at a 1:1000 and 1:1250 dilutions, respectively.
median frequency), the analysis was performed with a two-way analysis of variance (ANOVA) with repeated measurements of two factors (controls versus immobilization and time). Changes in cross-sectional area (CSA) and in weight ratio were tested using a one-way ANOVA. Pairwise tests using Tukey analyses were used as a post hoc test to determine the significance between groups. Statistical significance was accepted at P < 0.05.
Results Muscle fiber cross-sectional area Data of CSA measurement are shown in Table 1. Chronic immobilization did not affect significantly these fiber-type distribution patterns whatever the muscle studied. In soleus muscles, the largest percentage was represented by type I. We found few type IIB fibers in the three groups. In peroneus fast muscles, type II fiber remained clearly the predominant fiber type even after immobilization. At day 30, the mean CSA of type I fibers was significantly reduced in the left fixed soleus muscles and also reduced in the left fixed peroneus muscles. Type II fibers were not affected by 4 weeks immobilization. At day 60, the tendency reversed for both the slow and the fast muscle. Except for soleus whose type I fiber CSA returned roughly to its control value, a significant fiber hypertrophy appeared in the 3 types of fibers of the 2 muscles. Muscle/body (w/w) ratio
Statistical analysis All results were given as means and standard deviations. For EMG parameters (burst duration, mean amplitude,
Immobilization induced a significant decrease of this ratio in both soleus and peroneus longus muscles at 4 weeks that is consistent with the decrease in CSA (P < 0.05). At 8 weeks,
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
data showed a significant rise of ratios corresponding to the recovery of CSA control values for type I fibers (Table 2). Protein content The time evolution pattern of the amount of total proteins was very similar to the evolution of the fiber CSA for both soleus and peroneus longus muscles. The protein content decreased in immobilized soleus at the 4th week (concomitantly with the decrease in the fiber CSA) and increased at the 8th week (slightly higher than the control value). For the fast muscle, total amount of protein did not show any obvious decrease at week 4 but showed a great increase at week 8 that pointed out the relative fiber hypertrophy (Table 2). Connective tissue In peroneus longus muscle, the percentage of connective tissue was quite steady over the whole experimental period: 10 T 1%, 11 T 1%, and 11 T 2% in controls, at 4 and 8 weeks, respectively. In immobilized soleus, we noticed a transient conjunctive proliferation muscle at week 4 (19 T 2% versus 11 T 2% in controls) with return to control value at week 8 (11 T 4%). General appearance of electromyographic activity Samples of raw EMG and corresponding rectifiedfiltered signals for the immobilized soleus are represented in Fig. 2A. A walking sequence EMG activity of the immobilized peroneus longus and its corresponding rectified-filtered signal are shown in Fig. 2B. The soleus was activated during most of the stance phase of walking while the peroneus longus displayed a double burst of EMG activity per step cycle during treadmill locomotion (a short early burst in the stance phase and a long one during most of the swing phase). The respective duration of soleus EMG bursts was 379.2 T 46.8 ms in control animals and was not different in fixed muscles (377.9 T 60.2 ms). No difference existed between the controls and the fixed muscles as
499
regards the peroneus longus mean EMG bursts: 313.0 T 20.6 ms in controls versus 326.9 T 42.5 ms in fixed muscles for the long burst, 69.3 T 15 ms in controls versus 66.6 T 1.4 ms in fixed muscles for the short one. The total duration of the peroneus longus muscle activity (including both bursts) was not different between the controls and the immobilized muscles. As shown in Fig. 3, and for the four groups (control soleus, fixed soleus, control peroneus longus, and fixed peroneus longus), muscle activation duration was almost constant over the 60 days and there was no statistical difference between the first and the last recording session. EMG amplitude A slight and nonsignificant increasing trend within each of the four groups (soleus or peroneus, fixed or controls) is observed in EMG amplitude between the 6th day and the 60th day (Fig. 4). At day 6, the mean absolute amplitude of the fixed soleus muscles was slightly higher than that of the controls (0.087 T 0.03 mV versus 0.074 T 0.005 mV). The amplitude of the fixed soleus gradually increased from day 6 (0.087 T 0.03 mV) (Fig. 4A) up to a maximum on day 36 (0.16 T 0.06 mV, 184% of initial value, P < 0.05) and returned at day 60 to a value which was not statistically different from that observed on day 6. On the contrary, EMG amplitude for the control soleus muscles was almost constant over the 60 days. The time course of the fixed soleus muscles amplitude was statistically different from that observed in control soleus muscles from day 26 to day 48 (P < 0.05). As regards the peroneus longus muscle (Fig. 4B), its behavior was quite different. At day 6, the mean absolute amplitude of the fixed muscle group was close to that of the control homonymous muscle group (0.0508 T 0.01 mV versus 0.0425 T 0.01 mV). Control peroneus muscles EMG amplitude did not change throughout the same experimental period. For the immobilized peroneus muscles, the data showed that EMG amplitude did not show any significant time course changes and was not statistically different from that of the controls all along the measurement period. Power density spectrum (PDS)
Table 2 Evolution of muscle weight/body weight ratio and protein content after immobilization Ratio [muscle weight (mg)/ body weight (g)] Soleus
Peroneus longus
Controls 4 weeks 8 weeks Controls 4 weeks 8 weeks
0.4 0.33 0.37 0.38 0.33 0.40
T T T T T T
0.07 0.02 0.03 0.02 0.01 0.02
(n (n (n (n (n (n
= = = = = =
10) 6)* 5)* 10) 6)* 5)*
Protein content [protein (Ag)/ muscle (mg)] 94.8 78.4 109.4 80.2 91.7 139.2
T T T T T T
9 6 24.7 16 8 5.2
Data are expressed as mean T SD. Numbers in parentheses represent no. of animals. * Statistical difference (P < 0.05) from control value.
The analysis of PDS was performed in the four groups of rats (control and fixed soleus, control and fixed peroneus). For the soleus, median frequency control values were close to 274 Hz. No difference was observed between the first and the following recording sessions. In the fixed group, MF value was close to 268 Hz. We also observed a decreasing trend in MF in this fixed group but this trend was not significant. No difference was found between the control and the fixed muscle groups (Fig. 5A). Conversely, we observed a significant difference in median frequency in the immobilized peroneus muscles starting and remaining at a higher steady-state level than that of the controls (459.78 Hz versus 304.23 Hz, respectively), all along the experimental
500
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
Fig. 2. Electromyographic activity of soleus and peroneus longus muscles during locomotion. Raw EMG sample (lower trace) and the corresponding rectifiedfiltered EMG sequence (upper trace) of a fixed soleus (panel A) and a fixed peroneus longus muscle (panel B). Activity is aligned on paw contact (vertical arrows).
period (max difference = 212.19 Hz at day 12, min difference = 104.31 Hz at day 60) (Fig. 5B). No significant time course change was observed within each peroneus group throughout the experimental period.
Discussion A single muscle immobilization does not change the proportions of muscle fiber types, and our results are in agreement with those of Armstrong and Phelps (1984) concerning the estimation of each fiber-type population in the two muscles under study. Furthermore, histological analysis did not show either muscle necrosis on the control and immobilized muscles or any conjunctive tissue proliferation. Changes in muscle/body weight ratios and the fiber CSA are consistent with CSA analysis which shows that muscular atrophy is present after 4 weeks of immobilization. This atrophy mainly affects type I fibers and
notably those of the soleus. We also observed a loss of muscle weight in both the soleus and peroneus longus. Our data are in agreement with Petit and Gioux (1993) who observed transient atrophy involving mainly slow motor units, slow-twitch muscles, and subsequently type I fibers, and a return toward control values at 8 weeks with an increase in protein content. In these experiments, we also noticed this phenomenon at 8 weeks. The soleus ratio increased (but not up to control value) and the peroneus ratio indicated a hypertrophy related to the increase in type II fiber CSA. This is in agreement with the results reported by Hortoba´gyi et al. (2000), who noted the many beneficial effects of exercise on immobilization-induced atrophy in humans. In fact, eccentric or mixed exercise training could prevent atrophy or even induce muscular hypertrophy, concomitantly with a strength gain. Similar conclusions were obtained in two other studies performed on the rat. Kannus et al. (1998) noted that the capillary density, the percentage of intramuscular connective tissue and the fiber
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
501
Fig. 3. Burst duration evolution during chronic immobilization. Soleus groups [controls (n = 3), open squares; fixed muscles (n = 7), filled squares] are represented in the upper graph (A) and peroneus longus groups in the lower graph (B) [controls (n = 3), open triangles; fixed muscles (n = 4), filled triangles]. Data are expressed as means T standard deviation (vertical bars).
size were restored almost to control levels in the immobilized soleus by low- or high-intensity treadmill running. Herbert et al. (1988) observed that muscle weight loss could be slowed down in animals which were submitted daily to short periods of repetitive climbing with an overload. In our experiment, the motor task imposed on the animals was quite different from physical training in the above-described data (intensive treadmill running, twice a day, with an uphill inclination of 10- to 30- in the Kannus study; climbing task with 75% body weight load strapped to the tail in Herbert’s study). Our results show that some effects of immobilization are transitory such as the decrease in fiber size. However, a small but regular activation of an immobilized muscle over a rather long period could be enough to maintain its trophicity and seems to be more efficient on type II than on type I muscle fibers whatever the muscle profile. This is in relation with a hypertrophy of the type II fibers which has been observed after isometric exercises (Aagaard et al., 2001). As shown in the example illustrated in Fig. 6, we noticed a similar evolution in fast myosin heavy chain revealed by immunoblotting. The fast MHC expression in the soleus muscle decreased dramatically at 4 weeks and then returned
to control level at 8 weeks. As regards the peroneus longus muscle, the amount of fast MHC did not diminish at 4 weeks; in fact, it increased at 8 weeks. The EMG signal which was recorded during walking in control animals, presented the same pattern of activation as that previously described in the rat (Westerga and Gramsbergen, 1994). The locomotor behavior of both control and immobilized animals remained identical. After muscle fixation, EMG signals (duration and amplitude) recorded at day 6 were very close to those of controls. In spite of iEMG amplitude and spectral content modifications, no difference was observed in iEMG burst duration and shape in fixed animals during the next 7 weeks. The voluntary motor pattern was unchanged during the locomotion task. In fact, some functional adaptations in the lumbar spinal cord circuitry have only been reported in spinal cord transected animals (Hodgson et al., 1994). All other experiments with motor or proprioceptive modifications limited to the hindlimb have highlighted the absence of plasticity of locomotor patterns. After tendinous transfer, Forssberg and Svartengren (1983) observed that the same gastrocnemius muscle motor pattern was retained in cats during locomotion
502
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
Fig. 4. Mean EMG amplitude during chronic immobilization. Amplitude is expressed in mean absolute values T standard deviation. For SD, only one bar is represented on the figures to make them more clearer. (A) Changes in the soleus muscle groups. Controls (n = 3), open squares; fixed muscles (n = 7), filled squares. (B) Changes in the peroneus muscle groups. Controls (n = 3), open triangles; fixed muscles (n = 4), filled triangles. *Statistical difference between immobilized and control group (P < 0.05).
although the muscle had an antagonistic position compared to the original. Other adaptations to hindlimb muscle impairments, such as partial denervations in the cat (Gritsenko et al., 2001), or hindlimb deafferentation (Grillner and Zanger, 1984) did not lead to modifications in spinal circuits. In immobilized muscles under spontaneous conditions of free movement, several authors have noted a drastic decrease in iEMG activity (Fischbach and Robbins, 1969; Fournier et al., 1983; Hnik et al., 1985). However, during a standardized motor task, we observed an increase in iEMG amplitude or MF at week 4, which could be interpreted as the adaptation of the muscular command to the decrease in the motor unit twitch tension and tetanic force (Duchateau and Hainaut, 1990; Petit and Gioux, 1993). Thus, in order to obtain the same muscular force required for standardized treadmill locomotion, several motor mechanisms could come into play depending primarily on the profile of the immobilized muscle. Firstly, we observed that fixed soleus iEMG amplitude increased and reached its maximal value at day 30, when the type I muscle fiber was at its smallest size. It could be suggested that fiber CSA atrophy could increase iEMG amplitude by bringing more fibers and action
potentials in the neighboring of recording electrodes but this is in part compensated by connective tissue proliferation observed at 4 weeks in soleus muscle. Thus, the number of motor units (MU) recruited in the soleus muscle probably increased as is the case with synergist muscle removal (Gardiner et al., 1986). Secondly, motor recruitment in the peroneus was very different: no significant iEMG increase was observed but the MF of the EMG burst PDS was higher from the onset than that of the controls and during the whole immobilization period. Consequently, the tonic and phasic muscle adaptations are different; two different ways and time courses were observed, which could be related to the functional role and the mechanical properties of each muscle. The soleus muscle has a postural function, with little amplitude shortening due to a short and resistant tendon. Its motor units are mainly slow-twitch, which do not need highfrequency recruitment. Thus, its adaptation to a force decrease at 4 weeks could involve more spatial recruitment as suggested by Rothwell (1994). These results are also in agreement with the motor unit firing frequencies recorded by Erim et al. (1996) during voluntary contraction with an increasing force. For the first recruited MU (slow-twitch ones
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
503
Fig. 5. Median frequency evolution during chronic immobilization. Data are expressed as mean T standard deviation. (A) Changes in the soleus muscle groups. Controls (n = 3), open squares; fixed muscles (n = 7), filled squares. (B) Changes in the peroneus muscle groups. Controls (n = 3), open triangles; fixed muscles (n = 4), filled triangles. *Statistical difference between immobilized and control group (P < 0.05).
Fig. 6. Examples of fast myosin heavy chain immunoblot. Data are expressed as the percentage of control densitometry for each muscle. (A) Soleus muscle. (B) Peroneus longus muscle.
according to the size principle), the muscular force increase is obtained by the spatial recruitment of an increasing number of motor units firing at low frequency. The spatial recruitment does not presumably need the very few fast units which are only involved at maximal force because of the great number and the high tetanic forces of slow motor units in soleus. Thus, the increase in force at 4 weeks is probably obtained by increasing the number of slow units recruited at low frequencies which do not lead to an increase in MF. This adaptation, presumably mediated by the signals coming from Golgi tendon organs, is not so necessary at 8 weeks because of the restoration of muscle contractile properties corresponding to the restoration of muscle force. The peroneus longus is composed of only one third of slow units, whose tetanic forces are very small compared both to the fast-twitch units of the same muscle and the slow ones of the soleus (Emonet-De´nand et al., 1988). Thus, the recruitment quickly involves a large number of fast-twitch units allowing fast muscle shortening. By increasing stimulation frequencies in experimental situations during movement, it is more efficient to increase muscular force of fast units than for
504
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
slow ones (Petit et al., 2003). This increase in the fast motor unit firing rate plays a role in enhancing high frequencies in the EMG power spectrum (Fuglsang-Frederiksen and Rønager, 1988). We observed the same evolution in the EMG power spectrum of a postural muscle whose motor unit contraction time became shorter, after its distal tendon had been transferred to a more compliant one (Giroux-Metges et al., 2003): the FFT of chronic EMG of the soleus muscle exhibited higher frequencies during the same movement, and this was observed during the 8-week EMG recording period. Thus, the motor command adaptation is different in time course (during 8 weeks) and pattern (MF instead of iEMG) between soleus and peroneus longus muscles. The PL muscle could be more sensitive to information coming from muscle spindle receptors regarding the lack of muscle shortening. But this enhancement of contraction force (with a contractile protein content increase at 8 weeks) is not very helpful to facilitate the shortening course of this immobilized muscle and MF remained high during the 8 weeks. The adaptation observed in motor unit frequencies persisted during the whole immobilization period. These nervous regulations could also explain the dependence of atrophy on the length at which the muscle is fixed, with the maximal weight loss at short position (Fournier et al., 1983; Hnik et al., 1985; Simard et al., 1982). Immobilization in a shortened position reduces discharge of spindle receptors or tendon organ, in which sensitivities remain intact during immobilization (Gioux and Petit, 1993; Nordstrom et al., 1995), leading to neural adaptations of either central or spinal muscle commands. In our experiments, proprioceptive signals coming from the immobilized muscle, indicating a lack of shortening or lengthening, lead to an adaptation of the motor unit drive, presumably at the spinal level. This adaptation in the immobilized muscle command, which depends on the muscle profile, could be concomitant with the restoration of muscle fiber CSA at 8 weeks, in spite of very short treadmill locomotion sessions. These results are also different from those obtained with global immobilization, which induces definitive atrophy, concomitantly with the disuse of the non-performant limb. In fact, immobilizing a single muscle does not disturb very much the motor programs involving hindlimbs, because the movements and postures still work, and because all other proprioceptive afferents coming from intact muscles are not impaired. Thus, during posture or locomotion, the immobilized muscle is still globally driven as before, except for some adaptations limited to its motor unit recruitment in order to try to compensate the muscular weakness or the lack of shortening.
Acknowledgments We thank Danielle Gillet for the histochemical analysis, and Jean-Franc¸ois Cle´ment for helping with the signal analysis software.
References Aagaard, P., Andersen, J.L., Dyhre-Poulsen, P., Leffers, A.M., Wagner, A., Magnusson, S.P., Halkjaer-Kristensen, J., Simonsen, E.B., 2001. A mechanism for increased contractile strength training: changes in muscle architecture. J. Physiol. 534.2, 613 – 623. Armstrong, R.B., Phelps, R.O., 1984. Muscle fiber type composition of the rat hindlimb. Am. J. Anat. 171, 259 – 272. Bradford, M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein dye binding. Anal. Biochem. 72, 248 – 254. Brooke, M.H., Kaiser, K.K., 1970. Muscle fiber types: how many and what kind? Arch. Neurol. 23, 369 – 379. Clarke, K.A., Parker, A., 1986. A quantitative study of normal locomotion in the rat. Physiol. Behav. 38, 345 – 351. Duchateau, J., Hainaut, K., 1990. Effects of immobilization on contractile properties, recruitment and firing rates of human motor units. J. Physiol. 422, 55 – 65. Emonet-De´nand, F., Hunt, C.C., Petit, J., Pollin, B., 1988. Proportion of fatigue-resistant motor units in hindlimb muscles of cat and their relation to axonal conduction velocity. J. Physiol. 400, 135 – 158. Erim, Z., De Luca, C.J., Mineo, K., Aoki, T., 1996. Rank-ordered regulation of motor units. Muscle Nerve 19, 5633 – 5673. Fischbach, G., Robbins, N., 1969. Changes in contractile properties of disused soleus muscles. J. Physiol. 201, 305 – 320. Forssberg, H., Svartengren, G., 1983. Hardwired locomotor network in cat revealed by a retained motor pattern to gastrocnemius after muscle transposition. Neurosci. Lett. 41, 283 – 288. Fournier, M., Roy, R.R., Perham, H., Simard, C.P., Edgerton, V.R., 1983. Is limb immobilization a model of muscle disuse? Exp. Neurol. 80, 147 – 156. Fuglsang-Frederiksen, A., Rønager, J., 1988. The motor unit firing rate and the power spectrum of EMG in humans. Electroencephalogr. Clin. Neurophysiol. 70, 68 – 72. Gardiner, P., Michel, R., Browman, C., Noble, E., 1986. Increased EMG of rat plantaris during locomotion following surgical removal of its synergists. Brain Res. 380, 114 – 121. Gioux, M., Petit, J., 1993. Effects of immobilizing the cat peroneus longus muscle on the activity of its own spindles. J. Appl. Physiol. 75, 2629 – 2635. Giroux-Metges, M.A., Pennec, J.P., Petit, J., Goanvec, C., Dorange, G., Gioux, M., 2003. Motor unit properties in the soleus muscle after its distal tendon transfer to the plantaris muscle tendon in the rat. J. Physiol. 553.3, 925 – 933. Greene, E.C., 1955. Anatomy of the rat. Trans. Am. Philos. Soc. vol. XXVII. Hafner Publishing Co., New York. Grillner, S., Zanger, P., 1984. The effect of dorsal root transection on the efferent motor pattern in the cat’s hindlimb locomotion. Acta Physiol. Scand. 120, 393 – 405. Gritsenko, V., Mushahwar, V., Prochazka, A., 2001. Adaptive changes in locomotor control after partial denervation of triceps surae muscles in the cat. J. Physiol. 533.1, 299 – 311. Herbert, M.E., Roy, R.R., Edgerton, V.R., 1988. Influence of one-week hindlimb suspension and intermittent high load exercise on rat muscles. Exp. Neurol. 102, 190 – 198. Hnik, P., Vejsada, R., Goldspink, D.F., Kasicki, S., Krekule, I., 1985. Quantitative evaluation of electromyogram activity in rat extensor and flexor muscles immobilized at different lengths. Exp. Neurol. 88, 515 – 528. Hodgson, J.A., Roy, R.R., Leon, R., Dobkin, B., Edgerton, V.R., 1994. Can the mammalian lumbar spinal cord learn a motor task? Med. Sci. Sports Exerc. 26, 1491 – 1497. Hortoba´gyi, T., Dempsey, L., Fraser, D., Zheng, D., Hamilton, G., Lambert, J., Dohm, L., 2000. Changes in muscle strength, muscle size and myofibrillar gene expression after immobilization and retraining in humans. J. Physiol. 524, 293 – 304. Kannus, P., Jozsa, L., Jarvinen, T.L.N., Kvist, M., Vieno, T., Jarvinen, T.A.H., Natri, A., Jarvinen, M., 1998. Free immobilization and lowto high-intensity exercise in immobilization-induced muscle atrophy. J. Appl. Physiol. 84, 1418 – 1424.
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505 Nordstrom, M.A., Enoka, R.M., Reinking, R.M., Callister, R.C., Stuart, D.G., 1995. Reduced motor unit activation of muscle spindles and tendon organs in the immobilized cat hindlimb. J. Appl. Physiol. 78, 901 – 913. Petit, J., Gioux, M., 1993. Properties of motor units after immobilization of cat peroneus longus muscle. J. Appl. Physiol. 74, 1131 – 1139. Petit, J., Giroux-Metges, M.A., Gioux, M., 2003. Power developed by motor units of the Peroneus tertius muscle of the cat. J. Neurophysiol. 90 (5), 3095 – 3104.
505
Rothwell, J.C., 1994. The motor unit. Control of Human Voluntary Movement. Chapman and Hall, London, pp. 35 – 85. Simard, C.P., Spector, S.A., Edgerton, V.R., 1982. Contractile properties of rat hind limb muscles immobilized at different lengths. Exp. Neurol. 77, 467 – 482. Talmadge, R.J., Roy, R.R., 1993. Electrophoretic separation of rat skeletal muscle myosin heavy-chain isoforms. J. Appl. Physiol. 75, 2337 – 2340. Westerga, J., Gramsbergen, A., 1994. Development of the EMG of the soleus muscle in the rat. Dev. Brain Res. 80, 233 – 243.
Effects of immobilizing a single muscle on the morphology and the activation of its muscle fibers Marie-Agne`s Giroux-Metgesa, Jean-Pierre Penneca, Julien Petitb, Julie Morela, He´le`ne Talarmina, Mickae¨l Drogueta, Germaine Dorangea, Maxime Giouxa,* a
Laboratoire de Physiologie, Faculte´ de me´decine, 22 Avenue Camille Desmoulins, CS 93837, 29238 BREST Cedex 3, France b Faculte´ des sciences du sport et de l’e´ducation physique, Avenue Camille Jullian, 33607 Pessac Cedex, France Received 10 August 2004; revised 10 January 2005; accepted 22 March 2005 Available online 25 April 2005
Abstract A single muscle of Wistar female rats, either soleus or peroneus longus, was immobilized by fixing its cut distal tendon to the bone during 8 weeks. We observed a transitory weight loss in both muscles; the mean fiber cross-sectional area (CSA) showed a reduction at day 30, followed by an increase at day 60. The time course of the activation of the immobilized muscle was evaluated by recording the chronic electromyographic (EMG) activity during short periods (1 min every other day) of treadmill locomotion. During immobilization, the integrated EMG amplitude of the soleus increased, reaching a maximum at 4 weeks, but remained close to control values during 8 weeks for the peroneus. The median frequency (MF) of the power density spectrum of the soleus EMG was not statistically different between immobilized and control muscles, while MF of the immobilized peroneus EMG was permanently higher than that of control muscles. This suggests two different modes of adaptation in motor unit command, depending on the muscle profile, which could be concomitant with the restoration of muscle fibers CSA after 8 weeks. D 2005 Elsevier Inc. All rights reserved. Keywords: Immobilization; Soleus muscle; Peroneus muscle; Electromyographic activity
Introduction Muscle immobilization could lead to several changes such as atrophy, or modifications in its motor drive. During extensive immobilization performed by plaster cast or joint pinning, muscle atrophy is presumably the consequence of the reduction in the motor drive. Previous studies have demonstrated that muscular atrophy was probably consequent to a reduction in the daily use of the muscle concerned, due to a decrease in its voluntary motor activity. In fact, muscular weight loss has been shown to be correlated with a decrease in electromyographic (EMG) activity. As regards the soleus fixed in a neutral position, Fournier et al. (1983) have demonstrated that integrated
* Corresponding author. Fax: +33 2 98016313. E-mail address: [email protected] (M. Gioux). 0014-4886/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2005.03.008
EMG amplitude was reduced to 50% of the control value after 4 weeks of immobilization. The data also showed that muscle weight/body weight ratio decreased to approximately 50% of the control value. Hnik et al. (1985) have also reported a great loss in muscle weight to about 67% of the control, with a simultaneous decrease in EMG activity (cumulative activity, number of counts per minute) to 10% of the control value on soleus immobilized in the shortest position. These different techniques often involve several joints within one single limb and sometimes within both limbs in the same animal. These models then lead to a quite complete disuse of the hindlimb. However, other models exist in which immobilization is limited to a single muscle. In the model described by Petit and Gioux (1993), immobilization is performed by the fixation of the cut distal muscular tendon of one single muscle to the adjacent bone. Since there is no joint immobilization or fixation, this does not
496
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
lead to major alterations in the behavior of the animals which are allowed to move freely and, moreover, this does not lead to limb disuse. However, it has been shown that this model leads to muscle atrophy that mainly affects type I fibers even if a partial recovery toward control values is observed in motor unit contractile properties at the 8th week of immobilization of the peroneus longus muscle. Thus, the question is: are these morphological modifications of muscle fibers related to some changes in their neural activation? The evolution of mean CSA of muscle fibers and EMG activity was analyzed after single muscle immobilization. Because most studies have demonstrated that some changes are, partly, dependent on the muscular profile, this experiment was conducted successively on two different muscles: a slow-twitch muscle, the Soleus, and a fast-twitch one, the Peroneus Longus. Muscle fiber cross-sectional area (CSA) after ATPase staining was measured at the end of the 4th and 8th weeks of immobilization, in order to estimate the amount of atrophy in the 3 main fiber types. Repeated recordings of EMG activity of the single immobilized muscle was performed in order to evaluate the modifications in muscle activation which could be the consequence of adaptations in the motor command to the lack of shortening of the immobilized muscle whereas global use of the hindlimb and joint mobility are retained. Because EMG characteristics are related to the mechanical characteristics of the contraction produced, variations in EMG of immobilized muscles could only be evaluated with a standardized motor task. Thus, we recorded the EMG activity in animals submitted to short periods of treadmill running, in both controls and animals with a single fixed muscle.
Methods Animal groups The experiments were performed on two groups of female Wistar rats weighing 389 T 36 g (Centre d’e´levage De´pre´, St. Doulchard, France) and 2 months old at the beginning of experiments. All rats had previously been selected according to their ability to walk on a motor-driven treadmill and conditioned to walk at constant speed (30.5 cm s 1), which is more easily obtained with female rats. This velocity was chosen because it corresponded to the spontaneous velocity range observed by authors who have studied normal locomotion in the rat (Clarke and Parker, 1986). Another session of experiments was conducted to study the EMG variations. Concerning EMG, data were recorded from 10 rats for Soleus (3 controls, 7 immobilized muscles) and from 7 rats for Peroneus Longus (3 controls, 4 immobilized muscles). Fiber cross-sectional areas were measured in another group of animals (for Peroneus Longus: 5 control muscles, 5 muscles immobilized during 4 weeks and 6 muscles immobilized during 8 weeks; for
Soleus: 5 control muscles, 4 muscles immobilized during 4 weeks and 5 muscles immobilized during 8 weeks). All procedures were performed according to our ethical regional committee recommendations. The experiments were authorized by a departmental agreement no. A29019-3 and performed according to the recommendations of the European Community directive no. 86/609. Muscle fixation For each surgical procedure, animals were anesthetized with an intraperitoneal injection of a mixture of Xylazine (10 mg kg 1) (Rompun 2%, Bayer) and Ketamine (100 mg kg 1) (Imalge`ne, Merial) which provided a 2-h sedation. Surgery was performed under aseptic conditions. Great care was taken not to damage blood vessels and nerve branches. For both series of experiments, the muscle under study was immobilized by securing its distal tendon to the bone. Firstly, the left leg was placed in an external device in order to maintain the ankle joint at 90-. Thus, the muscle was set to this resting length. Secondly, the distal tendon was dissected and, then, firmly attached to the bone with a nylon thread (gauge 6.0) which was passed through a hole (made with a drill) in the distal end of the fibula, before being cut from its distal insertion and just below the fixation. Soleus and Peroneus Longus are monoarticular muscles and act as ankle extensors. After this immobilization, both proximal and distal tendons were fixed to the same bone. Due to this fixation procedure, the immobilized muscles could not be passively stretched. Furthermore, the anatomical layout of neighboring muscle insertions could not lead to any stretching of the immobilized muscles. Neighboring muscles could only induce a very slight shortening of the immobilized muscles (Greene, 1955). The motor activation of immobilized muscles could lead to an isometric or a slightly concentric contraction because of the in-series elasticity of their tendons (Fig. 1). EMG electrode implantation Electrodes used for chronical recording were made from two Teflon-insulated multistranded stainless steel wires (75 Am diameter). The surgical procedure and anesthesia were the same as for muscle fixation. Firstly, the proximal ends of the electrodes were soldered to a connector secured to the skull by screws and dental cement. Secondly, the distal ends were inserted subcutaneously up to the muscle. The recording surfaces were standardized by removing a 1.5mm length of insulation and the ends were bent back to form a hook and, then, gently pushed between the muscle fibers; the inter-electrode distance was set at 5 mm. The wires were stitched to the fascia using a 7.0 Prolene suture. The same type of suture was used to close the fascia. At the end of each experiment, the location of the implanted devices was verified and we checked to ensure that the immobilized muscles remained firmly fixed. When the
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
497
analyses were performed for each sequence and the duration of the bursts was measured. Firstly, the amplitude of every burst in the sequence was obtained by dividing the area under the rectified and filtered EMG signal by the duration of the burst and then the mean burst amplitude for the sequence was calculated. Secondly, a power spectral analysis was performed using a fast Fourier transform of the raw EMG signal. For each burst in the sequence, the power density spectrum (PDS) and the median frequency (MF, which divides the PDS in 2 parts of equal energy content) were calculated. Then, the mean median frequency for the sequence was calculated (Spike 2 software, Cambridge Electronics Design). In order to clarify the graphs and because the variations in the mean parameters were found to be low, the values obtained during 3 consecutive recording sessions were averaged. At terminal surgery inspection, animals were excluded from the analysis when the electrodes were out of place and/ or the tendon fixation had slipped. Fig. 1. Surgical procedure for left muscle fixation. (1) Fibula. (2) Soleus or peroneus longus muscle. (3) Hole of 0.3 mm diameter made with a drill. (4) Calcaneum. After setting the muscle to its neutral length, the distal tendon was firmly attached to the distal end of the fibula with a strong suture (Prole`ne 6.0). One single muscle was, then, fixed at its two ends on the fibula in the left hindlimb.
electrodes were out of place, the animals were excluded from the analysis (3 rats from a total amount of 20 animals). A single muscle was recorded in each rat. Electrode implantation was performed in either the soleus or the peroneus longus of the left hindlimb with the muscle either fixed or used as control. Thus, we proceeded to collect EMG recordings in the 4 animal groups (control soleus, immobilized soleus, control peroneus, and fixed peroneus muscles) from the sixth day after implantation. Electromyogram recording and analysis Recordings began 6 days after surgery when the hindlimb had completely healed up and when the animals were able to walk in a normal way without any impairment. EMG was periodically recorded during a single period of 1-min regular walk, once a day, every other day, for up to 8 weeks. This task is different from training because of its short duration (1 min), low intensity (walking at 30.5 cm s 1) even though it is repetitive, and steady. The signal was amplified (Medelec MS6, band width: 8 Hz –8 kHz) and stored on a digital tape recorder (DTR 1803, Biologic France, sampling frequency: 48 kHz) for off-line analyses. For every recording, raw EMG was digitally rectified and low-pass filtered (50 Hz) using a Biopac Labpro software (MP 100 system, sampling frequency 5 kHz). The threshold used to define the start and the end of a burst was arbitrarily set to twice the background noise amplitude of the raw signal. For each recording session, a sequence which contained 15 T 3 successive bursts was selected. Two
Muscle trophicity Histological control Fiber cross-sectional area measurement. Because only the CSA of the 3 main fiber types was evaluated, as in the precedent study on the same model (Petit and Gioux, 1993), the ‘‘ATPase method’’ (Brooke and Kaiser, 1970) was used. Prior to this final experiment, each animal was killed with an overdose of pentobarbital (intraperitoneal injection). CSA was measured in 14 soleus muscles and 16 peroneus muscles. For each point of analysis and for each fiber type, the number of cells/number of muscles are shown in parentheses (Table 1). The muscles were quickly removed and immediately frozen at constant length in isopentane precooled with liquid nitrogen ( 150-C), then stored at 80-C until histochemical processing. Transverse sections of 8 Am thickness were cut, stained for myofibrillar ATPase activity after preincubation at pH 9.4, 4.63, or 4.35, and with Hematoxylin– Eosin. HE slides were checked by a histologist in order to verify the absence of necrotic structural alterations. Muscle fiber cross-sectional areas were measured on a microscope Olympus IX 70, with a camera using Labview software. The gross morphology of immobilized muscle fibers was similar to the corresponding control muscle fibers. The fibers were classified following their myofibrillar ATPase activity according to Brooke and Kaiser’s criteria (Brooke and Kaiser, 1970). Representative fascicles with fibers cut perpendicular to their long axes were then chosen for fiber area measurement. The mean cross-sectional area was calculated for each fiber type. Connective tissue. In order to evaluate the proportion of fibrosis, the ratio of connective tissue per muscle was calculated from the estimated surface of connective tissue
498
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
Table 1 Evolution of mean cross-sectional area in the immobilized muscles Fiber type
Soleus
Peroneus longus 2
Mean CSA (Am ) T SD (no. of cells/no. of muscles)
% Total number/group
Mean CSA (Am2) T SD (no. of cells/no. of muscles)
% Total number/group
I Controls 4 weeks 8 weeks
6775.8 T 2875.3 (698/5) 3738.6 T 1892.3 (907/4)** 5368.9 T 2790 (1028/5)**
92.8 97.3 96
2462.7 T 1066.3 (547/5) 2034.6 T 974.9 (367/5)** 2972.5 T 1229.5 (255/6)**
31.4 32.2 24.3
IIA Controls 4 weeks 8 weeks
2647.4 T 1822.2 (40/5) 2252 T 996.4 (16/4) 4510.5 T 2668.6 (33/5)**
5.3 1.7 3
2927.3 T 1420.3 (580/5) 2793.2 T 1198.6 (396/5) 4045.6 T 1748.5 (346/6)**
33.3 34.7 33
IIB Controls 4 weeks 8 weeks
3375.5 T 1018.47 (14/5) 3002.4 T 1476.4 (9/4) 5775.6 T 2207.7 (10/5)**
1.9 1 1
4590.2 T 3101.4 (616/5) 4161.9 T 1458.75 (378/5) 5417.5 T 2304.2 (448/6)**
35.3 33.1 42.7
Data are expressed as mean T SD. Numbers in parentheses represent no. of cells analyzed/no. of muscles. Fiber-type proportions are expressed in percentage of total cell number per group. ** Statistical difference (P < 0.01) between fiber CSA in immobilized muscles and corresponding control value.
obtained by subtracting the cumulated surfaces of fiber CSA from the whole surface of the slide examined. Muscle weight/body weight ratio was measured in 10 controls, 11 animals for immobilized soleus (6 after 4 weeks and 5 after 8 weeks immobilization), and 11 animals for fixed peroneus longus (6 after 4 weeks and 5 after 8 weeks immobilization). Protein content and immunoblotting of myosin heavy chains. Other samples of the muscles which were frozen for CSA measurement were homogenized in homogenization buffer (0.3 M sucrose, 20 mM HEPES, 1 mM azide sodium, pH 7.4). The protein content was measured with Bradford method (Bradford, 1976). The amount of protein per muscle was measured after extraction in control and immobilized muscles in order to evaluate the changes in protein synthesis at the same time points (4 and 8 weeks of immobilization). Proteins were diluted 1:1 with 2 Laemmli sample buffer (S3401, SIGMA), boiled and subjected to SDS-PAGE. The semi-quantitative analysis of myosin heavy chain (MHC) content was performed according to the protocol of Talmadge and Roy (1993). After separation by one-dimensional sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on 5% gels, proteins were transferred onto nitrocellulose membranes (Millipore). Subsequently, fast MHC were detected with monoclonal antibodies antifast skeletal myosin (M 4276, Sigma) and goat anti-mouse IgG-1 phosphatase alkaline conjugate (M32108, Tebu) used at a 1:1000 and 1:1250 dilutions, respectively.
median frequency), the analysis was performed with a two-way analysis of variance (ANOVA) with repeated measurements of two factors (controls versus immobilization and time). Changes in cross-sectional area (CSA) and in weight ratio were tested using a one-way ANOVA. Pairwise tests using Tukey analyses were used as a post hoc test to determine the significance between groups. Statistical significance was accepted at P < 0.05.
Results Muscle fiber cross-sectional area Data of CSA measurement are shown in Table 1. Chronic immobilization did not affect significantly these fiber-type distribution patterns whatever the muscle studied. In soleus muscles, the largest percentage was represented by type I. We found few type IIB fibers in the three groups. In peroneus fast muscles, type II fiber remained clearly the predominant fiber type even after immobilization. At day 30, the mean CSA of type I fibers was significantly reduced in the left fixed soleus muscles and also reduced in the left fixed peroneus muscles. Type II fibers were not affected by 4 weeks immobilization. At day 60, the tendency reversed for both the slow and the fast muscle. Except for soleus whose type I fiber CSA returned roughly to its control value, a significant fiber hypertrophy appeared in the 3 types of fibers of the 2 muscles. Muscle/body (w/w) ratio
Statistical analysis All results were given as means and standard deviations. For EMG parameters (burst duration, mean amplitude,
Immobilization induced a significant decrease of this ratio in both soleus and peroneus longus muscles at 4 weeks that is consistent with the decrease in CSA (P < 0.05). At 8 weeks,
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
data showed a significant rise of ratios corresponding to the recovery of CSA control values for type I fibers (Table 2). Protein content The time evolution pattern of the amount of total proteins was very similar to the evolution of the fiber CSA for both soleus and peroneus longus muscles. The protein content decreased in immobilized soleus at the 4th week (concomitantly with the decrease in the fiber CSA) and increased at the 8th week (slightly higher than the control value). For the fast muscle, total amount of protein did not show any obvious decrease at week 4 but showed a great increase at week 8 that pointed out the relative fiber hypertrophy (Table 2). Connective tissue In peroneus longus muscle, the percentage of connective tissue was quite steady over the whole experimental period: 10 T 1%, 11 T 1%, and 11 T 2% in controls, at 4 and 8 weeks, respectively. In immobilized soleus, we noticed a transient conjunctive proliferation muscle at week 4 (19 T 2% versus 11 T 2% in controls) with return to control value at week 8 (11 T 4%). General appearance of electromyographic activity Samples of raw EMG and corresponding rectifiedfiltered signals for the immobilized soleus are represented in Fig. 2A. A walking sequence EMG activity of the immobilized peroneus longus and its corresponding rectified-filtered signal are shown in Fig. 2B. The soleus was activated during most of the stance phase of walking while the peroneus longus displayed a double burst of EMG activity per step cycle during treadmill locomotion (a short early burst in the stance phase and a long one during most of the swing phase). The respective duration of soleus EMG bursts was 379.2 T 46.8 ms in control animals and was not different in fixed muscles (377.9 T 60.2 ms). No difference existed between the controls and the fixed muscles as
499
regards the peroneus longus mean EMG bursts: 313.0 T 20.6 ms in controls versus 326.9 T 42.5 ms in fixed muscles for the long burst, 69.3 T 15 ms in controls versus 66.6 T 1.4 ms in fixed muscles for the short one. The total duration of the peroneus longus muscle activity (including both bursts) was not different between the controls and the immobilized muscles. As shown in Fig. 3, and for the four groups (control soleus, fixed soleus, control peroneus longus, and fixed peroneus longus), muscle activation duration was almost constant over the 60 days and there was no statistical difference between the first and the last recording session. EMG amplitude A slight and nonsignificant increasing trend within each of the four groups (soleus or peroneus, fixed or controls) is observed in EMG amplitude between the 6th day and the 60th day (Fig. 4). At day 6, the mean absolute amplitude of the fixed soleus muscles was slightly higher than that of the controls (0.087 T 0.03 mV versus 0.074 T 0.005 mV). The amplitude of the fixed soleus gradually increased from day 6 (0.087 T 0.03 mV) (Fig. 4A) up to a maximum on day 36 (0.16 T 0.06 mV, 184% of initial value, P < 0.05) and returned at day 60 to a value which was not statistically different from that observed on day 6. On the contrary, EMG amplitude for the control soleus muscles was almost constant over the 60 days. The time course of the fixed soleus muscles amplitude was statistically different from that observed in control soleus muscles from day 26 to day 48 (P < 0.05). As regards the peroneus longus muscle (Fig. 4B), its behavior was quite different. At day 6, the mean absolute amplitude of the fixed muscle group was close to that of the control homonymous muscle group (0.0508 T 0.01 mV versus 0.0425 T 0.01 mV). Control peroneus muscles EMG amplitude did not change throughout the same experimental period. For the immobilized peroneus muscles, the data showed that EMG amplitude did not show any significant time course changes and was not statistically different from that of the controls all along the measurement period. Power density spectrum (PDS)
Table 2 Evolution of muscle weight/body weight ratio and protein content after immobilization Ratio [muscle weight (mg)/ body weight (g)] Soleus
Peroneus longus
Controls 4 weeks 8 weeks Controls 4 weeks 8 weeks
0.4 0.33 0.37 0.38 0.33 0.40
T T T T T T
0.07 0.02 0.03 0.02 0.01 0.02
(n (n (n (n (n (n
= = = = = =
10) 6)* 5)* 10) 6)* 5)*
Protein content [protein (Ag)/ muscle (mg)] 94.8 78.4 109.4 80.2 91.7 139.2
T T T T T T
9 6 24.7 16 8 5.2
Data are expressed as mean T SD. Numbers in parentheses represent no. of animals. * Statistical difference (P < 0.05) from control value.
The analysis of PDS was performed in the four groups of rats (control and fixed soleus, control and fixed peroneus). For the soleus, median frequency control values were close to 274 Hz. No difference was observed between the first and the following recording sessions. In the fixed group, MF value was close to 268 Hz. We also observed a decreasing trend in MF in this fixed group but this trend was not significant. No difference was found between the control and the fixed muscle groups (Fig. 5A). Conversely, we observed a significant difference in median frequency in the immobilized peroneus muscles starting and remaining at a higher steady-state level than that of the controls (459.78 Hz versus 304.23 Hz, respectively), all along the experimental
500
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
Fig. 2. Electromyographic activity of soleus and peroneus longus muscles during locomotion. Raw EMG sample (lower trace) and the corresponding rectifiedfiltered EMG sequence (upper trace) of a fixed soleus (panel A) and a fixed peroneus longus muscle (panel B). Activity is aligned on paw contact (vertical arrows).
period (max difference = 212.19 Hz at day 12, min difference = 104.31 Hz at day 60) (Fig. 5B). No significant time course change was observed within each peroneus group throughout the experimental period.
Discussion A single muscle immobilization does not change the proportions of muscle fiber types, and our results are in agreement with those of Armstrong and Phelps (1984) concerning the estimation of each fiber-type population in the two muscles under study. Furthermore, histological analysis did not show either muscle necrosis on the control and immobilized muscles or any conjunctive tissue proliferation. Changes in muscle/body weight ratios and the fiber CSA are consistent with CSA analysis which shows that muscular atrophy is present after 4 weeks of immobilization. This atrophy mainly affects type I fibers and
notably those of the soleus. We also observed a loss of muscle weight in both the soleus and peroneus longus. Our data are in agreement with Petit and Gioux (1993) who observed transient atrophy involving mainly slow motor units, slow-twitch muscles, and subsequently type I fibers, and a return toward control values at 8 weeks with an increase in protein content. In these experiments, we also noticed this phenomenon at 8 weeks. The soleus ratio increased (but not up to control value) and the peroneus ratio indicated a hypertrophy related to the increase in type II fiber CSA. This is in agreement with the results reported by Hortoba´gyi et al. (2000), who noted the many beneficial effects of exercise on immobilization-induced atrophy in humans. In fact, eccentric or mixed exercise training could prevent atrophy or even induce muscular hypertrophy, concomitantly with a strength gain. Similar conclusions were obtained in two other studies performed on the rat. Kannus et al. (1998) noted that the capillary density, the percentage of intramuscular connective tissue and the fiber
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
501
Fig. 3. Burst duration evolution during chronic immobilization. Soleus groups [controls (n = 3), open squares; fixed muscles (n = 7), filled squares] are represented in the upper graph (A) and peroneus longus groups in the lower graph (B) [controls (n = 3), open triangles; fixed muscles (n = 4), filled triangles]. Data are expressed as means T standard deviation (vertical bars).
size were restored almost to control levels in the immobilized soleus by low- or high-intensity treadmill running. Herbert et al. (1988) observed that muscle weight loss could be slowed down in animals which were submitted daily to short periods of repetitive climbing with an overload. In our experiment, the motor task imposed on the animals was quite different from physical training in the above-described data (intensive treadmill running, twice a day, with an uphill inclination of 10- to 30- in the Kannus study; climbing task with 75% body weight load strapped to the tail in Herbert’s study). Our results show that some effects of immobilization are transitory such as the decrease in fiber size. However, a small but regular activation of an immobilized muscle over a rather long period could be enough to maintain its trophicity and seems to be more efficient on type II than on type I muscle fibers whatever the muscle profile. This is in relation with a hypertrophy of the type II fibers which has been observed after isometric exercises (Aagaard et al., 2001). As shown in the example illustrated in Fig. 6, we noticed a similar evolution in fast myosin heavy chain revealed by immunoblotting. The fast MHC expression in the soleus muscle decreased dramatically at 4 weeks and then returned
to control level at 8 weeks. As regards the peroneus longus muscle, the amount of fast MHC did not diminish at 4 weeks; in fact, it increased at 8 weeks. The EMG signal which was recorded during walking in control animals, presented the same pattern of activation as that previously described in the rat (Westerga and Gramsbergen, 1994). The locomotor behavior of both control and immobilized animals remained identical. After muscle fixation, EMG signals (duration and amplitude) recorded at day 6 were very close to those of controls. In spite of iEMG amplitude and spectral content modifications, no difference was observed in iEMG burst duration and shape in fixed animals during the next 7 weeks. The voluntary motor pattern was unchanged during the locomotion task. In fact, some functional adaptations in the lumbar spinal cord circuitry have only been reported in spinal cord transected animals (Hodgson et al., 1994). All other experiments with motor or proprioceptive modifications limited to the hindlimb have highlighted the absence of plasticity of locomotor patterns. After tendinous transfer, Forssberg and Svartengren (1983) observed that the same gastrocnemius muscle motor pattern was retained in cats during locomotion
502
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
Fig. 4. Mean EMG amplitude during chronic immobilization. Amplitude is expressed in mean absolute values T standard deviation. For SD, only one bar is represented on the figures to make them more clearer. (A) Changes in the soleus muscle groups. Controls (n = 3), open squares; fixed muscles (n = 7), filled squares. (B) Changes in the peroneus muscle groups. Controls (n = 3), open triangles; fixed muscles (n = 4), filled triangles. *Statistical difference between immobilized and control group (P < 0.05).
although the muscle had an antagonistic position compared to the original. Other adaptations to hindlimb muscle impairments, such as partial denervations in the cat (Gritsenko et al., 2001), or hindlimb deafferentation (Grillner and Zanger, 1984) did not lead to modifications in spinal circuits. In immobilized muscles under spontaneous conditions of free movement, several authors have noted a drastic decrease in iEMG activity (Fischbach and Robbins, 1969; Fournier et al., 1983; Hnik et al., 1985). However, during a standardized motor task, we observed an increase in iEMG amplitude or MF at week 4, which could be interpreted as the adaptation of the muscular command to the decrease in the motor unit twitch tension and tetanic force (Duchateau and Hainaut, 1990; Petit and Gioux, 1993). Thus, in order to obtain the same muscular force required for standardized treadmill locomotion, several motor mechanisms could come into play depending primarily on the profile of the immobilized muscle. Firstly, we observed that fixed soleus iEMG amplitude increased and reached its maximal value at day 30, when the type I muscle fiber was at its smallest size. It could be suggested that fiber CSA atrophy could increase iEMG amplitude by bringing more fibers and action
potentials in the neighboring of recording electrodes but this is in part compensated by connective tissue proliferation observed at 4 weeks in soleus muscle. Thus, the number of motor units (MU) recruited in the soleus muscle probably increased as is the case with synergist muscle removal (Gardiner et al., 1986). Secondly, motor recruitment in the peroneus was very different: no significant iEMG increase was observed but the MF of the EMG burst PDS was higher from the onset than that of the controls and during the whole immobilization period. Consequently, the tonic and phasic muscle adaptations are different; two different ways and time courses were observed, which could be related to the functional role and the mechanical properties of each muscle. The soleus muscle has a postural function, with little amplitude shortening due to a short and resistant tendon. Its motor units are mainly slow-twitch, which do not need highfrequency recruitment. Thus, its adaptation to a force decrease at 4 weeks could involve more spatial recruitment as suggested by Rothwell (1994). These results are also in agreement with the motor unit firing frequencies recorded by Erim et al. (1996) during voluntary contraction with an increasing force. For the first recruited MU (slow-twitch ones
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
503
Fig. 5. Median frequency evolution during chronic immobilization. Data are expressed as mean T standard deviation. (A) Changes in the soleus muscle groups. Controls (n = 3), open squares; fixed muscles (n = 7), filled squares. (B) Changes in the peroneus muscle groups. Controls (n = 3), open triangles; fixed muscles (n = 4), filled triangles. *Statistical difference between immobilized and control group (P < 0.05).
Fig. 6. Examples of fast myosin heavy chain immunoblot. Data are expressed as the percentage of control densitometry for each muscle. (A) Soleus muscle. (B) Peroneus longus muscle.
according to the size principle), the muscular force increase is obtained by the spatial recruitment of an increasing number of motor units firing at low frequency. The spatial recruitment does not presumably need the very few fast units which are only involved at maximal force because of the great number and the high tetanic forces of slow motor units in soleus. Thus, the increase in force at 4 weeks is probably obtained by increasing the number of slow units recruited at low frequencies which do not lead to an increase in MF. This adaptation, presumably mediated by the signals coming from Golgi tendon organs, is not so necessary at 8 weeks because of the restoration of muscle contractile properties corresponding to the restoration of muscle force. The peroneus longus is composed of only one third of slow units, whose tetanic forces are very small compared both to the fast-twitch units of the same muscle and the slow ones of the soleus (Emonet-De´nand et al., 1988). Thus, the recruitment quickly involves a large number of fast-twitch units allowing fast muscle shortening. By increasing stimulation frequencies in experimental situations during movement, it is more efficient to increase muscular force of fast units than for
504
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505
slow ones (Petit et al., 2003). This increase in the fast motor unit firing rate plays a role in enhancing high frequencies in the EMG power spectrum (Fuglsang-Frederiksen and Rønager, 1988). We observed the same evolution in the EMG power spectrum of a postural muscle whose motor unit contraction time became shorter, after its distal tendon had been transferred to a more compliant one (Giroux-Metges et al., 2003): the FFT of chronic EMG of the soleus muscle exhibited higher frequencies during the same movement, and this was observed during the 8-week EMG recording period. Thus, the motor command adaptation is different in time course (during 8 weeks) and pattern (MF instead of iEMG) between soleus and peroneus longus muscles. The PL muscle could be more sensitive to information coming from muscle spindle receptors regarding the lack of muscle shortening. But this enhancement of contraction force (with a contractile protein content increase at 8 weeks) is not very helpful to facilitate the shortening course of this immobilized muscle and MF remained high during the 8 weeks. The adaptation observed in motor unit frequencies persisted during the whole immobilization period. These nervous regulations could also explain the dependence of atrophy on the length at which the muscle is fixed, with the maximal weight loss at short position (Fournier et al., 1983; Hnik et al., 1985; Simard et al., 1982). Immobilization in a shortened position reduces discharge of spindle receptors or tendon organ, in which sensitivities remain intact during immobilization (Gioux and Petit, 1993; Nordstrom et al., 1995), leading to neural adaptations of either central or spinal muscle commands. In our experiments, proprioceptive signals coming from the immobilized muscle, indicating a lack of shortening or lengthening, lead to an adaptation of the motor unit drive, presumably at the spinal level. This adaptation in the immobilized muscle command, which depends on the muscle profile, could be concomitant with the restoration of muscle fiber CSA at 8 weeks, in spite of very short treadmill locomotion sessions. These results are also different from those obtained with global immobilization, which induces definitive atrophy, concomitantly with the disuse of the non-performant limb. In fact, immobilizing a single muscle does not disturb very much the motor programs involving hindlimbs, because the movements and postures still work, and because all other proprioceptive afferents coming from intact muscles are not impaired. Thus, during posture or locomotion, the immobilized muscle is still globally driven as before, except for some adaptations limited to its motor unit recruitment in order to try to compensate the muscular weakness or the lack of shortening.
Acknowledgments We thank Danielle Gillet for the histochemical analysis, and Jean-Franc¸ois Cle´ment for helping with the signal analysis software.
References Aagaard, P., Andersen, J.L., Dyhre-Poulsen, P., Leffers, A.M., Wagner, A., Magnusson, S.P., Halkjaer-Kristensen, J., Simonsen, E.B., 2001. A mechanism for increased contractile strength training: changes in muscle architecture. J. Physiol. 534.2, 613 – 623. Armstrong, R.B., Phelps, R.O., 1984. Muscle fiber type composition of the rat hindlimb. Am. J. Anat. 171, 259 – 272. Bradford, M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein dye binding. Anal. Biochem. 72, 248 – 254. Brooke, M.H., Kaiser, K.K., 1970. Muscle fiber types: how many and what kind? Arch. Neurol. 23, 369 – 379. Clarke, K.A., Parker, A., 1986. A quantitative study of normal locomotion in the rat. Physiol. Behav. 38, 345 – 351. Duchateau, J., Hainaut, K., 1990. Effects of immobilization on contractile properties, recruitment and firing rates of human motor units. J. Physiol. 422, 55 – 65. Emonet-De´nand, F., Hunt, C.C., Petit, J., Pollin, B., 1988. Proportion of fatigue-resistant motor units in hindlimb muscles of cat and their relation to axonal conduction velocity. J. Physiol. 400, 135 – 158. Erim, Z., De Luca, C.J., Mineo, K., Aoki, T., 1996. Rank-ordered regulation of motor units. Muscle Nerve 19, 5633 – 5673. Fischbach, G., Robbins, N., 1969. Changes in contractile properties of disused soleus muscles. J. Physiol. 201, 305 – 320. Forssberg, H., Svartengren, G., 1983. Hardwired locomotor network in cat revealed by a retained motor pattern to gastrocnemius after muscle transposition. Neurosci. Lett. 41, 283 – 288. Fournier, M., Roy, R.R., Perham, H., Simard, C.P., Edgerton, V.R., 1983. Is limb immobilization a model of muscle disuse? Exp. Neurol. 80, 147 – 156. Fuglsang-Frederiksen, A., Rønager, J., 1988. The motor unit firing rate and the power spectrum of EMG in humans. Electroencephalogr. Clin. Neurophysiol. 70, 68 – 72. Gardiner, P., Michel, R., Browman, C., Noble, E., 1986. Increased EMG of rat plantaris during locomotion following surgical removal of its synergists. Brain Res. 380, 114 – 121. Gioux, M., Petit, J., 1993. Effects of immobilizing the cat peroneus longus muscle on the activity of its own spindles. J. Appl. Physiol. 75, 2629 – 2635. Giroux-Metges, M.A., Pennec, J.P., Petit, J., Goanvec, C., Dorange, G., Gioux, M., 2003. Motor unit properties in the soleus muscle after its distal tendon transfer to the plantaris muscle tendon in the rat. J. Physiol. 553.3, 925 – 933. Greene, E.C., 1955. Anatomy of the rat. Trans. Am. Philos. Soc. vol. XXVII. Hafner Publishing Co., New York. Grillner, S., Zanger, P., 1984. The effect of dorsal root transection on the efferent motor pattern in the cat’s hindlimb locomotion. Acta Physiol. Scand. 120, 393 – 405. Gritsenko, V., Mushahwar, V., Prochazka, A., 2001. Adaptive changes in locomotor control after partial denervation of triceps surae muscles in the cat. J. Physiol. 533.1, 299 – 311. Herbert, M.E., Roy, R.R., Edgerton, V.R., 1988. Influence of one-week hindlimb suspension and intermittent high load exercise on rat muscles. Exp. Neurol. 102, 190 – 198. Hnik, P., Vejsada, R., Goldspink, D.F., Kasicki, S., Krekule, I., 1985. Quantitative evaluation of electromyogram activity in rat extensor and flexor muscles immobilized at different lengths. Exp. Neurol. 88, 515 – 528. Hodgson, J.A., Roy, R.R., Leon, R., Dobkin, B., Edgerton, V.R., 1994. Can the mammalian lumbar spinal cord learn a motor task? Med. Sci. Sports Exerc. 26, 1491 – 1497. Hortoba´gyi, T., Dempsey, L., Fraser, D., Zheng, D., Hamilton, G., Lambert, J., Dohm, L., 2000. Changes in muscle strength, muscle size and myofibrillar gene expression after immobilization and retraining in humans. J. Physiol. 524, 293 – 304. Kannus, P., Jozsa, L., Jarvinen, T.L.N., Kvist, M., Vieno, T., Jarvinen, T.A.H., Natri, A., Jarvinen, M., 1998. Free immobilization and lowto high-intensity exercise in immobilization-induced muscle atrophy. J. Appl. Physiol. 84, 1418 – 1424.
M.-A. Giroux-Metges et al. / Experimental Neurology 194 (2005) 495 – 505 Nordstrom, M.A., Enoka, R.M., Reinking, R.M., Callister, R.C., Stuart, D.G., 1995. Reduced motor unit activation of muscle spindles and tendon organs in the immobilized cat hindlimb. J. Appl. Physiol. 78, 901 – 913. Petit, J., Gioux, M., 1993. Properties of motor units after immobilization of cat peroneus longus muscle. J. Appl. Physiol. 74, 1131 – 1139. Petit, J., Giroux-Metges, M.A., Gioux, M., 2003. Power developed by motor units of the Peroneus tertius muscle of the cat. J. Neurophysiol. 90 (5), 3095 – 3104.
505
Rothwell, J.C., 1994. The motor unit. Control of Human Voluntary Movement. Chapman and Hall, London, pp. 35 – 85. Simard, C.P., Spector, S.A., Edgerton, V.R., 1982. Contractile properties of rat hind limb muscles immobilized at different lengths. Exp. Neurol. 77, 467 – 482. Talmadge, R.J., Roy, R.R., 1993. Electrophoretic separation of rat skeletal muscle myosin heavy-chain isoforms. J. Appl. Physiol. 75, 2337 – 2340. Westerga, J., Gramsbergen, A., 1994. Development of the EMG of the soleus muscle in the rat. Dev. Brain Res. 80, 233 – 243.