INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

DURING A SUSTAINED LOW-FORCE contraction, the development of fatigue is associated with an increase in surface electromyogram (EMG) amplitude that can be similar among individual synergist muscles (11). Such an increase suggests that more motor units are activated and that the firing frequency of motor units increases to compensate for the impairment of contractility of the fatigued muscles (13). However, when muscle contraction with low force production is sustained for a long time, the individual synergistic muscles are not continuously activated but alternate between periods of activity and silence, and these synergists most likely rotate in a complementary pattern to maintain the constant force of knee extensors (19), plantar flexors (18, 21), and elbow flexors (16, 17). This phenomenon has been called alternate muscle activity in synergistic muscles (21). Prior studies in which fine-wire electrodes were used have shown that alternating recruitment among motor units occurs during prolonged contraction of the biceps brachii (7) and the trapezius muscle (22). Enoka and Stuart (6) suggested that alternate muscle activity would be effective for minimizing fatigue. Furthermore, Semmler et al. (16, 17) suggested the possibility that elongated muscle endurance after immobilization is due to the observed alternate muscle activity among synergists. When these findings are considered, it is reasonable to assume that one of the causes of alternate muscle activity is related to the process of muscle fatigue. However, this phenomenon has not been thoroughly quantified, and the physiological mechanisms behind it remain as yet unknown. Therefore, the present study first aimed to quantify its features, i.e., frequency, timing, and combination of alternate muscle activity observed between synergists of the quadriceps muscle during low-level sustained knee extension (experiment 1).

Several previous studies on alternate muscle activity have utilized various experimental conditions with a single force production level, such as knee extension at 5% of maximal voluntary contraction (MVC) (19), plantar flexion at 10 or 50% of MVC (18, 21), and elbow flexion at 15% of MVC (16, 17). It is unclear whether alternate muscle activity depends on the intensity of the muscle contraction, given that most of the studies have examined the responses only at a single intensity. There is only one study that has demonstrated motor unit rotation during sustained elbow flexion at 10% of MVC but not at 40% of MVC (7). In addition, Sirin and Patla (18) examined the synergisms (coactivation or trade-off) of individual plantar flexors during sustained contraction under different conditions, and demonstrated that the trade-off of EMG activities between individual muscles of the plantar flexors were more evident in the knee-extended position than in the knee-flexed position. They postulated that alternate activity is related to the load placed on the muscle and to muscle effectiveness. With these findings taken into account, it is hypothesized that emergence of alternate muscle activity is dependent on the intensity of muscle contraction. To test this hypothesis, we investigated synergistic EMG activities of knee extensor muscles during sustained isometric contractions at different force production levels (experiment 2).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects. Eleven healthy male subjects [age 25.6 ± 2.0 (SD) yr, height 169.7 ± 2.7 cm, and body mass 70.1 ± 10.3 kg] voluntarily participated. They had no history of neurological disorders and had not participated in any programs of regular exercise. All subjects gave their written, informed consent to participate in the study after receiving a detailed explanation of the purposes, potential benefits, and associated risks. This study was approved by the office of the Department of Life Sciences, The University of Tokyo, and its protocol was consistent with their requirements for human experimentation.

General procedure and equipment. Each subject was required to perform a static unilateral knee extension exercise in a seated position with hip and knee joint angles of 100 and 90° flexed, respectively. The subject's upper body was firmly set against the chair, which had an adjustable seat belt to prevent any movement of the hip joint. The isometric force was measured by a strain-gauge force transducer (model 274II, Minebea, Tokyo, Japan), which was coupled with a strain amplifier and attached by a strap to the subject's ankle. The linearity of this measurement system was ensured by the manufacturer from 0 to 200 N. The force was displayed on the storage oscilloscope in front of the subject to provide visual feedback of the produced force and target. Surface EMGs from the muscle belly of the rectus femoris (RF), vastus lateralis (VL), vastus medialis (VM), and biceps femoris long head (BF) were recorded by using bipolar Ag-AgCl electrodes with a diameter of 5 mm and an interelectrode distance of 20 mm. The reference electrode for the EMG was placed on the iliac crest. The electrodes were connected to a preamplifier and differential amplifier with a bandwidth of 5 Hz to 1 kHz (model 1253A, NEC Medical Systems, Tokyo, Japan). For later analysis, all signals were stored on the hard disk of a personal computer by using a 12-bit analog-to-digital converter (Lab-PC+, National Instruments, Austin, TX) with a sampling frequency of 1 kHz.

Experiment 1. For 1 wk before the experiment, all subjects practiced sustaining a knee extension contraction at 2.5% of MVC so that they could produce stable force throughout the task. The MVC for knee extension was determined to be equal to the largest force of three trials that could be sustained for at least 2 s. After a sufficient rest period (~10 min) after completion of the MVC measurement, the subject performed sustained contraction for 60 min at the prescribed force production level corresponding to 2.5% of MVC for 60 min.

Experiment 2. In addition to performing the task at 2.5% of MVC in experiment 1, six of the subjects also performed sustained knee extensions at three more intensities: 5.0, 7.5, and 10.0% of MVC. The experimental setup was identical to that in experiment 1 except in terms of the level of force. In the tasks at 5.0% of MVC, subjects were instructed to maintain the force for 60 min, and all subjects could complete the tasks without volitional exhaustion. With respect to the higher intensity tasks, subjects were instructed to maintain the prescribed force level for 60 min or up to the point of volitional exhaustion, if it happened before 60 min. In the task at 7.5% of MVC, three of the six subjects could not complete the 60 min. The mean duration for this task was 36.5 ± 6.9 min. In the task at 10.0% of MVC, no subject could complete the 60 min, and the mean duration resulted in 16.2 ± 3.1 min. The MVC after the exercises at 2.5, 5.0, 7.5, and 10.0% of MVC fell to 88.5 ± 5.9, 85.2 ± 5.3, 71.9 ± 7.4, and 72.3 ± 6.4% of the initial MVC measurements without fatigue, respectively. No feedback or encouragement by the investigators was provided throughout performance of the tasks to prevent any intentional changes in muscle contraction. Each subject performed only one task per day, and all tasks were performed on separate days with at least 1 wk in between tasks. The order of the tasks was pseudorandomized.

Data analysis. In the sustained knee extension at 2.5% of MVC, alternate EMG activity was observed between RF and both VL and VM (Fig. 1) in all subjects. The EMG of BF, the antagonist muscle for knee extension contraction, was substantially smaller than that of the quadriceps muscle group. We focused on the marked changes in the EMG sequences between the quadriceps muscles to quantify the alternate muscle activity among synergistic muscles. The EMG signals were full-wave rectified and integrated over 15 s to yield an integrated EMG (iEMG) (Fig. 2A, top). A calculation period of 15 s was employed based on the fact that it required at least 10 s for each muscle to develop a significant change during the alternate muscle activity. The torque at the knee joint was calculated from the measured force times the length of the lower leg and then averaged every 15 s. The iEMG and torque during sustained contraction were expressed as a percentage of the corresponding values obtained during the MVC.

View larger version (10K): [in this window] [in a new window]   Fig. 1.   Typical example of torque and integrated electromyogram (iEMG) over 1-s period of rectus femoris (RF), vastus lateralis (VL), vastus medialis (VM), and biceps femoris long head (BF) during sustained knee extension at 2.5% of maximal voluntary contraction (MVC) for 60 min.

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Successive stages of data processing in obtaining the alternate muscle activity among the heads of the quadriceps muscle. A: data of RF muscle are shown as an example. Top and middle: iEMG and smoothed iEMG during sustained contraction at 2.5% of MVC, respectively. Bottom: time differentiation of filtered iEMG sequences (iEMGdiff/dt) calculated from smoothed iEMG sequence. Dotted lines, 3 SDs of iEMGdiff/dt for 3 min after onset of exercise. , Outliers of the iEMG sequence. B: extracted positive (+) and negative () outliers of RF (top), VL (middle), and VM (bottom) muscles. *, Case in which an opposite relation of outliers (negative and positive) was found among the muscles. C: iEMG sequences of RF (top), VL (middle), and VM (bottom) during low-level sustained contraction are indicated for purposes of comparing the objective assessment (B) and visual inspection (C). %max, Percentage of maximum. See text for further explanation.

Tests for possible natural cause of alternating EMG activity. A supplementary test was conducted to examine the possibility of involvement of changes in the hip joint angle or in the direction of the force that could produce drastic changes in EMGs similar to that seen in alternate EMG activity in knee extensor muscles. In addition, it was investigated whether the alternate EMG activity was consistently observed across the entire portion of each muscle. A triaxial force transducer (type 9251A, Kistler, Winterthur, Switzerland) (23) was attached to the ankle, and three force vectors, in the medial-lateral (F_x), upward-downward (F_y), and anterior-posterior (F_z) directions around the ankle, were measured (Fig. 3A). Two identical EMG electrode sets were attached to each muscle; one was at the proximal and the other at the distal portion. The subject performed sustained knee extension at 2.5% of MVC for 60 min, during which time the following instructions were given. First, the subject was asked to change the hip joint angle intentionally between 100 and 80° while maintaining the target force of the knee extension. Thereafter, the subject was asked to perform flexion, adduction-abduction, or external-internal rotation of the hip joint angle concurrently with the sustained knee extension. Additionally, sustained knee extension at 10.0% of MVC was performed under the same setup to compare EMGs from different portions and to monitor for possible changes in force direction.

View larger version (28K): [in this window] [in a new window]   Fig. 3.   Simultaneous recordings of medial-lateral (Fx), upward-downward (Fy), and anterior-posterior (Fz) force of a right leg, hip joint angle, and rectified electromyogram (EMG) obtained from RF, VL, VM, and BF during sustained knee extension at 2.5% of MVC. Two EMG recordings, from the proximal (P) and distal (D) sites of each knee extensor muscle, are shown. A: experimental setup for measurement of Fx, Fy, and Fz of the right leg. B: patterns of forces, angle, and rectified EMGs around the time at which alternate muscle activity emerged. C: forces, angle, and rectified EMGs when hip joint angle was changed intentionally from 100 to 80°. D-F: forces, angle, and rectified EMGs during hip flexion (D), hip adduction-abduction (E), and hip external-internal rotation (F). Horizontal scales are the same among the figures.

Statistical analyses. Values are given as means ± SE. For experiment 1, a two-way ANOVA with repeated measures (3 × 6, combinations of alternating activity × time) with a Tukey's post hoc test was used. For experiment 2, a two-way ANOVA with repeated measures (4 × 6, contraction levels × time) with a Tukey's post hoc test was used. In each statistical analysis, the level of significance was set at P < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Tests for possible natural cause of alternating EMG activity. The force vectors around the ankle and the changes in EMG from the proximal and distal portion of the muscles were examined during sustained knee extension 2.5% of MVC (Fig. 3, B-F) and 10% of MVC (Fig. 4). As demonstrated in Fig. 3B, no apparent changes were seen in the force of any vectors at the time when the alternating EMG activities emerged among the heads of the quadriceps muscle. Furthermore, even when the hip joint angle was changed intentionally between 100 and 80° during sustained knee extension, no marked changes in the EMG activity were observed (Fig. 3C). When the subject intentionally changed the force of the direction of flexion (Fig. 3D), adduction-abduction (Fig. 3E), and external-internal rotation (Fig. 3F) of the hip joint during sustained knee extension, small fluctuations in EMG amplitude could be observed, but these small changes were not systematic and quantitative changes were substantially smaller than those drastic changes seen in the alternate muscle activity (Fig. 3B). Furthermore, the subject could hardly maintain the target level of knee extension force when the direction of force was changed (F_z in Fig. 3, D-F), indicating that changing the direction of force while maintaining the target force in knee extension is less likely under natural conditions.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Typical recordings of force directions, hip joint angle, and EMG activity of knee extensor and flexor muscles during sustained knee extension at 10.0% of MVC.

Experiment 1. The MVC was 193.0 ± 13.2 N · m, and 2.5% of MVC was equivalent to 4.8 ± 0.9 N · m. During the sustained contraction at 2.5% of MVC for 60 min, alternate muscle activity in the quadriceps muscle was found in all 11 subjects. The frequency of the alternate muscle activity between RF and either VL or VM was significantly higher (P < 0.05) than that between VL and VM throughout the sustained contraction (Fig. 5). The frequencies of alternation between RF and either VL or VM increased progressively with time after exercise had commenced. Consequently, the total number of alternations for 60 min between RF and VL, RF and VM, and VL and VM was 6.9 ± 0.5, 7.4 ± 0.8, and 0.7 ± 0.3, respectively.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Mean value of emergent frequency for every 10 min of alternate muscular activity between RF and VL, RF and VM, and VL and VM during sustained contraction at 2.5% of MVC for 60 min.  Significant differences from the RF-VL relation, P < 0.05. * Significant differences from the RF-VM relation, P < 0.05.

Experiment 2. Six of the eleven subjects participated in an additional experiment employing different exercise intensities. The MVC value of these subjects was 185.9 ± 13.1 N · m. Calculated values of 2.5, 5.0, 7.5, and 10.0% of MVC was equivalent to 4.6 ± 0.3, 9.3 ± 0.7, 13.9 ± 1.0, and 18.6 ± 1.3 N · m, respectively. Typical examples of knee extension torque and iEMG for knee extensors (RF, VL, and VM) and flexor (BF), and hip joint angle during sustained contraction at four different intensities are shown in Fig. 6. During contraction at 7.5 and 10.0% of MVC, no marked differences were found in the EMG behaviors among the three heads of the quadriceps muscle. As the task intensity decreased to 2.5 and 5.0% of MVC, the behavior of iEMG for knee extensors, especially for the RF, became dissimilar among individual heads despite constancy of the force. In the task at 2.5% of MVC, the iEMG of knee extensors seemed to increase or decrease more frequently than that in the task at 5.0% of MVC.

View larger version (24K): [in this window] [in a new window]   Fig. 6.   Typical recording of torque, iEMG over 1-s period obtained from RF, VL, VM, BF, and hip joint angle during sustained knee extension at 2.5 (A), 5.0 (B), 7.5 (C), and 10.0% (D) of subject's MVC.

View larger version (14K): [in this window] [in a new window]   Fig. 7.   Values (means ± SE) from 6 subjects of frequency every 10 min of alternate muscle activity between RF and VL (A), RF and VM (B), and VL and VM (C) during sustained knee extension at 2.5, 5.0, 7.5, and 10.0% of MVC.  Significant differences from the sustained contraction at 5.0% of MVC, P < 0.05. * Significant differences from the sustained contraction at 7.5% of MVC, P < 0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Possible natural cause of alternate activity at the whole muscle level. A previous study on the triceps surae muscle investigated whether alternation of EMG activities may have occurred by repositioning the ankle or toe and by small changes in the force direction (21). We examined whether alternate muscle activity of the quadriceps muscle is due to changes in the hip joint angle and/or in the force vectors. Because changes in the hip joint angle could theoretically affect the activity of the RF, this effect was examined first. As a result, the amplitudes of EMG were almost constant despite drastic changes in the hip joint angle between 80 and 100° (Fig. 3C). Furthermore, the amplitudes of EMG of the quadriceps muscle were substantially small during flexion (Fig. 3D), adduction-abduction (Fig. 3E), and external-internal rotation of the hip (Fig. 3F) compared with those seen during alternate muscle activity (Fig. 3B). It is noteworthy that the subject could hardly maintain the knee extension force when these changes in force direction had to be made. From these findings, it appears most likely that the alternate muscle activity was not due to changes in force vector or the adjustment of posture but rather to complex physiological factors within the synergists.

Alternate muscle activity between bi- and monoarticular muscles. In the present study, alternate muscle activity was found between RF and either VL or VM, but not between VL and VM. In a previous study that employed sustained plantar flexion (21), it seemed that the alternate muscle activity mainly emerged between the biarticular gastrocnemius medialis and the monoarticular soleus muscle. These findings suggest that the alternate muscle activity in synergistic muscles occurs between bi- and monoarticular muscles. Biomechanical and electromyographic studies by Buchanan et al. (2, 3) have indicated that, during static contraction of muscles across the elbow joint in various directions, the lack of a correlation between the EMG activities of the biceps brachii as a biarticular muscle and any other monoarticular muscle was found. Gritti and Schieppati (9) demonstrated a decreased amplitude of the soleus H reflex induced by the conditioning stimulus to the gastrocnemius medialis nerve. They found that prolonged vibration of the Achilles tendon abolished the inhibition of the soleus H reflex and that inhibition of the soleus H reflex was decreased by isometric leg flexion (activation of gastrocnemius muscles but not the soleus muscle). On the basis of these results, they suggested the existence of an inhibitory effect of Ia afferents from the gastrocnemius muscle on the soleus motoneuron pool. Schieppati et al. (15) further suggested the possibility that the gastrocnemius and soleus muscles were not necessarily synergistic under all conditions but could be functionally antagonistic. These previous findings, taken together, suggest that it is likely that mono- and biarticular muscles function not only in synergistic but also in antagonistic ways. With these points taken into account, it is likely that the emergence of alternate muscle activity of synergistic muscles is related to the reciprocal complex physiological relations between bi- and monoarticular muscles.

Frequency of alternate muscle activity increases with time. The frequencies of alternate muscle activities in the quadriceps muscle progressively increased with time. Sjøgaard et al. (19, 20) pointed out that fatigue during sustained knee extension at 5% of MVC for 60 min was associated with a decline in the muscle cell excitability, which was induced by a loss of potassium from the cells. This is in line with the present result that the impaired muscle contractility with time could be related to factors inducing alternate muscle activity during low-level sustained contraction. Tamaki et al. (21) suggested that the synaptic inputs to -motoneurons among synergists might be modified by small-diameter afferents belonging to groups III and IV. These afferents are mostly polymodal and are sensitive to metabolites and chemicals associated with fatigue (10). Furthermore, Aymard et al. (1) evoked selective muscle fatigue of the biceps brachii and found a significant decrease in the inhibitory effect from fatigued muscle on the flexor carpi radialis motoneuron pool, which includes the synergistic muscles for elbow flexion. Similarly, Duchateau and Hainaut (5), who examined long-latency reflexes of the human hand muscles, suggested that selective muscle fatigue might enhance the descending supraspinal drive to -motoneurons of nonfatigued neighboring finger muscles. According to these studies that examined the effects of muscle fatigue on the activation of synergistic muscles, development of impaired muscle contractility would induce heterogeneity of neuromuscular activity among synergists to compensate for the declined excitability of -motoneurons of the fatigued muscles.

Alternate muscle activity depends on contraction intensity. The present examination at various contraction levels (experiment 2) showed that alternate muscle activity was observed in contractions with force production levels 5% of MVC by the quadriceps muscle. Previous studies on alternate muscle activity have employed various muscles and conditions with a given level of force production, such as knee extensors at 5% of MVC (19), plantar flexors at 10-50% of MVC (18, 21), elbow flexors at 10-15% of MVC (4, 7, 16, 17), or trapezius muscle at ~4% of MVC (22). Fallentin et al. (7) compared changes in action potentials of single motor units from the biceps brachii muscle between sustained contraction at 10 and 40% of MVC. They found an apparent lack of motor unit rotation when the higher intensity was employed. Sirin and Patla (18) demonstrated that the alternate EMG activities between individual plantar flexor muscles were more marked with the knee-extended position than with the knee-flexed position when the required force was identical between positions. Furthermore, Semmler et al. (17) reported that alternate EMG activities occurred during sustained elbow flexion after only 4 wk of limb immobilization that induced reductions in MVC. According to these previous findings, it is likely that the emergence of alternate muscle activity also depends on the contraction intensity.