| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Exercise and Sport Science and 2Department of Medical Physiology, Panum Institute, University of Copenhagen, Copenhagen, Denmark
Submitted 3 September 2007; accepted in final form 12 October 2007
 |
ABSTRACT |
|---|
|
|
|
INTRODUCTION |
|---|
|
; Nielsen and Kagamihara 1992) or during learning of a new motor task (Llewellyn et al. 1990; Smith 1981). Several studies have demonstrated that the transmission in different spinal pathways is specifically modulated to accommodate the functional demands met by the motor system during the cocontraction task (review by Nielsen 1998). Indeed, presynaptic inhibition of Ia afferents is increased and as a result the SOL H-reflex size is depressed during cocontraction of antagonistic ankle muscles (Nielsen and Kagamihara 1993). In addition, recurrent inhibition from Renshaw cells activated by motor axon collaterals is increased during cocontraction (Nielsen and Pierrot-Deseilligny 1996). This likely occurs to ensure that disynaptic reciprocal inhibition is maintained low and thus facilitate the simultaneous activation of the antagonistic muscles (Nielsen and Kagamihara 1992). These adjustments in segmental reflex pathways are likely caused by changes in the supraspinal control of the involved interneuronal populations. Both monkey and human experiments have suggested that different populations of corticospinal cells are responsible for the descending control during cocontraction compared with isolated flexion and extension movements (Aimonetti et al. 2002; Fetz and Cheney 1987; Humphrey and Reed 1983; Johannsen et al. 2000; Nielsen et al. 1993b).
Cocontraction between antagonistic ankle muscles is extensively practiced by ballet dancers. Indeed, these individuals have smaller H-reflex amplitudes compared with those of other trained individuals (Goode and Van Hoven 1982; Nielsen et al. 1993a). It has been demonstrated previously that different types of motor training induce a task-specific depression of the SOL H-reflex (Gruber et al. 2007; Hess et al. 2003; Perez et al. 2005; Schneider and Capaday 2003; Taube et al. 2007). Therefore we hypothesized that if the H-reflex depression observed in ballet dancers reflects a training-related adaptation in spinal cord reflexes (Nielsen et al. 1993a) it is possible that a training paradigm that involves cocontraction of antagonistic muscles may induce such adaptations.
To address this question, we explored whether repetition of a motor task involving cocontraction of antagonistic ankle dorsi- and plantarflexor muscles would lead to changes in the SOL H-reflexes and SOL and TA MEPs recruitment curves in different sessions. As control tasks, we trained individuals to perform isolated dorsi- or plantarflexion movements. In addition, by using subthreshold TMS, we measured TMS-elicited suppression of the SOL EMG, which likely reflects activation of inhibitory mechanisms within the primary motor cortex (Davey et al. 1994; Di Lazzaro et al. 1998; Petersen et al. 2001) and corticomuscular coherence, which provides information about the synaptic drive to spinal motoneurons during voluntary movement (Farmer et al. 1997; Halliday et al. 1995).
|
|
METHODS |
|---|
|
Ten healthy volunteers (age of 24 ± 8 yr; four female, six male) participated in this experiment. All of them gave their informed consent to the experimental procedure, which was approved by the local ethics committee. The study was performed in accordance with the Declaration of Helsinki. The study consisted of three sessions separated by at least 1 wk. One session involved training of cocontraction between ankle dorsi- and plantarflexor muscles. The other two (control) sessions involved repetition of tonic isolated dorsiflexion and plantarflexor movements. During all sessions subjects were seated in an armchair with the examined leg flexed in the hip (120°), the knee (160°), and the ankle in 110° of plantarflexion. The foot was attached to a footplate, which was connected to a strain gauge force transducer (Fig. 1A). The strain gauge data were captured using the CED 1401 system [Signal and Spike2 software; Cambridge Electronic Devices (CED), Cambridge, UK]. All contractions were performed isometrically and the maximal voluntary contraction (MVC) in the dorsi- and plantarflexor directions was measured at the beginning of each session.
|
All subjects participated first in a single training session involving cocontraction of antagonistic ankle muscles (n = 10). The cocontraction training period lasted for 30 min and the total duration of the session was around 1.5 h. The SOL H-reflex recruitment curves were recorded at the beginning and at the end of training in all subjects. Eight of the 10 subjects returned for an additional cocontraction training session in which the soleus (SOL) and tibialis anterior (TA) motor-evoked potentials (MEPs) and TMS-elicited suppression of the SOL EMG (Davey et al. 1994; Petersen et al. 2001) were tested. During training the SOL electromyogram (EMG) and the torque level were displayed as continuous lines on the oscilloscope for on-line feedback (Fig. 1A) and stored in a computer for later off-line analysis. Each subject was asked to perform a tonic plantarflexion (10% of MVC) and then, while maintaining the same EMG in the plantar flexor muscles, bring the torque back to zero by contracting the dorsiflexor muscles. The subjects had to maintain both a net torque level of 0 Nm and a similar EMG level in the SOL muscle before and after training. Therefore the cocontraction was "balanced" in absolute torque measures; i.e., dorsiflexor and plantarflexor muscles exerted the same level of torque. However, the cocontraction was "unbalanced" in the sense that the dorsiflexors are much weaker than the plantarflexor muscles. A training session consisted of 10 sets x 3 min with 2 min of rest in between the sets (Fig. 1B). Motor performance was quantified by comparing the difference between the maximum and minimum displacements of the torque line in 1 min before and after training by using Spike2 software. This procedure was done off-line by positioning two cursors at the beginning and the end of the middle minute of the performance of the cocontraction task. Then, values for the maximum and minimum torque displacements were extracted for each individual subject and the difference was compared before and after training. In addition, we quantified the SD of the torque in the same period of time before and after training. This measurement provides another estimate of motor performance in this task. Additional analysis was performed to compare motor performance changes within the first and last minutes of practice.
Control sessions
Seven of the 10 subjects who participated in the cocontraction training returned to complete two control sessions. The order of the control sessions was randomized. One session involved repetition of isolated dorsiflexion movements and the other one repetition of isolated plantarflexion movements. In each session the training period lasted for 30 min. The total duration of each session was around 1.5 h. The SOL H-reflex recruitment curves were recorded at the beginning and at the end of each session. During isolated dorsi- and plantarflexion training sessions subjects were asked to perform 10% of MVC in either direction. During each session the rectified and integrated EMGs for the TA and/or SOL were displayed as continuous lines on the oscilloscope for on-line feedback (Fig. 1A) and stored in a computer for later off-line analysis. A training session consisted of 10 sets x 3 min with 2 min of rest in between the sets. Motor performance was quantified by using the same procedure as in the cocontraction training session (see earlier text).
Stimulation and recording
Surface electrodes were used for stimulation and recording EMG activity. EMG activity was recorded from the SOL and TA muscles by bipolar surface electrodes (interelectrode distance, 2 cm). The amplified EMG signals were filtered (band-pass, 25 Hz to 1 kHz), sampled at 2 kHz, and stored on a PC for off-line analysis (CED 1401+ with Signal and Spike2 software; CED). For measurements of corticomuscular coherence the amplified EMG signals were filtered (band-pass, 5 Hz to 1 kHz) and sampled at 2 kHz.
SOL H-reflex
The SOL H-reflex was evoked by stimulating the posterior tibial nerve through a monopolar electrode (1-ms rectangular pulse) in the popliteal fossa using a constant-current stimulator (model DS7A; Digitimer, Hertfordshire, UK). The indifferent electrode was positioned above the patella. The reflex response was measured as peak-to-peak amplitude of the nonrectified reflex response. Three SOL H-reflex recruitment curves were averaged before each session to establish a baseline. Two recruitment curve measurements were taken after the training and averaged. Each recruitment curve was tested with 1–2 min of rest in between. Recruitment curves were assessed by averaging five responses at each stimulus intensity. The interval between each response was 5 s. The stimulus intensity was increased in steps of 0.05 mA, starting below H-reflex threshold and increasing up to supramaximal intensity to measure the maximal motor response (M-max). To ensure that M-wax values were reached, at the end of each recruitment curve the stimulus intensity was increased until a plateau was observed in the M-max and all values were recorded. In different sessions recruitment curves were assessed during cocontraction, isolated dorsi, and plantarflexion movements. Because measurements were obtained on separate days we did not directly compare the shape of curves between training sessions. Our main comparison is the difference between recruitment curves taken before and after each session.
Transcranial magnetic stimulation (TMS)
TMS was applied before and immediately after the cocontraction training session (n = 8). MEPs in the SOL and TA muscles were evoked by TMS of the contralateral motor cortical leg area by using a magnetic stimulator (Magstim 200; Magstim, Dyfed, UK) with the capability to deliver a magnetic field for 100 µs through the figure-of-eight coil (loop diameter, 9 cm; type no. 8809). The coil was positioned over the leg area and secured by straps to the head of the subject to ensure that the same area of the cortex was stimulated during the total time of the training period. Measures of cortical excitability, tested during cocontraction, included active motor threshold (AMT), MEP-max, and recruitment curves. Recruitment curves of increasing intensities in 5% steps were obtained with five trials per step starting below the intensity required to evoke a MEP and randomly increasing up to were the MEP size did not show further increase. MEP areas were measured and averaged off-line. SOL and TA MEPs were expressed as a percentage of the SOL and TA M-max. The common peroneal nerve was stimulated with supramaximal stimuli by bipolar surface electrodes (1-ms rectangular pulse) to obtain M-max in the TA muscle. The M-max in the SOL muscle was obtained during the SOL H-reflex testing (see SOL H-reflex). SOL and TA MEP measurements were obtained before and immediately after the cocontraction training session.
Subthreshold TMS
TMS at an intensity below MEP threshold applied during voluntary contraction has been shown to suppress EMG activity (Davey et al. 1994; Petersen et al. 2001). It is thought that the suppression of EMG activity may reflect activation of inhibitory mechanisms within the motor cortex (Davey et al. 1994; Di Lazzaro et al. 1998; Petersen et al. 2001). We use the term "MEP threshold" to refer to the intensity at which an MEP was observed in three of five consecutive trials. The EMG suppression was measured in the SOL muscle during cocontraction at similar levels of EMG activity before and after the cocontraction training (n = 8). The same stimulus intensity was used before and after training. Three measurements were averaged before the cocontraction training to establish the baseline level of EMG suppression. Each measurement included 75 sweeps with stimulation and 75 without stimulation. The interstimulus interval for TMS pulses was 1 s. Two measurements were taken after the training and averaged.
Corticomuscular coherence
Coherence between electroencephalographic (EEG) and EMG (TA and SOL) activity was assessed before and after the cocontraction training session. Only subjects who showed significant coherence between the leg motor cortex and the TA and/or SOL muscle during performance of a cocontraction were included in this part of the study. At baseline, four subjects showed a significant level of coherence (>95% confidence interval) between the EEG and TA EMG around 15- to 35-Hz frequency during cocontraction. Two of these subjects also showed significant EEG–SOL EMG coherence during cocontraction before training. EEG activity was recorded through a pair of bipolar silver electrodes. One electrode was placed at the vertex (Cz) and the other one 5 cm frontal to Cz. EEG signals were amplified (x50,000), filtered (1–1,000 Hz), and stored for later analysis. Corticomuscular coherence is abolished during movement and is greater during steady periods of contraction (Baker et al. 1997; Kilner et al. 1999). Therefore measurements were obtained while the subjects performed a tonic cocontraction of ankle dorsi- and plantarflexor muscles for about 2 min. The procedures for calculation of coherence between two signals have been described in detail in previous publications (Halliday et al. 1995).
Data analysis
During measurements of SOL H-reflex and TA and SOL MEP recruitment curves, three curves were averaged before each session to establish a baseline. Five peak-to-peak H-reflexes and MEPs area were averaged at each stimulus intensity.
H-REFLEX RECRUITMENT CURVE ANALYSIS. The AMT of the H-reflex and M-wave was calculated by using linear regression analysis. This was determined by the interaction between the x-intercept and the mean baseline using the linear regression formula: y = a + bx. Mean baseline activity for the H and M recruitment curves was calculated and values 1SD above the baseline were included in the regression line. All stimulus intensities in the recruitment curve were expressed to the M-wave AMT and measurements were expressed as a percentage of the SOL M-max. Measurements were binned at each stimulus intensity (range: 0.5–0.599). The same binning procedure was done at each stimulus intensity. To ensure a proper fit of the regression line all curves were visually inspected. In cases were a sudden steep rise was observed in the curve, precaution was taken that the value previous to the sudden rise was the first value included in the regression line.
MEP RECRUITMENT CURVE ANALYSIS. The AMT of the TA and SOL MEPs was calculated by using linear regression analysis with the same procedure described in H-REFLEX RECRUITMENT CURVE ANALYSIS. Mean baseline activity for each TA and SOL recruitment curve was calculated and values 1SD above the baseline were included in the regression line. All measurements were expressed to the TA and SOL AMT intensity and expressed as a percentage of the TA and SOL M-max, respectively. Paired t-test was used to compare the AMT and the slope of the ascending and descending part of the recruitment curves before and after training.
Two-way repeated-measures ANOVA test was used to determine the effect of 30 min of cocontraction on SOL H-reflex and MEP recruitment curves with TIME (pre- and posttraining) and stimulus INTENSITY as factors in all subjects that participated in the study (n = 10). The same statistical test was used to compare the effect of isolated dorsi- (n = 7) and plantarflexion (n = 7) training on H-reflex recruitment curves. An additional two-way repeated-measures ANOVA was performed on the seven subjects who participated in all sessions with TRAINING (cocontraction, dorsiflexion, and plantarflexion) and TIME (pre- and posttraining) as factors. A Bonferroni post hoc test was used for multiple comparisons. Motor performance in the first, second, and third minutes of the cocontraction task was compared by using repeated-measures ANOVA. Pearson correlation analysis was used to test correlations as needed. Paired t-test was used to compare the mean torque level, mean integrated EMG, and median power frequency before and after the cocontraction training. The median power frequency (0–1,000 Hz) was calculated by using a custom Spike2 script.
For quantification of the effect of subthreshold TMS on EMG activity the onset and end of the EMG suppression were estimated by visual inspection of the recordings. The mean level of EMG activity was measured between two cursors placed at the onset and end of the suppression (Fig. 5A). EMG activity in the particular time window was compared with and without stimulation. The EMG suppression was measured only for averages where no MEP was observed. Paired t-test was performed on the group data to determine the statistical significance of the EMG suppression. The mean and SE were calculated for each condition.
|
. This test is applied to the coherency estimates (where the coherence is the magnitude squared of the coherency), which are first transformed using tanh–1 to stabilize the variance. Differences of these transformed coherence estimates were tested for significance using a standardized normal variate, with variance 1/L, at a 5% significance level, where L is the number of segments averaged in the spectral estimates (see Rosenberg et al. 1989 for further details). |
|
RESULTS |
|---|
|
Figure 2 illustrates the effect of cocontraction training on motor performance. Motor performance, defined as the difference between the maximum and minimum torque displacements within 1 min, was significantly improved after 30 min of cocontraction training in all subjects (P 
0.001, n = 10; Fig. 2C). The difference between the maximum and minimum torque displacements was 8.5 Nm before and 3.5 Nm after the cocontraction training session. In the same period of time, the SD of the torque was decreased after training (pre = 0.76 ± 0.26; post = 0.39 ± 0.2, P 0.01), also indicating an improvement in motor performance. The mean torque level was similar before and after the training (pre = 1.4 ± 2.2 Nm; post = 0.9 ± 1.8 Nm, P = 0.2, Fig. 2D), indicating that similar torque was exerted during execution of the cocontraction task before and after training. No differences were observed in the mean integrated EMG in TA (pre = 0.05 ± 0.02 mV; post = 0.06 ± 0.03 mV, P = 0.3) and SOL (pre = 0.037 ± 0.007 mV; post = 0.033 ± 0.006 mV, P = 0.2) before and after the cocontraction training, indicating that similar effort was performed before and after training. Power spectrum analysis revealed no differences in TA median power frequency before and after the cocontraction training (pre = 103 ± 3.3; post = 103.3 ± 4.6, P = 0.9) and SOL (pre = 114 ± 10; post = 110 ± 8.9, P = 0.5). Subjects who participated in a second cocontraction training session after 1 wk showed similar improvements in the difference between the maximum and minimum torque displacements in the first session (50.8% of improvement; P 0.001) and in the second session (43.6% of improvement; P 0.01). No significant difference in the magnitude of the difference between the maximum and minimum torque displacements was found between the first and the second sessions (P = 0.4). The SD of the torque was decreased in the first session (42.3% of improvement; P 0.01) and in the second session (34.1% of improvement; P = 0.02). No significant difference in the changes in the SD of the torque was found between the first and the second sessions (P = 0.2).
|
0.01) and third (P = 0.023) minutes after training. Furthermore, the SD of the torque was decreased in the first (P 0.01) and third (P = 0.03) minutes after training, suggesting an overall training effect. The difference between the maximum and minimum torque displacements was unchanged after 30 min of dorsi- (P = 0.7, n = 7) and plantarflexion training (P = 0.4, n = 7). Furthermore, the SD of the torque was unchanged after 30 min of dorsi- (P = 0.8, n = 7) and plantarflexion training (P = 0.31, n = 7). These results indicate that the subject did not show any additional improvement in motor execution of these control tasks.
SOL H-reflex
Figure 3 illustrates the effect of cocontraction training (Fig. 3A), dorsiflexion (Fig. 3C) and plantarflexion (Fig. 3E) on the SOL H-reflex recruitment curve in all subjects. The SOL H-reflex recruitment curve was significantly depressed after the cocontraction training (ANOVA, df 9, 1, 12, F = 5.61, P = 0.04, n = 10). Post hoc test showed a significant depression at 0.9 (P 0.01) and 1.0 (P 0.01) normalized stimulus intensity. We observed no changes in the SOL H-reflex recruitment curve after the dorsiflexion (ANOVA, F = 0.5, P = 0.4, n = 7) and plantarflexion (ANOVA, F = 0.02, P = 0.8, n = 7) training sessions. Slopes of the ascending (P = 0.03) and descending (P 0.01) parts of the SOL H-reflex recruitment were significantly depressed after the cocontraction training but not after dorsiflexion (ascending, P = 0.2; descending, P = 0.4) and plantarflexion (ascending, P = 0.3; descending, P = 0.8) training sessions. H-max/M-max values were significantly depressed after the cocontraction (Fig. 3B, P 0.01, n = 10), but not dorsiflexion (Fig. 3D, P = 0.6, n = 7) or plantarflexion (Fig. 3F, P = 0.4, n = 7) training. No changes in the H-reflex threshold were observed before and after the cocontraction (P = 0.1), dorsiflexion (P = 0.5), and plantarflexion (P = 0.3) training. The size of the M-max was unchanged before and after the cocontraction (pre = 8.6 ± 2.9; post = 8.3 ± 3.1 mV, P = 0.3), dorsiflexion (pre = 8.2 ± 4.8; post = 8.7 ± 5.1 mV, P = 0.2), and plantarflexion (pre = 10 ± 4.4; post = 9.4 ± 4.1 mV, P = 0.3) training. No significant changes were observed in the M-max values reported at 1.7 AMT and at a higher stimulus intensity before and after the cocontraction (before, P = 0.8; after, P = 0.7), dorsiflexion (before, P = 0.6; after, P = 0.8), and plantarflexion (before, P = 0.7; after, P = 0.4) training.
|
0.01), but not after dorsiflexion (pre = 0.43 ± 0.2; post = 0.39 ± 0.2, P = 0.6) and plantarflexion (pre = 0.64 ± 0.2; post = 0.69 ± 0.1, P = 0.4) training. Two-factor repeated-measures ANOVA showed no effect of TRAINING (ANOVA, F = 0.3, P = 0.7), TIME (ANOVA, F = 0.07, P = 0.8), or interaction (ANOVA, F = 0.5, P = 0.6) on the M-max. Pearson correlation analysis showed a significant correlation between changes in H-reflex size at 0.9 normalized stimulus intensity and improvements in the difference between maximum and minimum torque displacements after the cocontraction training (r = 0.65, P = 0.03). A moderate but not significant correlation was found between changes in H-max/M-max ratio and improvements in the difference between maximum and minimum torque displacements after the cocontraction training (r = 0.55, P = 0.09). Motor-evoked potentials (MEPs)
Figure 4 illustrates the effect of cocontraction training on the TA and SOL MEP recruitment curves. Repeated-measures ANOVA showed a significant decrease in the size of TA (ANOVA, F = 19.8, P 0.01; n = 8; Fig. 4, A and C) MEP recruitment curve after the cocontraction training. Post hoc test showed a significant depression at 1.2 (P 0.01), 1.5 (P 0.01), and 1.6 (P 0.01) normalized stimulus intensity (Fig. 4A). Repeated-measures ANOVA also showed a significant decrease in the size of SOL (ANOVA, F = 19, P 0.01; n = 8; Fig. 4, B and D) MEP recruitment curve after the cocontraction training. Post hoc test showed a significant depression at 1.4 (P 0.01) normalized stimulus intensity (Fig. 4B). The MEP AMT was unchanged in the TA (pre = 58.6%; post = 57% of stimulator output, P = 0.4) and SOL (pre = 62.3%; post = 60% of stimulator output, P = 0.5) before and after training. Areas of the TA MEP-max ("TA MEP-max" pre = 4.4% of M-max, post = 2.9% of M-max, P 0.01) and SOL ("SOL MEP-max" pre = 0.4% of M-max, post = 0.27% of M-max, P = 0.01) MEP-max were significantly depressed after training. No correlation was found between changes in TA (r = 0.4, P = 0.3) and SOL (r = 0.3, P = 0.5) MEP-max and improvements in torque displacement after the cocontraction training. Also no correlation was found between changes in H-reflex size (at 0.9 normalized stimulus intensity) and changes in TA (r = 0.2, P = 0.6) and SOL (r = 0.4, P = 0.4) MEP-max after the cocontraction training.
|
Figure 5 shows the effect of cocontraction training on the TMS-elicited suppression of the SOL EMG (n = 8). After training, a small but significant increase in the amount of the SOL EMG suppression (from 88 to 79.9%; P = 0.01; Fig. 5B) was observed. No significant changes were observed in the SOL EMG suppression latency after training (pre = 41 ms; post = 39.8 ms). The background SOL EMG was continuously monitored to ensure that the contraction level was comparable before and after the cocontraction training session. Indeed, the SOL mean EMG was not different before and after the cocontraction training (P = 0.7).
Corticomuscular coherence
The difference-of-coherence test (Rosenberg et al. 1989) showed that after the cocontraction training session, three of the four subjects showed a significant increase in the magnitude of EEG–TA EMG coherence in the 15- to 35-Hz frequency range (before: 0.16 ± 0.06, range 0.05–0.25; after: 0.25 ± 0.07, range 0.09–0.38). The peak of coherence before the training session was observed between 20 and 28 Hz and shifted to lower frequencies around 19 Hz after the training session. The two subjects who had EEG–SOL EMG coherence before training showed a significant increment in the magnitude of coherence in the 15- to 35-Hz frequency range (before: 0.15 ± 0.01, range: 0.13–0.18; after: 0.23 ± 0.03, range: 0.18–0.28) after the cocontraction training.
Figure 6 illustrates the changes in EEG–TA EMG coherence in a single subject before (Fig. 6A) and after (Fig. 6B) the cocontraction training. We also illustrate the cumulant density function from the same subject calculated for the EEG and EMG before (Fig. 6C) and after (Fig. 6D) the cocontraction training. It is seen that after the cocontraction training session the largest negative peak, observed at a lag around 28 ms, increased. This lag is consistent with transmission in corticospinal pathways (Farmer et al. 2004). These changes in coherence and the cumulant density function were not accompanied by any changes in the EEG (Fig. 6, E and F) and EMG (Fig. 6, G and H) power. In both EEG and EMG power spectra characteristic peaks around 10 and 20 Hz were observed.
|
|
|
DISCUSSION |
|---|
|
Relation to improved performance?
Previous studies have reported that motor skill training is accompanied by changes in motor cortical representation (Pascual-Leone et al. 1994 1995), corticospinal excitability (Jensen et al. 2005; Perez et al. 2004), and H-reflex size (Perez et al. 2005) when measurements are obtained at rest before and after the training session. Although these changes in all likelihood reflect plastic changes in neuronal excitability and connectivity in relation to the acquisition of new motor tasks their significance for the actual performance of the motor task is less clear. In the present study we report changes in the H-reflex, MEP, and TMS-induced EMG suppression measured during the cocontraction task, which was also the motor task for which subjects trained. It seems likely that the changes we observed in these parameters are related directly to the improved performance of the cocontraction task. This is supported by our finding of a significant correlation between the improved performance in the cocontraction task and the reduced size of the H-reflex across all subjects. Nielsen and Kagamihara (1993) suggested that the reduction of the H-reflex size during cocontraction might be necessary to prevent the agonist–antagonist stretch reflex system from breaking into oscillations, thus making the level of cocontraction and the stability of the joint position difficult to control. The progressive reduction of the H-reflex size with increased performance is well in line with this.
In some previous studies depression of the H-reflex has been observed as a nonspecific change after repetitive exercise, possibly related to fatigue (Bulbulian and Darabos 1986; deVries et al. 1981; Motl and Dishman 2003). It is unlikely that the H-reflex depression observed in the present study should reflect a similar nonspecific change because it was ensured that the subjects were given ample rest periods in between the training sets and because no changes in the size of the SOL H-reflex were observed after ankle dorsi- and plantarflexion movements with a similar intensity and duration as the cocontraction training.
Mechanisms contributing to the H-reflex and MEP depression
There are several possible mechanisms that may have contributed to the reduction in the H-reflex and MEPs observed in our study after the cocontraction training (for review see Nielsen 1998). The most straightforward possibility would be a reduction in the excitability of spinal motoneurons shared by the two responses—if not for the fact that there were no changes in the level or frequency content of the background EMG activity before and after the training session. This makes it unlikely that any major changes in the excitability level of the pool of motoneurons or the discharge of the active motor units took place after the training. A change in the recruitment gain of the motoneuronal pool could theoretically also explain the depression of the responses without any apparent change in the overall activity and excitability level of the pool (Kernell and Hultborn 1990; Nielsen and Kagamihara 1993b). However, because such changes do not occur during cocontraction compared with other tasks (Nielsen and Kagamihara 1993a), it is unlikely that they should occur after repeated performance of cocontraction.
In our study the main effects on H-reflex size were observed in the later portion of the ascending part of the H-reflex recruitment curve. This is in agreement with the study by Crone et al. (1990) showing that the sensitivity of the SOL H-reflex to inhibition and facilitation is maximal for reflex sizes between 20 and 50% of the M-max. The changes observed in the descending part of the recruitment curve are more difficult to interpret because in this part of the recruitment curve the reflex response is little sensitive to excitation or inhibition (Pierrot-Deseilligny and Burke 2005). The H-reflex is mainly mediated by the monosynaptic Ia afferent pathway to the spinal motoneurons, although other less direct pathways such as Ib inhibition or recurrent inhibition may also contribute (Burke et al. 1983, 1984; Marchand-Pauvert et al. 2002; Nielsen and Pierrot-Deseilligny 1996). Although there is thus a possibility that the depression of the H-reflex may be caused by changes in the excitability of interneurons interposed in these indirect pathways, it seems more likely that increased presynaptic inhibition of the Ia afferents is involved. Increased presynaptic inhibition, measured by the long-latency depression (elicited by common peroneal nerve stimulation) and the femoral nerve facilitation of the SOL H-reflex, was observed recently to be involved in the depression of the SOL H-reflex after a visuomotor training task (Perez et al. 2005). The long-latency depression of the H-reflex reflects the level of presynaptic inhibition of SOL Ia afferents evoked by peripheral nerve stimulation (Mizuno et al. 1971), whereas the size of the femoral nerve facilitation reflects the size of the monosynaptic excitatory postsynaptic potential in the SOL motoneurons evoked by the activation of Ia afferents from the quadriceps muscle, and changes in its size are considered to indicate changes in the ongoing level of presynaptic inhibition of the Ia afferents (Hultborn et al. 1987a,b). Furthermore, Nielsen and Kagamihara (1993) demonstrated that the depression of the H-reflex during cocontraction, compared with isolated plantarflexion, was also caused by increased presynaptic inhibition of the Ia afferents. The magnitude of presynaptic inhibition acting on Ia afferent terminals, measured by the femoral nerve facilitation of the SOL H-reflex, is increased in the preparatory phase and during the cocontraction movement (Nielsen and Kagamihara 1993). Further studies are needed to address whether training effects might be present in the preparatory phase of the cocontraction movement and their contribution to H-reflex size on the cocontraction task.
The interneurons that mediate presynaptic inhibition of Ia afferents do not affect the terminals of corticospinal tract fibers (Jackson et al. 2006; Nielsen and Petersen 1994; Rudomin et al. 1975, 1981), and the depression of the MEPs therefore cannot be explained by this mechanism. The efficiency of the synapse between the corticospinal tract fibers and the spinal motoneurons is most certainly modulated by other mechanisms, but little is known of this and it is therefore also unclear to what extent such mechanisms might contribute to the observed depression of the MEPs. The MEPs are also influenced by the excitability of spinal interneurons contacted by the corticospinal tract cells (Iles and Pisini 1992; Nielsen et al. 1993b) and it cannot be excluded that changes in the excitability of these neurons contribute to the depression of the MEPs. However, at least in the case of the interneurons that mediate disynaptic reciprocal inhibition, this seems unlikely because reciprocal inhibition has been shown to be reduced during cocontraction (Nielsen and Kagamihara 1992) to allow simultaneous activation of the two antagonistic motoneuronal pools. It is then more likely that the depression of the MEPs is caused by changes in the excitability of the corticospinal tract cells. Nielsen et al. (1993b) and Aimonetti et al. (2002) provided evidence that the cortical excitability is reduced during cocontraction tasks and it would not be unlikely that improved performance of the task would be associated with still larger depression of the excitability. The increase in the TMS-induced suppression of the EMG activity may support this. Previous studies have demonstrated that this suppression reflects activation of cortical interneurons, which inhibit the activity of the corticospinal tract cells (Davey et al. 1994; Petersen et al. 2001) and several previous studies have demonstrated motor-learning–related changes in intracortical inhibitory circuits within the primary motor cortex (Classen et al. 1999; Liepert et al. 1998; Nordstrom and Butler 2002; Perez et al. 2004). However, the suppression in EMG activity is also influenced by changes in the corticospinal drive to the motoneurons and any such change might explain the observed increase of the suppression independent of any change in the inhibitory interneurons. Indeed, previous studies have suggested that different populations of cortical cells are involved in cocontraction compared with flexion–extension movements (Cheney and Fetz 1985; Humphrey et al. 1983; Nielsen et al. 1993b) and it would not be unlikely that improved performance of cocontraction would result in further alterations in the corticospinal drive. Our result of an increase in the magnitude of corticomuscular coherence around 15–35 Hz after the cocontraction training supports the view of changes in corticospinal drive to spinal motoneurons after the cocontraction training. These changes possibly reflect adaptations in cortex and muscle integration processes related to the acquisition of the cocontraction task.
Functional considerations
As mentioned earlier, the observed changes in the H-reflex may be related to the necessity of suppressing stretch reflex activity during cocontraction to prevent the stretch reflexes in the antagonist muscles from breaking into oscillations. It has indeed been demonstrated that the stiffness around a joint is mainly regulated by increasing the descending drive to the antagonistic motoneuron pools acting on the joint, whereas the reflex-mediated stiffness plays only a very minor role and is actively suppressed during tasks involving cocontraction of antagonistic muscles (see review by Nielsen 1998). The observation of a still further depression of the reflex activity with improved performance of cocontraction suggests that this depression is trainable and that repeated performance of tasks involving cocontraction may lead to a sustained suppression of the reflex activity.
The reduction in the MEP (and the changes in corticomuscular coherence and EMG suppression) likely reflect alterations in the corticospinal control of the spinal interneurons and motoneurons, but the exact nature of this alteration and thus its functional significance is unclear. It is undoubtedly an oversimplification to interpret these findings as reflecting simply "more or less" corticospinal control, but it seems reasonable to assume that the changes are related to the ease with which the subjects are able to perform the cocontraction task after the training. It also appears likely that the changes in reflex activity are secondary to the changes in the corticospinal control, although this is difficult to prove. The observed alterations at spinal and cortical levels may thus be seen together as the neuronal adaptations, which are necessary to accomplish a steady cocontraction around a joint and thus ensure joint stability. To what an extent these observations in relation to a deliberate voluntary cocontraction, which is heavily influenced by visual feedback control, may be extrapolated to the cocontraction that takes place in relation to normal motor behavior, such as balancing on a beam, is unclear and would require further study.
In conclusion, this study provides evidence for a task-specific depression of the SOL H-reflex after cocontraction training involving antagonistic ankle muscles in healthy humans. An increased presynaptic inhibition of Ia afferents is a likely mechanism for the SOL H-reflex depression, whereas an increased intracortical inhibition may contribute to the decreased excitability of corticospinal tract cells observed after the cocontraction training. These results indicate that the depression in H-reflex size observed during a cocontraction task can be trained and that repetition of motor tasks involving cocontraction of antagonistic muscles may lead to prolonged changes in reflex and corticospinal excitability.
|
|
GRANTS |
|---|
|
|
|
ACKNOWLEDGMENTS |
|---|
|
|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: J. B. Nielsen, Department of Exercise and Sport Science and Department of Medical Physiology, Panum Institute, University of Copenhagen, Blegdamsvej 3, 2200 Copenhagen N, Denmark (E-mail: j.b.nielsen{at}mfi.ku.dk)
|
|
REFERENCES |
|---|
|
Baker SN, Olivier E, Lemon RN. Coherent oscillations in monkey motor cortex and hand muscle EMG show task-dependent modulation. J Physiol 501: 225–241, 1997.
Bulbulian R, Darabos BL. Motor neuron excitability: the Hoffmann reflex following exercise of low and high intensity. Med Sci Sports Exerc 18: 697–702, 1986.
Burke D, Gandevia SC, McKeon B. The afferent volleys responsible for spinal proprioceptive reflexes in man. J Physiol 339: 535–552, 1983.
Burke D, Gandevia SC, McKeon B. Monosynaptic and oligosynaptic contributions to human ankle jerk and H-reflex. J Neurophysiol 52: 435–448, 1984.
Classen J, Liepert J, Hallett M, Cohen LG. Plasticity of movement representation in the human motor cortex. Electroencephalogr Clin Neurophysiol Suppl 51: 162–173, 1999.[Medline]
Crone C, Hultborn H, Mazieres L, Morin C, Nielsen J, Pierrot-Deseilligny E. Sensitivity of monosynaptic test reflexes to facilitation and inhibition as a function of the test reflex size: a study in man and the cat. Exp Brain Res 81: 35–45, 1990.[Web of Science][Medline]
Davey NJ, Romaiguere P, Maskill DW, Ellaway PH. Suppression of voluntary motor activity revealed using transcranial magnetic stimulation of the motor cortex in man. J Physiol 477: 223–235, 1994.
deVries HA, Wiswell RA, Bulbulian R, Moritani T. Tranquilizer effect of exercise. Acute effects of moderate aerobic exercise on spinal reflex activation level. Am J Phys Med 60: 57–66, 1981.[Web of Science][Medline]
Di Lazzaro V, Restuccia D, Oliviero A, Profice P, Ferrara L, Insola A, Mazzone P, Tonali P, Rothwell JC. Magnetic transcranial stimulation at intensities below active motor threshold activates intracortical inhibitory circuits. Exp Brain Res 119: 265–268, 1998.[CrossRef][Web of Science][Medline]
Farmer SF, Harrison LM, Mayston MJ, Parekh A, James LM, Stephens JA. Abnormal cortex–muscle interactions in subjects with X-linked Kallmann's syndrome and mirror movements. Brain 127: 385–397, 2004.
Fetz EE, Cheney PD. Functional relations between primate motor cortex cells and muscles: fixed and flexible. Ciba Found Symp 132: 98–117, 1987.[Medline]
Goode DJ, Van Hoven J. Loss of patellar and Achilles tendon reflexes in classical ballet dancers (Letter). Arch Neurol 39: 323, 1982.
Gruber M, Taube W, Gollhofer A, Beck S, Amtage F, Schubert M. Training-specific adaptations of h- and stretch reflexes in human soleus muscle. J Mot Behav 39: 68–78, 2007.[CrossRef][Web of Science][Medline]
Halliday DM, Rosenberg JR, Amjad AM, Breeze P, Conway BA, Farmer SF. A framework for the analysis of mixed time series/point process data—theory and application to the study of physiological tremor, single motor unit discharges and electromyograms. Prog Biophys Mol Biol 64: 237–278, 1995.[CrossRef][Web of Science][Medline]
Hess F, Van Hedel HJ, Dietz V. Obstacle avoidance during human walking: H-reflex modulation during motor learning. Exp Brain Res 151: 82–89, 2003.[CrossRef][Web of Science][Medline]
Hultborn H, Meunier S, Morin C, Pierrot-Deseilligny E. Assessing changes in presynaptic inhibition of Ia fibres: a study in man and the cat. J Physiol 389: 729–756, 1987a.
Hultborn H, Meunier S, Pierrot-Deseilligny E, Shindo M. Changes in presynaptic inhibition of Ia fibres at the onset of voluntary contraction in man. J Physiol 389: 757–772, 1987b.
Humphrey DR, Reed DJ. Separate cortical systems for control of joint movement and joint stiffness: reciprocal activation and coactivation of antagonist muscles. Adv Neurol 39: 347–372, 1983.[Medline]
Iles JF, Pisini JV. Cortical modulation of transmission in spinal reflex pathways of man. J Physiol 455: 425–446, 1992.
Jackson A, Baker SN, Fetz EE. Tests for presynaptic modulation of corticospinal terminals from peripheral afferents and pyramidal tract in the macaque. J Physiol 573: 107–120, 2006.
Jensen JL, Marstrand PC, Nielsen JB. Motor skill training and strength training are associated with different plastic changes in the central nervous system. J Appl Physiol 99: 1558–1568, 2005.
Johannsen P, Christensen LO, Sinkjaer T, Nielsen JB. Cerebral functional anatomy of voluntary contractions of ankle muscles in man. J Physiol 535: 397–406, 2001.
Kernell D, Hultborn H. Synaptic effects on recruitment gain: a mechanism of importance for the input-output relations of motoneurone pools? Brain Res 507: 176–179, 1990.[CrossRef][Web of Science][Medline]
Kilner JM, Baker SN, Salenius S, Jousmaki V, Hari R, Lemon RN. Task-dependent modulation of 15–30 Hz coherence between rectified EMGs from human hand and forearm muscles. J Physiol 516: 559–570, 1999.
Liepert J, Classen J, Cohen LG, Hallett M. Task-dependent changes of intracortical inhibition. Exp Brain Res 118: 421–426, 1998.[CrossRef][Web of Science][Medline]
Llewellyn M, Yang JF, Prochazka A. Human H-reflexes are smaller in difficult beam walking than in normal treadmill walking. Exp Brain Res 83: 22–28, 1990.[Web of Science][Medline]
Marchand-Pauvert V, Nicolas G, Burke D, Pierrot-Deseilligny E. Suppression of the H reflex in humans by disynaptic autogenetic inhibitory pathways activated by the test volley. J Physiol 542: 963–976, 2002.
Mizuno Y, Tanaka R, Yanagisawa N. Reciprocal group I inhibition on triceps surae motoneurons in man. J Neurophysiol 34: 1010–1017, 1971.
Motl RW, Dishman RK. Acute leg-cycling exercise attenuates the H-reflex recorded in soleus but not flexor carpi radialis. Muscle Nerve 28: 609–614, 2003.[CrossRef][Web of Science][Medline]
Nielsen J, Crone C, Hultborn H. H-reflexes are smaller in dancers from The Royal Danish Ballet than in well-trained athletes. Eur J Appl Physiol Occup Physiol 66: 116–121, 1993a.[CrossRef][Web of Science][Medline]
Nielsen J, Kagamihara Y. The regulation of disynaptic reciprocal Ia inhibition during co-contraction of antagonistic muscles in man. J Physiol 456: 373–391, 1992.
Nielsen J, Kagamihara Y. The regulation of presynaptic inhibition during co-contraction of antagonistic muscles in man. J Physiol 464: 575–593, 1993.
Nielsen J, Petersen N. Is presynaptic inhibition distributed to corticospinal fibres in man? J Physiol 477: 47–58, 1994.
Nielsen J, Petersen N, Deuschl G, Ballegaard M. Task-related changes in the effect of magnetic brain stimulation on spinal neurones in man. J Physiol 471: 223–243, 1993b.
Nielsen J, Pierrot-Deseilligny E. Evidence of facilitation of soleus-coupled Renshaw cells during voluntary co-contraction of antagonistic ankle muscles in man. J Physiol 493: 603–611, 1996.
Nielsen JB. Co-contraction of antagonistic muscles in man. Dan Med Bull 45: 423–435, 1998.[Web of Science][Medline]
Nordstrom MA, Butler SL. Reduced intracortical inhibition and facilitation of corticospinal neurons in musicians. Exp Brain Res 144: 336–342, 2002.[CrossRef][Web of Science][Medline]
Pascual-Leone A, Grafman J, Hallett M. Modulation of cortical motor output maps during development of implicit and explicit knowledge. Science 263: 1287–1289, 1994.
Pascual-Leone A, Nguyet D, Cohen LG, Brasil-Neto JP, Cammarota A, Hallett M. Modulation of muscle responses evoked by transcranial magnetic stimulation during the acquisition of new fine motor skills. J Neurophysiol 74: 1037–1045, 1995.
Perez MA, Lungholt BK, Nielsen JB. Presynaptic control of group Ia afferents in relation to acquisition of a visuo-motor skill in healthy humans. J Physiol 568: 343–354, 2005.
Perez MA, Lungholt BK, Nyborg K, Nielsen JB. Motor skill training induces changes in the excitability of the leg cortical area in healthy humans. Exp Brain Res 159: 197–205, 2004.[CrossRef][Web of Science][Medline]
Petersen NT, Butler JE, Marchand-Pauvert V, Fisher R, Ledebt A, Pyndt HS, Hansen NL, Nielsen JB. Suppression of EMG activity by transcranial magnetic stimulation in human subjects during walking. J Physiol 537: 651–656, 2001.
Pierrot-Deseilligny E, Burke D. The Circuitry of the Human Spinal Cord: Its Role in Motor Control and Movement Disorders. Cambridge, UK: Cambridge Univ. Press, 2005, p. 1–62.
Rosenberg JR, Amjad AM, Breeze P, Brillinger DR, Halliday DM. The Fourier approach to the identification of functional coupling between neuronal spike trains. Prog Biophysics Mol Biol 53: 1–31, 1989.[CrossRef]
Rudomin P, Engberg I, Jimenez I. Mechanisms involved in presynaptic depolarization of group I and rubrospinal fibers in cat spinal cord. J Neurophysiol 46: 532–548, 1981.
Rudomin P, Nunez R, Madrid J. Modulation of synaptic effectiveness of Ia and descending fibers in the spinal cord. J Neurophysiol 37: 413–429, 1975.[Web of Science]
Schneider C, Capaday C. Progressive adaptation of the soleus H-reflex with daily training at walking backward. J Neurophysiol 89: 648–656, 2003.
Smith AM. The coactivation of antagonist muscles. Can J Physiol Pharmacol 59: 733–747, 1981.[Web of Science][Medline]
Taube W, Gruber M, Beck S, Faist M, Gollhofer A, Schubert M. Cortical and spinal adaptations induced by balance training: correlation between stance stability and corticospinal activation (Abstract). Acta Physiol 189: 303, 2007.
Voigt M, Chelli F, Frigo C. Changes in the excitability of soleus muscle short latency stretch reflexes during human hopping after 4 weeks of hopping training. Eur J Appl Physiol Occup Physiol 78: 522–532, 1998.[CrossRef][Web of Science][Medline]
This article has been cited by other articles:
|
|
 |
|
  M. Hodapp, J. Vry, V. Mall, and M. Faist Changes in soleus H-reflex modulation after treadmill training in children with cerebral palsy Brain, January 1, 2009; 132(1): 37 - 44. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. A. Riley, S. Baudry, and R. M. Enoka Reflex Inhibition in Human Biceps Brachii Decreases With Practice of a Fatiguing Contraction J Neurophysiol, November 1, 2008; 100(5): 2843 - 2851. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Lundbye-Jensen and J. B. Nielsen Immobilization induces changes in presynaptic control of group Ia afferents in healthy humans J. Physiol., September 1, 2008; 586(17): 4121 - 4135. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
TOP
ABSTRACT
INTRODUCTION
