| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
The Journal of Neurophysiology Vol. 88 No. 1 July 2002, pp. 124-132
Copyright ©2002 by the American Physiological Society
1School of Kinesiology, Simon Fraser University, Burnaby, British Columbia V5A 1S6, Canada; 2Department of Physical Education, Health and Recreation, Western Washington University, Bellingham, Washington 98225; and 3Department of Neurology, University of British Columbia, Vancouver, British Columbia, Canada
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
METHODS
RESULTS
DISCUSSION
CONCLUSIONS
REFERENCES
|
|---|
Bawa, P.,
G. R. Chalmers,
H. Stewart, and
A. A. Eisen.
Responses of Ankle Extensor and Flexor Motoneurons to
Transcranial Magnetic Stimulation.
J. Neurophysiol. 88: 124-132, 2002.
Transcranial magnetic
stimulation (TMS) of the motor cortex excites limb muscles of the
contralateral side of the body. Reports of poorly defined, or a
complete lack of systematic excitatory responses of soleus motoneurons
compared with those of tibialis anterior (TA) motoneurons has led to
the proposal that while all ankle flexor motoneurons receive strong
corticomotoneuronal connections, very few soleus motoneurons do. In
addition, the connections to these few motoneurons are weak. The nature
of corticomotoneuronal connections onto these two motoneuron pools was
re-evaluated in the following experiments. The leg area of the left
motor cortex was stimulated with a large double-cone coil using Magstim
200, while surface electromyographic (EMG) and single motor unit (SMU) responses were recorded from soleus and TA muscles of healthy adult
subjects. Under resting conditions, the onset (25-30 ms) and duration
of concomitantly recorded short latency motor evoked potentials (MEPs)
in surface EMG from both muscles were similar. The input-output
relationships of the simultaneously recorded soleus and TA EMG
responses showed much greater increases in TA MEPs compared with soleus
MEPs with identical increases in stimulus intensity. Under resting and
nonisometric conditions, a later peak with onset latency of
approximately 100 ms was observed in soleus. During isometric
conditions or with vibration of the TA tendon, the second soleus peak
was abolished indicating reflex origin of this peak. Recordings from 42 soleus and 39 TA motor units showed clear response peaks in the
peristimulus time histograms (PSTHs) of every unit. Two statistical
tests were done to determine the onset and duration of response peaks
in the PSTHs. With 
2 test, the duration was
6.9 ± 4.2 ms (mean ± SD) for soleus and 5.1 ± 2.1 ms
for TA. Using the criterion of discerning a peak by bin counts being
three SDs above background, the duration was 10.0 ± 4.4 ms for
soleus and 7.8 ± 2.6 ms for TA. Results of these experiments do
not suggest a lack of systematic corticomotoneuronal connections on
soleus motoneurons when compared with those on TA, though some
differences in the strengths of corticomotoneuronal connections onto
the two pools do exist.
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
CONCLUSIONS
REFERENCES
|
|---|
Stimulation of the
precentral motor cortex with transcranial magnetic or electrical
stimulators produces clear short latency responses in the target
muscles of contralateral limbs. The responses, or motor evoked
potentials (MEPs), may be observed with surface electromygraphic (EMG)
recordings or with peristimulus time histograms (PSTHs) constructed
from responses of single motor units of the muscle (Day et al.
1987
). These responses are suggested to result from activation
of fast corticospinal neurons that make monosynaptic connections with
spinal motoneurons (Palmer and Ashby 1992). The corticomotoneuronal (CM) connections activated by transcranial magnetic
stimulation (TMS) have a systematic synaptic pattern such that TMS
recruits motoneurons of a pool according to the size principle
(Bawa and Lemon 1993). Motoneurons that do not receive
meaningful monosynaptic connections from the cortex fail to exhibit
clear and sharp response peaks. Based on this assumption, the relative
strengths of CM connections have been mapped in the upper limb
motoneurons by Palmer and Ashby (1992). Based on the same premise, it has been argued that soleus motoneurons do not show
clear responses because they receive weak or no CM connections (Brouwer and Qiao 1996; Brouwer et al.
1992). The presence of only diffused, poorly defined peaks have
been reported in PSTHs of soleus motor units by Nielsen and Petersen
(1995), even though the same groups have reported clear
response peaks for motor units of the neighboring ankle flexor,
tibialis anterior (Brouwer and Qiao 1995; Morita
et al. 2000). Besides weak excitatory connections to soleus,
these authors report that only a very small percentage of soleus
motoneurons receive excitatory connections (18%, Brouwer and Qiao
1995) while the majority of TA motoneurons are excited monosynaptically. Such random monosynaptic connections to a few soleus
motoneurons would lead to poor control of this large postural muscle.
The importance of these monosynaptic connections cannot be undermined
even if the reflex and polysynaptic descending connections are strong.
In the upper limb, the density of CM connections decline from distal to
proximal muscles (Palmer and Ashby 1992). Soleus and tibialis anterior do not have such a relationship as both act on the
ankle joint. However, soleus is a slow antigravity muscle while TA is
not. Would that explain the differences in responses observed to date?
The purpose of the following study was to examine the nature of both
population and single motor unit responses of soleus and TA motoneurons
in normal adult subjects. Portions of this work have been presented as
a poster (Bawa et al. 2001).
| |
METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
CONCLUSIONS
REFERENCES
|
|---|
Detailed data are reported from ten subjects in the age range 20-57 yr (3 female, 5 male) including three of the authors. Ten additional subjects were tested in one-time short experiments with surface EMG only. These experiments were approved by the Ethics Committee on Human Experiments at Simon Fraser University.
Surface EMG recording
The subject sat comfortably in a straight back chair with knees at approximately 150° and ankles at approximately 100°. Electromyographic (EMG) activities of soleus and tibialis anterior of the right leg were recorded with pairs of Ag-AgCl electrodes. For soleus, electrodes were positioned on the midline of the leg, just below where the two heads of gastrocnemius muscle end. Electrodes for the tibialis anterior were positioned at the same height from the sole of the foot as the soleus electrodes, although occasionally TA electrodes were positioned a bit higher over the middle of the muscle belly. These signals were preamplified (Grass P15 AC preamplifiers, U. S. A.) and filtered (30 Hz-3 KHz).
Single motor unit recording
Single motor unit (SMU) activity was recorded with a pair of stainless steel wires, 25-30 µm in diameter. These wires were embedded in a 25-gauge hypodermic needle, which acted as the carrier for the recording wires and floated freely in the muscle. Activity from these electrodes was preamplified (Grass P15 AC preamplifiers) and filtered (100 Hz-10 KHz).
H reflexes and M wave
To elicit H-reflexes in soleus, the posterior tibial nerve was stimulated at the popliteal fossa with bipolar stimulating electrodes using a Grass S88 stimulator and Grass SIU5 isolation unit. Pulse width of the stimulus was 1.0 ms, and a wide range of stimulus intensities were applied starting from that needed to obtain the threshold value of the H or the M wave to the highest value required for a maximum M wave (Mmax). This range of stimulus intensity also provided the maximum value of H reflex (Hmax). The normalized value of Hmax/Mmax was computed for each subject.
Transcranial magnetic stimulation
The area of the motor cortex representing the right lower leg was stimulated with Magstim 200 (Magstim Company Ltd., U. K.) placing the center of the large double cone coil slightly left of the midline. In a few initial experiments, a Dantec MagPro stimulator with a 14.5 cm diameter, cup shaped coil was used, but much higher stimulus intensities were required to elicit responses.
Procedures
To record MEPs of soleus and TA muscles in the surface EMG, the subject was asked to relax during stimulation of the left precentral cortex. Starting at 30% of the maximum output of the stimulator (1.4 Tesla is the maximum output with the large double cone coil), stimulus intensity was increased until a clear, short latency (onset 20-30 ms) response was observed in the soleus. Stimulus intensity was varied around this value for various parts of the experiment. For input/output relationships of MEPs (intensity versus response amplitude), stimulus intensity was increased from threshold to between 70 and 80%.
For SMU activity from soleus or tibialis anterior muscle, the
subject's foot was stabilized for isometric recordings. The subject
contracted the muscle under observation until a discernible motor unit
potential could be seen and heard. Starting with 30% intensity,
stimulus was applied to the cortex at a rate of 0.2/s. The timing of
the stimulus was random with respect to the spikes of the motor unit.
Since the shape and duration of the PSTH peak was unknown, random
stimulation was preferred over spike triggered stimulation
(Olivier, Bawa and Lemon 1995). The stimulus intensity was adjusted to allow clear responses of the tonically firing unit to
be observed. The goal was to obtain approximately 100 responses at one
single stimulus intensity. Responses of some motor units were examined
at more than one stimulus intensity.
Data recording
All data comprising surface EMG, SMU activity and stimulus
triggers were recorded on a video tape (Vetter digital PCM Recorder, Model 4000A, Rebersburg, PA) for off-line analysis. The bandwidth of
the recorder for the EMG channels was DC-7 KHz and that for the SMU
channel was DC 
15 KHz. Data were analyzed off-line using SPIKE2 and
SIGNAL software from Cambridge Electronic Design Ltd. (U.K.)
Data analysis
Each motor unit was converted into a TTL pulse with two in-series Bak time-voltage window discriminators (Bak, U. S. A.). Both for SIGNAL and SPIKE2, surface EMG was sampled at 1.0 KHz while single motor unit data were acquired at 10 KHz. SIGNAL was used for averaging MEPs for experiments with surface EMG recordings only. For the rest of the experiments, raw single motor unit records, associated TTL pulses, stimulus triggers, and surface EMG records from TA and soleus muscles were digitized simultaneously using SPIKE2 software. Once data were acquired on the computer, spike shapes were compared with corresponding TTL pulses. Wrongly discriminated pulses were corrected. Any spurious pulses on stimulus channel were also deleted. Surface EMG records were rectified and averaged using either SPIKE2 or SIGNAL. For spike activity, peristimulus time histograms (PSTHs) were constructed between the spikes and stimulus triggers using a 1.0 ms binwidth (except for Fig. 7). To determine firing rate, a first order interval histogram was constructed for each motor unit using 1.0 ms binwidth. The peak or the mode value has been provided as the interspike interval (ISI) in the Results
Statistics
To establish the presence and duration of a PSTH peak, two
statistical methods were used so that we could compare our values with
those in the literature. The first was the 2
test (2 test). For each of the 1 ms bins in
the PSTH between 20 and 60 ms (post stimulus time), it was determined
if the observed counts were significantly (P < 0.05)
greater than the average count per bin in the background period 50 to
2 ms prior to stimulus. The width of the response peak was taken from
the time of the first significant bin to the last significant bin
irrespective of whether the in-between bins had significant activity.
The reason for this is that due to the presence of subpeaks with TMS
there are bins with zero or insignificant activity, but the total peak
width includes these bins. The second test for significance of the peak was similar to the one use by Brouwer and Qiao (1995)
and will be referred to as 3SD test. For this test, the mean and the SD of the background activity (50 to 2 ms prestimulus time) were computed for each histogram. Bin activity was considered significant if
the number of counts in a bin were greater than the mean plus three SDs
of the background counts per bin. Again, the peak width was calculated
from the first significant bin to the last significant bin irrespective
of the significance of the in-between bins. The 2 test for bin counts is very stringent
whereas the 3SD test occasionally gave false positive results.
Therefore we first determined the peak with 2
test, did the 3SD test for 5 bins before and 5 bins extending beyond
the peak determined by 2 test. Once peak width
was determined, the response probability (Pr) was
calculated using the method given in Bawa and Lemon
(1993) for both the peak defined by the
2 test and the peak defined by the 3SD test.
Response probability is defined as the number of counts in the response
peak per stimulus.
Corrections for onset of PSTH peak
The motor unit potential recorded with intra-muscular
microelectrode samples only a few muscle fibers of the motor unit and, hence, has a very short-duration. There can be substantial delays within the muscle between when a motor unit is activated and when the
microelectrode picks up the activity. This delay depends on where in
the muscle the tip of the microelectrode is located. The delay between
the start of the electrical activity of the whole motor unit and the
TTL used to construct the PSTH was calculated by spike trigged
averaging of the unrectified surface EMG with respect to the TTL
(Milner-Brown and Stein 1975). The averaged activity is
the compound action potential of the whole motor unit. Once the onset
of the peak was determined by applying the statistical methods, it was
corrected by the amount of delay between the onset of the motor unit
action potential and the TTL. These delays ranged from 1 to 18 ms. The
peak onset time for each PSTH is reported after this correction.
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
CONCLUSIONS
REFERENCES
|
|---|
Surface EMG responses
Detailed observations of MEPs from surface EMG are presented from nine subjects. The stimulus threshold to observe soleus responses ranged from 30% to 70% of the maximum available stimulus intensity. The stimulus intensity was rarely increased beyond 75% because of discomfort to the subject.
An example of concomitantly recorded MEPs of soleus and tibialis anterior muscles, when the subject was completely at rest, are shown in Fig. 1 for three different stimulus intensities. The onset and duration of MEPs were the same in both muscles at all three intensities; this was true for three other intermediate intensities tested. As can be seen from this figure, when the subject was at rest, the amplitude of TA responses were larger than those of soleus (note different scales for the two muscles).
|
To compare relative amplitudes of TA and soleus responses, MEPs were recorded from five subjects at five to eight different stimulus intensities. Individual and mean values of the MEPs for all five subjects are shown in Fig. 2 using different scales for the two muscles. For each subject the slope of the input/output curves were always higher for the TA than for soleus. The same is true for the population curves.
|
Very frequently a second peak was observed in soleus with an onset latency of around 90 to 100 ms, the onset latency of the second peak varied considerably depending on the subject and the posture of the subject. The amplitude of the second peak increased with increasing stimulus intensity as can be seen by comparison of panels A through C in Fig. 1.
To determine the origin of the second (late) peak, namely whether it is
reflex in origin or results from descending pathways, additional tests
were carried out. The peak was seen only under nonisometric conditions
as shown in Fig. 3 (A,
C, E). When the subject's foot was fixed to make
the contraction of ankle flexors and extensors isometric, the second
peak was abolished (Fig. 3B). In a second test, the tibialis
anterior tendon was vibrated with a 120 Hz physiotherapy vibrator. The
primary soleus and TA peaks were not affected much, while the second
soleus peak was either reduced or abolished (Fig. 3D). In
other tests not shown in the figure, when stimulus was applied with the
ankle flexed passively thus stretching the soleus muscle, the second
peak increased in magnitude. On the other hand, when soleus was
shortened passively, the peak was reduced or abolished. In one subject,
electrical stimulation of the common peroneal nerve, distal to the neck
of the fibula, produced a soleus peak at 70 ms. The second peak in soleus with TMS also appears at 70 ms after the onset of TA MEP, thus
linking this peak to stretch reflex of soleus (Valls-Solé et al. 1994).
|
Single Motor Unit Responses
Data are reported from a total of 81 motor units from 5 subjects, 42 from soleus and 39 from tibialis anterior. Some units in each muscle were examined with multiple stimulus intensities resulting in 51 PSTHs for soleus and 45 PSTHs from TA. Stimulus intensities ranged from 40 to 75% for soleus motor units and from 38 to 47% for TA motor units. In one subject, stimulus intensities for evoking clear responses in TA were much lower than those required to produce similar responses in soleus motor units, while in the other four subjects the intensities used for the two muscles were in the same range.
The number of stimuli used to construct each of the 51 soleus PSTHs
ranged from 27 to 154 (mean ± SD; 103 ± 34), and for the 45 TA PSTHs the number ranged from 46 to 192 (117 ± 30). The number of stimuli varied for several reasons: the unit was lost during a ten
minute run, an additional unit started to discharge halfway through the
run of a targeted unit or several of the stimuli had to be ignored when
the responses were not clear. The onset and duration of peaks were
different depending on the statistic used to define the peak. Using the
2 test: onset of soleus peaks ranged from 24 to 42 ms (30.9 ± 4.6), onset of TA peaks ranged from 22 to 41 ms
(27.3 ± 3.6); peak duration for soleus ranged from 1 to 20 ms
(6.9 ± 4.2), and for TA it ranged from 1 to 9 ms (5.1 ± 2.1); response probability P for soleus ranged from 0.05 to
0.80 (0.43 ± 0.18) and for TA it ranged from 0.1 to 0.77 (0.45 ± 18). Using the 3SD criterion: onset of soleus peaks
ranged from 22 to 38 ms (29.7 ± 4.3), for TA it ranged from 21 to
38 (25.9 ± 3.4); peak duration for soleus ranged from 2 to 22 ms
(10.0 ± 4.4), for TA it ranged from 3 to 15 ms (7.8 ± 2.6);
response probability Pr for soleus ranged from
0.18 to 0.93 (0.52 ± 0.18), and for TA it ranged from 0.19 to
0.83 (0.52 ± 0.17). Similar Pr values for
soleus and TA indicates that both motoneuron populations were tested
with comparable amplitudes of EPSPs.
Clear response peaks were observed for all TA motor units. But contrary
to what has been reported in the literature (Brouwer and Qiao
1995; Morita et al. 2000; Nielsen and
Petersen 1995), we also observed clear responses for each of
the soleus motor units that we recorded. Figure
4 illustrates examples of response peaks
from soleus motor units from four subjects. Each peak stands clearly
above the background activity and is segmented into two sub-peaks.
These sub-peaks were not always clearly demarcated for every PSTH but
are shown here to illustrate the nature of sub-peaks (when observed) in
soleus responses. The lower horizontal bar indicates the duration of
the peak according to the 2 criterion while
the upper bar is according to the 3SD criterion.
|
Having observed clear PSTH response peaks for soleus motor units, the
question arose whether each motor unit in the muscle, in both soleus
and TA, responds to the same stimulus. Figure
5 shows data from three simultaneously
(but randomly picked by the electrode) recorded units from soleus and
Fig. 6 shows similar data from three TA
motor units. Each of the three units in soleus and each of the three
units in TA exhibited clear response peaks in corresponding PSTHs. The
mean firing rates (impulses/s) of the three soleus units were 5.9 (unit
1), 3.0 (unit 2) and 5.3 (unit 3); the corresponding peak widths were 7 ms, 8 ms and 14 ms respectively according to 3SD criterion, and 5 ms, 4 ms and 9 ms respectively by 2 criterion. For
the TA motor units in Fig. 6, the mean firing rates in impulses/s were
7.5 (unit 1), 6.8 (unit 2) and 9.7 (unit 3); peak widths were 9 ms, 7 ms and 7 ms respectively by 3 SD criterion, and 2 ms, 6 ms and 4 ms
respectively by 2 criterion. These data
suggest that soleus and TA motor units behave very similarly to
cortical stimulation, and every unit within each muscle respond to TMS
in a qualitatively similar fashion. Concomitantly firing motor units
from soleus were analyzed for eight more sets (2-3 units per set),
results were similar to those reported in Fig. 5. Two additional sets
were analyzed for TA with similar results. A few of the units analyzed
in these sets responded only phasically during the response peak. Such
units were clearly higher threshold units compared with the ones that
the subject discharged tonically with voluntary effort.
|
|
In the upper limb, the response probability increases with increase in
stimulus (Bawa and Lemon 1993). The same was found to be
true for the 5 soleus and 5 TA motor units tested for multiple stimulus
intensities in this study. PSTHs of a soleus motor unit and
corresponding averaged surface EMG responses are illustrated for four
stimulus intensities in Fig. 7. For the
range of stimulus intensities tested for this motor unit (58% to
65%), the duration of the surface EMG peak (D1) did not change with
intensity. Each excitatory peak was followed by depression. The
duration of this post peak depression (D2) decreased with decreasing
stimulus intensity. For the single motor unit, the corrected onset of
the peak was 27 ms for all four intensities, but the duration of the
peak had a trend to first increase and then decrease with decreasing
stimulus intensity (10 ms, 14 ms, 11 ms, and 6 ms by 3 SD criterion for 65%, 60%, 58%, and 55%, respectively). As seen for surface EMG, the
postpeak silent period was clearly longest and deepest for the
strongest stimulus.
|
SOLEUS RESPONSES TO TMS AND 1a AFFERENTS. In addition to the detailed experiments above, a few short experiments were conducted on a larger number of subjects. A total of eighteen subjects were tested for soleus responses to TMS (Magstim 200). Three of the 18 subjects tested did not show any short latency MEPs, even with background voluntary facilitation of soleus. In the rest of the subjects, responses were observed even under resting conditions, but the threshold to evoke MEPs varied considerably.
Soleus is a slow postural muscle with strong stretch reflexes and H-reflexes. Is there an inverse relationship between the threshold for response to TMS and the Hmax/Mmax ratio? In ten subjects between the ages of 20 and 30 yr, we recorded the Hmax/Mmax ratio and the threshold for response to TMS. Age limits were put on this group since the threshold to TMS responses increases with age (personal observations) and the Hmax/Mmax ratio decreases with age (Chalmers and Knutzen 2000). Two of the three
subjects who produced no clear soleus responses, had a
Hmax/Mmax >0.9 indicating a greater percentage of 1a afferent fibers with large
diameters. On the other hand, a third subject who did not show a soleus
response also had a very small H reflex
(Hmax/Mmax <0.1). Some
subjects had a very low threshold for response to TMS (<30%) and high
Hmax/Mmax values. Thus no
relationship between
Hmax/Mmax and threshold for response to TMS was found. These limited data suggest that there is no
relationship between the strength of monosynaptic cortical and reflex
control of the postural muscle. However, this conclusion should be
taken with precaution. A clear answer can be obtained, not with
H-reflexes, but by examining the gain of stretch reflexes.
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
CONCLUSIONS
REFERENCES
|
|---|
Transcranial magnetic stimulation studies on human subjects have
clearly shown excitation of limb muscles on the contralateral side. The
list of various muscles tested is compiled in Rothwell et al.
(1991). There is one exception, and that is the soleus muscle. Most of the studies demonstrated excitation of all upper limb
motoneurons, excitation of the ankle flexor, tibialis anterior, and
inhibition of the ankle extensor, soleus. However, a few studies have
shown clear excitation of soleus with surface EMG recordings (Capaday et al. 1999; Maertens de Noordhout et
al. 1999; Valls-Solé et al. 1994). Our
observations agree with these latter studies. The onset latency and
duration of MEPs are shown to be similar for TA and soleus muscles
(Morita et al. 2000; Valls-Solé et al.
1994). Valls- Solé et al. (1994) have
shown that soleus MEPs have a slightly shorter onset latency during
voluntary activation of soleus, and TA peaks can be slightly earlier
when TA is voluntarily activated. If the short latency MEPs in TA are
assumed to be of corticomotoneuronal origin, the same might apply for
soleus from the MEP studies. Since surface EMG can pick up activity of
several neighboring muscles, single motor unit studies would provide a clearer answer.
Single motor unit results
CORTICOMOTONEURONAL CONNECTIONS TO SOLEUS MOTONEURONS.
Brouwer et al. (1992), Brouwer and Qiao
(1995), and Nielsen and Petersen (1995)
recorded responses from soleus and tibialis anterior motor units. While
response peaks in PSTHs of TA motor units were very clear and very
similar to those reported in the upper limbs, response peaks of soleus
motor units were either not observed or were poorly defined (diffused)
in appearance. Of all the motor units tested, only a small percentage
of the motor units showed excitatory peaks: 18% of 62 units studied by Brouwer and Qiao (1995), and 37% of 27 soleus units
reported by Nielsen and Petersen (1995). From these
observations, the authors concluded that soleus essentially lacks
corticomotoneuronal connections, and that cortical connections to
soleus involve mostly polysynaptic pathways. In our experiments, no
difference was observed between the responses of TA or soleus motor
units as indicated by very similar Pr values.
Analysis of all sets of concomitantly firing units showed that if one
unit responded with an excitatory peak, all other simultaneously
recorded units did as well. That is, all soleus units responded with an
excitatory peak, which is contrary to the results reported by Brouwer
and Qiao (1995) and Nielsen and Petersen
(1995). Most of the units analyzed in our study were low
threshold (<10% MVC), but a few reached thresholds 
30% MVC. The
units responding just phasically during the response peak were also of
relatively higher threshold, thus providing us with the nature of CM
connections to a good sample of the soleus pool. If all motoneurons of
this large muscle did not receive weighted excitatory input as
other motoneuron pools do, it would cause serious control problems for
this postural muscle.
made a strong argument for the presence of
corticomotoneuronal connections to all muscles tested, which included
soleus. Since TES produces clear D wave and no I waves, the authors
were able to estimate the rise times of monosynaptic EPSPs in all
muscles tested, including the TA and soleus. The values were 1.13 ms
for TA and 1.14 ms for soleus, indicating similar positioning of
corticomotoneuronal synapses onto the two populations of motoneurons.
On the other hand, of all the muscles tested, soleus had the smallest
response. This observation can be interpreted as soleus requiring
higher stimulus intensity to produce a response compared with its
anatomical neighbor TA, as observed by us. Therefore it is proposed
that the poorly defined PSTH peaks, or absent soleus responses
(Brouwer and Qiao 1995; Morita et al.
2000; Nielsen and Petersen 1995) may be due to
weak stimulus intensities used in these studies. Morita et al.
(2000) reported observing short latency peaks in soleus
motor units at very strong voluntary background contractions. In our
study, most of the peaks were constructed for low threshold units
firing at the lowest possible rates and background contraction was
generally very low (<10% MVC as perceived by the subject). An
important difference to point out is the way stimulus was applied with
respect to motor unit spikes. The random mode of stimulation used by us gives a true profile of the PSTH peak, whereas when the applied stimulus is triggered by the preceding motor unit spike (Morita et al. 2000), it affects the shape of the earlier part of the peak, particularly the wide peaks observed for the lower limb motor
units. Low stimulus intensities combined with spike-triggered stimulus
may have resulted in narrow or nonexistent peaks.
Inhibition of soleus reported by other authors is not surprising either
(Rothwell et al. 1991). Inhibition of voluntary activity is observed at low stimulus intensities even for the upper limb muscles. Only when intensity is increased does one see the excitatory peak (Davey et al. 1994). Since higher intensities are
required for soleus compared with those for TA, inhibition or a lack of clear excitation of soleus motoneurons would result if the same stimulus intensity were used for the two muscles. In fact, we did
observe inhibition of soleus units with low intensity stimulus. With
adequate stimulus, when each response was examined, we saw no
difference in the pattern of responses between the upper limb (Bawa and Lemon 1993) and lower limb motoneurons. Clear
inhibition, indicated by a very long silent period after the stimulus,
was frequently observed for motoneurons belonging to all motoneuron pools, even though the averaged response for each motoneuron was excitatory.
PEAK WIDTHS OF SOLEUS AND TA PSTH PEAKS.
The peak widths reported in the upper limbs are approximately 5 ms
(4.6 ± 1.7 ms for intrinsic hand muscles, Miller 1991). Using 3SD
test, Brouwer and Qiao reported TA peaks in the range 2.4 ± 0.1 ms (59 units), and soleus peaks in the range 3.3 ± 0.4 (62 units). Using 2 test, Nielsen and Petersen
(1995) reported soleus peak width in the range 2-5 ms
for their 27 motor units. In the present work, the mean duration with
3SD test was 7.8 ± 2.6 ms for TA and 10.0 ± 4.4 ms for
soleus. These values are much higher than those reported by Brouwer and
Qiao. Using 2 test, our values are higher than
those given by Nielsen and Petersen. These values are much higher than
those for the upper limb motoneurons in normal human subjects. When the
soleus response was segmented into sub-peaks, we never observed more
than two clear sub-peaks (Fig. 4). In the upper limbs, sub-peaks are
<2 ms in duration (Miller 1991). If we consider
sub-peaks in Fig. 4, a sub-peak could be as long as 8 ms (second peak
for Subject 1). As for the upper limb, we assume that the sub-peaks
originate from D and I waves. If D and I waves were all to travel in
the fast CM axons, then the peak width should not be different in the
upper and lower limbs, there should be very little dispersion between
successive waves from cervical to the lumbar regions. How do we explain
the differences in the peak widths and widths of sub-peaks for the lower limb motoneurons? Probably several factors contribute to the peak
width. According to Edgley et al. (1997), fast CM axons have a lower threshold for D waves while the slow axons have a lower
threshold for I waves. If D and I waves travel in axons of different
conduction velocities, the dispersion between successive waves would
increase with distance from the cortex. This is proposed to be the
reason for the longer duration of response peaks in PSTHs constructed
for lumbar motoneurons. Factors which could cause longer peaks in
soleus compared with those in TA motoneurons are: (1) weaker
connections onto soleus motoneurons would produce weaker EPSPs, which
in turn result in longer duration of PSTH peaks (Fetz and
Gustaffson 1983), (2) slower motoneurons produce slower EPSPs
and hence longer PSTH peaks, and (3) polysynaptic pathways may
contribute to soleus peaks (Nielsen and Petersen 1995).
LATENCY AND DURATION OF PSTH PEAKS. In Fig. 7 we reported a slight decrease in peak width at the highest stimulus intensity. This should not be taken as a universal observation. The peak latency and duration depends on the motoneuron firing rate, the afterhyperpolarisation (AHP) of motoneuron, the strength and duration of the synaptic volley, and excitability of the cortex and spinal cord. Therefore to compare the effect of stimulus strength on the latency or duration between motoneurons of different subjects, or of different pools or even within the same pool, is difficult. With the presence of multiple sub-peaks of a PSTH, the response probability Pr will always increase (unless it reaches maximum) with increasing stimulus intensity. But a shift in latency and change in peak width (increase or decrease) depends on many factors. When stimulus intensity is increased, one observes all possibilities such as: an increase in Pr without shift in onset latency or change in duration; an increase in Pr with an increase in duration and decrease in latency; an increase Pr with no decrease in latency, but a decrease in duration. For example, if a small intensity evokes a D wave and the motoneuron responds to this D wave, then we already may have the shortest peak latency, and there may not be a further shift in latency with increasing stimulus intensity. In fact, the latency may stay the same, and as Pr increases, the duration will decrease because the motoneuron will respond more frequently to the D wave, and therefore less to the subsequent I waves. If on the other hand the weak stimulus does not evoke a D wave, or the motoneuron is not responsive to the D wave, then the peak onset is late. However, if slightly higher stimulus intensity produces a response to D wave, then one will observe a shortening of the peak latency while Pr increases. Because of the myriad of possibilities, this question cannot be answered with this study.
Surface EMG
Figure 2 demonstrated the presence of a second soleus peak that
has also been reported by Valls-Solé et al.
(1994). These authors demonstrated that this peak is
absent during isometric contractions or when soleus is shortened
passively suggesting that this peak results from stretch reflex of
soleus muscle when TA contracts in response to TMS. Our results confirm
their findings with the additional test using vibration. Vibration of
TA tendon drives Ia spindle afferents of the ankle flexor; this high Ia activity produces primary afferent depolarization in the terminals of
soleus Ia afferents causing a reduction of soleus stretch reflex by
presynaptic inhibition (Hultborn et al. 1987). The
question arises: if both TA and soleus are active at the same time, why does only soleus show this activity. There are two possible reasons: first, that TA is always activated more strongly than the soleus muscle
(Fig. 3), thus TA contracts more strongly flexing the ankle and
stretching soleus; the second reason is suggested to be a higher gain
stretch reflex in soleus than in TA. Such peaks may be extremely
important in interpretation of data from other experimental paradigms.
For example, if such a peak is subliminal in conditioning-testing experiments, a large test response may simply be the result of superposition of test and (subliminal) reflex activity, and hence, misinterpreted.
POSTEXCITATORY INHIBITION.
Figure 7 shows that the postexcitatory inhibition increased pari
passu with the excitatory peak. The mechanisms underlying this
inhibition have been suggested to be mainly cortical (Fuhr et
al. 1991). One argument is that post MEP inhibition is observed in hand muscles where Renshaw cell inhibition is not present. Second,
RC inhibition does not last 150-200 ms and the post MEP inhibitory
period does. Third, one can elicit H-reflexes during this inhibitory
period, therefore inhibition cannot be spinal. We argue that a
substantial contribution to this reduced activity comes from
refractoriness of motoneurons, and that this refractoriness can be made
almost twice the duration of the ISI (interspike interval). It is true
that RC inhibition, when it exists, will affect only the initial parts
of this period. However, refractoriness of motoneurons will contribute
to the postpeak depression considerably up to twice the duration of
AHP. When a stimulus arrives at random times with respect to the spikes
of motoneurons, the stimulus can discharge some of the motoneurons even
when the EPSP is produced during the AHP. At weak stimulus intensities
only those motoneurons, which are in late phases of AHP, will respond
to the small EPSP. When the CM volley is extremely strong, it can
generate large enough EPSPs to discharge even those motoneurons that
just fired 15-20 ms before (Jones et al. 1995). Such
motoneurons, after responding to the CM volley, will be refractory for
almost twice the duration of AHP. A strong CM volley, therefore can
produce a silent period of 150-200 ms in motoneurons discharging at 10 imp/s. When just about all tonically firing motoneurons respond to the
CM volley, the MEP is large, and so is the depth and duration of the
post MEP depression. The reason why one can elicit H-reflexes during this depression is that with appropriate stimulus strength applied during motoneuron refractoriness, one can discharge the motoneuron again (Jones et al. 1995). A clear indication of
motoneurons being able to fire again during this period is the second
soleus peak shown in Figs. 1 and 3. An argument for segmental
contribution to the post MEP depression does not in any way diminish
the importance of the contribution of the cortex to this depression.
| |
CONCLUSIONS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
CONCLUSIONS
REFERENCES
|
|---|
Soleus receives corticomotoneuronal connections like other
motoneuron pools innervating the limb muscles. The difference is that
the connections are relatively weak in many subjects, however all
motoneurons receive excitatory connections for systematic recruitment
as motoneurons in other muscles do. The existence of a large range of
stimulus intensities to elicit MEPs in subjects 20-30 yr of age
suggests a large variability in the strength of CM connections onto
soleus motoneurons. The onset and duration of MEPs in the ankle
extensor soleus and the ankle flexor TA were very similar suggesting
qualitatively similar cortical connections. On the other hand,
input-output relations of concomitantly recorded MEPs indicates much
stronger connections onto the ankle flexor TA motoneurons, which is
opposite of what has been shown for the wrist muscles. Fetz and Cheney
(1980) showed stronger CM connections onto wrist
extensors than onto wrist flexors. Functionally, this is not the
opposite; wrist flexors and ankle extensors are antigravity muscles and
may depend more on segmental reflexes. Single motor unit PSTH peaks in
the lower limb muscles are of longer duration than in the upper limbs,
otherwise no clear difference was observed in their input-output
properties, post peak depression, or any other activation patterns.
| |
ACKNOWLEDGMENTS |
|---|
This work was supported by a grant from Natural Sciences and Engineering Research Council of Canada.
| |
FOOTNOTES |
|---|
Address for reprint requests: P. Bawa, School of Kinesiology, Simon Fraser University, Burnaby, BC V5A 1S6, Canada (E-mail: bawa{at}sfu.ca).
Received 10 January 2002; accepted in final form 28 February 2002.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
CONCLUSIONS
REFERENCES
|
|---|
-motoneurones following single spikes and doublets. In Alpha and Gamma Motor Systems, edited by
Taylor A,
Gladden MH, and
Durbaba R. New York: Plenum Press, 1995, p. 103-105.This article has been cited by other articles:
|
|
 |
|
  P. G. Martin, J. E. Butler, S. C. Gandevia, and J. L. Taylor Noninvasive Stimulation of Human Corticospinal Axons Innervating Leg Muscles J Neurophysiol, August 1, 2008; 100(2): 1080 - 1086. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Del Balso and E. Cafarelli Adaptations in the activation of human skeletal muscle induced by short-term isometric resistance training J Appl Physiol, July 1, 2007; 103(1): 402 - 411. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Z. Ross, N. Middleton, R. Shave, K. George, and A. Nowicky Human, Environmental & Exercise: Corticomotor excitability contributes to neuromuscular fatigue following marathon running in man Exp Physiol, March 1, 2007; 92(2): 417 - 426. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. Taube, M. Schubert, M. Gruber, S. Beck, M. Faist, and A. Gollhofer Direct corticospinal pathways contribute to neuromuscular control of perturbed stance J Appl Physiol, August 1, 2006; 101(2): 420 - 429. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. W. Stinear and T. G. Hornby Stimulation-induced changes in lower limb corticomotor excitability during treadmill walking in humans J. Physiol., September 1, 2005; 567(2): 701 - 711. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R.-V. Ung, M.-A. Imbeault, C. Ethier, L. Brizzi, and C. Capaday On the Potential Role of the Corticospinal Tract in the Control and Progressive Adaptation of the Soleus H-Reflex During Backward Walking J Neurophysiol, August 1, 2005; 94(2): 1133 - 1142. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
