Exp Brain Res (1984) 53:400-408

( Springer-Verlag 1984 9

Changes in Reciprocal Ia Inhibition During Voluntary Contraction in Man

M. Shindo 1, H. ttarayama 1'3, K. Kondo t, N. Yanagisawa 1, and R. Tanaka 2
1 Dept. of Medicine (Neurology), Shinshu University School of Medicine, Asahi 3-1-1, Matsumoto, Japan 2 Dept. of Neurobiology, Tokyo Metropolitan Institute for Neurosciences, 2-6 Musashidai, Fuchu City, Tokyo, Japan

Summary. Reciprocal Ia inhibition from ankle flexors to extensors was studied during voluntary tonic isometric dorsiflexion and plantar flexion in five normal subjects. The Ia inhibition was examined as the short-latency suppression of the soleus H-reflexes by stimulation of the low-threshold afferents in the common peroneal nerve (Mizuno et al. 1971). At rest, weak la inhibition was demonstrated in four subjects out of five, the maximal amount being 14.1 + 5.0% suppression of the control H-reflex. The absolute amount of inhibition, which was calculated by subtracting the mean size of the conditioned Hreflex from that of the control H-reflex and expressed as a percentage of the maximal M-response, increased during ankle dorsiflexion, and decreased or disappeared during plantar flexion in parallel with the amount of contraction. The neural mechanisms for facilitation of the Ia inhibitory pathway during dorsiflexion were considered to support the hypothesis of "a-y-linkage in reciprocal inhibition", i.e. combined facilitatory effects on the Ia inhibitory interneurone from the supraspinal centers directly and indirectly via the y motoneurone - Ia afferent route. The mechanism for inhibition of the pathway during plantar flexion was considered to be inhibition of the Ia interneurone of the flexor side by Ia interneurone of antagonist extensors. A quantitative aspect of activity in the reciprocal Ia inhibitory pathway on the performance of voluntary movement is revealed in this study.

Introduction
Reciprocal Ia inhibition is considered to be one of the most important neural mechanisms in the performance of movements, and it has been extensively investigated in animals (Eccles 1969; Lundberg 1970; Hultborn 1972; Baldissera et al. 1981). It has been established that the interneurone mediating disynaptic Ia inhibition is under the control of various supraspinal and segmental systems, and it has been suggested that this interneurone plays the role of an integrative center in reciprocal innervation. In man, the Ia inhibition was revealed initially in subjects with certain neurological disorders (Mizuno et al. 1971; see also Yanagisawa et al. 1976; Yanagisawa and Tanaka 1978; Yanagisawa 1980). It appeared as a very short latency inhibition of the Hreflex in the ankle extensors following stimulation of the low-threshold afferents in the antagonist peroneal nerve. Ia inhibition of ankle flexors by tibial nerve stimulation was also demonstrated. Ia inhibition from the ankle flexors to the extensors has been observed in only a small fraction of normal subjects at rest, and has always been very weak (Tanaka 1974, 1980). Tanaka, however, demonstrated it to appear constantly during voluntary dorsiflexion of the foot. The threshold for this inhibition was lower in stronger contraction than in weaker contraction. On the basis of these observations he postulated a positive relationship between the excitability of the Ia inhibitory pathway and the amount of contraction. In order to investigate this more precisely we studied the quantitative relationship between amounts of Ia inhibition and voluntary contraction; Ia inhibition on the ankle extensors from the flexors was studied at rest and during several strength stages of tonic voluntary dorsiflexion and plantar flexion of the foot.

Key words: H-reflex - Ia irrhibition - Tonic voluntary contraction - Ankle muscles - Man

3 Present address: Dept. of Neurology, Brain Research Institute, Niigata University, Asahimachi 1, Niigata City, Japan Offprint requests to: Dr. M. Shindo (address see above)

M. Shindo et al.: Changes in Reciprocal Ia Inhibition During Voluntary Contraction in Man oSCillOSCOpe

401

~--------------~ line tarot

~Or~

~J ~l-wave rectit~l [~_gated integrator I

~ - ~ ]

recticorder 91data recorder

Fig. 1. Schematic illustration of experimental setting

Methods
The study was made on five healthy male volunteers (ages between 26 and 33). Most subjects were examined 2-4 times on different dates. The subject was comfortably seated in a reclining armchair. The leg was fixed to an immobile foot-plate which was connected to a torque meter. The thigh was fixed to the chair and the foot to the foot-plate with leather bands. The angles of the knee and ankle joints were kept at about 120 and 100 deg, respectively. The experimental setting is shown in Fig. 1. Electromyogram was recorded bipolarly with surface electrodes. The paired electrodes were placed 3--4 cm apart longitudinally over the soleus, just below the gastrocnemius muscle belly, and the tibialis anterior muscles. The EMGs were amplified with conventional amplifiers (T.C. = 0.05 s, high cut = 1 kHz), and were recorded on an inkwriting recticorder (Nihon Kohden, WI-640G) and on a data recorder (Sony, DFR-3515). Conditioning stimuli were applied monopolarly to the common peroneal nerve at the level of the caput fibulae. A needle or surface electrode was used as cathode, and a surface plate electrode was placed anterior to the caput fibulae as anode. The stimulating pulse had a duration of 1 ms, and the strength was 0.9-1.2 times the threshold for the M-response (XMT) in the tibialis anterior muscle. Single or 2-3 pulses with intervals of 3.5 ms were delivered. The position of the cathode electrode was adjusted in order to evoke contraction of the tibialis anterior muscle, but not the peroneal muscles, at the lowest motor threshold. For evoking test H-reflexes in the soleus muscle the tibial nerve was stimulated in the popliteal fossa with surface electrodes. The stimulus was a single pulse of 1 ms duration, the strength being adjusted to obtain the maximal H-reflex and also Hreflexes with amplitudes between % and % of the maximum. A small M-response was obtained with the maximal H-reflex. The stability of the stimulating conditions was monitored by this small M-response throughout the experiment. It was essential to ensure that stimulation of one nerve did not spread to its antagonist, and for this purpose the electrode settings were adjusted so that the conditioning stimuli to the flexor nerve never evoked M- or H-response in the extensor muscles even with a strength of three times the threshold for the flexor M-response, and, similarly, the test stimuli to the extensor nerve did not excite the flexor muscles even when applied at a strength of 1.5 times that needed to evoke the maximum H-response in the extensors.

Recording was done at rest, and during tonic voluntary dorsiflexion or plantar flexion of the foot. The test stimuli were applied every 3--4 s. An interstimulus interval of more than 10 s has been recommended to avoid cumulative depression of the H-reflex by preceding stimuli (Desmedt 1973). But since a longer interstimulus interval would result in an unfavourable anticipation of the electrical shocks (Yanagisawa 1980) and require an unfavourably long period for the experiment, we adopted the present interval as a reasonable compromise. The first several reflexes in each session of the experiment were discarded to minimize the change in cumulative effect of inhibition by the preceding stimuli. Preceding conditioning stimuli were combined with them in random sequences st) that the subject could not anticipate when the conditioning stimuli were to be applied. The conditioned Hreflex of the soleus muscle was first recorded with different conditioning-test stimulus intervals from 0 to 20.0 ms in order to reveal the time-course of the conditioning effect. Then, to study the change of the conditioned H-reflex with respect to the strength of voluntary dorsiflexion or plantar flexion, recordings were made with a fixed conditioning-test stimulus interval set between 1.0 and 2.0 ms, which showed optimal effect. The threshold of the conditioning stimulus was checked during voluntary" contraction as well as at rest, and was verified to be constant. The size of the test reflex was measured by peak-to-peak amplitude in earlier experiments, but in later experiments the H-reflex was subjected to fullwave rectification and integration, and its size was then read out through a digital voltmeter. Evoked EMG responses were also recorded on an ink-writer through a memory device (8-bit A/I) input, lk words, Kawasaki Electronica TM-1410), the sampling time and the readout time being 100 las/word and 2 ms/word, respectively, and were used for checking stability in the experiment. The muscle contraction was performed nearly isometrically in order to minimize the change of muscle length. Although the leg was fixed to the immobile foot-plate, the contraction was not completely isometric and the angle of the foot could change up to about 5 deg at the maximal dorsiflexion or plantar flexion. However, the angle-change in actual experiments (up to 30% of the maximal contraction) was much smaller, and did not seem to affect the results seriously. The amount of contraction was monitored by the torque meter and was expressed as a percentage of the torque force at the maximal contraction. The subject was requested to adjust the strength of contraction in several degrees, which were indicated on the monitor oscilloscope as a target line. The amount of contraction requested was within 30% of the maximum for the following reasons: (1) The antagonist stretch from incomplete fixation of the leg, although inevitable, should be minimized during voluntary contraction. (2) Co-contraction of the antagonists could occur with stronger contraction. (3) Tile unconditioned control H-reflex might decrease very much in size with stronger contraction of pretibial muscles, which would make it difficult to assess the conditioning effects. (4) The subject was sometimes unable to maintain stronger contraction for a long enough period of time. (5) Relative linearity between EMG and torque was well maintained in this range. Since the torque is a parameter of the force applied mechanicaUy to the foot-plate we were uncertain whether it would directly represent the physiological phenomenon. To examine the relationship between the mechanical effect and the physiological phenomenon, we compared the electromyogram of the lower leg and the torque around the ankle. In all subjects the amount of contraction measured by the torque meter related almost linearly with that measured by the integrated EMG in the range below 40% of the maximal contraction measured by torque, which was the range used in the present study. The slope of the regression line differed somewhat between dorsiflexion and plantar flexion, and among the subjects. Co-contraction of the antagonists was not

402

M. Shindo et al.: Changes in Reciprocal Ia Inhibition During Voluntary Contraction in Man inhibition in normal persons. The amount of inhibition, however, differed according to occasion even in the same subject. For example, in the subject shown in Fig. 2, the amounts of Ia inhibition (mean value) by a single conditioning pulse with conditioning-test stimulus interval of 2.5 ms were 20.8%, 19.4% and 11.0% of the control H-reflex in the first, second and third sessions of experiment, which were carried out at intervals of more than one month. Generally the amount of inhibition at rest was rather small, and tended to decrease with repetition of experiments. The maximal amount of inhibition at rest among the four subjects was 14.4 ___5.0 (mean and SD) % of the control H-reflex. During tonic dorsiflexion of the foot this inhibition became more remarkable. In the case shown in Fig. 2, the amount of inhibition increased from 19.5% to 51.1% of its unconditioned value at 2.5 ms conditioning-test stimulus interval (open circles), but it almost disappeared during tonic plantar flexion (diamonds). We examined the relationship between the amount of inhibition of the conditioned H-reflex and the amount of tonic voluntary dorsiflexion and plantar flexion. Figures 3 and 5 are data from the subject in Fig. 2.

demonstrated in the present series beyond potential spread from the antagonists. It took one to two hours for one series of experiments per subject, during which time the laboratory was kept quiet and the lighting constant. We tried to keep the subject alert as well as relaxed, and to keep his posture constant throughout the recording session.

--~120[

~ +6(2 oN o~ 4o

""'""

1"

~ ...........

controlO 0'.5 1:0 1.5 2.'0 15 3'.0 3.5 4'.0 Conditioning-test stimulus interval (ms)
Fig. 2. Time course of Ia inhibition. At rest inhibition began between 1.0 and 1.5 ms after the conditioning stimulus (filled circles). The amount of inhibition increased during dorsiflexion (open circles), and almost disappeared during plantar flexion (diamonds). Both contractionswere 5% of the maximum. A single conditioning stimulus with intensity of 1.00 XMT was used. Each symbol represents the mean and vertical bars show the standard error obtained from 5 to 13 trials. The abscissa shows the conditioning-test stimulus interval (ms), and the ordinate the size of H-reflex (% of the unconditioned value)

Ia Inhibition on Ankle Extensors During Tonic Contraction of Ankle Flexors

Figure 3 shows the peroneal inhibitory e f f e c t t e s t e d at the optimal interval of 2.0 ms, while the strength of ankle dorsiflexion was set at several different stages. The size of the conditioned H-reflex expressed as a percentage of its unconditioned control value decreased in parallel with the amount of tonic voluntary dorsiflexion from 88.8% at rest to 51.5%, 25.2%, 23.3% as the amount of contraction increased by 10%, 20%, 30% of the maximum, respectively (Fig. 3A, open circles). A t first sight, this might seem to indicate that Ia inhibition of soleus motoneurones increased in accordance with the strength of voluntary dorsiflexion. However, the control H-reflex itself also decreased in size during dorsiflexion: from 25.6% of the maximal M-response at rest to 9.9%, 7.7%, 8.9% at 10%, 20%, 30% of the maximal contraction (Fig. 3A, filled circles; see also Tanaka 1974). If the sizes of the control H-reflex were markedly different, the same inhibitory effect could result in different amounts of change, and it would be difficult to correlate directly the relative amount of inhibition, as expressed by a percentage of its unconditioned H-reflex, with the strength of contraction. Fortunately, in this case, the control Hreflexes were almost the same at various strengths of

Results

Figure 2 shows the time course of the early suppression of the soleus H-reflex by conditioning single stimuli with a strength just liminal for the direct Mresponse of the tibialis anterior muscle in one subject. In a resting state the H-reflex began to be suppressed at a conditioning-test stimulus interval between 1.0 and 1,5 ms, the maximal amount of suppression being 19.5% of the unconditioned value of the H-reflex at 2.0 ms (filled circles). This suppression was demonstrated even with a strength of 0.9 XMT. The characteristics of short latency and low threshold (less than 1 XMT) indicate that the suppression is reciprocal Ia inhibition (Mizuno et al. 1971; Tanaka 1974). This Ia inhibition on soleus motoneurones at rest was observed in four subjects out of five in contrast to the previous study (Tanaka 1974, 1980), which showed a very low incidence of Ia

M. Shindo et al.: Changes in Reciprocal Ia Inhibition During Voluntary Contraction in Man

A
"1-

403

B "~ 20

A ~15 E

B
"6

! ,ot
~-6o
t~

/
I I I I

8
"6

10

c'-

~50-

~ 40 -40
"r'30

Y
-5
I I

.E "65

~3o
i
I 0 I
I I d

,,- 20
O

i-O O

~820i20
-510 co
I I I I I

rest 10 20 30 rest rest 10 20 30 rest sttn~ stknl Amount of dorsiflexion ( % of max.contr.)

rest

10

20

30

rest

10

20

30

Amount of dorsiflexion ( % of max. contr. )

Fig. 3A, B. Inhibition of H-reflex during dorsiflexion. A Open circles show the inhibition, expressed as % of the unconditioned control H-reflex, which increased as contraction strengthened. Filled circles show the control H-reflex, expressed as % of the maximal M-response, which decreased on contraction. When a smaller test H-reflex was used, the inhibition did not increase (right-hand side). Each symbol represents the mean of five data and the standard error. B The abohite amount of inhibition (% of M max), which was calculated by subtracting the mean size of the conditioned H from that of the control H, increased in parallel with the amount of contraction. Each vertical strip indicates the range of expected standard deviation of the difference of means

Fig. 4A, B. The change in the control H-reflex size (A) and the absolute amount of Ia inhibition (B) at rest and during tonic dorsiflexion of the foot in all subjects. The absolute amount o f inhibition was calculated as in Fig. 3B. The control H-reflex decreased in size and the Ia inhibition increased with respect to the amount of contraction. The strength of conditioning stimuli was between 0.9-1.2 XMT. The abscissa shows the amount of contraction (% of the maximum), and the ordinate the size of the control H-reflex (% of M max, A) and the absolute amount of inhibition (% of M max, B). * and A show the data from the same two subjects examined on different dates

contraction and, therefore, the increase of inhibition at stronger contraction, particularly from 10% to 20%, seems meaningful. Further, even when weaker stimuli were used at rest so as to obtain the same size of test reflex as used during contraction, the amount of Ia inhibition was very small, as was the case in the larger control H-reflex at rest (compare left-hand and right-hand sides in Fig. 3A). It seems justifiable, therefore, to think that the activity of the Ia inhibitory pathway increases as the contraction becomes stronger. However, the unconditioned H-reflex did not always remain constant at different amounts of antagonist contraction. In fact, in the majority of experiments, the control H-reflex tended to be depressed in parallel with the amount of dorsiflexion; the stronger the contraction, the smaller the size of the H-reflex (Fig. 4A). Thus, we cannot readily compare the relative value of inhibition expressed as a percentage of its control value between the resting state and various stages of contraction. In this case, the size of test H-reflexes could be adjusted at rest to be the same as that in different experimental conditions; i.e. during tonic voluntary contraction. This

procedure, however, has some disadvantages. For instance, the detailed shape of the H-reflex during active dorsiflexion is not necessarily the same as that of the H-reflex at rest even if the sizes of the two are adjusted so as to be the same. This means that the sampled motoneurones in the test H-reflexes in both conditions would be partly different. Besides, it takes much time to adjust on each occasion the size of the control H-reflex at rest to the same amplitude as in various strengths of contraction. We, therefore, decided to calculate the absolute size of soleus motoneurones that were inhibited by the conditioning stimuli. This was done by subtracting the mean value of the absolute size of the conditioned H-reflex from that of the unconditioned control H-reflex, and expressing this as a percentage of the maximal M-response. The value calculated by this procedure represents the absolute size of motoneurones which were inhibited by the conditioning volley within the motoneurones partially inhibited by dorsiflexion. Here we call this value the absolute amount of Ia inhibition. The results were an increase in the absolute amount of Ia inhibition by increase of dorsiflexion (Figs. 3B and 4B). Figure 3B demon-

404

M. Shindo et al.: Changes in Reciprocal Ia Inhibition During Voluntary Contraction in Man

strates a successive increase in inhibition f r o m 3.0% at rest to 4 . 8 % , 5 . 7 % , 6 . 8 % of the maximal Mresponse at 10%, 2 0 % , 30% of maximal dorsiflexion, respectively. It should be noted that this procedure is apparently an underestimation of the excitability change in the Ia inhibitory pathway, for the following reasons. Firstly, the soleus motoneurones, which provided part of the control H-reflex at rest and were suppressed by active dorsiflexion (Figs. 3A, 4A), would be more susceptible to inhibition, including Ia inhibition, than the ones suppressed by the conditioning volley. If this is the case, the former group of motoneurones were already excluded from the test sample during contraction, and did not contribute to the part suppressed further by a conditioning volley, resulting in a smaller estimation of Ia inhibition than actually occurred. Secondly, increase in Ia discharges from the ankle flexors during active dorsiflexion (Vallbo 1970) might cause an occlusion in the afferent pathway, and reduce the size of conditioning Ia volleys and thus of Ia inhibition. Since the Ia discharges increase more in stronger contraction (Vallbo 1974), this factor would play a larger role in stronger contractions. Figure 4 summarizes the change in size of the unconditioned control H-reflex ( A ) as well as the absolute a m o u n t of Ia inhibition by the conditioning stimuli (B) at rest and during tonic dorsiflexion in all subjects. A l t h o u g h the control H-reflex decreased in size as contraction s t r e n g t h e n e d , the absolute a m o u n t of Ia inhibition increased almost parallel with the a m o u n t of contraction. This indicates an actual increase in the soleus m o t o n e u r o n e s which were inhibited via the Ia inhibitory p a t h w a y as the contraction increased. E v e n t h o u g h the increased a m o u n t of inhibition by dorsiflexion could c h a n g e in different sessions d o n e on different days with the same subject, this relationship was p r e s e r v e d ; this is presented in Fig. 4B as a parallel shift of a line in two subjects, and shows the reproducbility of the experiment.

!1oo
i 80

"5 15

i,o[
~ac

.
I I I I

i
I I | I

rest

10

20

30

rest

10

20

30

Amount of plantar flexion ( o~ of max. contr.)

Fig. 5A, B. Inhibition of H-reflex during plantar flexion. Similar illustration as Fig. 3. A The inhibition decreased, while the control H-reflex increased as voluntary contraction strengthened. B The absolute amount of inhibition decreased in parallel with the mount of contraction

A
"~15 E

~1o

~5 5

w
~5 20
0
I I

Ia Inhibition on Ankle Extensors During Tonic Contraction of Ankle Extensors

Figure 5 shows the results f r o m a case in which the a m o u n t of inhibition of the c o n d i t i o n e d H-reflex decreased from 50.1% o f its u n c o n d i t i o n e d value at resVto 9 . 4 % , - 2 . 9 % , 0 . 3 % at a m o u n t s of contraction, of 10%, 2 0 % , 30% of the m a x i m u m (Fig. 5 A , o p e n circles). O n the o t h e r h a n d , the u n c o n d i t i o n e d control H-reflex itself increased successively f r o m 28.9% of the maximal M - r e s p o n s e at rest to 4 2 . 1 % , 4 3 . 4 % , 48.0% at 10%, 2 0 % , 30% of the maximal contraction (filled circles). T h e absolute a m o u n t of inhibition, which was calculated as in the series on dorsiflexion, decreased from 14.4% at rest to 3 . 9 % , - 1 . 3 % , 0 . 2 % of the maximal M - r e s p o n s e at 10%, 2 0 % , 30% of the maximal plantar flexion (Fig. 5B).

I=,

rest

10

2O

3O

rest

10

20

30

Amount of plantar flexion

( % of max. contr. )

Fig. 6A, B. The change in the control H-reflex (A) and the absolute amount of Ia inhibition (B) at rest and during tonic plantar flexioff'in all subjects. Similar iUustr~ion as Fig: 4. The control H-reflex increased in size during contraction, and the absolute amount of Ia inhibition decreased almost parallel with the amount of contraction. The stimulating condition was the same as in Fig. 4 Figure 6 summarizes the c h a n g e of the control Hreflex and the absolute a m o u n t of inhibition with respect to the a m o u n t of tonic plantar flexion in all five subjects. T h e control H-reflex increased in size with stronger voluntary contraction, a n d at the same

M. Shindo et al.: Changes in Reciprocal Ia Inhibition During Voluntary Contraction in Man

405

[ d~
amount of

i I =fk o, I

la inhibition

oo o o o o o o o oooeooo

oo oooo

o~ oe o e o o o o o eoooooeo o oooooooo oo o o o o o o e e o o o o o o o e oooooooooooo~oo ooooooooooooo oo o o o o o o o o o o o o ooooooeooo~ oo o o o o o o o o o o o oooooooooooo oo o o o o o o o ooooooo oo o9 OoOo~ o

. o oo o e oo o o oo o ooo oo % o o ~9 o o,o o ,o o * o o
ooo oo oo

~o

Fig. 7. A Schematicillustration of the relationship between the amount of Ia inhibition (ordinate) and the amount of contraction (abscissa) based on the present data. The excitabilityof Ia inhibitorypathwaychanges continuouslyfrom dorsiflexionthrough rest to plantar fexion. B Neuronal connection concerning reciprocal Ia inhibition. For explanations see text. (E, excitation; I, inhibition; ct, ct motoneurone; y, ? motoneurone; Ia, Ia inhibitory interneurone; R, Renshaw cell)

time the absolute amount of Ia inhibition decreased almost in parallel with the amount of contraction. In one exceptional subject (indicated by an asterisk), the amount of inhibition increased during plantar flexion as well as dorsiflexion, but no co-contraction of pretibial muscles was demonstrated during plantar flexion.

Discussion

Ia Inhibition at Rest The Ia inhibition of extensor soleus motoneurones from the flexor nerve was revealed at rest in four subjects out of five in the present study. The relatively high incidence of Ia inhibition at rest is noteworthy, because it was previously reported to have been recognized only in a few normal subjects (Mizuno et al. 1971; Tanaka 1974, 1980). One factor in the high incidence of Ia inhibition in the present

study could be that the present experiment was done on subjects sitting in a chair, while those in the previous experiments were carried out in the prone position. In the latter situation the subject's knee was in an almost extended position, with the foot protruding freely from the edge of the bed and thus being pulled downwards somewhat by the force of gravity. This position naturally causes a certain stretching of the triceps surae muscle, which could result in facilitation of extensor Ia inhibitory interneurones via increased extensor Ia discharges and lower the excitability of the f e x o r Ia inhibitory pathway. Another factor to be considered is that all the subjects were young and athletic and showed mild to moderate hyperreflexia by tendon tap in the lower extremities; in some subjects the H-reflex was demonstrated even in pretibial muscles at rest. Tanaka (1974) discerned Ia inhibition on the ankle extensor motoneurones from the flexors at rest in one subject in whom the H-reflex was demonstrated in pretibial muscles, and he discussed the shift in

406

M. Shindoet al.: Changesin Reciprocal Ia InhibitionDuring VoluntaryContractionin Man accordance with the amount of contraction. This is the main finding disclosed in the present study.

equiliblium between the ankle flexor and extensor excitabilities to the flexor side. Since Ia inhibition at rest was recognized most clearly in the first session of experiment in the present study, the psychological factor should not be disregarded either. Although Ia inhibition was observed at rest in most subjects, the maximal amount of inhibition was small and the inhibition itself disappeared readily on plantar flexion and also tended to become less clear with repetition of experiments in all the subjects.

The Neural Mechanisms Controlling la Inhibitory Pathway

There are several known mechanisms which control the excitability of the Ia inhibitory pathway. Two possible mechanisms of increase in la inhibition during dorsiflexion can be considered; (1) facilitation of Ia inhibitory interneurone directly by descending routes and/or indirectly by increased activity of Ia afferents by way of the ~,-loop, and (2) disinhibition of Ia interneurone by inhibition of Renshaw-cell activity. Figure 7B shows the neuronal connections for these mechanisms. The first mechanism, which is properly described as the "ct-~/-linkage in reciprocal inhibition" (Hongo et al. 1969; Lundberg 1970), has been discussed in detail elsewhere (Tanaka 1974, 1976). In brief, the reciprocal Ia inhibitory interneurone is facilitated directly from supraspinal structures, e.g. the corticospinal tract in the monkey (Jankowska and Tanaka 1974; Jankowska et al. 1976). Such a descending facilitation was demonstrated indirectly in human experiments combined with voluntary movements (Simoyama and Tanaka 1974). Another source which facilitates Ia interneurones is group Ia afferents from homonymous and synergistic muscles. These afferents have been shown in human experiments to increase in discharge-frequency during isometric voluntary contraction (Vallbo 1970). Vallbo further showed that the frequency increased as the contraction intensified (Vallbo 1974). These observations would suggest the first mechanism to be the most likely one. Concerning the second possible mechanism, since the Renshaw cell has inhibitory connection to the Ia inhibitory interneurone (Hultborn et al. 1971), the latter's activity can be facilitated by suppression of the former. Although the strongest excitatory input so far known for the Renshaw cell is the axoncollateral of the ct motoneurone, the Renshaw cell is also under control of segmental or supraspinal centers (cf. Baldissera et al. 1981). Recently the excitability of the recurrent inhibitory pathway in man has been investigated during voluntary contraction with an indirect method (Hultborn and Pierrot-Deseilligny 1979). The investigators demonstrated that during tonic plantar flexion the excitability of Renshaw cells on the agonist side decreased progressively in 40%, 60% and 80% of the maximal contraction, and they suggested that this decrease would be due to an inhibitory control (spinal and/or suprasegmental) acting on Renshaw cells and would be

Quantitative Evaluation of la Inhibition at Rest and During Tonic Voluntary Contraction

Ia inhibition on soleus motoneurones was demonstrated during dorsiflexion of the foot in all subjects, which confirms the results in a previous report (Tanaka 1974). Furthermore, the present study disclosed a quantitative relationship between the amount of inhibition and the amount of voluntary contraction. The absolute amount of soleus motoneurones which were inhibited by the conditioning volley increased during dorsiflexion and decreased or disappeared during plantar flexion, in parallel with the amount of tonic voluntary contraction. As the conditioning effect on the H-reflex may vary according to the size of the test H-reflex (Desmedt 1973), it was imperative to examine whether a smaller H-reflex might reveal a larger Ia inhibition in the present study. There was no virtual increase of Ia inhibition at rest when tested on a smaller H-reflex (Fig. 3A, right-hand side). On the other hand, it was disclosed (Kuno 1959) in a relationship between the size of the test monosynaptic reflex and its recurrent inhibition in the spinal cat, that the amount of inhibition by the same recurrent volley increased as the test reflex increased its size up to 40% of the total motoneurone pool. The effect remained constant thereafter up to 65% of the test size (Fig. 2 in Kuno 1959). In the present study, the size of the test H-reflex was expressed as a percentage of the maximal M-response, which is the same principle as Kuno's method, and it was less than 60% of the maximal response. Although it is uncertain whether the relationship between the size of monosynaptic reflex and the amount of inhibition might be applied to the Ia inhibition in man or not, the present study showed that the amount of Ia inhibition increased in spite of a decrease in test H-reflex size during dorsiflexion, which is in the opposite direction from Kuno's results. We may conclude that the Ia inhibitory pathway from ankle flexor to extensor is facilitated during tonic voluntary dorsiflexion and is depressed during voluntary plantar flexion, in

M. Shindo et al.: Changes in Reciprocal Ia Inhibition During Voluntary Contraction in Man

407

favourable for reciprocal Ia inhibition (see Fig. 7B). However, they could not show such inhibition of Renshaw-cell activity during weak contraction (10%). Furthermore, it is unknown whether this is also the case in the flexor Renshaw cells during tonic dorsiflexion. The present study was performed on rather weak contraction between 5 and 30% of the maximal force, and no qualitative difference in the increase of Ia inhibition was observed in this range. Therefore, we tentatively suggest that the Renshawcell mechanism would not contribute much to facilitation of the Ia inhibitory pathway during weaker dorsiflexion as in the present study. Some contribution from the Renshaw-cell mechanism would be expected at a stronger contraction. The decrease in excitability of the Ia inhibitory pathway during tonic plantar flexion (Figs. 5 and 6) would beexplained most simply by the inhibition of Ia interneurones on the flexor side by Ia interneurones on the extensor side, which was facilitated in the same way as the flexor Ia interneurones during dorsiflexion described above (Fig. 7B). In the cat, Ia inhibitory interneurones receive disynaptic IPSP from Ia afferents from the antagonist (Hultborn et al. 1971), and this may also be true in man. Another possibility, which remains to be demonstrated, is that the Ia inhibitory interneurone is inhibited or disfacilitated directly by supraspinal centers during antagonist contraction.

and pathological conditions. This is a purposeful or even crucial neural mechanism for maintaining a standing position with the soles of the feet planted on the ground, and must have developed through a long course of phylogenesis as well as ontogenesis. For voluntary movement of the foot, however, this balance has to be altered with respect to the direction as well as the amount of movement. This alteration could be achieved at least partly by controlling the excitability of the reciprocal Ia inhibitory pathway in combination with agonist a motoneurones. The results of the present study may provide a physiological basis for this mechanism.
Acknowledgement. The authors wish to express their thanks to Mr.
P.E. Davenport for scrutinizing their English.

References
Baldissera F, Hultborn H, Illert M (1981) Integration in spinal neuronal systems. In: Brooks V (ed) Handbook of physiology, Section I: Nervous system II. American Physiological Society, Bethesda, pp 509-595 Creed RS, Denny-Brown D, Eccles JC, Liddel EGT, Sherrington CS (1932) Reflex activity of the spinal cord. Reprinted with annotations by Lloyd DPC (1972) Oxford University Press, London, pp 148-151 Desmedt JE (1973) A discussion of the methodology of the triceps surae T- and H-reflexes. In: Desmedt JE (ed) New developments in electromyography and clinical neurophysiology, vol 3. Karger, Basel, pp 773-780 l~cdes JC 0969) Th~ inhibitory pathway of the central nervous system. Liverpool University Press, Liverpool Hongo T, Jankowska E, Lundberg A (1969) The rubrospinal tract. II. Facilitation of interueuronal transmission in reflex paths to motoneurones. Exp Brain Res 7:365-391 Hultborn H, Jankowska E, Lindstrom S (1971) Recurrent inhibition from axon collaterals of transmission in the Ia inhibitory pathway to motoneurones. J Physiol (Lond) 215:591-612 Hultborn H (1972) Convergence on interneurones in the reciprocal Ia inhibitory pathway to motoneurones. Acta Physiol Scand [Suppl] 375:1--42 Hultborn H, Pierrot-Deseilligny E (1979) Changes in recurrent inhibition during voluntary soleus contractions in man studied by an H-reflex technique. J Physiol (Lond) 297: 229-251 Jankowska E, Tanaka R (1974) Neuronal mechanism of the disynaptic inhibition evoked in primate spinal motoneurones from the corticospinal tract. Brain Res 75:163-166 Jankowska E, Padel Y, Tanaka R (1976) Disynaptic inhibition of spinal motoneurones from the motor cortex in the monkey. J Physiol (Lond) 258:467-487 Kuno M (1959) Excitability following antidromic activation in spinal motoneurones supplying red muscles. J Physiol (Lond) 149:374-393 Lundberg A (1970) The excitatory control of the Ia inhibitory pathway. In: Anderson P, Jansen JKS (eds) Excitatory synaptic mechanisms. Universitetsforlaget, Oslo, pp 333-340 Mizuno Y, Tanaka R, Yanagisawa N (1971) Reciprocal group I inhibition of triceps surae motoneurones in man. J Neurophysiol 34:1010-1017

Functional Significance of Reciprocal Ia Inhibitory Pathway

Tanaka (1974, 1976) suggested a close parallelism between the excitability of agonist t~ motoneurones and the excitability of the Ia inhibitory pathway to the antagonist. He also discussed extensor predominance in the excitability balance between flexors and extensors of the foot from the results on Ia inhibition. The present result provides further support to this. In normal man extensor predominance exists, as the amount of Ia inhibition from flexor to extensor at rest was not so marked and there was no constant Hreflex in pretibial muscles. The presence of anklejerk and failure to provoke tendon-jerk in pretibial muscles are common findings in normal man to support the idea of extensor predominance in Iamediated monosynaptic reflex in leg muscles. With lesions in the central nervous system, this extensor predominance in the lower extremities develops markedly, as in capsular hemiplegia in man and decerebrate rigidity in animals (Creed et al. 1932). Thus the extensor predominance in the leg seems to be fundamental in neural connections in both normal

408

M. Shindo et al.: Changes in Reciprocal Ia Inhibition During Voluntary Contraction in Man Vallbo AB (1974) Human muscle spindle discharge during isometric voluntary contractions. Amplitude relations between spindle frequency and torque. Acta Physiol Scand 90:319-336 Yanagisawa N, Tanaka R, Ito Z (1976) Reciprocal Ia inhibition in spastic hemiplegia of man. Brain 99:555-574 Yanagisawa N, Tanaka R (1978) Reciprocal Ia inhibition in spastic paralysis in man. In: Cobb WA, Duijn H (eds) Contemporary clinical neurophysiology. Elsevier, Amsterdam (EEG Suppl, No 34, pp 521-526) Yanagisawa N (1980) Reciprocal reflex connections in motor disorders in man. In: Desmedt JE (ed) Spinal and supraspinal mechanisms of voluntary motor control and locomotion. Karger, Basel (Prog Clin Neurophysiol, vol 8, pp 129-141)

Simoyama M, Tanaka R (1974) Reciprical Ia inhibition at the onset of voluntary movements in man. Brain Res 82:334--337 Tanaka R (1974) Reciprocal Ia inhibition during voluntary movements in man. Exp Brain Res 21:529--540 Tanaka R (1976) Reciprocal Ia inhibition and voluntary movements in man. In: Homma S (ed) Understanding of the stretch reflex. Elsevier, Amsterdam Oxford New York (Prog Brain Res, vol 44, pp 291-302) Tanaka R (1980) Inhibitory mechanism in reciprocal innervation in voluntary movements. In: Desmedt JE (ed) Spinal and supraspinal mechanisms of voluntary motor control and locomotion. Karger, Basel (Prog Clin Neurophysiol, vol 8, pp 117-128) Vallbo AB (1970) Discharge patterns in human muscle spindle afferents during isometric voluntary contractions. Acta Physiol Scand 80:552-566

Received February 22, 1983