The Journal of Neurophysiology Vol. 78 No. 6 December 1997, pp. 2975-2984
Copyright ©1997 by the American Physiological Society

Stretch of Quadriceps Inhibits the Soleus H Reflex During Locomotion in Decerebrate Cats

John E. Misiaszek and Keir G. Pearson

Department of Physiology, University of Alberta, Edmonton, Alberta T6G 2H1, Canada

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Misiaszek, John E. and Keir G. Pearson. Stretch of quadriceps inhibits the soleus H reflex during locomotion in decerebrate cats. J. Neurophysiol. 78: 2975-2984, 1997. Previously, it has been demonstrated that afferent signals from the quadriceps muscles can suppress H reflexes in humans during passive movements of the leg. To establish whether afferent input from quadriceps contributes to the modulation of the soleus H reflex during locomotion, the soleus H reflex was conditioned with stretches of the quadriceps muscle during bouts of spontaneous treadmill locomotion in decerebrate cats. We hypothesized that 1) in the absence of locomotion such conditioning would lead to suppression of the soleus H reflex and 2) this would be retained during periods of locomotor activity. In the absence of locomotion, slow sinusoidal stretches (0.2 Hz, 8 mm) of quadriceps cyclically modulated the amplitude of the soleus H reflex. The H reflex amplitude was least during the lengthening of the quadriceps and greatest as quadriceps shortened. Further, low-amplitude vibrations (48-78 µm) applied to the patellar tendon suppressed the reflex, indicating that the muscle spindle primaries were the receptor eliciting the effect. During bouts of locomotion, ramp stretches of quadriceps were applied during the extensor phase of the locomotor rhythm. Soleus H reflexes sampled at two points during the stance phase were reduced compared with phase-matched controls. The background level of the soleus electromyographic activity was not influenced by the applied stretches to quadriceps, either during locomotion or in the absence of locomotion. This indicates that the excitability of the soleus motoneuron pool was not influenced by the stretching of quadriceps, and that the inhibition of the soleus H reflex is due to presynaptic inhibition. We conclude that group Ia afferent feedback from quadriceps contributes to the regulation of the soleus H reflex during the stance phase of locomotion in decerebrate cats. This afferent mediated source of regulation of the H reflex, or monosynaptic stretch reflex, would allow for rapid alterations in reflex gain according to the dynamic needs of the animal. During early stance, this source of regulation might suppress the soleus stretch reflex to allow adequate yielding at the ankle and facilitate the movement of the body over the foot.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

It is now generally recognized that reflexes can be modified in a task-dependent manner, and during the execution of a specific task (Rossignol 1996). The modification of a reflex response can be in the form of a change in strength (Brooke et al. 1997) or a change in sign (Pearson 1995a). The dynamic regulation of reflexes is clearly an important aspect of motor control and sensorimotor integration.

Recently, the modulation of the H reflex amplitude during locomotion has received much attention. The amplitude of the H reflex in ankle extensors modulates over a step cycle, being large during stance and reduced during swing. This has been demonstrated in humans (Brooke et al. 1991; Capaday and Stein 1986, 1987; Crenna and Frigo 1987) and cats (Akazawa et al. 1982; Duenas et al. 1990). There are two mechanisms by which the amplitude of this reflex can be modified: 1) variations in the excitability of the motoneurons and 2) alterations in the amount of transmitter released by the afferent terminals as a result of presynaptic inhibition. Although much of the modulation of reflex amplitude can be attributed to the normal ebb and flow of motoneuron activity associated with locomotion (Shefchyk et al. 1984), the amplitude of the reflex does not always parallel the background recruitment level (Capaday and Stein 1987; Duenas et al. 1990). This has led to the conclusion that presynaptic inhibition is of importance in the regulation of the strength of the reflex during locomotion.

The origins of the presynaptic inhibition are not yet clear. One possibility is that primary afferent depolarization (PAD) generated by activation of the central rhythm generator for locomotion (commonly referred to as the "CPG") produces presynaptic inhibition in large diameter muscle afferents (Stein and Capaday 1988). During fictive locomotion the PAD in primary afferents is largest during flexor activity (swing phase) with smaller depolarizations occurring during extensor activity (stance phase) (Gossard et al. 1991). However, Gossard (1996) recently demonstrated that the PAD generated by the CPG does not influence the magnitude of the monosynaptic excitatory postsynaptic potential (EPSP), indicating that the CPG is not the source of the presynaptic inhibition modulating the H reflex during walking.

Alternatively, the source of the presynaptic inhibition regulating the H reflex might arise from peripheral afferent feedback. In humans, the amplitude of the soleus H reflex is phasically modulated during passive-pedaling movements of the legs (McIlroy et al. 1992; Misiaszek et al. 1995b) or segments of the legs (Brooke et al. 1993). During such manipulations the soleus H reflex is depressed throughout the movement cycle, but is markedly more reduced when the leg is being flexed. When the excitability of the soleus motoneuron pool is stabilized with a tonic contraction of the soleus muscle, this pattern of inhibition is retained (Misiaszek et al. 1995b). This suggests that the afferent feedback associated with the passive movement of the legs leads to suppression of the soleus H reflex via presynaptic inhibition.

One source of the afferent signals is probably stretch-sensitive receptors of the extensor muscles of the leg. First, the soleus H reflex is most depressed during the flexion phase of the movement when extensors are stretching. When the movement is slow the depression of the H reflex becomes isolated to the flexion phase of the movement (Cheng et al. 1995a; Misiaszek et al. 1995b). Second, increasing the rate of passive-pedaling movement further suppresses the soleus H reflex (McIlroy et al. 1992; Misiaszek et al. 1995b). This rate-dependent influence on the amplitude of the H reflex is proportioned to the estimated rate of stretch of the extensor muscles of the leg (Cheng et al. 1995a). Third, soleus H reflexes conditioned with a patellar tendon tap are substantially depressed (Cheng et al. 1995b).

These findings implicating the stretch-sensitive receptors in extensors as the source of the soleus H reflex depression in humans were supported by observations in the dog. Misiaszek et al. (1995a) found that an H reflex in a small muscle of the foot of the dog was depressed by passive rotation about the knee. This movement-induced attenuation of the H reflex was retained after the deactivation of the cutaneous receptors and joint receptors of the knee, but was abolished when the quadriceps muscles were tenotomized. Thus it appears that afferent feedback arising from the stretching of quadriceps muscles can lead to the attenuation of H reflexes in ankle and foot muscles during passive movements of the legs in both humans and dogs.

Additional evidence that extensor muscle afferents can suppress the monosynaptic reflex in ankle extensor muscles comes from early observations on presynaptic inhibition in the cat. Eccles et al. (1962) demonstrated in decerebrate or spinal, anesthetized cats that, as a general rule, stimuli applied to extensor muscle afferents had little effect on monosynaptic reflex (MSR) amplitudes. The one exception found in that study was when stimuli were applied to the quadriceps nerve. However, subsequently Decandia et al. (1967) demonstrated that stimulation of the lateral gastrocnemius-soleus nerve produced substantial presynaptic inhibition of the MSR of medial gastrocnemius. This result was later confirmed by Barnes and Pompeiano (1970b), who found that vibration of lateral gastrocnemius-soleus also produced a depression of the MSR of medial gastrocnemius, by presynaptic inhibition. Therefore there is currently considerable evidence that the muscle afferents from extensors can lead to suppression of the H reflex and MSR in ankle extensors in nonlocomoting animals.

An unanswered question is as follows. Does this muscle afferent-induced presynaptic inhibition of the Ia monosynaptic reflex contribute to the modulation of the H reflex (and MSR) during locomotion? The purpose of the present study was to address this question by comparing H reflex amplitudes in the decerebrate walking cat with and without imposed stretches of the quadriceps muscle. It was hypothesized that stretching quadriceps during the extension phase of the step cycle would lead to suppression of the soleus H reflex. We also examined the effects of stretching the quadriceps muscles on the soleus H reflex in the absence of locomotion. This was done to ensure that the passive stretching of quadriceps led to qualitatively similar effects in the decerebrate cat as has been demonstrated for the anesthetized dog or the intact human, which would suggest that similar mechanisms are involved. Further, we attempted to establish whether input from group Ia afferents from quadriceps are involved in inhibiting the soleus H reflex by vibrating the quadriceps muscle.

	METHODS

Abstract Introduction Methods Results Discussion References

Six mature cats (2.2-3.5 kg), of either sex, were used in these experiments. The procedures were approved by the University of Alberta Health Sciences Laboratory Animal Welfare Committee. Under halothane anesthetic (delivered with 95% O₂-5% CO₂), the trachea of each animal was cannulated for continued administration of the anesthetic. One of the carotid arteries was ligated distally and cannulated proximally to monitor blood pressure. The other carotid artery was ligated to reduce the blood loss during decerebration. A cannula was also inserted into a jugular vein for the administration of fluids and drugs. Maintenance of the animal's body temperature was assisted by the use of a heating pad during the surgical procedures.

After the initial procedures, the right hind leg was extensively dennervated by transecting the following nerves: saphenous, obturator, sartorius, hamstrings, medial gastrocnemius, lateral gastrocnemius, distal tibial, and common peroneal. A bipolar cuff electrode was placed around the tibial nerve, at the popliteal fossa, to act as a stimulating electrode. A similar electrode was placed around the sciatic nerve to act as a recording electrode and occasionally, as a stimulating electrode. The patellar tendon was cut just proximal to its tibial insertion and freed of tissue to the patella. A hole was drilled into the patella to allow attachment of the patella to a muscle puller with heavy surgical silk (no. 4 silk, Surgicos). A pair of Teflon-coated, stainless steel wires (Cooner Wire, AS632), bared at the ends, were inserted into the belly of soleus with a longitudinal separation of ~5 mm to record the electromyographic (EMG) activity of soleus.

The animal was placed above a treadmill with the use of a stereotaxic device to secure the head. The right hind leg of the animal was immobilized with the use of ankle and knee clamps. Immobilization of the hip was achieved by fixing the position of the pelvis by passing a heavy wire through both iliac crests and a centrally placed steel flange. This apparatus was then secured within a pool of dental acrylic. The flange was then attached to the mounting frame with the use of a custom clamp. Each joint was fixed at ~90°. A schematic of the experimental setup is shown in Fig. 1A. The length of quadriceps was set so that at the minimum length the muscle remained just taught; this was referred to as "0" length. A temperature probe placed subcutaneously beside soleus was used to monitor the temperature of the limb, and radiant heat was applied as required to maintain a temperature of 37°C.

View larger version (20K): [in this window] [in a new window]   FIG. 1. Schematic of the experimental setup for the fixed limb preparation in a decerebrate walking cat. A: premammillary decerebrated cats walked spontaneously on a treadmill with 3 limbs. The test limb was fixed and extensively dennervated. B: during bouts of locomotion, the patellar tendon was stretched with a ramp stretch 20 ms after the onset of the soleus electromyographic (EMG) burst. H reflexes were elicited at 2 points within stretch, during the plateau of the ramp and hold stretch and during the rise phase of the ramp. Phase-matched control reflexes were interdigitated with the reflexes sampled during the perturbed steps.

Once the animal was secured above the treadmill, a decerebration was performed by first removing the cortical tissue and then transecting the brain stem at a 50° angle from the rostral edge of the superior colliculi. The anesthetic was discontinued at this time. A bolus of 2-5 ml of Dextran was routinely administered shortly after decerebration. Within 1 h of the decerebration, the animals produced bouts of spontaneous walking on the moving treadmill for up to 4 h. During the bouts of walking, the EMG activity of the soleus of the fixed right hind limb was rhythmically coordinated with the movements of the three free limbs.

Soleus H reflexes were elicited by delivering a 0.2-ms square-wave pulse (Grass S88 stimulator) to the tibial nerve. The strength of the applied stimulus was measured in terms of multiples of the first detection of the afferent volley (multiples of threshold, times threshold) recorded in the sciatic nerve. Stimuli ranged in intensity from 1.2 to 2 times threshold, but were consistent within a block of data. The stimulus threshold was periodically checked throughout an experiment to ensure stability of the stimulus strength. The sciatic volley was analyzed post hoc to assure consistency of the applied stimulus.

Stretch of quadriceps in the absence of locomotion

During periods in which the animal was not walking, two forms of stretch were applied to the quadriceps muscles: 1) slow sinusoidal stretches (0.1 or 0.2 Hz with a peak-to-peak amplitude of 8 mm) and 2) vibrations (150 Hz, with amplitudes ranging from 48 to 78 µm). Four of the animals were used in the first paradigm, and three in the second. The quadriceps muscle group was stretched or vibrated by a puller that was controlled by computer (IBM compatible) with custom software. H reflexes were evoked at a constant rate ranging between 0.5 and 0.7 Hz to allow for sampling throughout the cycle of rhythmic stretching and at various points during and after the vibration. Data were obtained for various levels of tonic background EMG activity. Different levels of activity were achieved either spontaneously or by applying gentle perineal stimulation.

Stretch of quadriceps during locomotion

During bouts of spontaneous locomotion, an 8-mm ramp stretch of quadriceps was periodically delivered during periods of extensor activity, as indicated by the soleus EMG. The duration of the rise and plateaus of the applied stretches was varied to approximate the duration of the extensor burst. The onset of the stretch typically occurred 20 ms after the onset of the soleus EMG burst. The rectified soleus burst was sampled by the controlling computer and used to activate the puller. Stretches of quadriceps were applied during every second soleus burst such that a control step was achieved between each trial. A typical section of data (Fig. 1B) displays the rectified and filtered soleus EMG (top trace) with the quadriceps length (middle trace) and a stimulus marker (bottom trace).

H reflexes were sampled for both the control steps and the perturbed steps at two points within the duration of the extensor burst. One point corresponded with the midpoint of the rise phase of the ramp stretch of quadriceps, whereas the other point corresponded with the plateau of the stretch, 100 ms after reaching the final length (see Fig. 1B). Three animals were used in this paradigm.

Data acquisition and analysis

The soleus EMG and sciatic electroneurogram (ENG) were amplified (30-10,000 Hz bandpass, Grass P511) before being recorded to magnetic tape (VHS) using a Vetter 4000A PCM recording unit. Additionally, the length output from the muscle puller and a stimulus marker were stored. Sections of data were later digitized at a sampling rate of 3 kHz (10 kHz for afferent volley data) and stored to disk using an Axotape data acquisition system (Axon Instruments). At this time, both the raw EMG and a rectified and filtered version were stored.

With the use of custom software, the peak-to-peak amplitude of the H reflex was measured. H reflexes obtained during the sinusoidal stretching of quadriceps in the absence of locomotion were grouped into eight bins equally spaced across the cycle of stretching. The amplitudes of at least 10 individual samples were recorded for each bin, from each animal. For the vibration paradigm, the reflex amplitude was measured along with the latency from the onset of the vibration that the reflex was evoked.

Data from the locomotion paradigm were grouped and averaged based on whether a stretch was applied or not and when in the extensor burst the stimulus was delivered. A minimum of 20 samples, and often more, was collected for each condition for each animal. The average prestimulus, rectified EMG was measured to ensure that the application of the stretch did not alter the excitability of the soleus motoneuron pool.

Detection of significant differences between conditions was achieved using a one-way analysis of variance (ANOVA) for data arising from a single animal or a two-way ANOVA for pooled data. Prior-planned t-tests were then employed to describe any detected differences. In all instances, differences were deemed significant if P < 0.05.

	RESULTS

Abstract Introduction Methods Results Discussion References

Attenuation of soleus H reflexes in the nonwalking cat

In the nonwalking decerebrate cat, slow sinusoidal stretches applied to the quadriceps tendon produced substantial modulation of the peak-to-peak amplitude of the soleus H reflex. This is illustrated in Fig. 2 for stretches of quadriceps of ~0.2 Hz. The middle trace in Fig. 2A shows a series of H reflexes sampled during the sinusoidal stretching of quadriceps, with quadriceps length depicted in the top trace. The modulation of the reflex is clear in the raw data, with reflexes being most suppressed when the quadriceps muscles are lengthening and greatest when the muscles are shortening. The raw data were grouped into eight bins according to the phase of the cycle of quadriceps stretching, beginning at the peak of the stretch. Sample average EMG traces, for each bin, are depicted in Fig. 2B (left). The corresponding neurographic data, illustrating the afferent volley recorded in the sciatic nerve, are shown in Fig. 2B (right). The clear modulation in reflex amplitude was not reflected in the amplitude of the afferent volley, indicating that the stretching of the quadriceps was not compromising the ability to stimulate the afferents of the tibial nerve. Therefore these effects were not arising from variations in applied stimuli. In Fig. 2C the peak-to-peak measures of the average traces above are depicted with their standard errors highlighting the consistency with which these observations were obtained within an animal. The data from this animal displayed a significant phase modulation of the amplitude of the H reflex, whereas the amplitude of the afferent volley was not affected by the stretching. These results were consistent between animals. In all four animals used in this paradigm, the pattern of H reflex modulation was similar. An analysis of the pooled data again revealed a significant phase modulation. A similar pooled analysis of the afferent volley amplitude was not possible because technical problems prevented the post hoc analysis of the ENG from two of the animals used in this protocol.

View larger version (31K): [in this window] [in a new window]   FIG. 2. Slow sinusoidal stretches of quadriceps modulate the amplitude of the soleus H reflex. A: quadriceps length (top trace) and raw soleus EMG (bottom trace) traces during sinusoidal stretching of quadriceps. The amplitude of the soleus H reflex modulates despite the presence of a sustained tonic contraction. B: average soleus EMG (left) and sciatic electroneurogram (right) traces for each of 8 bins across the cycle of stretching. Bin 1 begins at the left arrow in A and bin 8 ends at the right arrow. The H reflex modulates, despite the stable afferent volley amplitude. For amplitude reference, see C. C: histograms depicting the average peak-to-peak amplitudes for the data shown in B, but also illustrating the standard errors for each average. Each bar is the average of at least 10 samples. All data depicted are from 1 animal.

It is also important to note that the slow sinusoidal stretching of the quadriceps did not affect the amplitude of the tonic EMG activity in soleus. In Fig. 3A, rectified and filtered soleus EMG activity (middle trace of each section) is shown for a 60-s period; the bottom section is the direct continuation of the top section. The tonic contraction was initiated by gentle perineal stimulation that lasted ~3 s. In the top section, there were no stimuli delivered to the tibial nerve, and it can be seen that the slow stretching of quadriceps (0.1 Hz for these traces) did not noticeably alter the tonic EMG level in a phase-related manner. In the bottom section, the tibial nerve stimulation was introduced. Although the tonic contraction level of soleus did not modulate, the H reflex is clearly modulated by the slow sinusoidal stretch of quadriceps. This stability of the tonic contraction level of soleus was critically evaluated for the data that were collected during the 0.2-Hz slow sinusoidal stretches of quadriceps. The rectified soleus EMG was averaged for the 30 ms before the delivery of each tibial nerve stimulus. The average EMG was then standardized to the maximal contraction recorded in that animal (%max). The pooled data from four animals, for each of the eight bins, are depicted in Fig. 3B. No significant effect due to the slow stretching of quadriceps was detected for this pooled data.

View larger version (26K): [in this window] [in a new window]   FIG. 3. Slow sinusoidal stretches of quadriceps do not modulate the level of ongoing tonic contraction in soleus. A: 60-s length of data depicting the rectified and filtered soleus EMG (middle trace) during an 8-mm, 0.1-Hz sinusoidal stretch of quadriceps (top trace). The tibial nerve stimulation was introduced in the bottom section, and it can be clearly seen that the H reflex is phasically modulated. The bottom section is the direct continuation of the data of the top section. B: average prestimulus EMG, pooled across animals used in the analysis of data for the 0.2-Hz sinusoidal stretch of quadriceps. Data were standardized to the maximal contraction level achieved in each animal. Error bars indicate 1 SE.

In three cats, soleus H reflexes were conditioned by applying brief vibrations to the patellar tendon. Vibrations ranging from 48 to 78 µm in amplitude, with durations between 550 and 900 ms, were routinely used. Sample data arising from one cat are shown in Fig. 4. The top panels of Fig. 4, A and B, depict EMG data, with each spike in the top trace depicting an H reflex. During the periods of patellar vibration, the amplitude of the soleus H reflex is dramatically suppressed. This is true for vibrations of 78 µm as well as lower amplitude vibrations of 48 µm. In the middle panels, the peak-to-peak amplitudes for each individual H reflex have been plotted against the time after vibration onset. The shaded bar indicates when the vibration was applied. It is obvious in both instances that with patellar tendon vibration the soleus H reflex is quickly attenuated, remains attenuated for the duration of the vibration, and then slowly returns to control levels.

View larger version (25K): [in this window] [in a new window]   FIG. 4. Vibration of the quadriceps tendon attenuates the soleus H reflex. Data from 1 cat are depicted for 2 amplitudes of applied vibration. A: the amplitude of vibration was 78 µm, and the tibial nerve stimulus was at M threshold. B: the amplitude of vibration was 48 µm, and the tibial nerve stimulus was below M threshold. Top panel of each depicts raw data, with the soleus EMG in the top trace and the quadriceps length in the bottom trace. The peak-to-peak amplitude for each H reflex sampled are depicted in the middle panels, plotted against the time following the onset of the vibration. Shaded horizontal bar indicates the time that the vibration was applied. Notice that, after the termination of the vibration, the return to control levels of the reflex amplitude was slow and could be described by an exponential equation. In the bottom panel of B, the data following the offset of the vibration were bin averaged (50-ms bins) and expressed as a percentage of inhibition. For this data set the time constant was estimated to be 500 ms. In contrast, the afferent volley was not affected by vibration of the patellar tendon as depicted in the bottom panel of A.

This slow recovery of reflex amplitude could be described by fitting an exponential curve to the data. In the bottom panel of Fig. 4B, the data of the above panel, after the termination of vibration, were grouped into 50-ms bins and expressed as a percent inhibition (where 0% would indicate control amplitudes and 100% would indicate that the reflex could not be measured). For this particular data set, the time constant for the exponential curve was estimated to be ~500 ms. For the data set in Fig. 4A, the time constant was also ~500 ms. Although there did not appear to be any clear influence of vibration duration on the length of the time constant, we did not have sufficient data sets to adequately investigate this possibility. In addition, none of the data sets had a sufficient number of samples early in the vibration to estimate the timing of the onset of the inhibition. However, in all instances, reflexes sampled 60 ms after the onset of the vibration or later were below the 95% confidence band for the control values (the average of reflexes sampled up to 2,000 ms before the onset of vibration).

The data in Fig. 2, and those used in the analysis across animals for the influence of slow sinusoidal stretching of quadriceps, were obtained with stimuli that were subthreshold for the M wave. Originally, a stimulus intensity that produced a small, stable M wave was believed to be best suited for the needs of the present experiment. Such a stimulus intensity is common in the literature and is useful in allowing the easy on-line monitoring of stimulus intensity during the experiment. However, in our initial experiments (2 animals) we observed that when such a stimulus was used the modulation of the soleus H reflex was progressively diminished with increasing background levels of EMG. One possible explanation is that the command that produced the increasing EMG level might also alter the inhibitory pathways producing the modulation. Another is that at the higher level of motoneuron pool excitability the modulated reflex input might always be sufficiently strong to produce a near maximal reflex EMG. In one animal, we systematically addressed this issue by varying the stimulus intensity used to evoke the H reflex in soleus at different background levels of EMG.

The results are summarized in Fig. 5 and arise from the same animal from which data were presented in Fig. 2. In Fig. 5, A and B, reflexes were evoked with a stimulus that was above M threshold and elicited a maximal H reflex in the quiescent animal. At this stimulus intensity, clear modulation of the soleus H reflex could be observed when there was no detectable EMG activity in the muscle (Fig. 5A). However, when the tonic background level of EMG was high, roughly equivalent to the level produced during locomotion, the modulation of the reflex was absent (Fig. 5B). In contrast, when the stimulus intensity used to evoke the reflexes was below M threshold, the stretch-induced modulation of the reflex occurred, even when the tonic background EMG activity was high (Fig. 5D). This suggests that the lack of modulation for the combination of conditions depicted in Fig. 5B likely arises from the second of the possibilities described above. That is, the afferent input that elicits the reflex is modulated, but because of the high excitability of the motoneuron pool, high-intensity stimuli are always sufficient to elicit a maximal response. Consequently, the experiments involving the testing of soleus H reflex amplitude during locomotion were always performed with stimuli subthreshold for the M wave.

View larger version (31K): [in this window] [in a new window]   FIG. 5. Large amplitude H reflexes do not modulate with high levels of background EMG. For each histogram the data are the average peak-to-peak amplitude of the soleus H reflex recorded in 1 cat (the same animal as Fig. 2). A and B: these data are derived from reflexes elicited with stimuli suprathreshold of the M wave. In A, there was no detectable EMG activity in soleus when the reflexes were evoked. In contrast, the data in B were gathered during sustained high levels of background EMG activity. C and D: these data are derived from reflexes elicited with stimuli below M threshold. Notice that in D, when the background EMG is high, the H reflex is clearly modulated by sinusoidal stretching of quadriceps. The "high" level of EMG activity was similar for the 2 stimulus intensities depicted, and represents activity close to the level achieved during locomotion for this cat. Each bar is the average of at least 10 samples, and the error bars depict 1 SE.

Attenuation of soleus H reflexes with quadriceps stretch during locomotion

The contribution of afferent feedback from quadriceps in regulating the amplitude of the soleus H reflex during locomotion was assessed in three animals. A ramp stretch (8 mm) applied to the quadriceps reduced the amplitude of the soleus H reflex (Fig. 6). This was observed at two points within the extensor phase of locomotion (indicated by the soleus EMG burst). In Fig. 6A, data from one animal are illustrated. The raw soleus EMG is depicted in the top trace. The H reflexes elicited in four successive steps are shown above the complete trace on an expanded time scale and amplified. H reflexes elicited during the ramp stretch of quadriceps, or soon after reaching the plateau of the stretch, were reduced compared with phase-matched controls. The effect of stretching quadriceps was consistent within an animal as well as between animals. This is summarized in Fig. 6B, which depicts the mean peak-to-peak amplitude of the H reflex averaged across animals. The data were standardized as a percentage of control for each time point, for each animal. Analysis of the pooled data revealed that stretching quadriceps significantly attenuated the soleus H reflex. Further, both the reflexes sampled during the rise phase and the plateau phase of the applied stretch were significantly lower than phase-matched controls.

View larger version (13K): [in this window] [in a new window]   View larger version (11K): [in this window] [in a new window]   FIG. 6. Stretching quadriceps inhibits soleus H reflexes during locomotion. A: a section of raw data depicting a number of steps including 1 of each test and control step. The traces are as follows: top trace, raw soleus EMG; middle trace, quadriceps length; bottom trace, stimulus marker. The raw soleus EMG traces have been expanded (above the raw EMG trace) to allow the visualization of the H reflex from 4 successive steps. Reflexes sampled during the applied quadriceps stretch were depressed, for both points tested. B: mean peak-to-peak H reflex amplitudes, averaged across animals and expressed as a percentage of the phase-matched control. Error bars indicate 1 SE.

The amplitude of the afferent volley recorded in the sciatic nerve was not affected by the applied stretch to quadriceps (Fig. 7A). Therefore it can be assumed that the stimulus applied to the tibial nerve was consistent between conditions and cannot account for the suppression of the soleus H reflex. Another important observation was that stretching of quadriceps had no obvious effect on the EMG activity level of the soleus muscle during locomotion (Fig. 7B). The left panel of Fig. 7B shows the average rectified EMG traces from one cat in the absence of tibial nerve stimulation during stretches of quadriceps (dark trace) and in the absence of stretch (light trace). It is also apparent in Figs. 2A and 6A that stretching quadriceps during the step cycle did not affect the EMG activity of soleus. This observation is more critically evaluated in the right panel of Fig. 7B. In this panel, the mean background EMG level in the 30 ms before tibial nerve stimulation is depicted for each condition, averaged across cats. No significant effect on the level of EMG activity, due to stretch of the quadriceps, could be detected. This indicates that the stretching of quadriceps during locomotion does not influence the excitability of the soleus motoneuron pool. This result was somewhat surprising as Guertin et al. (1995) demonstrated that activation of quadriceps group I afferents leads to polysynaptic excitation of the ankle extensors during fictive locomotion in cats. However, we consistently failed to observe this effect in decerebrate walking preparations (see also Hiebert 1997; Whelan et al. 1995). It is likely that the different preparations (decerebrate walking vs. fictive locomotion) can account for this discrepancy.

View larger version (22K): [in this window] [in a new window]   FIG. 7. A: average sciatic nerve recordings from 1 animal are depicted in the left panel. Each trace is 6 ms in duration and is the average of at least 20 samples. Right panel depicts the mean peak-to-peak amplitude of the afferent volley averaged across animals, expressed as a percentage of the phase-matched control. B: the stretching of quadriceps did not noticeably alter the soleus EMG burst pattern in the walking cat. In the left panel, average (n = 10) rectified and integrated soleus EMG traces are shown for control steps (top trace) and perturbed steps (middle trace). The quadriceps length trace is shown in the bottom trace. In the right panel, the mean prestimulus soleus EMG levels, averaged across animals, are depicted with standard errors.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

Previous studies have demonstrated that passive movements of the legs leads to suppression of H reflexes in humans (Brooke et al. 1993; Cheng et al. 1995a; Collins et al. 1993; McIlroy et al. 1992; Misiaszek et al. 1995b) and dogs (Misiaszek et al. 1995a, 1996). From these studies, it was concluded that afferent feedback associated with the stretching of the quadriceps muscles significantly contributed to the attenuation of the reflex. In the present study, we demonstrate that the soleus H reflex in the cat is similarly attenuated by stretching of quadriceps, and importantly, that this source of H reflex suppression remains during locomotion.

In cats, as in humans, the amplitude of the soleus H reflex is strongly modulated during locomotion (Akazawa et al. 1982). Although much of the pattern of reflex modulation follows the pattern of motoneuron pool activity, the changes in reflex amplitude cannot be completely explained by this mechanism. Indeed, in humans, the soleus H reflex is generally smaller during running when compared with walking for similar background EMG levels (Capaday and Stein 1987). It was suggested that an important mechanism in reflex control is presynaptic inhibition of the afferent terminals mediating the reflex. From the results of the present study, it is apparent that afferent feedback is one source from which this presynaptic inhibition can be initiated during locomotion in cats. Afferent discharge, introduced by the stretching of the quadriceps muscle, led to a reduction of the soleus H reflex amplitude during locomotion (Fig. 6). Furthermore, this peripheral source of reflex regulation is likely mediated via presynaptic mechanisms because the background level of soleus EMG was not influenced by the stretching of quadriceps (Fig. 7B).

It is unlikely that these results arise from postactivation (homosynaptic) depression (Hultborn et al. 1996). In the present experiments, the soleus muscle is held isometric, and the contraction levels are similar between conditions. Thus one can assume that the spindle discharge of the soleus muscle is similar between conditions. Therefore any influence due to postactivation depression is likely to be similar in all instances.

Further evidence implicating presynaptic inhibition is the slow recovery of the soleus H reflex following high-frequency vibration of the quadriceps tendon (Fig. 4). Vibration depressed the soleus H reflex, and the average time constant of recovery from depression was ~500 ms. This slow recovery is similar in time course of recovery of monosynaptic Ia EPSPs in medial gastrocnemius motoneurons following electrical stimulation of heteronymous group I afferents in cats (Schmidt 1971). This long time course of the inhibition is unlikely to arise from postsynaptic events because postsynaptic inhibition does not persist beyond ~30 ms (Eccles et al. 1962). A similarly long-lasting reduction in monosynaptic Ia EPSP amplitude in gastrocnemius-soleus motoneurons has been produced by vibration of flexors or extensors in cats (Barnes and Pompeiano 1970a,b). In these latter studies, the reduction of the EPSP amplitude was associated with the generation of primary afferent depolarization in the afferents of the gastrocnemius-soleus nerve. Thus, although in the present study there was no direct evidence that the vibration of quadriceps led to presynaptic inhibition of the soleus H reflex, the similarity in the time course of H reflex recovery in our study with the time course of recovery of EPSP amplitudes described in other studies suggests that similar mechanisms were involved, namely presynaptic inhibition.

The demonstration that slow sinusoidal stretches of quadriceps phasically modulated the amplitude of the soleus H reflex suggests that the inhibition is initiated by activation of quadriceps group Ia afferents. These slow stretches, when applied in the absence of background force, are not likely to recruit Golgi tendon organs. Further, in Figs. 2 and 3 it is apparent that the modulation of the soleus H reflex was phase advanced with the quadriceps length. This phase advance in the pattern of modulation suggests that the muscle spindle primary endings, responding to the dynamic stretch of the muscle, are responsible for initiating the inhibition. Such a phase advance would not be expected to be generated by the secondary endings of the muscle spindles, which respond more to the instantaneous length of the muscle. In addition, vibration of the quadriceps muscles produced a dramatic depression of the soleus H reflex. Small-amplitude vibrations, such as those used in the present experiment, selectively activate only primary spindle endings of the muscles vibrated (Brown et al. 1967). From our results we cannot be certain that the inhibition produced by the vibration did not arise as a result of spread of the vibration to soleus itself. However, the results arising from the slow sinusoidal stretching of quadriceps would support an argument against such an effect. Together, these results suggest that the peripheral afferent source of the reflex inhibition is the primary spindle ending of the quadriceps muscles.

One criticism of the present study might be that the Ia afferents from quadriceps exhibit a considerable amount of activity during the stance phase of walking (Loeb et al. 1985) and that the application of an 8-mm stretch would produce abnormally high discharge rates. This certainly would be true of a freely walking preparation. However, in the present study the test limb of the decerebrate animals was fixed and the quadriceps muscle held at isometric length for the control steps. Therefore it is more likely that when the quadriceps contracted during the stance phase the spindles were unloaded and the discharge rate fell (Hiebert 1997; Taylor et al. 1985). Application of the stretch would then be expected to activate the Ia afferents. It is unclear whether the application of the stretch would produce a discharge rate greater, equal, or less than the discharge rate normally experienced during walking or perturbations. Regardless, it is clear that increasing the discharge rate of the quadriceps Ia afferents decreases the amplitude of the soleus H reflex during locomotion.

It is well established that there are heteronymous monosynaptic projections from the quadriceps Ia afferents to triceps surae motoneurons (Eccles et al. 1962; Hultborn et al. 1987). This heteronymous monosynaptic projection would be expected to influence the results of the present study in two ways. The stretching of quadriceps would be expected to 1) increase the EMG activity of soleus and consequently 2) facilitate the soleus H reflex. We cannot speak to the second of these potential influences. In cats (Eccles et al. 1962; Hultborn et al. 1987) and humans (Cheng et al. 1995b) the facilitation of the soleus H reflex produced by stimulation of quadriceps group I afferents is short lived and soon followed by a prolonged inhibition. In the present experiment soleus H reflexes were always sampled well after the initiation of the inhibition was expected to occur. Thus any facilitatory effect on the soleus H reflex was likely masked by the potent presynaptic inhibition also generated by this afferent source. If we were able to sample more soleus H reflexes immediately after the onset of the vibration used in the present study, it is likely that this heteronymous facilitation would have been witnessed. Unfortunately this was not the case.

It was somewhat surprising that stretching of quadriceps did not alter the soleus EMG activity in either walking or nonwalking preparations. Previous studies have described a monosynaptic excitatory pathway from quadriceps Ia afferents to soleus motoneurons (Eccles et al. 1957). Why we did not see this facilitation of the soleus EMG is not clear. One possibility is that this heteronymous monosynaptic connection from quadriceps to soleus is itself presynaptically inhibited either in response to the stretching or due to the activation of the CPG. In humans and cats, Hultborn et al. (1987) demonstrated that this connection is subject to presynaptic influences. However, Faist et al. (1996) recently demonstrated that during walking in man the heteronymous facilitation from quadriceps is largest during stance. If the heteronymous monosynaptic pathway is regulated in a similar fashion between species, then we would also expect that the facilitation would be greatest during stance phase for cats. We did not see evidence of this when we stretched quadriceps during the extensor phase of locomotion in cats. This suggests that there is a species difference in the regulation this pathway.

Another possibility is that the facilitation of the soleus motoneurons by the quadriceps monosynaptic connections was present, but compensated for by a decrease in soleus homonymous monosynaptic activation of the soleus pool. It has been suggested that a portion of the soleus EMG activity produced during stance phase of locomotion arises from the homonymous stretch reflex (Pearson 1995b). From the present paper it is clear that the homonymous stretch reflex would be presynaptically suppressed with a stretch of quadriceps. Therefore it is equally surprising that the stretching of quadriceps did not result in a decrease in the ongoing EMG. However, this lack of a decrease in background EMG can be explained if there was a concomittent facilitatory effect arising from the heteronymous pathway from quadriceps. That is, it is possible that both the facilitatory and inhibitory effects on soleus EMG activity were present after the stretching of quadriceps. However, with one cancelling the effect of the other, no net change in EMG activity was observed. Such a mechanism would be an attractive method by which the selective suppression of the soleus stretch reflex could be achieved, without also suppressing the excitability of the soleus motoneuron pool. Barnes and Pompeiano (1970c) demonstrated that such a dual heteronymous manipulation of the monosynaptic reflex of medial gastrocnemius was produced by vibration of the lateral gastrocnemius. In that study, vibration of the lateral gastrocnemius produced a facilitation of the monosynaptic reflex of the medial gastrocnemius during the actual vibration, which was followed by a prolonged inhibition once the vibration ceased. The facilitation of the reflex was achieved by a postsynaptic influence, whereas the prolonged inhibition was associated with PAD. The PAD of the medial gastrocnemius primary afferents was present during the vibration, but the postsynaptic facilitation was sufficiently strong so as to mask the presynaptic inhibiton.

Functional implications

The results of the present study clearly demonstrate that afferent feedback contributes to the regulation of the soleus H reflex amplitude during locomotion in the decerebrate cat. This afferent mediated reflex inhibition could play two important roles. First, the modulation of the soleus H reflex, evoked by stretching quadriceps, might reduce the stretch reflex of soleus during early stance. Suppression of the soleus stretch reflex in early stance would allow the ankle to yield at this point in the cycle and thus facilitate the movement of the body over the foot. The importance of this mechanism might be more apparent when other forms of locomotion are compared. For instance, when walking up a grade there is little yield at either the knee or ankle (Smith and Carlson-Kuhta 1995), whereas during galloping the yield at both the ankle and knee increases (Goslow et al. 1973; Smith et al. 1993). Thus during galloping one would expect that the soleus H reflex would be further suppressed. Such a comparison does not yet exist for the cat; however, Capaday and Stein (1987) demonstrated that the soleus H reflex in humans was reduced during running compared with walking. Second, by providing an afferent-mediated source of reflex regulation, the gain of the reflex can be quickly altered according to the dynamic needs of the animal. There is evidence that this afferent-induced attenuation of the reflex pathway is achieved via spinal mechanisms (Brooke et al. 1995; Misiaszek et al. 1996). Thus, during locomotion the strength of the soleus H reflex can rapidly respond to sudden perturbations by further suppressing the reflex if the afferent input from this source increases, or releasing the reflex if the afferent input decreases.

It is now quite clear that muscle afferent input from a number of sources influences the amplitude of the soleus H reflex (Brooke et al. 1997). From the results of our study, it is clear that the stretch of quadriceps is one of these sources. Contralateral afferent feedback has also been demonstrated to modulate H reflexes (Brooke et al. 1995; Collins et al. 1993; Koceja and Kamen 1992; Misiaszek et al. 1996). It has also been shown that cutaneous inputs can influence the level of presynaptic inhibition at Ia afferents by inhibiting the presynaptic inhibitory pathway (Nakashima et al. 1990). Although CPG-related PAD does not influence the magnitude of the MSR (Gossard 1996), other central structures cannot be ruled out. Recently, some central structures have been shown to produce primary afferent depolarization of Ia afferents of triceps surae (Enriquez et al. 1996) and thus might contribute to the presynaptic inhibition of the H reflex. In the future, an important problem to solve will be the relative contribution of each of these sources of reflex modulation when it becomes part of a montage of influences acting on the reflex arc.

	ACKNOWLEDGEMENTS

  The authors thank R. Gramlich for excellent technical assistance.

  This work was supported by a grant from the Medical Research Council of Canada to K. G. Pearson and an Alberta Heritage Foundation Fellowship to J. E. Misiaszek.

	FOOTNOTES

  Address reprint requests to J. E. Misiaszek.

  Received 7 May 1997; accepted in final form 14 July 1997.

	REFERENCES

Abstract Introduction Methods Results Discussion References

• AKAZAWA, K., ALDRIDGE, J. W., STEEVES, J. D., STEIN, R. B. Modulation of stretch reflexes during locomotion in the mesencephalic cat. J. Physiol. (Lond.) 329: 553-567, 1982.[Abstract/Free Full Text]
• BARNES, C. D., POMPEIANO, O. Inhibition of monosynaptic extensor reflex attributable to presynaptic depolarization of the group Ia afferent fibers produced by vibration of flexor muscle. Arch. Ital. Biol. 108: 233-258, 1970a.[Medline]
• BARNES, C. D., POMPEIANO, O. Presynaptic and postsynaptic effects in the monosynaptic reflex pathway to extensor motoneurons following vibration of synergic muscles. Arch. Ital. Biol. 108: 259-294, 1970b.[Medline]
• BARNES, C. D., POMPEIANO, O. Effects of muscle vibration on the pre- and postsynaptic components of the extensor monosynaptic reflex. Brain Res. 18: 384-388, 1970c.[Medline]
• BROOKE, J. D., CHENG, J., COLLINS, D. F., MCILROY, W. E., MISIASZEK, J. E., STAINES, W. R. Sensori-sensory afferent conditioning with leg movement: gain control in spinal reflex and ascending paths. Prog. Neurobiol. 51: 393-421, 1997.[Medline]
• BROOKE, J. D., COLLINS, D. F., BOUCHER, S., MCILROY, W. E. Modulation of human short latency reflexes between standing and walking. Brain Res. 548: 172-178, 1991.[Medline]
• BROOKE, J. D., MCILROY, W. E., COLLINS, D. F., MISIASZEK, J. E. Mechanisms within the human spinal cord suppress fast reflexes to control the movement of the legs. Brain Res. 679: 255-260, 1995.[Medline]
• BROOKE, J. D., MISIASZEK, J. E., CHENG, J. Locomotor-like rotation of either hip or knee inhibits soleus H reflexes in humans. Somatosens. Mot. Res. 10: 357-364, 1993.[Medline]
• BROWN, M. C., ENGBERG, I., MATTHEWS, P.B.C. The relative sensitivity to vibration of muscle receptors of the cat. J. Physiol. (Lond.) 192: 773-780, 1967.[Abstract/Free Full Text]
• CAPADAY, C., STEIN, R. B. Amplitude modulation of the soleus H relfex in the human during walking and standing. J. Neurosci. 6: 1308-1313, 1986.[Abstract]
• CAPADAY, C., STEIN, R. B. Difference in the amplitude of the human soleus H reflex during walking and running. J. Physiol. (Lond.) 392: 513-522, 1987.[Abstract/Free Full Text]
• CHENG, J., BROOKE, J. D., MISIASZEK, J. E., STAINES, W. R. The relationship between the kinematics of passive movement, the stretch of extensor muscles of the leg and the change induced in the gain of the soleus H reflex in humans. Brain Res. 672: 89-96, 1995a.[Medline]
• CHENG, J., BROOKE, J. D., STAINES, W. R., MISIASZEK, J. E., HOARE, J. Long-lasting conditioning of the human soleus H reflex following quadriceps tendon tap. Brain Res. 681: 197-200, 1995b.[Medline]
• COLLINS, D. F., MCILROY, W. E., BROOKE, J. D. Contralateral inhibition of soleus H reflexes with different velocities of passive movement of the opposite leg. Brain Res. 603: 96-101, 1993.[Medline]
• CRENNA, P., FRIGO, C. Excitability of the soleus H reflex arc during walking and stepping in man. Exp. Brain Res. 66: 49-60, 1987.[Medline]
• DECANDIA, M., PROVINI, L., TABORIKOVA, H. Presynaptic inhibition of the monosynaptic reflex following the stimulation of nerves to extensor muscles of the ankle. Exp. Brain Res. 4: 34-42, 1967.
• DUENAS, S. H., LOEB, G. E., MARKS, W. B. Monosynaptic and dorsal root reflexes during locomotion in normal and thalamic cats. J. Neurosci. 63: 1467-1476, 1990.
• ECCLES, J. C., ECCLES, R. M., LUNDBERG, A. The convergence of monosynaptic excitatory afferents on to many different species of alpha motoneurones. J. Physiol. (Lond.) 137: 22-50, 1957.
• ECCLES, J. C., SCHMIDT, R. F., WILLIS, W. D. Presynaptic inhibition of the spinal monosynaptic reflex pathway. J. Physiol. (Lond.) 161: 282-297, 1962.
• ENRIQUEZ, M., JIMENEZ, I., RUDOMIN, P. Segmental and supraspinal control of synaptic effectiveness of functionally identified muscle afferents in the cat. Exp. Brain Res. 107: 391-404, 1996.[Medline]
• FAIST, M., DIETZ, V., PIERROT-DESEILLIGNY, E. Modulation, probably presynaptic in origin, of monosynaptic Ia excitation during human gait. Exp. Brain Res. 109: 441-449, 1996.[Medline]
• GOSLOW, G. E. JR, REINKING, R. M., STUART, D. G. The cat step cycle: hind limb joint angles and muscle lengths during unrestrained locomotion. J. Morphol. 141: 1-41, 1973.[Medline]
• GOSSARD, J.-P. Control of transmission in muscle group Ia afferents during fictive locomotion in the cat. J. Neurophysiol. 76: 4104-4112, 1996.[Abstract/Free Full Text]
• GOSSARD, J.-P., CABELGUEN, J.-M., ROSSIGNOL, S. An intracellular study of muscle primary afferents during fictive locomotion in the cat. J. Neurophysiol. 65: 914-926, 1991.[Abstract/Free Full Text]
• GUERTIN, P., ANGEL, M. J., PERREAULT, M. C., MCCREA, D. A. Ankle extensor group I afferents excite extensors throughout the hindlimb during MLR evoked fictive locomotion in the cat. J. Physiol. (Lond.) 487: 197-209, 1995.[Medline]
• HIEBERT, G. W. Regulation of Muscle Activity During Walking (PhD thesis). Alberta, Canada: Univ. of Alberta, 1997.
• HULTBORN, H., ILLERT, M., NIELSEN, J., PAUL, A., BALLEGAARD, M., WIESE, H. On the mechanism of the post-activation depression of the H-reflex in human subjects. Exp. Brain Res. 108: 450-462, 1996.[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. (Lond.) 389: 729-756, 1987.[Abstract/Free Full Text]
• KOCEJA, D. M., KAMEN, G. Contralateral influences on triceps surae motoneuron excitability. Electroencephalogr. Clin. Neurophysiol. 85: 177-182, 1992.[Medline]
• LOEB, G. E., HOFFER, J. A., PRATT, C. A. Activity of spindle afferents from cat anterior thigh muscles. I. Identification and patterns during normal locomotion. J. Neurophysiol. 54: 549-564, 1985.[Abstract/Free Full Text]
• MCILROY, W. E., COLLINS, D. F., BROOKE, J. D. Movement features and H-reflex modulation. II. Passive rotation, movement velocity and single leg movements. Brain Res. 582: 85-93, 1992.[Medline]
• MISIASZEK, J. E., BARCLAY, J. K., BROOKE, J. D. Inhibition of canine H reflexes during locomotor-like rotation about the knee arises from muscle mechanoreceptors in quadiceps. J. Neurophysiol. 73: 2499-2506, 1995a.[Abstract/Free Full Text]
• MISIASZEK, J. E., BARCLAY, J. K., BROOKE, J. D. Mechanisms within the spinal cord are involved in the movement-induced attenuation of an H reflex in the dog. J. Neurophysiol. 76: 3589-3592, 1996.[Abstract/Free Full Text]
• MISIASZEK, J. E., BROOKE, J. D., LAFFERTY, K. B., CHENG, J., STAINES, W. R. Long-lasting inhibition of the human soleus H reflex pathway after passive movement. Brain Res. 677: 69-81, 1995b.[Medline]
• NAKASHIMA, K., ROTHWELL, J. C., DAY, B. L., THOMPSON, P. D., MARSDEN, C. D. Cutaneous effects on presynaptic inhibition of flexor Ia afferents in the human forearm. J. Physiol. (Lond.) 426: 369-380, 1990.[Abstract/Free Full Text]
• PEARSON, K. G. Reflex reversal in walking systems of mammals and arthropods. In: Neural Control of Movement, edited by W. R. Ferrell, and U. Proske . London: Plenum, 1995a, p. 135-141
• PEARSON, K. G. Proprioceptive regulation of locomotion. Curr. Opin. Neurobiol. 5: 786-791, 1995b.[Medline]
• , sect. 12, p. 173-216. ROSSIGNOL, S. Control of sterotypic limb movements. In: Handbook of Physiology. Exercise: Regulation and Integration of Multiple Systems., . Bethesda, MD: Am. Physiol. Soc., 1996
• SCHMIDT, R. F. Presynaptic inhibition in the vertebrate central nervous system. Ergeb. Physiol. 63: 20-101, 1971.[Medline]
• SHEFCHYK, S. J., STEIN, R. B., JORDAN, L. M. Synaptic transmission from muscle afferents during fictive locomotion in the mesencephalic cat. J. Neurophysiol. 51: 986-997, 1984.[Abstract/Free Full Text]
• SMITH, J. L., CARLSON-KUHTA, P. Unexpected motor patterns for hindlimb muscles during slope walking in the cat. J. Neurophysiol. 74: 2211-2215, 1995.[Abstract/Free Full Text]
• SMITH, J. L., CHUNG, S. H., ZERNICKE, R. F. Gait-related motor patterns and hindlimb kinetics for the cat trot and gallop. Exp. Brain Res. 94: 308-322, 1993.[Medline]
• STEIN, R. B., CAPADAY, C. The modulation of human reflexes during functional motor tasks. Trends Neurosci. 11: 328-332, 1988.[Medline]
• TAYLOR, J., STEIN, R. B., MURPHY, P. R. Impulse rates and sensitivity to stretch of soleus muscle spindle afferent fibers during locomotion in premammillary cats. J. Neurophysiol. 53: 341-360, 1985.[Abstract/Free Full Text]
• WHELAN, P. J., HIEBERT, G. W., PEARSON, K. G. Stimulation of the group I extensor afferents prolongs the stance phase in walking cats. Exp. Brain Res. 103: 20-30, 1995.[Medline]

0022-3077/97 $5.00 Copyright ©1997 The American Physiological Society