Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2000 Jan 1;522(Pt 1):125–135. doi: 10.1111/j.1469-7793.2000.t01-1-00125.xm

Distribution of presynaptic inhibition on type-identified motoneurones in the extensor carpi radialis pool in man

Jean-Marc Aimonetti 1, Jean-Pierre Vedel 1, Annie Schmied 1, Simone Pagni 1
PMCID: PMC2269738  PMID: 10618157

Abstract

  1. The question was addressed as to whether the magnitude of Ia presynaptic inhibition might depend on the type of motor unit activated during voluntary contraction in the wrist extensor muscles. For this purpose, we investigated the effects of applying electrical stimulation to the median nerve on the responses of 25 identified motor units to radial nerve stimulation delivered 20 ms after a conditioning stimulation.

  2. The reflex responses of the motor units yielded peaks in the post-stimulus time histograms with latencies compatible with monosynaptic activation. Although median nerve stimulation did not affect the motoneurone net excitatory drive assessed from the mean duration of the inter-spike interval, it led to a decrease in the contents of the first two 0.25 ms bins of the peak. This decrease may be consistent with the Ia presynaptic inhibition known to occur under these stimulation conditions.

  3. In the trials in which the median nerve was being stimulated, the finding that the response probability of the motor units, even in their monosynaptic components, tended to increase as their force threshold and their macro-potential area increased and as their twitch contraction time decreased suggests that the median nerve stimulation may have altered the efficiency with which the Ia inputs recruited the motoneurones in the pool.

  4. These effects were consistently observed in seven pairs of motor units each consisting of one slow and one fast contracting motor unit which were simultaneously tested, which suggests that the magnitude of the Ia presynaptic inhibition may depend on the type of motor unit tested rather than on the motoneurone pool excitatory drive.

  5. The present data suggest for the first time that in humans, the Ia presynaptic inhibition may show an upward gradient working from fast to slow contracting motor units which is able to compensate for the downward gradient in monosynaptic reflex excitation from ‘slow’ to ‘fast’ motor units. From a functional point of view, a weaker Ia presynaptic inhibition acting on the fast contracting motor units may contribute to improving the proprioceptive assistance to the wrist myotatic unit when the contraction force has to be increased.

Since it was first discovered (Frank & Fuortes, 1957), Ia presynaptic inhibition has been generally assumed to contribute importantly to the sensorimotor integration processes by selectively altering the efficiency of the motoneuronal sensory inputs (Lomeli et al. 1998). Ia presynaptic inhibition has been previously investigated in humans by assessing the changes in the H-reflex amplitude recorded during various locomotor activities (Capaday & Stein, 1987a; Faist et al. 1996), during the contraction of antagonist muscles (Meunier & Morin, 1989), and during the co-contraction of antagonist muscles (Nielsen & Kagamihara, 1993; Schmied et al. 1997b).

Hultborn et al. (1987) have developed a technique for exploring Ia presynaptic inhibition at the motor unit level in humans. The motor unit response to stimulation applied to homonymous Ia muscle spindle afferents shows a peak at a latency compatible with monosynaptic activation in the post-stimulus time histograms (PSTHs). Upon making comparisons with animal data, Hultborn et al. (1987) observed that the first 0.5 ms part of the reflex peak can be taken to be purely monosynaptic, since it cannot be contaminated by any oligosynaptic components. Any changes in the number of counts during this first 0.5 ms can therefore be safely taken to reflect changes in the Ia excitatory post-synaptic potential caused by presynaptic inhibition. However, for methodological reasons, most of the previous studies that have been performed at the single motor unit level on humans have been conducted on muscles at rest. None of the previous human studies and only a couple of animal studies have therefore dealt with the magnitude of the Ia presynaptic inhibition depending on the type of motor unit tested. In the cat triceps surae muscle, for example, the Ia presynaptic inhibition does not alter all the individual motoneurones in the same way (Cook & Cangliano, 1972; Rudomín et al. 1975). There may rather exist an upward gradient of the presynaptic inhibition estimated in terms of the changes in the amplitude of the Ia EPSPs, working from the fast to slow contracting motor units (Zengel et al. 1983).

In the present study, in which the purely monosynaptic components of the motor unit responses were analysed, the reflex responses of single motor units elicited by electrically stimulating the radial nerve were recorded in the extensor carpi radialis muscles while they were being voluntarily activated by performing isometric wrist extension. In order to induce Ia presynaptic inhibition, the responses of the extensor motor units were conditioned by applying electrical stimulation to the median nerve 20 ms before the radial nerve stimulation (see Burke et al. 1994). This is the first time the question has been addressed as to whether the magnitude of the Ia presynaptic inhibition might depend in humans on the type of motor unit tested. The motor units were identified on the basis of their force thresholds, their macro-potentials and their twitch contraction times.

METHODS

Experiments were performed on seven healthy male right-handed subjects aged 20–30 years, with the approval of the Ethics Committee of the local Medical University (CCPPRB-Marseille I, approval N8 92/74). All the subjects gave their informed written consent to the experimental procedure as required by the Declaration of Helsinki (1964).

Instructions to subjects

The subjects were seated in an adjustable armchair. Their right forearm was placed in a cushioned groove, and the distal end of their forearm was immobilized in a U-shaped device, leaving the wrist joint free and maintaining the hand in a semi-prone position, flexed at an angle of 10 deg. The subjects were asked to selectively activate their wrist extensor muscles by pushing with the back of their hand against a force transducer device, keeping their fingers relaxed. The subjects were asked to maintain a sufficiently strong level of muscle contraction to ensure that the motor unit tested kept firing tonically during the whole recording sequence, each of which lasted for about 5 min. To control the motor unit activity, the subjects used the auditory and visual feedback provided by a loudspeaker and an oscilloscope placed in front of them. Rest periods of 2 min were allowed to elapse between the sequences.

Data recording

The net force produced by the wrist extension, calibrated in Newtons, was recorded as a direct (DC) and filtered signal (AC; bandpass, 0.1-1000 Hz). The overall EMG activity of the extensor carpi radialis and that of the flexor carpi radialis muscles was recorded by means of a pair of non-polarizable surface electrodes (Ag-AgCl) placed 2 cm apart. The amplified EMG signals were filtered with a 30 Hz to 10 kHz bandpass, and full-wave rectified and integrated by means of a Paynter filter (PF-I, time constant 0.05 s; BAK Electronics).

The action potentials generated by the activity of single motor units were recorded in the extensor carpi radialis muscles by means of transcutaneously inserted single-use metal microelectrodes (impedance 12 MΩ tested at 1 kHz; Frederick Haer and Co.). The microelectrode and the surface electrodes were connected to amplifiers (P11 Amplifier, Grass Instruments) through probes with an isolated ground to ensure optimum subject protection (current leakage less than 3 μA). The bandwidth of the amplifier was set at 300 Hz to 3 kHz for the intramuscular recording.

The motor units activated were recorded at various EMG activity levels, depending on their force thresholds. In some instances, motor unit pairs composed of one low threshold unit and one higher threshold unit were concurrently recorded. The paired recordings were performed to check that the effects observed at various EMG activity levels were consistent when both motor unit types were tested with the same motoneurone pool excitatory drive, i.e. with the same EMG activity corresponding to the contraction necessary to keep the higher force threshold motor unit firing. The signals were stored on an 8-channel digital tape recorder (DTR-1800; Biologic).

Stimulation protocol

To elicit reflex responses in the extensor carpi radialis muscles, the radial nerve was electrically stimulated (Stimulator S88, Grass Instruments; shock duration, 1 ms; frequency, 0.3 Hz) via a constant current unit using a pair of spherical surface electrodes (diameter, 1.3 cm). The cathode was placed 8–10 cm above the elbow on the external side of the upper arm between the biceps brachialis and triceps brachii muscles, and the anode was placed opposite the cathode. Since the Ia afferent stimulation threshold necessary to activate a motor unit is known to increase with its recruitment threshold according to the size principle (Henneman, 1977; Schmied et al. 1997a), it was necessary to standardize the stimulation intensity from one experiment to another and from one subject to another. At the beginning of each recording, after starting at a very low force threshold motor unit (< 0.2 N), the stimulation intensity was gradually increased until 100 % motor unit response rate was obtained. Upon reducing the stimulation by 8–10 %, the response rate of the motor units with low force thresholds was about 70 %. The radial nerve stimulation intensity was then kept constant in all types of motor units, the amplitude of the M-wave being used as a monitor of the stability of the stimulation. The advantage of this procedure was that it kept the strength of the group I afferent burst as constant as possible. In order to limit the effects of any post-activation depression (Hultborn et al. 1996), 10 stimuli were applied before starting the recordings, since the effects of post-activation depression are known to disappear within 15 s in humans (Wood et al. 1996).

To evoke Ia presynaptic inhibition in the extensor carpi radialis muscles, the median nerve was electrically stimulated (Stimulator S88, Grass Instruments; shock duration, 1 ms) via a constant current unit using a pair of spherical surface electrodes (diameter, 1.3 cm). The median nerve stimulation was applied 20 ms before the radial nerve stimulation (see Burke et al. 1994). The cathode was placed on the internal side of the forearm just above the distal tendons of the biceps muscle in the elbow fossa, and the anode was placed opposite the cathode. The voltage of the median nerve stimulation was set at 0.8 × the motor threshold (MT) in the flexor carpi radialis muscle, and its stability was carefully monitored using a current probe (P6022 Tektronix Inc.).

The sequences contained all 100 stimuli applied to the radial nerve. Each motor unit was tested first while stimulating the radial nerve alone, and then while stimulating both the median and radial nerves. In some experiments, this order was reversed, and no differences were observed. The data obtained with the fixed and the reversed order of presentation were therefore combined in the figures.

Data analysis

The action potentials of the motor units tested were discriminated using a dual window discriminator (RP-I, BAK Electronics). The firing patterns of the motor units were characterized in terms of the mean duration of the inter-spike intervals. Their responses to the radial nerve stimulation were analysed in post-stimulus time histograms giving the firing probability of each motor unit in 0.25 ms bins, during 30 ms after the radial nerve stimulation. The motor unit reflex activation was assessed from its probability of responding to the radial nerve stimulation in the narrow peak occurring with a latency compatible with a monosynaptic loop (21.74 ± 1.94 ms; Table 1). The peak characteristics were automatically determined using the cumulative sum procedure, (cusum; Ellaway, 1978) which was computed with regards to the mean value of the motor unit tonic activity (baseline mean) estimated in a time window of 15 ms after the radial nerve stimulation, as illustrated in Fig. 1. With each cusum curve, the sharp rising phase and its confidence limit were used to determine the first bin count significantly different from the baseline mean. According to cusum rising phase, two time limits were positioned at the onset and offset of the peak. The peak duration and its latency and the motor unit response probability in the whole peak above the baseline mean were then automatically measured. The purely monosynaptic components of the reflex response of each motor unit was assessed from the contents of the first two 0.25 ms bins in the reflex peak in the post-stimulus time histogram (see Hultborn et al. 1987).

Table 1.

Motor unit characteristics

Response probability (imp/trigger) Monosynaptic component(imp/trigger) Peak latency (ms) Peak duration (ms) Inter-spike interval (ms) Force threshold (N) Macro-MUP area (mV ms) Twitch contraction time (ms)
Radial nerve stimulation 0.62 ± 0.11 0.19 ± 0.07 21.74 ± 1.94 1.83 ± 0.42 99.7 ± 11.9 1.58 ± 1.38 0.35 ± 0.24 51.0 ± 21.7
0.45–0.78 0.10–0.43 18.25–27.00 0.75–2.75 82.0–126.9 0.05–4.29 0.10–0.86 19.5–89.3
Radial and median nerve stimulation 0.32 ± 0.10 0.08 ± 0.03 21.70 ± 1.98 1.75 ± 0.40 98.7 ± 12.1 1.55 ± 1.40 0.33 ± 0.22 50.6 ± 20.4
0.11–0.51 0.03–0.18 18.25–27.25 1.00–2.75 74.7–117.9 0.04–4.42 0.09–0.85 18.6–89.7
Open in a new tab

Values are means ± S.D., with ranges given below.

Figure 1. Analysis of the motor unit reflex peak using the cumulative sum procedure.

Figure 1

Open in a new tab

The reflex response of a motor unit to the radial nerve stimulation yielded a narrow peak in the post-stimulus time histograms. The characteristics of the peak were automatically determined using the cumulative sum procedure (cusum; Ellaway, 1978). The sharp rising phase in the cusum and its confidence limit were used to determine the first bin count significantly different from the baseline mean. By positioning time limits at the onset and offset of the peak, it was possible to automatically measure the peak duration and latency and the motor unit response probability in the whole peak above the baseline mean and in its purely monosynaptic component (first 0.5 ms).

In the trials in which the median nerve was being stimulated, the responses of the motor units were quantified as actual values or as relative changes in the whole peak and in the first 0.5 ms with respect to the respective value measured when the radial nerve was stimulated alone (response with median nerve stimulation and response without median nerve stimulation/response without median nerve stimulation × 100). Changes in the peak latency by 0.25 ms (shortening or lengthening) were often observed while applying median nerve stimulation. These changes were found to be closely correlated to changes in the inter-spike interval variability induced either by changes in the firing control of the motor unit by the subjects or by the median nerve stimulation. These shifts in the peak latency were clearly observed when using cutaneous stimulation as in the following paired paper (Aimonetti et al. 2000). In view of this observation, the peak onset was always considered as the first bin significantly different from the baseline mean using the cusum confidence limit.

The level of muscle force necessary to keep a motor unit firing steadily around 10 Hz during wrist extension reflex trials was measured, averaged over each recording session and taken as an index of its force threshold.

The macro-potential extracted by applying the spike-triggered averaging procedure to the surface EMG activity was expressed in terms of its area (mV ms). To ensure that the same motor unit had been tested with and without median nerve stimulation, the shape and area of the macro-potentials obtained during each recording sequence were carefully compared.

The force change (twitch) selectively associated with the activity of each motor unit was extracted by applying the spike-triggered averaging procedure (Stein et al. 1972) to the net filtered extension force, and the contraction time (ms) of the twitch was measured. To minimize the distortion possibly resulting from the partial fusion of successive twitches, any spikes occurring with an inter-spike interval of less than 90–140 ms (both before and after the spike), depending on the motor unit tested, were excluded from the analysis. When two motor units were concurrently recorded, their twitches were extracted during sequences in which the subjects had to maintain each motor unit firing at low frequency discharge, one after the other. Whenever possible, spike-triggered averaging was performed on two or three different parts of the recording period, including at least 100 action potentials. The mean contraction time of the two or three twitch profiles inferred for each motor unit was calculated and taken as an index of its contraction time.

Statistical analyses

The characteristics of the motor unit responses were compared between the recordings obtained with and without median nerve stimulation by making paired comparisons using an ANOVA procedure for repeated measures (Statview 4.11 for Macintosh, Abacus concept). The significance level was set at P < 0.05. The relationships between the response probability (or the changes in the response probability) and each of the motor unit's functional parameters were analysed by performing linear regression analyses.

RESULTS

A total number of 25 motor units was tested, each of which was characterized by the mean duration of the inter-spike intervals, the level of its force threshold, the area of its macro-potential, the twitch contraction time, and the amplitude, duration and latency of the PSTH peak evoked by applying electrical stimulation to the radial nerve. Table 1 gives the statistics (mean ±s.d., and range) on each of these functional parameters. The data given here are the pooled data obtained from all seven subjects.

The main result to emerge from the present study was that the responses of the slow contracting motor units were more strongly depressed by the Ia presynaptic inhibition than those of the faster contracting motor units. This led to the possibility that the Ia presynaptic inhibition may affect the gain with which the Ia inputs recruited the motoneurones in the pool.

Paired recordings

Figure 2 illustrates the patterns of response of a slow contracting motor unit (Fig. 2A and B) and a simultaneously tested faster contracting motor unit (Fig. 2E and F). The subjects were asked to adjust their wrist muscle contraction force to the level necessary to maintain the highest force threshold unit firing steadily. When the radial nerve was being stimulated alone, the response probability of the motor unit (Fig. 2a, 0.53 impulses per trigger) activated at a low force threshold (0.92 N), i.e. that having a small macro-potential area (Fig. 2C, 0.28 mV ms) and a long twitch contraction time (Fig. 2D, 74.21 ms), was found to be greater than the response probability of the motor unit (Fig. 2E, 0.45 impulses per trigger) activated at a higher force threshold (4.2 N), i.e. that having a larger macro-potential area (Fig. 2G, 0.72 mV ms) and a faster twitch contraction time (Fig. 2H, 19.61 ms). Applying stimulation to the median nerve decreased the response probability of the slow contracting motor unit by 50.9 % (Fig. 2B, 0.26 impulses per trigger), without affecting its macro-potential area (0.27 mV ms) or its twitch contraction time (76.2 ms). The median nerve stimulation decreased the response probability of the faster contracting motor unit of the two activated by only 26.7 % (Fig. 2F, 0.33 impulses per trigger), without affecting its macro-potential area (0.74 mV ms) or its twitch contraction time (20.2 ms).

Figure 2. Paired recording of slow and fast contracting motor units.

Figure 2

Open in a new tab

The activity of a slow contracting motor unit was recorded concurrently with that of a faster contracting motor unit. The subjects were asked to adjust their wrist muscle contraction force to the level necessary to keep the highest force threshold unit firing steadily. In the recording in which the radial nerve was being stimulated alone, the response probability of the motor unit (A, 0.53 impulses per trigger) activated at a low force threshold (0.92 N), i.e. that having a small macro-potential area (C) and a long twitch contraction time (CT; D), was greater than the response probability of the motor unit (E, 0.45 impulses per trigger) activated at a higher force threshold (4.2 N), i.e. that having a larger macro-potential area (G) and a faster twitch contraction time (H). Contrary to this, while concomitantly stimulating the median nerve, the response probability of the fast contracting motor unit (F, 0.33 impulses per trigger; −26.7 %) was greater than that of the slow contracting motor unit (B, 0.26 impulses per trigger; −50.9 %). These data suggest that the magnitude of the presumed Ia presynaptic inhibition may depend on the type of motor unit tested rather than on the motoneurone pool excitatory drive, and that the median nerve stimulation may have altered the strength with which the Ia inputs recruited the motoneurones in the pool. Twitches of the two motor units were extracted during sequences in which the subjects had to maintain each motor unit firing at low frequency discharge, one after the other. The selected inter-spike intervals chosen for the triggering impulses are noted in D and H.

Similar findings were obtained in the case of six other pairs consisting of one fast and one simultaneously tested slow contracting motor unit. When the radial nerve was being stimulated alone, the response probability of the slow contracting motor units (0.62 ± 0.09 impulses per trigger; mean duration of the inter-spike intervals, 48.69 ± 3.63 ms) with low force thresholds (0.57 ± 0.25 N), small macro-potential areas (0.34 ± 0.07 mV ms), and long twitch contraction times (63.65 ± 11.13 ms) was greater than that of the fast contracting motor units (0.56 ± 0.11 impulses per trigger; mean duration of the inter-spike intervals: 103.94 ± 14.93 ms) with higher force thresholds (3.45 ± 0.66 N), larger macro-potential areas (0.60 ± 0.07 mV ms), and shorter twitch contraction times (25.56 ± 3.73 ms). Contrary to this, in the trials in which the median nerve was being stimulated, the response probability of the fast contracting motor units (0.38 ± 0.09 impulses per trigger; −32.90 ± 7.13 %) was greater than that of the slow contracting motor units (0.24 ± 0.06 impulses per trigger; −58.54 ± 5.77 %). Two important points emerge from these findings. First, the magnitude of the Ia presynaptic inhibition may depend on the type of motor unit tested rather than on the motoneurone pool excitatory drive. Second, the median nerve stimulation may have been able to compensate for the downward gradient in monosynaptic reflex excitation from ‘slow’ to ‘fast’ motor units, since the ‘fast’ motor units were more responsive to the radial nerve stimulation than the ‘slow’ motor units.

Relationships within motor unit properties

Regression analyses were performed between the response probability and each of the motor unit's functional parameters during recordings both with and without median nerve stimulation.

During the recordings performed without median nerve stimulation, consistent trends were observed between each motor unit's response probability and its functional parameters. The responses of the motor units tended to decrease as their force thresholds (Fig. 3A, r =−0.66, P < 0.001) and their macro-potential areas (Fig. 3B, r =−0.58, P = 0.003) increased, and as their twitch contraction times (Fig. 3C, r = 0.63, P = 0.002) decreased. The present data are in keeping with the size principle (Henneman, 1977; Awiszus & Feistner, 1993; Schmied et al. 1997a), which suggests that the motor unit reflex responsiveness increases gradually from the fast to the slow contracting motor units.

Figure 3. Relationships between the reflex responses in the whole peak (actual data) and the functional parameters of the motor units.

Figure 3

Open in a new tab

While stimulating the radial nerve alone, the response probability of the motor units tended to decrease as their force thresholds (A) and their macro-potential areas (B) increased, and as their twitch contraction times decreased (C), in keeping with the size principle. In the recordings with median nerve stimulation, the response probability of the motor units tended to increase as their force thresholds (D) and their macro-potential areas (E) increased, and as their twitch contraction times decreased (F). In the recordings with median nerve stimulation, there may therefore exist an upward gradient in the proprioceptive Ia monosynaptic assistance to the extensor motor units, working from slow to fast contracting motor units. The median nerve stimulation may have been able to compensate for the downward gradient in monosynaptic reflex excitation from ‘slow’ to ‘fast’ motor units.

During the recordings with median nerve stimulation, the motoneurone net excitatory drive, as estimated by assessing the mean duration of the inter-spike interval, did not differ significantly from that observed in the recordings without median nerve stimulation (Table 1). Likewise, the mean duration and mean latency of the PSTH peaks were not found to differ between the recordings with and without median nerve stimulation (Table 1).

In the recordings in which the median nerve was being stimulated, consistent trends were observed between the response probability and each of the motor unit's functional parameters. The responses of the motor units tended to increase as their force thresholds (Fig. 3D, r = 0.54, P = 0.002) and their macro-potential areas (Fig. 3E, r = 0.45, P = 0.05) increased, and as their twitch contraction times decreased (Fig. 3F, r =−0.45, P = 0.04). The present results suggest that the median nerve stimulation was able to compensate for the downward gradient in monosynaptic reflex excitation from slow to fast motor units.

The actual magnitude of the inhibition estimated by measuring the difference between the response probabilities of each motor unit tested with and without median nerve stimulation was significantly correlated (r = 0.41, P = 0.03) with the response probability measured while the radial nerve was being stimulated alone (data not shown), which is in keeping with the results of previous animal studies (Rudomín et al. 1975; Zengel et al. 1983).

The purely monosynaptic components of the motor unit reflex response to radial nerve stimulation were investigated by assessing the contents of the first two bins in the reflex peak (Hultborn et al. 1987). In the recordings with median nerve stimulation, the contents of the first two bins in the reflex peak (0.08 ± 0.03 impulses per trigger) were significantly smaller (P < 0.001) than those recorded without median nerve stimulation (0.19 ± 0.07 impulses per trigger). The median nerve stimulation can therefore be said to have induced presynaptic inhibitory effects on the endings of Ia muscles spindle afferents originating from the extensor carpi radialis muscles.

In the trials in which the median nerve was being stimulated, consistent trends were observed between the changes in the purely monosynaptic components of the reflex response and each of the motor unit's functional parameters. The decrease in the purely monosynaptic components of the responses of the motor units due to Ia presynaptic inhibition tended to weaken as their force thresholds (Fig. 4A, r = 0.68, P < 0.001) and their macro-potential areas (Fig. 4B, r = 0.58, P = 0.004) increased, and as their twitch contraction times decreased (Fig. 4C, r =−0.61, P < 0.001). The present data therefore provide further evidence supporting the hypothesis that in human wrist extensor muscles, the magnitude of the Ia presynaptic inhibition may increase from fast to slow contracting motor units.

Figure 4. Relationships between the changes in the purely monosynaptic components of the responses and the functional parameters of the motor units.

Figure 4

Open in a new tab

In the recordings with median nerve stimulation, the decrease in the first 0.5 ms of the peak responses of the motor units tended to be weaker as their force thresholds (A) and their macro-potential areas (B) increased, and as their twitch contraction times (C) decreased. These data confirm that there may exist an upward gradient in the magnitude of the presynaptic inhibition, working from the fast to the slow contracting motor units.

The effects of the size of the unconditioned response on the strength of Ia presynaptic inhibition

In the previous studies in which Ia presynaptic inhibition has been estimated by assessing changes in the H-reflex amplitude, it has been reported that the susceptibility of the monosynaptic reflex to inhibition depended on the size of the test H reflex (see Crone et al. 1990). In keeping with these findings, it was necessary to determine if the sensitivity of the various types of motor units tested may depend on the size of the unconditioned responses. While keeping the intensity of the median nerve constant at 0.8 MT, a total number of 10 motor units was tested using various radial nerve stimulation intensities. Five motor units characterized by low force thresholds (< 1 N), small macro-MUP areas and slow twitch contraction times and five others characterized by higher force thresholds (> 3 N), larger macro-MUP areas and shorter twitch contraction time were first tested using a radial nerve stimulation (stimulation 1), which was adjusted in order to induce an approximately 60–70 % test response probability of a very low threshold unit. The same motor units were then tested while changing the intensity of the radial nerve stimulation (stimulation 2). In the case of the low force threshold motor units, the intensity of the stimulation was reduced in order to obtain lower unconditioned responses, whereas in the case of the high force threshold motor units, the intensity of the stimulation was increased in order to obtain greater unconditioned responses.

Figure 5 illustrates the responses of a low and a high threshold motor unit tested under these conditions. In the case of the low threshold motor unit, applying median nerve stimulation decreased its unconditioned response by 58.2 % (0.67 impulses per trigger) while using stimulation 1 at 1.1 MT (Fig. 5A and B) and its unconditioned response by 54.2 % (0.48 impulses per trigger) while using stimulation 2 at 0.9 MT (Fig. 5C and D). In the case of the high threshold motor unit, applying median nerve stimulation decreased its unconditioned response by 23.8 % (0.42 impulses per trigger) while using stimulation 1 at 1.1 MT (Fig. 5E and F) and its unconditioned response by 25.4 % (0.59 impulses per trigger) while using stimulation 2 at 1.2 MT (Fig. 5G and H). Similar data were obtained on the eight other motor units tested under these conditions, as summarized in Table 2. These observations confirm that the strength of the Ia presynaptic inhibition may depend on the type of motor unit tested rather than on the size of the unconditioned responses of the motor units. The magnitude of the Ia presynaptic inhibition was found to be greater in the low force threshold motor units than in the high threshold motor units, whatever the size of their unconditioned response.

Figure 5. The size of the unconditioned response did not alter the strength of the Ia presynaptic inhibition.

Figure 5

Open in a new tab

While keeping the intensity of the median nerve constant at 0.8 MT, one low force threshold and one higher force threshold motor unit was tested either while using a constant radial nerve stimulation intensity (stimulation 1) or while adjusting the intensity of the radial nerve stimulation to obtain comparable size of unconditioned response (stimulation 2). In the case of the low threshold motor unit, applying median nerve stimulation decreased to a similar extent its unconditioned response while using stimulation 1 at 1.1 MT (A and B) or stimulation 2 at 0.9 MT (C and D). Likewise in the case of the high threshold motor unit, applying median nerve stimulation decreased to a similar extent its unconditioned response while using stimulation 1 (E and F) or stimulation 2 at 1.2 MT (G and H). These observations confirm that the strength of the Ia presynaptic inhibition may depend on the type of motor unit tested rather than on the size of the unconditioned responses of the motor units.

Table 2.

The effects of changing the intensity of the radial nerve stimulation

Low force threshold MUs High force threshold MUs
Test stimulation 1 1.1 × MT Test stimulation 2 0.9 × MT Test stimulation 1 1.1 × MT Test stimulation 2 1.2 × MT
Response probability without median nerve stimulation (impulses per trigger) 0.62 ± 0.09 0.41 ± 0.07 0.39 ± 0.09 0.58 ± 0.05
Response probability with median nerve stimulation (impulses per trigger) 0.24 ± 0.05 0.17 ± 0.04 0.27 ± 0.09 0.39 ± 0.05
Changes in response probability (%) −60.83 ± 4.43 −57.83 ± 4.84 −30.76 ± 14.86 −32.41 ± 9.48
Open in a new tab

Low force threshold MUs: n =5; force threshold, 0.71 ± 0.23 N; macro-MUP area, 0.37 ± 0.09 mV ms; CT, 59.95 ± 10.65 ms. High force threshold MUs: n = 5; force threshold, 3.57 ± 0.70 N; macro-MUP area, 0.62 ± 0.07 mV ms; CT, 32.42 ± 5.68 ms.

DISCUSSION

In human extensor carpi radialis muscles, the magnitude of the Ia presynaptic inhibition was investigated during voluntary contraction, depending on the type of motor unit tested. The results showed that the magnitude of the Ia presynaptic inhibition depended on the mechanical and electrophysiological properties of the motor units tested, increasing from the fast to the slow contracting motor units.

In humans, the reliability of motor unit functional characterization is known to depend on physiological parameters: the reliability of the data on twitch contraction time extracted using the spike-triggered averaging method can be affected by the synchronization of the motor unit discharges (Milner-Brown et al. 1973). In a previous study in which motor unit pairs were recorded in the extensor carpi radialis muscles (Schmied et al. 1994), a very small number of action potentials were found to occur synchronously without any consistent effects on the twitch extraction.

The size principle during the test stimulation

In the recordings in which the radial nerve was being stimulated alone, the reflex response tended to decrease as the motor unit's force threshold and macro-potential area increased, and as its twitch contraction time decreased. These findings are in keeping with the size principle established in studies on the triceps surae motoneurones of anaesthetised cats, in line with what was previously suggested by the results of motor unit recordings on the human second dorsal interosseous (Buller et al. 1980), soleus (Awiszus & Feistner, 1993) and extensor carpi radialis (Schmied et al. 1997a) muscles.

In other human studies dealing with the distribution of Ia monosynaptic inputs within the motoneurone pool, completely different data have been obtained on the tibialis anterior muscle (Semmler & Türker, 1994), the abductor digiti minimi muscle (Mazzocchio et al. 1995) and the masseter muscle (Scutter & Türker, 1998). In addition to methodological reasons, these discrepancies might also be attributable to the existence of muscle-related differences in the patterns of distribution of Ia monosynaptic EPSPs to the motoneurones (Mazzocchio et al. 1995), in the presynaptic inhibitory processes (Meunier & Morin, 1989; Schmied et al. 1997a) and in the post-synaptic inhibitory processes (Katz et al. 1993).

Evidence for the involvement of Ia presynaptic inhibition

The conditioning stimulation applied to the median nerve can be said to have preferentially activated the largest group I muscular afferents, i.e. the Ia muscle spindle afferents homonymous to the flexor carpi radialis muscle, since the stimulation intensity was set at 0.8 MT (see Burke et al. 1984). In human wrist flexor radialis muscle, the Ia presynaptic inhibition induced by activating the extensor Ia muscle spindle afferents has been previously reported to peak when a 20 ms conditioning-test interval was used (Berardelli et al. 1987). Burke et al. (1994) have reported that after applying conditioning stimulation to the median nerve, the Ia presynaptic inhibition depressed the amplitude of the H reflex recorded in the extensor carpi radialis muscles with a similar delay and efficiency in some subjects, in whom it was possible to elicit an H reflex in the wrist extensor muscles at rest.

In keeping with these previous observations, we made the assumption here that the decrease in the motor unit responses observed while stimulating the median nerve was due to Ia presynaptic inhibition. This assumption was strongly supported by the analysis of the single motor unit reflex responsiveness based on the post-stimulus time histograms, focusing in particular on the early monosynaptic component (the first 0.5 ms) of the reflex peak, which is known not to be contaminated by any polysynaptic events (Hultborn et al. 1987).

The magnitude of Ia presynaptic inhibition and the type of motor unit involved

In the recordings in which the median nerve was being stimulated, the responses of the motor units were significantly correlated with the functional parameters of the units. The higher the motor unit force threshold, the larger its macro-potential area became, and the shorter its contraction time, the greater the motor unit response probability was found to be. In the recordings with median nerve stimulation, there may therefore exist an upward gradient in the proprioceptive Ia monosynaptic assistance from which extensor motor units benefit, working from the slow to fast contracting motor units. The present data suggest that the median nerve stimulation may be able to compensate for the downward gradient in monosynaptic reflex excitation from ‘slow’ to ‘fast’ motor units, since the distribution of the monosynaptic Ia effects on the various types of motor unit tested was in keeping with the size principle in the recordings in which the radial nerve alone was stimulated.

The involvement of Ia presynaptic inhibition in the compensation for the downward gradient in monosynaptic reflex excitation from slow to fast contracting motor units, was confirmed by the results of the regression analyses, which showed that the changes in the purely monosynaptic components of the motor units’ responses were correlated with their functional parameters. The Ia presynaptic inhibition has been suggested in simulation studies to alter the relationships between the reflex output and the level of excitation of a motoneurone pool (Capaday & Stein, 1987b). Another simulation study has strengthened the idea that synaptic effects might act differentially on the various types of cells which compose a given motoneurone pool (Kernell & Hultborn, 1990). The present data are in keeping with the result of these two studies, since the slow contracting motor units strongly affected by the Ia presynaptic inhibition turned out to be less responsive to the radial nerve stimulation than the fast contracting motor units, which were less strongly affected by Ia presynaptic inhibition. It is worth noting that the changes in the gain with which the Ia inputs recruited the motoneurones in the wrist extensor pool caused by Ia presynaptic inhibition were consistently observed in pairs consisting of one fast and one concurrently tested slow contracting motor unit. These findings suggest that the Ia presynaptic inhibition may affect the gain with which the Ia inputs recruited the wrist extensor motoneurone, depending on the type of motor unit tested, rather than on the motoneurone pool excitatory drive.

The results of previous animal studies have shown that the presynaptic inhibition of Ia EPSPs is greater in some motoneurones than in others. In the cat hindlimb motor nuclei, about one-fourth of the motoneurones were not sensitive to Ia presynaptic inhibition after conditioning stimulation was applied to the posterior biceps-semitendinous nerve (Cook & Cangliano, 1972). Zengel et al. (1983) have observed that in the cat triceps surae muscles, the Ia presynaptic inhibition affected the slow contracting motor units, corresponding to motoneurones with large EPSPs, more strongly than the fast contracting motor units, corresponding to motoneurones with small EPSPs. In the latter study, however, the magnitude of the Ia presynaptic inhibition, as estimated in terms of the percentage change in the amplitude of the Ia EPSPs, was found to be around 25 % in all the various types of motor units tested. In our study, the proportional attenuation in the magnitude of presumed Ia inhibition decreased from slow- to fast-twitch motor units. Besides differences between the species studied, it is worth noting that the present experiments were performed during voluntary contractions in which a tonic presynaptic inhibition may interact with the presynaptic inhibition induced by the median nerve stimulation, whereas Zengel et al. (1983) recorded the activity of motoneurones in deeply anaesthetized cats.

In the present study, the upward gradient in the magnitude of the Ia presynaptic inhibition was strong enough to actually compensate for the downward gradient in monosynaptic reflex excitation from ‘slow’ to ‘fast’ motor units, since the ‘slow’ units were less responsive in the recordings with median nerve stimulation than the ‘fast’ units.

It is worth noting that the strength of the Ia presynaptic inhibition was not found to depend on the size of the unconditioned responses of the motor units to the radial nerve stimulation. Changing the intensity of the radial nerve stimulation clearly changed the size of the unconditioned responses but did not alter the efficiency of the Ia presynaptic inhibition induced by stimulating the median nerve. This is in line with the results of previous studies in which the Ia presynaptic inhibition was estimated by assessing changes in the H-reflex amplitude (see Crone et al. 1990). The small H reflexes, in which ‘slow’ units are preferentially recruited, were strongly depressed by Ia presynaptic inhibition, whereas the large H reflexes, in which ‘fast’ units are preferentially recruited, were weakly depressed by Ia presynaptic inhibition.

Functional implications

The Ia presynaptic inhibition induced by electrically stimulating the antagonist group I afferents was found here to selectively inhibit single motor units tested in the human wrist extensor muscles, depending on the type of motor unit involved. The present data suggest for the first time in humans that there may exist an upward gradient in Ia presynaptic inhibition, working from ‘fast’ to ‘slow’ motor units. These effects were consistently observed in pairs of motor units consisting of one low threshold and one concurrently tested high threshold motor unit. This difference cannot therefore be explained in terms of the motoneurone excitatory drive. The magnitude of the Ia presynaptic inhibition might thus depend on the type of motor unit involved.

A weaker Ia presynaptic inhibition acting on the fast contracting motor units – those producing the greatest twitch forces – might contribute to enhancing the efficiency of the proprioceptive assistance. This assumption is strengthened by the fact that the Ia presynaptic inhibition was found here to be able to compensate for the downward gradient in monosynaptic reflex excitation from slow to fast motor units. The Ia presynaptic inhibition may thus be able to selectively modulate the proprioceptive assistance to specific types of motoneurones homonymous to the wrist myotatic unit, thus contributing to adapting sensory feedback activity to fit the requirements of the ongoing motor task.

Acknowledgments

We are grateful to Dr J. Blanc for correcting the English manuscript. This research was supported by grants from the Association Française contre les Myopathies (A.F.M.), the Fondation pour la Recherche Médicale (F.R.M.), and the Direction des Recherches, Etudes et Techniques (D.R.E.T.-D.G.A.).

References

  1. Aimonetti JM, Vedel JP, Schmied A, Pagni S. Mechanical cutaneous stimulation alters Ia presynaptic inhibition in human wrist extensor muscles: a single motor unit study. The Journal of Physiology. 2000;522:137–145. doi: 10.1111/j.1469-7793.2000.0137m.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Awiszus F, Feistner H. The relationship between estimates of Ia-EPSP amplitude and conduction velocity in human soleus motoneurons. Experimental Brain Research. 1993;95:365–370. doi: 10.1007/BF00229795. [DOI] [PubMed] [Google Scholar]
  3. Berardelli A, Day BL, Marsden CD, Rothwell JC. Evidence favouring presynaptic inhibition between antagonist muscle afferents in the human forearm. The Journal of Physiology. 1987;391:71–83. doi: 10.1113/jphysiol.1987.sp016726. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Buller NP, Garnett R, Stephens JA. The reflex responses of single motor units in human hand muscles following muscle afferent stimulation. The Journal of Physiology. 1980;303:337–349. doi: 10.1113/jphysiol.1980.sp013289. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Burke D, Gandevia SC, Mckeon B. Monosynaptic and oligosynaptic contributions to human ankle jerk and H-reflex. Journal of Neurophysiology. 1984;52:435–448. doi: 10.1152/jn.1984.52.3.435. [DOI] [PubMed] [Google Scholar]
  6. Burke D, Gracies JM, Mazevet D, Meunier S, Pierrot-Deseilligny E. Non-monosynaptic transmission of the cortical command for voluntary movement in man. The Journal of Physiology. 1994;480:191–202. doi: 10.1113/jphysiol.1994.sp020352. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Capaday C, Stein RB. Difference in the amplitude of the human soleus H reflex during walking and running. The Journal of Physiology. 1987a;392:513–522. doi: 10.1113/jphysiol.1987.sp016794. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Capaday C, Stein RB. A method for simulating the reflex output of a motoneurone pool. Journal of Neuroscience Methods. 1987b;21:91–104. doi: 10.1016/0165-0270(87)90107-5. [DOI] [PubMed] [Google Scholar]
  9. Cook WA, Cangliano A. Presynaptic and postsynaptic inhibition of spinal neurons. Journal of Neurophysiology. 1972;35:389–403. doi: 10.1152/jn.1972.35.3.389. [DOI] [PubMed] [Google Scholar]
  10. Crone C, Hultborn H, Mazieres L, Morin C, Nielsen J, Pierrot-Deseilligny E. Sensitivity of monosynaptic test reflexes to facilitation and inhibition as a function of the test reflex size: a study in man and the cat. Experimental Brain Research. 1990;81:35–45. doi: 10.1007/BF00230098. [DOI] [PubMed] [Google Scholar]
  11. Ellaway PH. Cumulative sum technic and its application to the analysis of peristimulus time histograms. EEG and Clinical Neurophysiology. 1978;45:302–304. doi: 10.1016/0013-4694(78)90017-2. [DOI] [PubMed] [Google Scholar]
  12. Faist M, Dietz V, Pierrot-Deseilligny E. Modulation, probably presynaptic in origin, of monosynaptic Ia excitation during human gait. Experimental Brain Research. 1996;109:441–449. doi: 10.1007/BF00229628. [DOI] [PubMed] [Google Scholar]
  13. Frank K, Fuortes MGF. Presynaptic and postsynaptic inhibition of monosynaptic reflexes. Federation Proceedings. 1957;16:39–40. [Google Scholar]
  14. Henneman E. Functional organization of motoneurons pools: The size principle. Proceedings of International Union Physiological Science. 1977;12:50. [Google Scholar]
  15. 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. Experimental Brain Research. 1996;108:450–462. doi: 10.1007/BF00227268. [DOI] [PubMed] [Google Scholar]
  16. Hultborn H, Meunier S, Morin C, Pierrot-Deseilligny E. Assessing changes in presynaptic inhibition of Ia fibres: a study in man and the cat. The Journal of Physiology. 1987;389:729–756. doi: 10.1113/jphysiol.1987.sp016680. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Katz R, Mazzocchio R, Pénicaud A, Rossi A. Distribution of recurrent inhibition in the human upper limb. Acta Physiologica Scandinavica. 1993;149:183–198. doi: 10.1111/j.1748-1716.1993.tb09611.x. [DOI] [PubMed] [Google Scholar]
  18. Kernell D, Hultborn H. Synaptic effects on recruitment gain: a mechanism of importance for the input-output of motoneurone pools? Brain Research. 1990;507:176–179. doi: 10.1016/0006-8993(90)90542-j. [DOI] [PubMed] [Google Scholar]
  19. Lomeli J, Quevedo J, Linares P, Rudomín P. Local control of information flow in segmental and ascending collaterals of single afferents. Nature. 1998;395:600–604. doi: 10.1038/26975. [DOI] [PubMed] [Google Scholar]
  20. Mazzocchio R, Rothwell JC, Rossi A. Distribution of Ia effects onto human hand muscle motoneurones as revealed using an H-reflex technique. The Journal of Physiology. 1995;489:263–273. doi: 10.1113/jphysiol.1995.sp021048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Meunier S, Morin C. Changes in presynaptic inhibition of Ia fibres to soleus motoneurones during voluntary dorsiflexion of the foot. Experimental Brain Research. 1989;76:510–518. doi: 10.1007/BF00248907. [DOI] [PubMed] [Google Scholar]
  22. Milner-Brown HS, Stein RB, Yemm R. The contractile properties of human motor units during voluntary isometric contractions. The Journal of Physiology. 1973;228:285–306. doi: 10.1113/jphysiol.1973.sp010087. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Nielsen J, Kagamihara Y. The regulation of presynaptic inhibition during co-contraction of antagonistic muscles in man. The Journal of Physiology. 1993;464:575–593. doi: 10.1113/jphysiol.1993.sp019652. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Rudomín P, Núñez R, Madrid J. Modulation of synaptic effectiveness of Ia and descending fibers in cat spinal cord. Journal of Neurophysiology. 1975;38:1181–1195. doi: 10.1152/jn.1975.38.5.1181. [DOI] [PubMed] [Google Scholar]
  25. Schmied A, Morin D, Vedel JP, Pagni S. The ‘size principle’ and synaptic effectiveness of muscle afferent projections to human extensor carpi radialis motoneurones during wrist extension. Experimental Brain Research. 1997a;113:214–229. doi: 10.1007/BF02450320. [DOI] [PubMed] [Google Scholar]
  26. Schmied A, Vedel JP, Calvin-Figuière S, Rossi-Durand C, Pagni S. Task-dependence of muscle afferent monosynaptic inputs to human extensor carpi radialis motoneurones. Electroencephalography and Clinical Neurophysiology. 1997b;105:220–234. doi: 10.1016/s0924-980x(97)00015-5. [DOI] [PubMed] [Google Scholar]
  27. Schmied A, Vedel JP, Pagni S. Human spinal lateralization assessed from motoneurone synchronization: dependence on handedness and on motor unit type. The Journal of Physiology. 1994;480:369–387. doi: 10.1113/jphysiol.1994.sp020367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Scutter SD, Türker KS. Ia afferent distribution to motoneurons in human masseter. Society for Neuroscience Abstracts. 1998;24:151. [Google Scholar]
  29. Semmler JG, Türker KS. Compound group I excitatory input is differentially distributed to motoneurons of the human tibialis anterior. Neuroscience Letters. 1994;178:206–210. doi: 10.1016/0304-3940(94)90760-9. [DOI] [PubMed] [Google Scholar]
  30. Stein RB, French AS, Mannard A, Yemm R. New methods for analysing motor function in man and animals. Brain Research. 1972;40:187–192. doi: 10.1016/0006-8993(72)90126-6. [DOI] [PubMed] [Google Scholar]
  31. Wood SA, Gregory JE, Proske U. The influence of muscle spindle discharge on the human H reflex and the monosynaptic reflex in the cat. The Journal of Physiology. 1996;497:279–290. doi: 10.1113/jphysiol.1996.sp021767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Zengel JE, Reid SA, Sypert GW, Munson JB. Presynaptic inhibition, EPSP amplitude, and motor unit type in triceps surae motoneurons in the cat. Journal of Neurophysiology. 1983;49:922–931. doi: 10.1152/jn.1983.49.4.922. [DOI] [PubMed] [Google Scholar]

Articles from The Journal of Physiology are provided here courtesy of The Physiological Society

RESOURCES