Skip to main content
NCBI home page
The Journal of Physiology logoLink to The Journal of Physiology
. 2016 Apr 24;594(10):2707–2717. doi: 10.1113/JP272164

Unexpected factors affecting the excitability of human motoneurones in voluntary and stimulated contractions

Serajul I Khan 1, Janet L Taylor 1, Simon C Gandevia 1,✉
PMCID: PMC4865580  PMID: 26940402

Abstract

Key points

  • The output of human motoneurone pools decreases with fatiguing exercise, but the mechanisms involved are uncertain. We explored depression of recurrent motoneurone discharges (F‐waves) after sustained maximal voluntary contractions (MVCs).

  • MVC depressed the size and frequency of F‐waves in a hand muscle but a submaximal contraction (at 50% MVC) did not.

  • Surprisingly, activation of the motoneurones antidromically by stimulation of the ulnar nerve (at 20 or 40 Hz) did not depress F‐wave area or persistence.

  • Furthermore, a sustained (3 min) MVC of a hand muscle depressed F‐waves in its antagonist but not in a remote hand muscle.

  • Our findings suggest that depression of F‐waves after voluntary contractions is not simply due to repetitive activation of the motoneurones but requires descending voluntary drive.  Furthermore, this effect may depress nearby, but not distant, spinal motoneurone pools.

Abstract

There are major spinal changes induced by repetitive activity and fatigue that could contribute to ‘central’ fatigue but the mechanisms involved are poorly understood in humans. Here we confirmed that the recurrent motoneuronal discharge (F‐wave) is reduced during relaxation immediately after a sustained maximal voluntary contraction (MVC) of an intrinsic hand muscle (abductor digiti minimi, ADM) and explored the relationship between motoneurone firing and the depression of F‐waves in three ways. First, the depression (in both F‐wave area and F‐wave persistence) was present after a 10 s MVC (initial decrease 36.4 ± 19.1%; mean ± SD) but not after a submaximal voluntary contraction at 50% maximum. Second, to evoke motoneurone discharge without volitional effort, 10 s tetanic contractions were produced by supramaximal ulnar nerve stimulation at the elbow at physiological frequencies of 25 and 40 Hz. Surprisingly, neither produced depression of F‐waves in ADM to test supramaximal stimulation of the ulnar nerve at the wrist. Finally, a sustained MVC (3 min) of the antagonist to ADM (4th palmar interosseous) depressed F‐waves in the anatomically close ADM (20 ± 18.2%) but not in the more remote first dorsal interosseous on the radial side of the hand. We argue that depression of F‐waves after voluntary contractions may not be due to repetitive activation of the motoneurones but requires descending voluntary drive. Furthermore, this effect may depress nearby, but not distant, spinal motoneurone pools and it reveals potentially novel mechanisms controlling the output of human motoneurones.

Key points

  • The output of human motoneurone pools decreases with fatiguing exercise, but the mechanisms involved are uncertain. We explored depression of recurrent motoneurone discharges (F‐waves) after sustained maximal voluntary contractions (MVCs).

  • MVC depressed the size and frequency of F‐waves in a hand muscle but a submaximal contraction (at 50% MVC) did not.

  • Surprisingly, activation of the motoneurones antidromically by stimulation of the ulnar nerve (at 20 or 40 Hz) did not depress F‐wave area or persistence.

  • Furthermore, a sustained (3 min) MVC of a hand muscle depressed F‐waves in its antagonist but not in a remote hand muscle.

  • Our findings suggest that depression of F‐waves after voluntary contractions is not simply due to repetitive activation of the motoneurones but requires descending voluntary drive.  Furthermore, this effect may depress nearby, but not distant, spinal motoneurone pools.

Abbreviations

ADM

abductor digit minimi

CMEP

cervicomedullary motor evoked potential

EMG

electromyographic activity

FDI

first dorsal interosseous

5‐HT

5‐hydroxytryptamine (serotonin)

Mmax

maximal compound muscle action potential

MVC

maximal voluntary contraction

4PI

fourth palmar interosseous

Introduction

The output of human motoneurone pools decreases during and after sustained voluntary isometric fatiguing contractions and the major mechanisms for this reduction include both supraspinal and spinal factors (Gandevia, 2001). Some fatigue‐related changes at the spinal motoneurones have been demonstrated with subcortical stimulation of the corticospinal tract at the level of the cervicomedullary junction (Gandevia, 2001; Butler et al. 2003). Cervicomedullary motor evoked potentials (CMEPs) are reduced in size during the latter half of a sustained 2 min maximal voluntary contraction (MVC) (Butler et al. 2003; Martin et al. 2006). More dramatic changes in CMEPs occur when a transcranial magnetic stimulus to the motor cortex is used to interrupt descending voluntary drive and cause transient disfacilitation of the motoneurone pool. During this transient disfacilitation, CMEPs are easily evoked prior to fatigue but are abolished after ∼20 s of maximal voluntary effort (McNeil et al. 2009, 2011 a,b). These changes in the CMEP during an MVC suggest altered motoneurone excitability which may reflect changes in the intrinsic properties of the motoneurones. However, changes in the size of the CMEP do not necessarily correspond to changes in the excitability of motoneurones because the CMEP can also be influenced by premotoneuronal changes presumably at the corticomotoneuronal synapse (Gandevia et al. 1999; Petersen et al. 2003).

In an attempt to assess changes produced in motoneurones more directly, we have previously measured the recurrent discharge of motoneurones (Renshaw, 1941; Eccles, 1955; McLeod & Wray, 1966; Trontelj, 1973), following repetitive activity produced by MVCs. These discharges are known as F‐waves when measured in the electromyogram in humans (McLeod & Wray, 1966; see also Espiritu et al. 2003). Animal studies indicate that when motoneurones are stimulated antidromically the recurrent discharge is generated in the non‐myelinated proximal segment of the axon initial segment (Coombs et al. 1957) or first node of Ranvier (Gogan et al. 1984). Our study showed that F‐waves were strongly depressed after MVCs in both an upper and a lower limb muscle and, furthermore, the depression was activity‐dependent, being greater after longer efforts (Khan et al. 2012). Reductions in F‐waves after MVCs were also reported by Rossi et al. (2012). We proposed that activity‐dependent hyperpolarisation at the soma or the axon initial segment depresses the recurrent discharge of human motoneurones. This proposal fits with many studies showing that repetitive activity hyperpolarises axons (e.g. Gasser, 1935; Bostock & Grafe, 1985; Kiernan et al. 2004; Milder et al. 2014), probably via altered activity of the Na+–K+ pump.

The aim of the current studies was to test the link between repetitive activity of motoneurones and depression of F‐waves in three ways. First, we determined whether the strength of a voluntary contraction of an intrinsic hand muscle affects the depression of F‐waves. For these muscles a voluntary contraction beyond about 50% maximum recruits most motoneurones so that a stronger contraction requires an increase in the discharge frequency of motoneurones (e.g. Kukulka & Clamann, 1981; De Luca et al. 1982). This would be expected to produce greater axonal hyperpolarisation and hence F‐wave depression. Second, we predicted that antidromic activation of motoneurones by nerve stimulation would produce similar depression in F‐waves to that observed after voluntary contractions, and that the depression would increase with the frequency of the tetanic stimulus. Finally, because we found that antidromic activation of motoneurones (unlike voluntary activity) did not depress F‐waves, we tested whether strong voluntary contraction of an anatomically ‘nearby’ muscle would depress F‐waves in the target muscle. A preliminary version of the results has been presented (Gandevia et al. 2014).

Methods

A total of 53 healthy subjects (23 females) participated in the study. Fifteen subjects (31 ± 7 years; mean ± SD; nine females) participated in the first study which examined the effect of contraction strength (50 and 100% MVC) on F‐wave depression in abductor digiti minimi (ADM). Seven of these subjects plus a further five (33 ± 6 years; seven females) participated in the second study which examined whether the post‐exercise depression of F‐waves required activation of the motoneurones through voluntary drive. Twenty‐seven subjects [ten from study 1 plus a further 17 (29 ± 7 years; ten females)] initially participated in the third study which examined F‐waves in ADM after a 3 min MVC of the fourth palmar interosseous (4PI) which adducts the little finger and is the functional antagonist to ADM. As indicated below, based on performance on the training and experiment days only 19 subjects were included in the analysis. Written informed consent to the experimental procedures was obtained from each subject. The study was approved by the local human research ethics committee and was conducted according to the Declaration of Helsinki (2008).

Experimental setup

In Studies 1 and 3, subjects sat in an adjustable chair with the right forearm and fingers rested on a table and the elbow flexed in a comfortable position. The forearm was pronated and secured by straps. The thumb rested in a midflexed position and the second to fourth fingers were secured by a strap. The little finger was placed firmly in a custom‐designed finger splint with an axis of rotation about the metacarpophalangeal joint in a horizontal plane. The splint was connected to a force transducer to measure isometric finger abduction and adduction force.

In Study 2, subjects were seated with their right forearm rested in a neutral position on a padded L‐shaped support which was affixed to a table and supported the right forearm and hand from the elbow to the fingertips. The forearm was secured tightly to the support with a strap at the elbow and the wrist to prevent any movement from stimulation. The thumb rested in a neutral position and a vacuum pillow secured the other fingers. A padded clamp on the wrist prevented ulnar deviation during tetanic stimulation.

Surface electromyographic activity (EMG) was recorded from ADM or both ADM and first dorsal interosseous (FDI) using Ag/AgCl disc electrodes (6 mm in diameter) filled with conducting gel. Recordings from FDI were required in Study 3 as we needed to record from a ‘remote’ ulnar‐innervated muscle on the thenar side of the hand. The cathode was placed over the motor‐point of each muscle and the anode 2 cm distal to the motor‐point. A large ground electrode was placed on the anterior aspect of the wrist. The EMG signals were recorded in two ways. First, EMG signals were amplified (×1000), band‐pass filtered (200–1000 Hz; two‐pole Bessel filter; CED 1902 amplifiers, Cambridge Electronic Design Ltd, Cambridge, UK) and sampled at 5 kHz (Fig. 1, lower panel; see also fig. 1 in Khan et al. 2012). The high‐pass filter was set at 200 Hz to remove the low frequency tail of the M wave to obtain a more stable baseline, so that it was easier to identify the onset of individual F‐waves and to measure their amplitude and area (filtered F‐waves; Fig. 1 B). Measurements made in this way are reported in the text and figures. Second, so that the M wave could also be assessed in the conventional way, the EMG signals were amplified (×300), band‐pass filtered (10–1000 Hz; two‐pole Butterworth filter; CED 1902 amplifiers) and sampled at 2 kHz (Fig. 1, top panel). The experimental interventions in the three studies (described below) did not significantly change the duration of the maximal M wave (data not shown). For Study 3, we quantified co‐contraction of ADM and FDI during the sustained MVC of 4PI using the EMG amplitudes derived from this less filtered signal.

Figure 1. ‘Unfiltered’ and ‘filtered’ F‐waves in ADM after a 10 s maximal voluntary contraction of ADM .

Figure 1

Open in a new tab

A, superimposed F‐waves (n = 30) recorded from ADM with a wide bandpass of 10–1000 Hz (‘unfiltered’ F‐waves) at rest before (left panel) and immediately after (right panel) a 10 s MVC. B, the same superimposed F‐waves heavily high‐pass filtered (200–1000 Hz) before contraction (left panel) and immediately after (right panel) the contraction. The rectangular boxes enclose the F‐waves that were used in the analysis.

Force data were sampled at 1000 Hz. All signals were stored on a computer (CED 1401 Plus) and Spike 2 software (version 6.13; CED).

Stimulation

In all studies, the ulnar nerve at the wrist was stimulated to evoke maximal compound muscle action potentials (M max) and F‐waves at rest in ADM. The best location for the cathode was identified by stimulation with a hand‐held electrode on the ulnar side of the forearm. Single electrical stimuli (100 μs duration) were delivered by a constant current stimulator (DS7AH, Digitimer Ltd, Welwyn Garden City, UK) through a custom‐designed bar electrode (interelectrode distance 4 cm) with the cathode over the ulnar nerve and just proximal to the wrist. With the subject at rest, the stimulus intensity was gradually increased until the compound muscle action potential failed to increase in peak‐to‐peak amplitude despite an increase in current. To maintain supramaximal stimulation despite any effect of activity‐induced hyperpolarisation of the motor axons during a sustained contraction, the intensity of stimulation to evoke F‐waves was set at 150% of the intensity that evoked M max (20–45 mA).

In Study 2, the ulnar nerve at the elbow was also stimulated using a constant‐current stimulator (Digitimer DS7AH) to activate ADM. Electrical stimuli (500 μs duration) were delivered through Ag–AgCl electrodes with the cathode and anode proximal and distal to the medial epicondyle of the humerus, respectively. The best position was identified using a hand‐held electrode to minimise stimulus current require to evoke the maximal compound muscle action potential (M max). The size of M max was determined and the intensity of stimulation was set at 150% of the intensity required to produce a maximal M‐wave (15–30 mA). Stimulation at the elbow was used (see below) because current levels for maximal stimulation were much lower at this site than at the wrist and hence prolonged tetanic stimulation was more comfortable.

Protocol

Study 1: F‐waves in ADM after 10 s submaximal and maximal contractions

This assessed the effect of the strength of a voluntary contraction of ADM on F‐wave depression. Initially, subjects performed three brief MVCs of 2–3 s each separated by at least 1 min of rest. The peak force of the three MVCs was measured and a target level of the rectified integrated EMG required to reach 50% MVC was set on a visual feedback display. Two control sets of F‐waves, separated by 1 min, were collected prior to the voluntary contraction. For each set, subjects received 30 supramaximal stimuli (150% M max) to the ulnar nerve at the wrist delivered at 0.5 Hz with the muscle relaxed. After another 1 min of rest, subjects performed a 10 s submaximal voluntary contraction to the target level (50% MVC). They relaxed immediately after the contraction, and starting after ∼ 2 s of relaxation, six post‐contraction sets of F‐waves, each separated by 30‐s rest, were collected. After a rest of 30 min, the protocol was repeated with a 100% MVC.

Study 2: F‐waves in ADM after a 10 s supramaximal tetanic stimulation of ulnar nerve

This assessed whether the depression of F‐waves required activation of the motoneurones through voluntary drive. F‐waves were examined in relaxation before and after supramaximal tetanic stimulation of ulnar nerve at the elbow for 10 s at 25 and 40 Hz. In brief, F‐waves were collected with sets of 30 supramaximal ulnar nerve stimuli (150% M max) at the wrist delivered at 0.5 Hz (i.e. as in Study 1). Initially, two control sets were collected, with a 1 min rest between sets. After 25 Hz tetanus subjects relaxed immediately, and six more sets of F‐waves were collected. The protocol was repeated with a 40 Hz tetanus after a rest of 30 min.

Study 3: F‐waves in ADM and FDI after contraction of the fourth palmar interosseous

Because the tetanic stimulation did not depress F‐wave area or persistence despite repetitive antidromic activation of all motoneurones supplying ADM (see Results for Study 2), we further explored the effects of voluntary activity on F‐waves. We measured F‐waves in ADM and in a remote ulnar‐innervated hand muscle, the FDI, after a sustained 3 min MVC of the fourth palmar interosseous (ADM's antagonist). There was a training and an experimental session, separated by at least 2 days. In the training session, 27 subjects practised maximal sustained adductions of the little finger so that co‐contraction of the functional antagonist (ADM) was low. This was judged from the surface EMG (root mean square with 100 ms time constant) with a level ≤ 10% that in a maximal contraction of ADM considered acceptable. Twenty‐three subjects were able to perform the task and continued to the experimental session, in which F‐waves were collected from ADM and FDI before and after a sustained 3 min MVC of the fourth palmar interosseous. As in Study 1, sets of 30 supramaximal ulnar nerve stimuli (150% M max) at the wrist were delivered at 0.5 Hz. Two control sets were collected before and ten sets after the sustained MVC. At the end of the protocol, subjects performed brief isometric MVCs of ADM and FDI. Data from 19 subjects were used for ADM: two subjects whose ADM EMG exceeded 10% that in a maximal contraction of ADM were excluded and two subjects withdrew as the supramaximal stimuli were too uncomfortable. Data from 15 subjects were used for FDI as four subjects had low persistence of F‐waves in this muscle (less than 50%).

Data analysis and statistics

During off‐line analysis, Signal software (version 3.05; Cambridge Electronic Design) was used to determine all measures (area, amplitude and persistence). To measure the area and the amplitude of F‐waves, responses to 30 supramaximal stimuli were overdrawn on the screen at high gain and the cursors were set at the beginning and the end of the responses that showed a clear deflection from the baseline. Maximal compound muscle action potentials (M max) were superimposed, and cursors were set at the start and the end of the responses to determine the area and peak‐to‐peak amplitude. For both F‐ waves and M max, measurements were made from individual potentials. To identify any EMG activity at rest, root mean square EMG was calculated over 50 ms immediately before each stimulus. The area and peak‐to‐peak amplitude of F‐waves measured from the heavily filtered EMG showed similar results, and area is reported throughout the text. Because of activity‐dependent changes in muscle fibre action potentials during voluntary efforts, the area of each F‐wave was normalised to the area of the corresponding M max (also heavily filtered) and then expressed as a percentage of mean control values (mean of control set 1 and 2) obtained prior to the MVC (Taylor et al. 1999; Khan et al. 2012). For each subject, the mean area of the 30 F‐waves in each set was used for statistical analysis. To illustrate in more detail the time course of changes, figures show the F‐wave area averaged for each ten consecutive stimuli (Espiritu et al. 2003). In addition, as F‐waves are not seen after every supramaximal stimulus, we measured the frequency with which F‐waves occurred in each set of 30 consecutive stimuli for ADM. This value (usually termed ‘persistence’) was expressed as a percentage. Assessment based on 30 trials can be considered sufficient (reviewed by Panayiotopoulos & Chroni, 1996). We identified an F‐wave as present if a response with an appropriate latency (minimum of 22 ms for ADM) had an amplitude ≥ 20 μV. Standard F‐wave characteristics at baseline from subjects used in studies 1, 2 and 3 are given in Table 1.

Table 1.

Baseline F‐wave parameters pooled across studies for ADM and FDI

Mean onset Mean amplitude Mean area Chronodispersion F/M amp
latency (ms) (mV) (μV.s) Persistence (%) (ms) ratio (%)
ADM (n = 46) 27.3 ± 1.8 0.54 ± 0.27 7.66 ± 3.5 91.7 ± 7.9. 3.2 ± 0.8 4.4 ± 1.7
FDI (n = 15) 29.1 ± 2.6 0.22 ± 0.07 3.95 ± 2.1 70.2 ± 13.6 3.5 ± 0.9 1.7 ± 0.6
Open in a new tab

Baseline data were obtained using the wide band pass (see Methods). Mean onset latency was determined for each subject (from two baseline sets of 30 stimuli). Chronodispersion is the time difference between the minimum and maximal onset latency. F/M amp ratio is the amplitude of the F‐wave relative to the amplitude of Mmax.

Statistical comparisons were made using repeated‐measures ANOVA (SYSTAT version 13; Systat Software Inc., San Jose, CA). For studies 1 and 2, F‐wave area was analysed with two‐way repeated measures ANOVA with time as one factor and contraction strength as the other. Because data for F‐wave persistence were not normally distributed, we rank‐transformed the data and performed a two‐way parametric repeated measures ANOVA. Whenever the ANOVA showed a significant main effect or a significant interaction, post hoc Student–Newman–Keuls tests were used to identify differences between the control and subsequent time points or between contractions. For Study 3, F‐wave area and persistence were analysed using one‐way repeated measures ANOVA to assess the main effect of time. These were followed by post hoc Student–Newman–Keuls tests to identify differences between the control and subsequent time points. Data are represented as mean ± SD in the text and as mean ± SEM in the figures. Statistical significance was set at P < 0.05.

Results

Study 1: F‐waves at rest after submaximal and maximal voluntary contractions of ADM

Stimulation of the ulnar nerve evoked F‐waves in the relaxed ADM muscle before and after a voluntary contraction of ADM at 50 and 100% maximum. Figure 2 A shows superimposed raw traces for a single subject. For the group of subjects (n = 15), F‐wave area showed a significant effect for contraction strength (F 1,14 = 5.270, P = 0.038), time (F 7,98 = 5.982, P < 0.001) and an interaction (F 7,98 = 2.610, P = 0.016). After a 10 s MVC, the F‐wave area initially decreased by 36.4 ± 19.1% of the control (mean ± SD), and post hoc testing showed that the depression lasted for ∼ 9 min (P < 0.017; Fig.2 B). Unlike the 10 s MVC, after a submaximal contraction at 50% maximum, F‐wave area was not depressed significantly (F 7,98 = 2.198, P = 0.091). Comparison of corresponding time points of the two contractions showed significant differences from immediately after the MVCs to ∼4 min (set 1, 2 and 3) and at 7 min (set 5) after contraction (P < 0.036).

Figure 2. F‐waves in ADM after submaximal and maximal contraction of ADM .

Figure 2

Open in a new tab

A, F‐waves recorded from ADM at rest before and after a 10 s 50% MVC (left) and 100% MVC (right) for a single subject. In each panel, top traces represent one set of F‐waves recorded before the MVC and are shown superimposed. Lower traces show superimposed responses for the first and second sets after the contractions. Each set consists of 30 F‐waves. Vertical calibration, 0.2 mV; horizontal calibration, 10 ms. F‐wave traces start at 23 ms after the stimulus. B, group data (n = 15; mean ± SEM) showing changes in the area of F‐waves after submaximal (open circles) and maximal sustained contraction for 10 s (filled circles). In this and subsequent figures, F‐wave area was corrected to the respective M max response and then normalised to the mean value of the control sets (C1 and C2). For illustration, each circle represents the average area of ten consecutive F‐waves from each subject. For statistical tests, F‐waves were grouped into sets of 30. C, group data showing F‐wave persistence in the relaxed muscle before and after submaximal (open circle) and maximal (filled circle) voluntary contraction lasting for 10 s. The persistence was calculated from a set of 30 consecutive stimuli. The shaded vertical bar denotes the sustained voluntary contraction (VC). The asterisks (*) indicate data points that are significantly lower than the baseline values after a 10 s MVC (P < 0.05). The daggers (†) indicate significant differences between contractions at specific times (P < 0.05).

F‐wave persistence (the frequency of occurrence of F‐waves) was calculated for each set of 30 consecutive stimuli. Baseline persistence in ADM was high and reproducible prior to contraction (average baseline persistence was 95.8 ± 4.5 and 96.9 ± 3.9% before maximal and submaximal contractions, respectively; Fig. 2 C). Two‐way ANOVA showed a significant effect of contraction strength (F 1,14 = 6.659, P = 0.023), and time (F 7,98 = 5.203, P < 0.001), but no interaction (F 7,98 = 1.576, P = 0.152). F‐wave persistence declined by 19.2 ± 16.3% of control after a 10 s MVC and on post hoc testing remained significantly decreased for ∼3 min (P < 0.001; Fig. 2 C). In contrast, F‐wave persistence was not reduced after a 50% MVC (P = 0.525; Fig. 2 C). Comparison of corresponding time points of the two contractions showed significant differences from immediately after the MVCs to ∼3 min (P < 0.023).

Study 2: F‐waves at rest after a 25 and 40 Hz tetanus produced by ulnar stimulation

In Study 2, we examined whether the post‐exercise depression of F‐waves required activation of the motoneurones through voluntary drive. F‐waves were evoked in ADM before and after supramaximal tetanic stimulation of the ulnar nerve for 10 s at 25 and 40 Hz. Baseline persistence in ADM was high and stable prior to the prolonged tetanus (average baseline persistence was 93.6 ± 7.6 and 93.9 ± 8.5%; Fig. 3 C). Superimposed F‐waves before and after the tetanus from a typical subject are shown in Fig. 3 A. This antidromic activation of the ADM motoneurone pool did not depress F‐wave area or persistence (Fig. 3 B and C). For the group of subjects (n = 12), two‐way ANOVA for area showed no effects of frequency (F 1,11 = 2.09, P = 0.176), time (F 7,77 = 2.636, P = 0.065), nor an interaction (F 7,77 = 1.161, P = 0.335). Similarly, analysis of persistence showed no effects (P > 0.05).

Figure 3. F‐waves in ADM after supramaximal tetanic stimulation of the ulnar nerve .

Figure 3

Open in a new tab

A, F‐waves recorded from ADM in relaxation before and after supramaximal tetanic stimulation of ulnar nerve at the elbow for 10 s at 25 Hz (left) and 40 Hz (right) from a single subject. In each panel, top traces represent F‐waves recorded before the tetanus and are shown superimposed. Lower traces show superimposed responses for the first and second sets after the tetanus. Each set consisted of 30 F‐waves. Vertical calibration, 0.2 mV. F‐wave traces start at 22 ms. B, group data (n = 12; mean ± SEM) showing changes in the area of F‐waves after a 10 s tetanus at 25 Hz (open circles) and 40 Hz (filled circles). For illustration, each circle represents the average area of ten consecutive F‐waves from each subject. For statistical tests, F‐waves were grouped into sets of 30. Tetanic stimulation did not depress F‐wave area. C, group data showing F‐wave persistence in the relaxed ADM muscle before and after a 25 Hz (open circles) and 40 Hz (filled circles) tetanus. F‐wave persistence was not altered by supramaximal tetanic stimulation.

Study 3: F‐waves at rest after a 3 min MVC in ADM antagonist

Study 3 examined F‐wave area and persistence in ADM and FDI muscles before and after a 3 min sustained MVC of 4PI (ADM's antagonist) in 19 subjects. Surprisingly, F‐waves in the non‐contracting ADM showed some depression after the contraction, whereas those in FDI did not. Figure 4 A and D shows superimposed raw traces for a single subject. Baseline persistence in ADM was high and stable prior to contraction (average baseline persistence was 87.7 ± 12.6 and 88.9 ± 11.5%, Fig. 4 C), whereas baseline persistence in FDI was lower (70.3.7 ± 19.5% and 68.9 ± 17.7%, Fig. 4 F). For the group of subjects, immediately after the 3 min MVC, there was a small reduction in F‐wave persistence in ADM (∼18%; F 11,187 = 7.87, P < 0.001; n = 19; Fig. 4 C). Post hoc testing showed that this depression lasted for ∼5 min (P < 0.01). A similar magnitude depression occurred in F wave area (∼15 %), but was not statistically significant (F 11,198 = 1.11, P = 0.38; Fig 4 B). Importantly, over the same time, F‐wave area (F 11,176 = 1.96, P = 0.34; Fig. 4 E) and persistence (F 11,154 = 2.03, P = 0.29; n = 15; Fig 4 F) in the more distant intrinsic hand muscle on the radial side of the hand (FDI) were unchanged. We quantified co‐contraction of ADM and FDI during the sustained MVC of 4PI. The mean EMG amplitude determined from the root mean square level over the 3 min contraction was 6.9 ± 3.8 and 5.0 ± 4.2% MVC for ADM and FDI, respectively.

Figure 4. F‐waves in ADM and FDI after 3 min maximal contraction of the fourth palmar interosseous (ADM's antagonist) .

Figure 4

Open in a new tab

A and D, F‐waves recorded from ADM (A) and FDI (D) at rest before and after a 3 min sustained contraction of the fourth palmar interosseous muscle for a single subject. Top traces represent F‐waves recorded at rest before the MVC and are shown superimposed. Lower traces show superimposed responses for the first and second sets after the MVC. Each set consisted of 30 F‐waves. Vertical calibration, 0.2 mV; horizontal calibration, 10 ms. F‐wave traces start at 21 and 23 ms for ADM and FDI muscles, respectively. B and E, group data (n = 19; mean ± SEM) showing changes in the area of F‐waves in ADM (B) and FDI (E) after a 3 min sustained MVC of ADM's antagonist muscle. For illustration, each circle represents the average area of ten consecutive F‐waves for each subject. For statistical tests, F‐waves were grouped into sets of 30. C and F, group data showing F‐wave persistence in the relaxed ADM (C) and FDI (F). In ADM, F‐wave area and persistence were depressed after a 3 min maximal contraction of ADM's antagonist muscle. In contrast, F‐wave area and persistence were unchanged in the more distant FDI muscle. Asterisks (*) indicate that values for the set of 30 F‐waves are significantly lower than the baseline value for the contraction (P < 0.05). The shaded vertical bar denotes the MVC.

Discussion

Because the recurrent discharge or F‐wave depends on the integrity and excitability of the axon initial segment of the motoneurone (e.g. Eccles, 1955; see Introduction), its measurement offers arguably the most direct way to examine the excitability of motoneurone pools in human subjects (for a review see McNeil et al. 2013). Our study builds on the observation that F‐waves are depressed by strong voluntary activity in intrinsic hand muscles (Rossi et al. 2010, 2012; Giesebrecht et al. 2011; Khan et al. 2012) and a lower limb muscle (Khan et al. 2012). Furthermore, the axon initial segment rather than a more distal part of the axon is probably the key site for the depression (Khan et al. 2012). Given that we have shown previously that this depression is greater with long compared with brief MVCs, the new finding that the depression is less when the intensity of the contraction is only 50% MVC is not surprising. Consistent with our prior findings, a higher impulse ‘load’ at the motoneurones due to the greater voluntary ‘effort’ driving the motoneurone pool would increase the depression of the F‐wave. However, results from Study 2 are surprising: they show that intense antidromic activation of the motoneurone pool fails to reproduce the depression seen with strong voluntary contractions. Despite a large impulse load, F‐waves were not reduced. Clearly, voluntary activity can change motoneuronal excitability in a way that is not mimicked by a non‐volitional impulse load. Results from Study 3 provide some insight into the type of process that could be involved in the depression of the F‐wave after strong voluntary contractions. When the test muscle is relaxed or minimally active, prolonged strong contraction of another hand muscle can depress F‐waves in the test muscle. This effect is apparent when the two muscles innervated by the intrinsic hand motoneurone pools are adjacent on each side of the fifth metacarpal but not when they are widely spaced across the hand.

Prior to the present study, the explanation for the depression of F‐waves with MVCs (e.g. Khan et al. 2012; Rossi et al. 2012) relied on a well‐established phenomenon in which repetitive axonal activity hyperpolarises the active axons. This has been measured in animal studies (e.g. Gasser, 1935; Bostock & Grafe, 1985) and confirmed in human studies (e.g. Vagg et al. 1998; Kiernan et al. 2004; Milder et al. 2014). However, the observation that brief 2 s maximal contractions reduce the size and frequency of F‐waves in an intrinsic hand muscle without an effect on excitability of motor axons measured at a distal site along the ulnar nerve implicated the axon initial segment as the site of origin for the depression (Khan et al. 2012). While this explanation seems plausible, it does not fit with two results from the current study. First, we delivered antidromic activity using supramaximal tetanic stimulation with an impulse load of 250–400 impulses over 10 s. This load is physiologically realistic based on recordings during MVCs from human motoneurones innervating hand muscles (e.g. Bellemare et al. 1983; Gandevia et al. 1990) and has been assessed indirectly by tetanic stimulation of the ulnar at the elbow (Fuglevand & Keen, 2003), and hence we expected that these impulses would reach the motoneurone pool and depress F‐waves (Khan et al. 2012). However, no depression occurred (Fig. 3) despite the fact that the effect of the antidromic impulses would be supplemented by synaptic events generated by stimulus‐evoked activity in muscle spindle and other afferents in the ulnar nerve. Second, even when the test muscle (ADM) was not contracted and few impulses occurred at the axon initial segment, voluntary activity of a nearby muscle (fourth palmar interosseous) produced differential depression of F‐waves in the test muscle without depression in another non‐contracting muscle further away on the radial side of the hand (FDI).

Possible mechanisms

Our finding that depression of F‐waves changes with the level of voluntary activity (Study 1) and duration of voluntary contraction (Khan et al. 2012) fits with an activity‐ and time‐dependent process. However, what is added by the findings in Studies 2 and 3 is that the process does not depend solely on the motoneurone itself. Intense non‐voluntary activity in the motoneurones does not depress their F‐waves (Study 2) while intense voluntary activity in some nearby motoneurones can depress them. Hence, we speculate that there is a mechanism directly related to descending voluntary activity which changes the excitability of the motoneurones independent of their repetitive firing per se and independent of their post‐synaptic activation by large‐diameter afferents. One type of mechanism could involve release of neuromodulators by descending motor pathways acting on motoneurones. Theoretically there are several pathways that could be involved (e.g. reviewed by Lemon, 2008) but none is yet known to have the properties suggested above.

One pathway known to operate widely within the spinal cord is the descending serotonergic system (e.g. Lemon, 2008; cf. Bowker et al. 1987; Bowker & Abbott, 1990), and it would have limited ability to act focally among small regions of active and nearby inactive motoneurones. This argues against it having a role in the effect described here. However, its actions on motoneurones provide evidence that activation of extra‐synaptic receptors by descending paths is a plausible way by which motoneurone excitability might be influenced. Descending serotonergic projections from raphe neurones fire during motor activities (Veasey et al. 1995; for a review see Jacobs et al. 2002) and released serotonin (5‐HT) can modify excitability at motoneuronal and other spinal sites (e.g. Jankowska et al. 2000; Schmidt & Jordan, 2000; Wei et al. 2014). In turtle spinal cord, intense activation of descending paths to motoneurones releases 5‐HT which initially facilitates plateau potentials via synaptic receptors in the dendrites, but after about 30 s its concentration is sufficient to spill over from the synapses to inhibit action potentials at the initial segment via extra‐synaptic 5‐HT1A receptors (Perrier & Cotel, 2008; Cotel et al. 2013). Such a type of mechanism produced by strong sustained voluntary drive to motoneurones would set up feedback akin to central fatigue that would reduce motoneurone output.

If there is a single mechanism leading to the depression of F‐waves with strong voluntary contraction, the effect on a test motoneurone pool is largest when its motoneurones are voluntarily active (∼48%), smaller when a nearby muscle is active (∼15%) and absent when a more remote muscle contracts. How might this operate? At the C8–T1 level of the human spinal cord, the motoneurone columns for 19 intrinsic hand muscles are contained within a cross‐sectional area of only 1 mm2 (Sengul et al. 2012). This high density of motoneurone pools within the ventral horn means that there will inevitably be intermingling of motoneurones innervating different muscles, a phenomenon observed directly in the monkey for motoneurones of intrinsic hand muscles (Jenny & Inukai, 1983). However, a degree of somatotopy, such as that for muscles on the ulnar across to the radial side of the hand, is preserved (C. Watson, personal communication) and it would be expected on developmental grounds (e.g. Dasen & Jessell, 2009). In this scenario, a neuromodulator or neurotransmitter released in an intensity‐ and time‐dependent manner during voluntary contractions of one hand muscle may influence motoneurones of close but not remote muscles in the hand. Such an effect should be less for motoneurone pools of proximal arm or leg muscles where intermingling among different motoneurone pools is likely to be less (e.g. Tosolini & Morris, 2012).

There are limits to voluntary exercise in terms of both its intensity and its duration. Some limits reside not within the muscle but within the CNS (e.g. Gandevia, 2001; Taylor & Gandevia, 2008). Centrally, there can be deficiencies in the level of descending drive, but there is growing evidence that spinal segmental factors can diminish or even abolish the output from motoneurone pools to large synchronised corticospinal volleys during strong efforts (e.g. McNeil et al. 2009, 2011 b). While these changes may represent premotoneuronal and postsynaptic effects at the voluntarily activated motoneurones (McNeil et al. 2011 a), the present work highlights the axon initial segment of the motoneurone as a potential site where excitability changes occur. Such changes occur not simply by activity‐dependent hyperpolarisation of active motoneurones but they can occur in resting or minimally active motoneurones by voluntary activation of nearby motoneurone pools, possibly by the local action of neuromodulators. This work reveals potentially novel mechanisms controlling the output of human motoneurones.

Additional information

Competing interests

None.

Author contributions

All authors were involved in the conception and design of the studies, the collection, analysis, and interpretation of the data. All helped with drafting and revision of the manuscript and all approved the final submitted version. All studies were conducted at Neuroscience Research Australia in Sydney.

Funding

This work was supported by the National Health and Medical Research Council (of Australia).

References

  1. Bellemare F, Woods JJ, Johansson R & Bigland‐Ritchie B (1983). Motor‐unit discharge rates in maximal voluntary contractions of three human muscles. J Neurophysiol 50, 1380–1392. [DOI] [PubMed] [Google Scholar]
  2. Bostock H & Grafe P (1985). Activity‐dependent excitability changes in normal and demyelinated rat spinal root axons. J Physiol 365, 239–257. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bowker RM & Abbott LC (1990). Quantitative re‐evaluation of descending serotonergic and non‐serotonergic projections from the medulla of the rodent: evidence for extensive co‐existence of serotonin and peptides in the same spinally projecting neurons, but not from the nucleus raphe magnus. Brain Res 512, 15–25. [DOI] [PubMed] [Google Scholar]
  4. Bowker RM, Reddy VK, Fung SJ, Chan JY, Barnes CD (1987). Serotonergic and non‐serotonergic raphe neurons projecting to the feline lumbar and cervical spinal cord: a quantitative horseradish peroxidase‐immunocytochemical study. Neurosci Lett 75: 31–37. [DOI] [PubMed] [Google Scholar]
  5. Butler JE, Taylor JL & Gandevia SC (2003). Responses of human motoneurons to corticospinal stimulation during maximal voluntary contractions and ischemia. J Neurosci 23, 10224–10230. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Coombs JS, Curtis DR & Eccles JC (1957) The interpretation of spike potentials of motoneurones. J Physiol 139, 198–231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cotel F, Exley R, Cragg SJ & Perrier JF (2013). Serotonin spillover onto the axon initial segment of motoneurons induces central fatigue by inhibiting action potential initiation. Proc Natl Acad Sci USA 110, 4774–4779. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dasen JS & Jessell TM (2009). Hox networks and the origins of motor neuron diversity. Curr Topics Dev Biol 88, 169–200. [DOI] [PubMed] [Google Scholar]
  9. De Luca CJ, LeFever RS, McCue MP & Xenakis AP (1982). Behaviour of human motor units in different muscles during linearly varying contractions. J Physiol 329, 113–128. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Eccles JC (1955). The central action of antidromic impulses in motor fibres. Pflügers Arch 260, 385–415. [DOI] [PubMed] [Google Scholar]
  11. Espiritu MG, Lin CS & Burke D (2003). Motoneuron excitability and the F wave. Muscle Nerve 27, 720–727. [DOI] [PubMed] [Google Scholar]
  12. Fuglevand AJ & Keen DA (2003). Re‐evaluation of muscle wisdom in the human adductor pollicis using physiological rates of stimulation. J Physiol 549, 865–875. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Gandevia SC (2001). Spinal and supraspinal factors in human muscle fatigue. Physiol Rev 81, 1725–1789. [DOI] [PubMed] [Google Scholar]
  14. Gandevia SC, Khan SI & Taylor JL (2014). The recurrent discharge of human motoneurones is reduced by voluntary but not antidromic activation. International Motoneuron Meeting, Halifax, Canada. http://www.motoneuron.org/wp‐content/uploads/2014/06/Oral‐abstracts1.pdf
  15. Gandevia SC, Macefield G, Burke D & McKenzie DK (1990). Voluntary activation of human motor axons in the absence of muscle afferent feedback. The control of the deafferented hand. Brain 113, 1563–1581. [DOI] [PubMed] [Google Scholar]
  16. Gandevia SC, Petersen N, Butler JE & Taylor JL (1999). Impaired response of human motoneurones to corticospinal stimulation after voluntary exercise. J Physiol 521, 749–759. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gasser HS (1935). Changes in nerve‐potentials produced by rapidly repeated stimuli and their relation to the responsiveness of nerve to stimulation. Am J Physiol 111, 35–50. [Google Scholar]
  18. Giesebrecht S, Martin PG, Gandevia SC & Taylor JL (2011). Altered corticospinal transmission to the hand after maximum voluntary efforts. Muscle Nerve 43, 679–687. [DOI] [PubMed] [Google Scholar]
  19. Gogan P, Gustafsson B, Jankowska E & Tyc‐Dumont S (1984). On re‐excitation of feline motoneurones: its mechanism and consequences. J Physiol 350, 81–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Jacobs BL, Martin‐Cora FJ & Fornal CA (2002). Activity of medullary serotonergic neurons in freely moving animals. Brain Res Rev 40, 45–52. [DOI] [PubMed] [Google Scholar]
  21. Jankowska E, Hammar I, Chojnicka B & Heden CH (2000). Effects of monoamines on interneurons in four spinal reflex pathways from group I and/or group II muscle afferents. Eur J Neurosci 12, 701–714. [DOI] [PubMed] [Google Scholar]
  22. Jenny AB & Inukai J (1983). Principles of motor organization of the monkey cervical spinal cord. J Neurosci 3, 567–575. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Khan SI, Giesebrecht S, Gandevia SC & Taylor JL (2012). Activity‐dependent depression of the recurrent discharge of human motoneurones after maximal voluntary contractions. J Physiol 590, 4957–4969. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kiernan MC, Lin CSY & Burke D (2004). Differences in activity‐dependent hyperpolarization in human sensory and motor axons. J Physiol 558, 341–349. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kukulka CG & Clamann HP (1981). Comparison of the recruitment and discharge properties of motor units in human brachial biceps and adductor pollicis during isometric contractions. Brain Res 219, 45–55. [DOI] [PubMed] [Google Scholar]
  26. Lemon RN (2008) Descending pathways in motor control. Ann Rev Neurosci 31, 195–218. [DOI] [PubMed] [Google Scholar]
  27. Martin PG, Gandevia SC & Taylor JL (2006). Output of human motoneuron pools to corticospinal inputs during voluntary contractions. J Neurophysiol 95, 3512–3518. [DOI] [PubMed] [Google Scholar]
  28. McLeod JG & Wray SH (1966). An experimental study of the F wave in the baboon. J Neurol Neurosurg Psychiatry 29, 196–200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. McNeil CJ, Butler JE, Taylor JL & Gandevia SC (2013). Testing the excitability of human motoneurons. Front Hum Neurosci 7, 152. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. McNeil CJ, Giesebrecht S, Gandevia SC & Taylor JL (2011. a). Behaviour of the motoneurone pool in a fatiguing submaximal contraction. J Physiol 589, 3533–3544. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. McNeil CJ, Giesebrecht S, Khan SI, Gandevia SC & Taylor JL (2011. b). The reduction in human motoneurone responsiveness during muscle fatigue is not prevented by increased muscle spindle discharge. J Physiol 589, 3731–3738. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. McNeil CJ, Martin PG, Gandevia SC & Taylor JL (2009). The response to paired motor cortical stimuli is abolished at a spinal level during human muscle fatigue. J Physiol 587, 5601–5612. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Milder DA, Sutherland EJ, Gandevia SC & McNulty PA (2014). Sustained maximal voluntary contraction produces independent changes in human motor axons and the muscle they innervate. PLoS ONE 9, e91754. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Panayiotopoulos CP & Chroni E (1996). F‐waves in clinical neurophysiology: a review, methodological issues and overall value in peripheral neuropathies. Electroencephalogr Clin Neurophysiol 101, 365–374. [PubMed] [Google Scholar]
  35. Perrier JF & Cotel F (2008). Serotonin differentially modulates the intrinsic properties of spinal motoneurons from the adult turtle. J Physiol 586, 1233–1238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Petersen NT, Taylor JL, Butler JE & Gandevia SC (2003). Depression of activity in the corticospinal pathway during human motor behaviour after strong voluntary contractions. J Neurosci 23, 7974–7980. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Renshaw B (1941). Influence of discharge of motoneurons upon excitation of neighboring motoneurons. J Neurophysiol 4, 167–183. [Google Scholar]
  38. Rossi A, Biasella A, Scarselli C, Piu P & Ginanneschi F (2010). Influence of activity‐induced axonal hypoexcitability on transmission of descending and segmental signals. Brain Res 1320, 47–59. [DOI] [PubMed] [Google Scholar]
  39. Rossi A, Rossi S & Ginanneschi F (2012). Activity‐dependent changes in intrinsic excitability of human spinal motoneurones produced by natural activity. J Neurophysiol 108, 2473–2480. [DOI] [PubMed] [Google Scholar]
  40. Schmidt BJ & Jordan LM (2000). The role of serotonin in reflex modulation and locomotor rhythm production in the mammalian spinal cord. Brain Res 53, 689–710. [DOI] [PubMed] [Google Scholar]
  41. Sengul G, Watson C, Tanaka I & Paxinos G (2012). Atlas of the Spinal Cord: Mouse, Rat, Rhesus monkey, Marmoset, and Human. Academic Press, San Diego. [Google Scholar]
  42. Taylor JL, Butler JE & Gandevia SC (1999). Altered responses of human elbow flexors to peripheral‐nerve and cortical stimulation during a sustained maximal voluntary contraction. Exp Brain Res 127, 108–115. [DOI] [PubMed] [Google Scholar]
  43. Taylor JL & Gandevia SC (2008). A comparison of central aspects of fatigue in submaximal and maximal voluntary contractions. J Appl Physiol 104, 542–550. [DOI] [PubMed] [Google Scholar]
  44. Tosolini AP & Morris R (2012). Spatial characterization of the motor neuron columns supplying the rat forelimb. J Neurosci 200, 19–30. [DOI] [PubMed] [Google Scholar]
  45. Trontelj JV (1973). A study of the F response by single fibre electromyography. New Dev Electromyogr Clin Neurophysiol 3, 318–322. [Google Scholar]
  46. Vagg R, Mogyoros I, Kiernan MC & Burke D (1998). Activity‐dependent hyperpolarization of human motor axons produced by natural activity. J Physiol 507, 919–925. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Veasey SC, Fornal CA, Metzler CW & Jacobs BL (1995). Response of serotonergic caudal raphe neurons in relation to specific motor activities in freely moving cats. J Neurosci 15, 5346–5359. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Wei K, Glaser JI, Deng L, Thompson CK, Stevenson IH, Wang Q, Hornby TG, Heckman CJ & Kording KP (2014). Serotonin affects movement gain control in the spinal cord. J Neurosci 34, 12690–12700. [DOI] [PMC free article] [PubMed] [Google Scholar]

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

RESOURCES