Journal Logo


Acute Effect of Noradrenergic Modulation on Motor Output Adjustment in Men


Author Information
Medicine & Science in Sports & Exercise: August 2018 - Volume 50 - Issue 8 - p 1579-1587
doi: 10.1249/MSS.0000000000001622
  • Free


The noradrenergic system is involved in the control of many higher functions such as arousal, attention, stress response, memory, sensory information processing, and long-term synaptic plasticity through its action on the synaptic transmission in the prefrontal cortex. Modulation of prefrontal functions by noradrenaline has been largely investigated, and perturbation of the noradrenergic signaling has been shown to be involved in the pathogenesis of many neuropsychiatric disorders (for review articles, see Refs. [1–3]). Surprisingly, much less is known regarding the noradrenergic actions on the motor system and the functional consequences in human.

Previous studies reported that increasing central noradrenergic concentration using reboxetine (REB), a noradrenergic reuptake inhibitor, improves motor function in hand muscles of healthy subjects (4–6) and stroke patients (7,8). The presence of a REB effect seemed, however, to be task specific and was mostly reported during visuomotor tasks (6). Concomitantly to the improved motor performance under REB, Plewnia et al. (4,5) and Kuo et al. (9) observed an increase in corticospinal excitability and a decrease in short-interval intracortical inhibition in healthy individuals by using transcranial magnetic stimulation (TMS). Furthermore, using functional magnetic resonance, Wang et al. (7) reported that the improvement in motor function under REB in paretic patients resulted from a rearrangement of cortical interactions, supporting a role for cortical-related mechanisms in these adaptations.

Although extensive animal literature demonstrated a significant role for noradrenergic projections in the modulation of spinal excitability, its role at the spinal level in humans is still unclear. By increasing intrinsic excitability of motor neurons, monoaminergic inputs have proved to play a role in adjusting motoneuronal excitability, allowing thereby motor units to discharge at a higher rate in response to ionotropic inputs (for reviews in animal, see Refs. [10,11]). On the basis of the results of animal studies, a noradrenergic modulation of spinal excitability in humans is likely and needs further investigations.

Therefore, given the paucity of the literature on the effect of noradrenergic modulation on motor control, the present work investigated the influence of REB on maximal force and force steadiness during submaximal isometric contractions (12,13). The latest task was chosen because, along with a sufficient maximal force, accurate control of force is necessary to perform adequately most functional tasks (14). Associated changes in spinal and corticospinal excitability were assessed by recording the Hoffmann (H) reflex and motor-evoked potential (MEP) induced by TMS in the knee extensor muscles, which play an important role in everyday activities. Because the size of the H reflex, induced by Ia-afferents activation, depends on the excitability of spinal motor neurons and modulation of Ia synaptic transmission (15), whereas MEP estimates both the excitability of cortical and spinal motor neurons, by comparing the concurrent changes of both responses, we aimed to identify the relative modulation in excitability at spinal and cortical levels (16). Contrary to previous studies, we recorded H reflexes and MEP responses in the contracting muscle which is more functionally relevant and enabled us to investigate the duration of TMS-induced intracortical inhibition by measuring the silent period (SP) in the ongoing EMG activity. This parameter has been shown to assess long-lasting cortical inhibition mediated by type B gamma-aminobutyric (GABAB) receptors (17) that ensure the stability of cortical network activity (18).



Eleven recreationally active male subjects (age, 23.0 ± 2.1 yr; height, 1.82 ± 0.08 m; mass, 80.1 ± 8.5 kg) completed this study. Before their participation in the study, all volunteers received written information regarding the nature and purpose of the experimental protocol. All subjects passed a medical screening and an electrocardiogram at rest to exclude any contraindication to REB and/or TMS. Subjects had the opportunity to ask questions before signing the written statement of consent. The experimental protocol was approved by the Ethical Committee of the Erasmus Hospital (Brussels).

Ergometric device

The neuromuscular parameters were recorded while the subjects sat on an adjustable chair such that the hip and knee joints were at 100° (angle between trunk and thigh) and 80° (full extension = 0°), respectively. The chair had a long back-rest and the head of the subject was secured in a custom-made headrest to ensure a stable position during the experiment. A force transducer was rigidly connected to the front of the chair by a custom-made fixing system (for detailed description of the setup, see Ref. [19]). The right leg was attached to the transducer using a velcro strap that was placed ~2 cm above the lateral malleolus. To limit trunk movement during contractions, the subject was secured to the chair by a harness. Headrest and transducer positioning were recorded for each subject to reproduce identical setup conditions in the two experimental sessions (19).

Torque and electromyographic recordings

The isometric torque produced by the knee extensor muscles was measured using a force transducer (linear range, 0–2500 N; U2000 load cell; Maywood Instruments Ltd, Basingstoke, UK). Torque was calculated by multiplying the output from the transducer by the lever arm measured as the distance between the center of the transducer and the axis of rotation of the knee joint. Voluntary and electrically evoked EMG activities were recorded from rectus femoris (RF), vastus medialis (VM), vastus lateralis (VL) and long head of the biceps femoris (BF) by means of self-adhesive bipolar electrodes (Ag–AgCl, 10-mm diameter; interelectrode center to center distance, 5 cm). The EMG electrodes were placed over the muscle belly of each muscle, whereas the reference electrodes were located over the lateral condyle of the tibia. The locations of EMG electrodes were marked with indelible ink to ensure similar recording conditions during the two experimental sessions. All EMG signals were amplified (×1000) and filtered (10 Hz–1 kHz) by a custom-made differential amplifier. All signals were acquired on a computer at a sampling rate of 2 kHz with a data-acquisition system (Model MP 150; Biopac Systems, Santa Barbara, CA) and analyzed offline with associated AcqKnowledge software.

Electrical stimulation

Single and paired rectangular pulses of 1-ms duration were delivered to the femoral nerve by a constant current stimulator (DS7A Digitimer, Welwyn Garden City, UK) triggered by a digital timer (Master-8; AMPI, Jerusalem, Israel) through self-adhesive electrodes (Ag–AgCl, 10-mm diameter). The cathode was positioned over the nerve in the femoral triangle, and the anode midway between the greater trochanter and the iliac crest (20).

The evoked EMG responses were the H reflex and M wave. Stimulus intensity was optimized for the VM because pilot experiments performed in our laboratory indicated that reflex responses were more stable in this muscle. This was confirmed by Doguet and Jubeau (21), who also observed a better interday reliability in VM than in VL.

Unlike previous works investigating the effect of noradrenergic modulation on corticospinal excitability at rest (4,9), we determined the H-reflex and M-wave recruitment curves during submaximal isometric contractions at 20% of maximal voluntary contraction (MVC). Because muscle activation favors a constant spinal synaptic transmission, such an approach reduces the intrasubject variation and assesses the corticospinal pathway in a more functional condition (22). Stimulus intensity was increased gradually by steps of 1 mA until the H reflex reached maximal amplitude (Hmax) and by steps of 2 mA thereafter until the M-wave amplitude reached a plateau (Mmax). Three to eight stimulations were induced at each step depending on the response variability.

Voluntary activation was tested by the superimposed stimulation method using paired supramaximal electrical stimuli (PES) delivered at 10-ms interval during the MVC plateau and at rest immediately after the end of the MVC (19,23). The stimulation intensity used to test voluntary activation was set 30% above that required to induce Mmax.

Transcranial magnetic stimulation

A double-cone coil (130-mm outer diameter) was positioned over the cortex to elicit MEP in the right knee extensors with a Magstim 200 stimulator (Magstim, Dyfed, UK). The junction of the double-cone coil was positioned 1–2 cm to the left of the vertex. During each experiment, the position of the coil was determined to induce the greatest MEP response for a given stimulus in VM (see protocol), and this position was marked on the scalp. The recruitment curve was performed at 20% MVC by increasing the stimulation intensity in steps of 3% of the maximal stimulator output (3–8 stimulations at each step) until the MEP amplitude reached a plateau (MEPmax).

To assess voluntary activation by TMS, the stimulator output was set to evoke the greatest superimposed response and MEP in VM with a minimal MEP size in the BF, while the subject was sustaining a 50% MVC with the knee extensors. The mean TMS intensity used to test voluntary activation corresponded to 58.3% ± 3.0% (range, 55%–63%) of the stimulator output.


Oral administration of REB, a selective and specific noradrenaline reuptake inhibitor, was used to increase basal extracellular noradrenaline levels in the nervous system (24). A single dose of 8 mg of REB was used in the present study, similarly to previous studies investigating the effect of REB on motor performance in hand muscles (5–7), to compare our data with results reported in upper limb muscles.

During the two experimental sessions, subject received a capsule with either 8 mg of REB or a placebo (PLA) consisting of 20 mg of lactose. Subjects ingested the capsule 85 min before the beginning of the recordings because plasma concentration was reported to be high and relatively stable between 85 and 140 min after the ingestion of a single oral dose of 8 mg of REB (4). REB and PLA were administrated in a randomized order, and both experimenters and subjects were unaware of the capsules’ content. The drug treatment was supervised by a medical doctor who was available in case of side effects during and after the experiments. The only side effects reported by three subjects under REB were as follows: perturbed sleeping the night after the experiment and feeling a little bit warmer than usual before the start of the experimental recordings. This last feeling was also reported by one subject during the session under PLA.

Experimental protocol

Subjects took part in three experimental sessions, one familiarization session and two experimental sessions spaced by 1 wk and conducted at the same time of the day to limit diurnal variation in muscle force (25). The familiarization session was dedicated to accustom subjects to the setting conditions and the protocol to minimize learning and anxiety effects. They practiced isometric contractions at different levels during the torque-matching task and experienced electrical motor nerve stimulation and TMS. The course of the familiarization session was similar to the real experimental sessions. During the experimental sessions, the protocol was planned so that all the recordings started 85 min after drug intake and completed within 1 h (see the reason mentioned previously [4]). Exercise, or caffeine or alcohol consumption was forbidden during the 24 h preceding each session.

After the ingestion of the drug (REB or PLA), subjects were equipped with a heart rate monitor (Polar H1 sensor and RS300X watch, Kempele, Finland) and EMG electrodes and installed on the experimental chair. Thereafter, they warmed up by performing a series of 10 submaximal contractions and two brief MVC. The optimal location and stimulation intensity for TMS and electrical stimulation were then determined during brief (~3 s) submaximal contractions (at 20% and 50% MVC) and at rest for electrical stimulation.

The experimental recordings (Fig. 1) started with two to three 3-s MVC, separated by 1 min of rest, during which paired electrical supramaximal stimuli were induced. They were followed by two to three PES evoked at rest to determine the resting potentiated mechanical response. After 3 min of recovery, subjects performed an MVC and two contractions at 75% and 50% MVC. TMS was applied during the torque plateau of each contraction. Each contraction was repeated two to three times and separated by at least 1 min of rest. The order of the two submaximal contractions was randomized across subjects but kept constant for each subject across all experimental sessions. A brief MVC was finally performed with a single superimposed electrical stimulation to record the superimposed Mmax that was used to normalize the MEP. After a rest period of 4 min, the experimental session continued with the torque-matching task. For each of the four levels (5%, 10%, 20%, and 50% MVC), subjects first performed a 5-s contraction to experience and match the required torque. This was followed by the recording of two 15-s isometric contractions maintained as steady as possible and separated by 30 s of rest. A visual feedback of the torque was provided on a 22-inch monitor, located in front of the subject. The gain of the displayed signal was adjusted so that the target line was always in the middle of the screen and torque fluctuations were visually similar for the different contraction levels. Target torques were presented in a random order for each subject. After a rest period of 4 min, the experimental session ended with the recording of the H-reflex, M-wave, and MEP recruitment curves.

Overview of the experimental protocol. The recordings started 85 min after drug intake with voluntary activation testing using electrical and TMS, followed by the torque matching task, the H-reflex and M-wave recruitment curve and ended with MEP recruitment curve. SES, single electrical stimulation.

Data analysis

MVC torque and associated average value of the rectified EMG (aEMG) of VM, VL, RF, and BF were determined for a 500-ms period during the plateau of the MVC. The coefficient of variation (CV) of the torque exerted during the torque-matching task and the corresponding aEMG were determined for the steadiest 5-s period. The values for the two contractions at each target were averaged.

The peak-to-peak amplitude and area of the H reflex, MEP, and M wave were measured. As the amplitude and area of the evoked responses exhibited similar features, only the former parameter is reported in the article. H-reflex and MEP responses were normalized to Mmax to account not only for potential changes in excitability of the muscle fiber membrane and slight variation in the position of the EMG electrodes between sessions, but also to reduce interindividual differences. Because active threshold, corresponding to the lowest intensity inducing a small response (9), is usually considered not sensitive enough to detect spinal and corticospinal excitability changes (4,9), only maximal amplitudes and slopes of the ascending part of the recruitment curves for H reflex, M wave, and MEP were analyzed. These responses represent a greater portion of the motor neuron pool and are usually considered as a more reliable measure of spinal and corticospinal excitability (9). Maximal response amplitude (Hmax, Mmax, and MEPmax; Fig. 2A) was determined by averaging the five largest responses obtained during the recruitment curve. The ascending limb of the recruitment curve for H reflex, M wave, and MEP were fitted by a Boltzmann sigmoid (Figs. 2B–D) (26,27) to determine the following parameters: 1) the stimulus intensities corresponding to 50% of the maximal amplitude of the evoked responses (H50, M50, and MEP50, respectively) and 2) the slope (Hslope, Mslope, and MEPslope) of the ascending limb of the recruitment curve (Figs. 2B–D). The slope was calculated for the part of the curve corresponding to the stimulus intensity associated with H50, M50, and MEP50 and was obtained using the following equation:

Changes induced by REB intake on spinal and corticospinal excitability in a representative subject. A, Traces of Hmax, MEPmax, and Mmax in PLA and REB conditions recorded in VM during an isometric contraction at 20% MVC. B–D, The recruitment curves for M wave, H reflex, and MEP, respectively, in PLA (open circles) and REB (filled circles) conditions. The thick lines represent the slope of the ascending part of recruitment curves occurring at the stimulus intensity associated with H 50, M 50, and MEP50, which are indicated by the dashed lines on the graphs.

where m is the inverse of the Boltzmann equation slope (22,26,27). Before calculating the slope, the amplitude of the responses was expressed as a percentage of Mmax and its intensity as a percentage of the stimulus intensity associated with M50 or MEP50. The r2 quantifying how the Boltzmann sigmoid fits the data was high for all recruitment curves and ranged between 0.93 and 0.99.

The duration of the cortical SP associated with either the MEP recorded during MVC or the MEP responses used to calculate MEPmax during the contraction at 20% MVC was measured to assess the level of intracortical inhibition. SP duration was determined in the contracting muscle as the interval between the TMS induced artifact and the return of continuous EMG activity. To reduce the intersubject variability and the possible confounding effect of a change in MEP amplitude on SP duration, the ratio between SP and the corresponding MEP was calculated (SP/MEP) (16,28).

The superimposed torque induced by TMS or PES during MVC was measured as the difference between the superimposed peak torque and MVC torque (20,23). Peak torque of the response induced by PES was measured at rest, whereas it was estimated rather than measured directly for TMS (23). For each subject, a linear regression between the amplitude of the superimposed torque evoked by TMS and voluntary torque was performed for intensities of 50%, 75%, and 100% MVC. The y-intercept was taken as the amplitude of the estimated resting response (23). Voluntary activation level (% of maximum), tested using either PES or TMS, was calculated according to the following equation: [1 − superimposed torque/torque induced at rest in response to PES (or estimated resting response for TMS)] × 100 (19,20,23).

Statistical analysis

Before comparing each dependent variable, the normality of the data was controlled using a Shapiro–Wilk normality test. Depending on the distribution of the data for each variable, paired Student’s t-test or Wilcoxon signed-rank test was used to compare all parameters between PLA and REB conditions, except those recorded during the torque-matching task. The CV for torque at 5%, 10%, 20%, and 50% MVC and associated aEMG were analyzed using a two-factor (condition–torque level) ANOVA with repeated measures. When a significant main effect was found, a Bonferroni’s post hoc test was used to compare differences between selected data points. Depending on the distribution of the data, Spearman rank correlation coefficient (rs) or Pearson correlation coefficient (rp) was calculated to explore the correlation between individual changes in voluntary activation induced by REB (as compared with PLA) and initial voluntary activation level tested either using TMS or PES under PLA condition. For all comparisons, the statistical level of significance was set at 0.05. Data are reported as means ± SD within the text, and means ± SEM in the figures.


MVC and electrically induced responses

The mean MVC torque was 9.5% greater in REB (267.4 ± 51.0 N·m) than in PLA (244.1 ± 43.6 N·m) condition (Fig. 3; P < 0.001). For all muscles, there was no difference in aEMG between conditions (all P values > 0.47). The aEMG values in PLA and REB conditions were, respectively, 619.9 ± 232.2 and 651.2 ± 233.0 μV for VM, 412.6 ± 121.0 and 427.8 ± 120.6 μV for VL, 384.1 ± 108.3 and 369.2 ± 122.4 μV for RF, and 60.3 ± 26.9 and 65.3 ± 28.3 μV for BF. The mean peak torque produced by PES at rest was similar in both conditions, 90.7 ± 16.4 and 92.2 ± 17.6 N·m for PLA and REB, respectively (P = 0.566).

Effect of REB on MVC torque. Individual changes (thin lines) and mean change (thick line) induced by REB intake are compared with PLA condition. Significant difference between REB and PLA condition: *P < 0.05.

Voluntary activation and MEP during MVC

Voluntary activation was significantly greater in REB than in PLA condition (Fig. 4) when tested either by TMS (3.0%; P < 0.001) or by PES (2.8%; P = 0.046). The increase in voluntary activation in REB condition and the subjects’ initial voluntary activation level were significantly correlated when tested either by TMS (rs = −0.62; P = 0.048) or by PES (rp = −0.86; P < 0.01).

Effect of REB on voluntary activation tested either by TMS (A) or PES (B). Individual changes (thin lines) and mean changes (thick lines) induced by REB intake are compared with PLA condition. Significant difference between REB and PLA condition: *P < 0.05.

The amplitude of the MEP recorded during MVC did not differ significantly (P = 0.270) between PLA (50.2% ± 11.6% Mmax) and REB (53.9% ± 15.7% Mmax) conditions. Mmax amplitude was similar (P = 0.415) in both conditions (13.3 ± 5.2 and 12.6 ± 5.5 mV in PLA and REB, respectively). There was no drug effect for SP duration (96.1 ± 8.1 and 95.5 ± 7.2 ms for PLA and REB, respectively; P = 0.565) or SP/MEP ratio (18.0 ± 8.9 and 18.7 ± 10.4 ms·mV−1 for PLA and REB, respectively; P = 0.755).

Torque variability

The ANOVA for torque variability indicated no interaction between drug condition and torque level (P = 0.709), but a significant main effect was found for torque level (P < 0.001) and drug condition (P < 0.001; Fig. 5A). The CV for torque was significantly lower in REB than in PLA condition for all torque levels (range for posttests, P values = 0.001–0.027). In addition, CV was greater at 50% MVC compared with the other torque levels (all posttests, P values < 0.001). The mean aEMG for agonist muscles (Fig. 5B) and BF (Fig. 5C), expressed as percent of their value recorded during knee extensors MVC, increased as a function of the torque level (ANOVA, P < 0.001). Regardless of the muscle, no drug effect (ANOVA, P values > 0.67) or drug and torque level interactions were found (P values > 0.62).

Effect of REB on torque variability and associated EMG activities during sustained submaximal contractions ranging between 5% and 50% MVC. A, CV of torque (%) at the four different contraction levels in REB and PLA conditions. B and C, The mean value for the EMG activity of the agonist muscles (average value of VM, VL, and RF aEMG) and of the BF, respectively. All EMG values are expressed as a percentage of their value recorded during knee extensors MVC. Significant difference between REB and PLA condition: *P < 0.05.

Recruitment curves

Results obtained from the analysis of the evoked potentials recorded during the recruitment curves are presented for one subject in Figure 2 and as mean values in Table 1. The Mmax and the slope (Mslope) of the ascending limb of M-wave recruitment curve were similar in both REB and PLA conditions (Table 1). In contrast, the Hmax and MEPmax as well as the slopes (Hslope and MEPslope) of their recruitment curves were significantly greater in REB than in PLA condition (Table 1). SP duration was similar in both conditions, whereas SP/MEP ratio was significantly less in REB than in PLA condition (Table 1).

M-wave, H-reflex, and MEP parameters recorded in VM in PLA and REB conditions at 20% MVC.


The purpose of the present study was to further explore the effect of noradrenergic neuromodulation on the maximal force and fine control of muscle force and associated changes in corticospinal and spinal excitability. To that end, we compared data obtained after the ingestion of 8 mg of REB, a reuptake inhibitor of noradrenaline, to those recorded in PLA condition. Our results showed an increase in both maximal force and force steadiness under REB. The motor performance improvements were accompanied by a greater corticospinal excitability which is at least partly related to an increase of the motoneuronal excitability of the spinal level.

Modulation of spinal and corticospinal excitabilities

The changes in corticospinal excitability associated with REB intake differ depending on the level of muscle activation. When assessed during MVC, REB had no effect on MEP amplitude and SP/MEP ratio. This finding agrees with our previous observations (19,20). In contrast, the analysis of the recruitment curve performed at 20% MVC indicated that REB substantially increased corticospinal excitability and decreased intracortical inhibition. The lack of REB effect during MVC is likely due to the fact that, at the time the superimposed stimulation is induced, motor neurons are either already activated or in a refractory state (29).

Similarly to the MEP, the analysis of the H reflex recorded at 20% MVC revealed that its maximal amplitude and the slope of the ascending part of its recruitment curve were increased by REB, indicating a greater spinal excitability and gain in the reflex loop (15,22). These results contrast with a previous study showing that REB increased MEP without affecting the maximal H reflex amplitude recorded in the abductor pollicis brevis at rest (4). This observation, together with the increased MEP amplitude, led the authors to conclude that REB mainly enhanced cortical excitability. The discrepancy with our results may be explained by the fact that voluntary activation brought motor neurons closer to their firing threshold (21,22), thereby augmenting the sensitivity of the method to detect small changes in spinal excitability.

The concurrent changes in SP/MEP ratio, MEP, and H reflex in the present study suggest that noradrenaline reuptake inhibition influences both spinal and cortical levels. At the cortical level, noradrenaline is known to modulate excitability in a task-specific way by regulating both glutamatergic facilitation and GABAergic inhibition through complex mechanisms that have been investigated in animal preparations (for a review, see Ref. [1]). A similar effect of noradrenergic modulation on cortical excitability in humans is supported by previous pharmacological studies that reported an increase in intracortical facilitation and a reduced short-interval intracortical inhibition under REB (4,9). Our results showing a decreased of the SP/MEP ratio also suggest a reduced long-latency intracortical inhibition. Because changes in this ratio may be partly driven by the increase in MEP, we additionally compared the SP duration associated with MEP of similar amplitude (expressed as percentage of Mmax) in PLA and REB conditions. This procedure enables to test a similar proportion of the motor neurons pool and avoid the influence of MEP size on SP duration. In that condition, the mean SP duration was significantly briefer (P = 0.001) in REB (126 ms) than in PLA (141 ms), which is consistent with the SP/MEP ratio results. At the spinal level, our results are in line with animal studies showing that noradrenaline increases the intrinsic excitability of motor neurons (11). The main mechanism by which noradrenaline is thought to facilitate motor neurons excitability is the induction of plateau potentials by the activation of L-type calcium channels (30) and by the depolarization of the resting membrane potential (31). Through these mechanisms, noradrenaline does not directly excite motor neurons, but potentiates the action of excitatory synaptic inputs (11). In addition to an increase in motor neuron excitability, an effect of noradrenergic modulation on presynaptic inhibition is not excluded (32), which may have facilitated muscle afferent feedback to the motor neuron pool, contributing to increase its responsiveness.

Maximal torque

The present results confirm those of our previous study regarding MVC torque (19). In that work, we administered two separate doses of 8 mg of REB, one the evening before the experiment and one upon arrival at the laboratory, and reported a 9% increase in MVC torque in the REB condition. In the present work, we administrated a single dose of REB to be able to compare our results with studies that investigated the effect of a similar dose on motor performance in hand muscles (5–7). A single 8-mg dose of REB seems to be sufficient to increase the MVC torque produced by the knee extensors. Because the torque evoked at rest by the PES was unchanged by REB, our data point toward voluntary activation enhancement being the main mechanism of torque increase. The absence of aEMG changes during MVC in the REB condition may be surprising at first, but it is known that surface EMG is not sensitive enough to detect small changes in motor unit discharge rate (33).

The observation of a greater voluntary activation in REB than in PLA condition is in line with the study of Wang and colleagues (7), who investigated the effect of REB on grip power in hemiparetic stroke patients and the associated changes in cortical activation using functional magnetic resonance imagery. They observed that REB increased grip power on the paretic side, whereas power was unchanged in the unaffected hand. The increase in power in the paretic side was associated with a partial normalization of cortical activity and cerebral connectivity that most probably resulted in an enhancement of the initially reduced motor output. Similar mechanisms may have contributed to the increase in MVC torque we observed in REB condition. Indeed, unlike hand muscles (25,34), most healthy individuals are unable to fully activate their knee extensor muscles (19,35). Knee extensors could therefore be more responsive to noradrenaline reuptake inhibition induced by REB. This view point is supported by the negative correlation observed between the gain in voluntary activation and subjects’ initial capacity.

Torque steadiness

The lower torque variability (expressed as CV) observed in REB compared with PLA condition suggests that noradrenergic neuromodulation also contributes to refine motor control. As mentioned in the Introduction section, studies investigating the effect of REB on fine motor control are sparse and controversial in healthy subjects. For example, it has been shown that REB has a positive effect on motor control during tasks involving upper limb movements and requiring adjustments of motor performance based on a visual feedback (5,6). In contrast, REB did not reinforce the performance or training-dependent improvements of simple and skilled finger movements performed without visual feedback (5,6). The task chosen in the present study had an important visuomotor component and required high attention to continuously adjust the torque level to match the target. In agreement with the previously mentioned observations, our results thus support a role for noradrenergic modulation in increasing accuracy during visuomotor tasks.

Torque fluctuation during steady submaximal contractions is influenced by the variability of motor units discharge rate and changes in common input to motor neurons (36–38) that are modulated at cortical and subcortical levels (36). Through its effect at cortical and spinal levels, noradrenaline may contribute to modulate the common input signal sent to the motor neuron pool and adjust the gain of motor neurons, improving thereby the effective neural input sent to the muscle (38). The greater torque steadiness observed in the present study under REB is most likely the result of complex mechanisms (see Introduction) that need to be addressed in further research works.

In conclusion, the present findings indicate that voluntary activation and accuracy in force control can be increased by an enhanced level of noradrenaline concentration. This improvement in motor performance is accompanied by changes located at both cortical and spinal levels.

The authors thank Arnaud Bourroux and Allan Verleye for their help in data collection and Dr. Luk Buyse for the medical supervision of the study.

This study was supported by a grant from the Research Council of the Université Libre de Bruxelles. Bart Roelands was supported by a Fund for Scientific Research Flanders (FWO). Publications costs were partly covered by the University Foundation, Belgium.

The authors report no conflict of interest. The results of the present study do not constitute endorsement by the American College of Sports Medicine. The findings are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.


1. Xing B, Li YC, Gao WJ. Norepinephrine versus dopamine and their interaction in modulating synaptic function in the prefrontal cortex. Brain Res. 2016;1641(Pt B):217–33.
2. Benarroch EE. The locus ceruleus norepinephrine system: functional organization and potential clinical significance. Neurology. 2009;73(20):1699–704.
3. Berridge CW, Waterhouse BD. The locus coeruleus–noradrenergic system: modulation of behavioral state and state-dependent cognitive processes. Brain Res Rev. 2003;42(1):33–84.
4. Plewnia C, Hoppe J, Hiemke C, Bartels M, Cohen LG, Gerloff C. Enhancement of human cortico-motoneuronal excitability by the selective norepinephrine reuptake inhibitor reboxetine. Neurosci Lett. 2002;330(3):231–4.
5. Plewnia C, Hoppe J, Cohen LG, Gerloff C. Improved motor skill acquisition after selective stimulation of central norepinephrine. Neurology. 2004;62(11):2124–6.
6. Wang LE, Fink GR, Dafotakis M, Grefkes C. Noradrenergic stimulation and motor performance: differential effects of reboxetine on movement kinematics and visuomotor abilities in healthy human subjects. Neuropsychologia. 2009;47(5):1302–12.
7. Wang LE, Fink GR, Diekhoff S, Rehme AK, Eickhoff SB, Grefkes C. Noradrenergic enhancement improves motor network connectivity in stroke patients. Ann Neurol. 2011;69(2):375–88.
8. Zittel S, Weiller C, Liepert J. Reboxetine improves motor function in chronic stroke. A pilot study. J Neurol. 2007;254(2):197–201.
9. Kuo HI, Paulus W, Batsikadze G, Jamil A, Kuo MF, Nitsche MA. Acute and chronic noradrenergic effects on cortical excitability in healthy humans. Int J Neuropsychopharmacol. 2017;20(8):634–43.
10. Heckman C, Enoka R. Physiology of the motor neuron and the motor unit. In: Eisen A, editor. Handbook of Clinical Neurophysiology. Vol 4. Amsterdam: Elsevier B.V.; 2004. pp. 119–47.
11. Heckman CJ, Mottram C, Quinlan K, Theiss R, Schuster J. Motoneuron excitability: the importance of neuromodulatory inputs. Clin Neurophysiol. 2009;120(12):2040–54.
12. Enoka RM, Christou EA, Hunter SK, et al. Mechanisms that contribute to differences in motor performance between young and old adults. J Electromyogr Kinesiol. 2003;13(1):1–12.
13. Carville SF, Perry MC, Rutherford OM, Smith IC, Newham DJ. Steadiness of quadriceps contractions in young and older adults with and without a history of falling. Eur J Appl Physiol. 2007;100(5):527–33.
14. Marmon AR, Pascoe MA, Schwartz RS, Enoka RM. Associations among strength, steadiness, and hand function across the adult life span. Med Sci Sports Exerc. 2011;43(4):560–7.
15. Pierrot-Deseilligny C, Burke D. The Circuitry of the Human Spinal Cord: Its Role in Motor Control and Movement Disorders. Cambridge (UK): Cambridge University Press; 2005.
16. Duclay J, Pasquet B, Martin A, Duchateau J. Specific modulation of corticospinal and spinal excitabilities during maximal voluntary isometric, shortening and lengthening contractions in synergist muscles. J Physiol. 2011;589(Pt 11):2901–16.
17. Rossini PM, Burke D, Chen R, et al. Non-invasive electrical and magnetic stimulation of the brain, spinal cord, roots and peripheral nerves: basic principles and procedures for routine clinical and research application. An updated report from an I.F.C.N. Committee. Clin Neurophysiol. 2015;126(6):1071–107.
18. Kohl MM, Paulsen O. The roles of GABAB receptors in cortical network activity. Adv Pharmacol. 2010;58:205–29.
19. Klass M, Duchateau J, Rabec S, Meeusen R, Roelands B. Noradrenaline reuptake inhibition impairs cortical output and limits endurance time. Med Sci Sport Exer. 2016;48(6):1014–23.
20. Klass M, Roelands B, Levenez M, et al. Effects of noradrenaline and dopamine on supraspinal fatigue in well-trained men. Med Sci Sports Exerc. 2012;44(12):2299–308.
21. Doguet V, Jubeau M. Reliability of H-reflex in vastus lateralis and vastus medialis muscles during passive and active isometric conditions. Eur J Appl Physiol. 2014;114(12):2509–19.
22. Devanne H, Lavoie BA, Capaday C. Input–output properties and gain changes in the human corticospinal pathway. Exp Brain Res. 1997;114(2):329–38.
23. Todd G, Taylor JL, Gandevia SC. Measurement of voluntary activation of fresh and fatigued human muscles using transcranial magnetic stimulation. J Physiol. 2003;551(Pt 2):661–71.
24. Hajos M, Fleishaker JC, Filipiak-Reisner JK, Brown MT, Wong EH. The selective norepinephrine reuptake inhibitor antidepressant reboxetine: pharmacological and clinical profile. CNS Drug Rev. 2004;10(1):23–44.
25. Martin A, Carpentier A, Guissard N, van Hoecke J, Duchateau J. Effect of time of day on force variation in a human muscle. Muscle Nerve. 1999;22(10):1380–7.
26. Klimstra M, Zehr EP. A sigmoid function is the best fit for the ascending limb of the Hoffmann reflex recruitment curve. Exp Brain Res. 2008;186(1):93–105.
27. Carroll TJ, Riek S, Carson RG. Reliability of the input–output properties of the cortico-spinal pathway obtained from transcranial magnetic and electrical stimulation. J Neurosci Methods. 2001;112(2):193–202.
28. Orth M, Rothwell JC. The cortical silent period: intrinsic variability and relation to the waveform of the transcranial magnetic stimulation pulse. Clin Neurophysiol. 2004;115(5):1076–82.
29. Martin PG, Gandevia SC, Taylor JL. Output of human motoneuron pools to corticospinal inputs during voluntary contractions. J Neurophysiol. 2006;95(6):3512–8.
30. Hounsgaard J, Hultborn H, Jespersen B, Kiehn O. Bistability of alpha-motoneurons in the decerebrate cat and in the acute spinal cat after intravenous 5-hydroxytryptophan. J Physiol. 1988;405:345–67.
31. Powers RK, Binder MD. Input–output functions of mammalian motoneurons. Rev Physiol Biochem Pharmacol. 2001;143:137–263.
32. Fung SJ, Manzoni D, Chan JY, Pompeiano O, Barnes CD. Locus coeruleus control of spinal motor output. Prog Brain Res. 1991;88:395–409.
33. Mottram CJ, Jakobi JM, Semmler JG, Enoka RM. Motor-unit activity differs with load type during a fatiguing contraction. J Neurophysiol. 2005;93(3):1381–92.
34. Phillips SK, Bruce SA, Newton D, Woledge RC. The weakness of old age is not due to failure of muscle activation. J Gerontol. 1992;47(2):M45–9.
35. Behm DG, Whittle J, Button D, Power K. Intermuscle differences in activation. Muscle Nerve. 2002;25(2):236–43.
36. Semmler JG, Sale MV, Meyer FG, Nordstrom MA. Motor-unit coherence and its relation with synchrony are influenced by training. J Neurophysiol. 2004;92(6):3320–31.
37. Jesunathadas M, Klass M, Duchateau J, Enoka RM. Discharge properties of motor units during steady isometric contractions performed with the dorsiflexor muscles. J Appl Physiol (1985). 2012;112(11):1897–905.
38. Farina D, Negro F, Muceli S, Enoka RM. Principles of motor unit physiology evolve with advances in technology. Physiology (Bethesda). 2016;31(2):83–94.


Copyright © 2018 by the American College of Sports Medicine