The influence of residual force enhancement on spinal and supraspinal excitability

Participants

A total of 11 healthy male participants with a mean ± standard deviation (SD) age of 24 ± 2 years, height of 177 ± 4 cm, and mass of 75 ± 7 kg were recruited from the university population for participation in the study. All had no prior history of neuromuscular disease or ankle joint injuries. Data were collected within a single testing session. Participants gave written informed consent prior to testing and all procedures were approved by the Human Research Ethics Board of the University of Guelph (REB: 15NV008).

Experimental set-up

The following methods are used frequently in our lab and there is overlap in the language used throughout this methods section to that of Sypkes et al. (2017). A HUMAC NORM dynamometer (CSMi Medical Solutions, Stoughton, MA, USA) was used for all torque, angular velocity, and position recordings. Each participant sat with their right hip and knee angles set at 110° and 140° (180°; straight), respectfully. Joint angles were measured using a goniometer. The right knee was immobilized with the dynamometer’s leg restraint (superior) and a malleable cushion (inferior), while movement at the torso was restricted with a four-point seatbelt harness. The right foot was fixed to a dorsi/plantar flexor adaptor with one inelastic strap placed over the ankle and another at the mid-distal portion of the metatarsals. The maximal ankle dorsiflexion and plantar flexion angles were set to 90° and 130° (90°; neutral), respectively, allowing for 40° of ankle excursion.

Locations for the EMG electrodes were prepared by shaving and cleaning the skin with alcohol swabs. Silver–silver chloride (Ag-AgCl) electrodes (1.5 × 1 cm: Kendall, Mansfield, MA, USA) were used for all recordings. The active electrode was placed over the tibialis anterior approximately seven cm inferior and two cm lateral to the tibial tuberosity, and a reference electrode was placed over the distal tendon of the tibialis anterior, at the level of the malleoli. To record antagonist muscle activation, the active electrode was placed on the soleus, along the midline of the leg approximately two cm inferior to the border of the heads of the gastrocnemii, and a reference electrode was placed on the calcaneal tendon. A single ground electrode was centered on the patella.

Surface EMG, torque, angular velocity, joint angle, and stimulus trigger data were converted to digital format using a 12-bit analog-to-digital converter (PowerLab System 16/35; ADInstruments, Bella Vista, NSW, Australia), and analyzed with Labchart software (Labchart, Pro Modules 2014, v. 8). Torque and EMG data were recorded at a sampling rate of 1,000 and 2,000, respectively. EMG data were band pass filtered using a digital filter (3–1,000 Hz). Figure 1 depicts the joint angle, integrated EMG (iEMG), and torque traces for a single trial.

Deep fibular and tibial nerve stimulation

To normalize voluntary EMG and evoked potentials, maximal compound muscle action potentials (M-waves) were recorded over the tibialis anterior (Fig. 2A) and soleus muscles by transcutaneously stimulating the deep fibular and tibial nerves, respectively, with a standard clinical bar electrode (Empi, St. Paul, MN, USA) coated in conductive gel. The deep fibular nerve was located by palpating the head of the fibula and moving posteroinferiorly until the nerve was intercepted. The tibial nerve, innervating the plantar flexor muscles, was found by locating the distal tendon of the semitendinosus muscle and moving laterally while palpating deep into the popliteal fossa. All peripheral nerve stimuli were delivered as a single pulse from a constant current, high voltage stimulator (model DS7AH; Digitimer, Welwyn Garden City, Hertfordshire, UK). Voltage was set to a maximum of 400 V and pulse width to 200 μs. Current was increased until a plateau was reached for the peak-to-peak amplitude of the M-wave (M_max). To ensure consistent activation of all motor units throughout the experiment, the current (60–150 mA) was adjusted to a supramaximal level, equivalent to 110% of that required to generate M_max.

Experimental procedures

Each isometric control trial (ISO) was followed by an RFE trial. Protocol A was followed by protocol B (details below) and this sequence was repeated a total of four times. Thus, four ISO trials and four RFE trials were performed for each of the two protocols for a total of 16 contractions matching an iEMG target of 40% (Fig. 1B). Participants were given visual feedback of the iEMG trace on the computer monitor and were verbally encouraged to match the target as closely as possible during all contractions. Three minutes of rest separated all submaximal contractions throughout the experiment.

Data analysis and statistics

The mean torque and root mean squared EMG (EMG_RMS) were calculated in a 500 ms window occurring prior to each stimulus. A paired t-test was performed to compare the torque and EMG data between RFE and ISO trials for each stimulus to validate the presence of RFE at the time of stimulation. With respect to RFE, non-responders were identified as participants who displayed no increase in isometric steady-state torque following active lengthening as compared with the purely isometric contraction (i.e., RFE). When no RFE occurred during MEP and CMEP trials, all corresponding steady-state isometric and purely isometric torque, raw and normalized tibialis anterior and soleus EMG, M_max, CMEP, and MEP potential data were removed from data analysis (n = 2 subjects). While non-responders are common in most investigations of RFE, it remains unclear why some individuals do not display RFE (Seiberl et al., 2015). It has, however, been hypothesized that differences in muscle fiber types, the magnitude of muscle length change during stretch, and titin isoforms may contribute to differences between responders and non-responders (Joumaa et al., 2008; Seiberl et al., 2015). Additionally, one subject was removed for experimenters failing to match MEP peak to peak amplitude to 40% of their M_max. To assess spinal excitability in the RFE and ISO states, CMEPs were normalized to M_max (CMEP/M_max) to control for possible changes in peripheral excitability. To measure supraspinal excitability in the RFE and ISO states, MEPs were normalized to CMEPs (MEP/CMEP) to control for any changes in both spinal and peripheral excitability. The EMG_RMS of the resting M_max recorded at the tibialis anterior and soleus was used to normalize voluntary EMG for each muscle. The EMG_RMS of the soleus M_max was also used to quantify antagonist coactivation.

In order to detect and subsequently remove outlier data for each participant, a mean value was created for normalized CMEPs and MEPs and any individual response which fell more than two SDs above or below the mean was rejected. A paired t test was used to test for differences in M_max (mV), CMEP (% M_max), and MEP (% CMEP) data between the RFE and ISO states to elucidate RFE-induced changes in peripheral, spinal, and supraspinal excitability, respectively. In order to assess any effects of fatigue during the experimental protocol, a paired t test was also used to compare torque produced during the MVC performed before and after the experiment. Effect sizes are reported as Cohen’s d. Descriptive data found in text are reported as means ± SD, and presented in figures as means ± standard error of the mean. Significance was determined based on a p-value of < 0.05.

Maximum voluntary contraction torque and voluntary activation

Pre-trial MVC torque was 24.4 ± 5.2 Nm, and all participants achieved near-maximal values for voluntary activation (99.6% ± 0.6%). Following the experimental protocol, MVC torque and voluntary activation were reassessed. The post-trial MVC torque (22.7 ± 6.0 Nm) was not significantly different from the pre-trial value (paired t: df (10), t-value (−2.01), p = 0.07, d_z = 0.60), and all participants remained capable of achieving near-maximal voluntary activation values (98.7% ± 1.7%).

Dorsiflexion torque and muscle activity in the RFE state following active lengthening

Following active lengthening, steady-state isometric torque was significantly greater (paired t: df (10), t-value (−5.89), p = 0.00015, d_z = 2.02) than the torque recorded during purely isometric contractions at the same muscle length (Fig. 3A), resulting in an average RFE across all participants of 10.0% ± 6.7%. Participants successfully maintained the iEMG target level such that EMG of the tibialis anterior did not differ prior to stimulation during the RFE and ISO contractions (paired t: df (10), t-value (−0.28), p = 0.79, d_z = 0.08), indicating that motor neuron output was similar in both the RFE and ISO states. Antagonist coactivation (i.e., soleus) was not different (paired t: df (10), t-value (−0.35), p = 0.73, d_z = 0.12) between the RFE and ISO contractions, respectively (Figs. 3C and 3D).

Evoked muscle responses in the residual force enhanced state following active lengthening

There was no significant difference (paired t: df (10), t-value (0.09), p = 0.93, d_z = 0.04) in M_max peak-to-peak amplitude between the RFE and ISO state (Fig. 3B). During the ISO state, the peak-to-peak amplitudes of CMEPs and MEPs were successfully matched to 43.3% ± 8.7% and 41.6% ± 11.9% of the resting M_max, respectively, and were not significantly different from each other (paired t: df(10), t-value(0.54), p = 0.60, d_z = 0.16). When CMEPs were normalized to M_max (CMEP/M_max), there was a significant 17.1% ± 24.8% decrease in peak-to-peak amplitude during the RFE state when compared with the ISO state (paired t: df (10), t-value (2.32), p = 0.04, d_z = 0.70; Fig. 3E). When MEPs were normalized to CMEPs (MEP/CMEP), there was no significant difference (paired t: df (10), t-value (−1.58), p = 0.15, d_z = 0.48) in the mean peak-to-peak amplitudes between RFE and ISO contractions (Fig. 3F).

CNS excitability in RFE state

Following active lengthening as compared with a purely isometric contraction, when matching force output, there is a reduction (range: 5–20%) in the activation required to maintain force production as assessed via surface EMG (Oskouei & Herzog, 2005; Seiberl et al., 2012; Jones, Power & Herzog, 2016). From a mechanical perspective, in the force enhanced isomeric steady-state, there is a relatively greater contribution of passive force to total force production, possibly owing to stiffening of the giant molecular spring titin (Herzog, 2018). As a result of this property of muscle, during torque- and activation-matching experiments, the RFE EMG-torque relationship is shifted to the right as compared with a purely isometric contraction (Paquin & Power, 2018), indicating lower activation required to achieve a given torque output. Reduced activation indicates a down-regulation of motor neuron output during the isometric steady-state following active muscle lengthening to achieve the same force level as a purely isometric contraction. This activation reduction in the RFE state has been attributed to a reduction in the number of active motor units (Altenburg et al., 2008) or a reduction in motor unit firing rate (Jakobi et al., 2018). Important to the present study is the use of an activation matching task (i.e., activation clamp). By matching activation, despite the same motor neuron output (40% EMG) for the purely isometric contraction and the isometric steady-state following active lengthening, we were able to show 10% RFE and a 17% reduction in spinal excitability. These findings provide support that the history-dependence of torque alters corticospinal excitability through a neuromechanical coupling (Hahn et al., 2012; Grant et al., 2017; Sypkes et al., 2017). However, there was no change in supraspinal excitability. These findings differ from Hahn et al. (2012) which showed, for the soleus muscle, an increase in supraspinal excitability and no change in spinal excitability following maximal effort lengthening contractions in the RFE state. The divergent findings of the present study and Hahn et al. (2012), may be attributed to the task-dependent nature of corticospinal excitability (For review see Kalmar, 2018). Corticospinal excitability has been shown to differ across contraction intensities (Martin, Gandevia & Taylor, 2006; Oya, Hoffman & Cresswell, 2008; McNeil et al., 2011) and for different muscle groups (Hoffman et al., 2009; Giesebrecht et al., 2010). Thus, it is possible there are task- (maximal vs. submaximal) and muscle-dependent (soleus vs. tibialis anterior) alterations to corticospinal excitability in the RFE state of varying contraction intensities, which requires further investigation.

While the results of the present study diverge from the findings of Hahn et al. (2012), they fall in line with results from other previous studies investigating the influence of altered muscle torque production capacity on corticospinal excitability. When torque production capacity is increased during submaximal lengthening contractions, Gruber et al. (2009) found that when compared to isometric contractions, there is a reduction in spinal excitability (i.e., CMEP amplitude). Lengthening contractions were also associated with increased supraspinal excitability (i.e., MEP/CMEP ratios). In contrast, following active muscle shortening in the torque depressed state, when the ability of the muscle to generate force is impaired, we recently demonstrated increased spinal excitability (i.e., normalized CMEP amplitude) and no change in supraspinal excitability (i.e., normalized MEP amplitude) (Sypkes et al., 2017). The present study, together with the previously mentioned investigations, provides strong evidence for a neuromechanical coupling that may alter corticospinal excitability in a manner that may be related to the capacity of the muscle to produce force.

Tension-dependent spinal inhibitory mechanisms

We observed a decrease in spinal excitability in the isometric steady-state following active lengthening contractions as compared with the purely isometric contraction. Given the task was activation matched at 40% EMG for both conditions, thus achieving a similar motor neuron output, the alteration to spinal excitability was most likely owing to peripheral sensory inputs. Ib afferents are anatomically located in series with the muscle and aponeurosis, and provide inhibitory sensory feedback to the agonist motor neuron pool during tonic, non-locomotor tasks (Conway, Hultborn & Kiehn, 1987; McCrea et al., 1995). Firing of Ib afferents is modulated in a tension-dependent manner. Therefore, in the present study, the smaller normalized CMEP in the RFE compared to ISO state may be attributed to greater Ib afferent firing due to the increased isometric force after muscle lengthening. Nevertheless, due to the nature of maintaining a constant level of muscle activation, spinal inhibition would be expected to be paired with an increase in supraspinal excitability (i.e., increased normalized MEP amplitude) to counteract inhibition from afferent muscle feedback. While this counterbalance was not observed, there are several subjects in the present study who show a negative relationship between their normalized CMEP and MEP amplitudes, such that an increase or decrease in CMEP amplitude was accompanied by an opposite change in MEP amplitude (Figs. 3E and 3F). Although not statistically significant, this relationship between supraspinal and spinal excitability is similar to what was reported for maximal (Grant et al., 2017) and submaximal torque depression (Sypkes et al., 2017) studies.