J Appl Physiol 116: 140 –148, 2014. First published December 5, 2013; doi:10.1152/japplphysiol.00906.2013.
Absence of lateral gastrocnemius activity and differential motor unit behavior in soleus and medial gastrocnemius during standing balance Martin E. Héroux,1 Christopher J. Dakin,1 Billy L. Luu,1 John Timothy Inglis,1,2,3 and Jean-Sébastien Blouin1,2,4 1
School of Kinesiology, University of British Columbia, Vancouver, British Columbia, Canada; 2Brain Research Center, University of British Columbia, Vancouver, British Columbia, Canada; 3International Collaboration for Repair Discoveries, University of British Columbia, Vancouver, British Columbia, Canada; and 4The Institute for Computing, Information and Cognitive Systems, University of British Columbia, Vancouver, British Columbia, Canada Submitted 6 August 2013; accepted in final form 2 December 2013
Héroux ME, Dakin CJ, Luu BL, Inglis JT, Blouin J-S. Absence of lateral gastrocnemius activity and differential motor unit behavior in soleus and medial gastrocnemius during standing balance. J Appl Physiol 116: 140 –148, 2014. First published December 5, 2013; doi:10.1152/japplphysiol.00906.2013.—In a standing position, the vertical projection of the center of mass passes in front of the ankle, which requires active plantar-flexor torque from the triceps surae to maintain balance. We recorded motor unit (MU) activity in the medial (MG) and lateral (LG) gastrocnemius muscle and the soleus (SOL) in standing balance and voluntary isometric contractions to understand the effect of functional requirements and descending drive from different neural sources on motoneuron behavior. Single MU activity was recorded in seven subjects with wire electrodes in the triceps surae. Two 3-min standing balance trials and several ramp-and-hold contractions were performed. Lateral gastrocnemius MU activity was rarely observed in standing. The lowest thresholds for LG MUs in ramp contractions were 20 –35 times higher than SOL and MG MUs (P ⬍ 0.001). Compared with MUs from the SOL, MG MUs were intermittently active (P ⬍ 0.001), had higher recruitment thresholds (P ⫽ 0.022), and greater firing rate variability (P ⬍ 0.001); this difference in firing rate variability was present in standing balance and isometric contractions. In SOL and MG MUs, both recruitment of new MUs (R2 ⫽ 0.59 – 0.79, P ⬍ 0.01) and MU firing rates (R2 ⫽ 0.05– 0.40, P ⬍ 0.05) were associated with anterior-posterior and medio-lateral torque in standing. Our results suggest that the two heads of the gastrocnemius may operate in different ankle ranges with the larger MG being of primary importance when standing, likely due to its fascicle orientation. These differences in MU discharge behavior were independent of the type of descending neural drive, which points to a muscle-specific optimization of triceps surae motoneurons. triceps surae; motor units; standing balance; human AS A PHYSICAL LOAD, THE STANDING
human body is an inverted pendulum and is thus inherently unstable. With the vertical projection of the body’s center of mass usually passing in front of the ankle joint and insufficient passive stiffness of the Achilles tendon and other tissues (18), active plantar-flexor torque is required to maintain standing balance (18, 20). In humans, plantar flexion is primarily accomplished by the triceps surae: the soleus muscle (SOL) and the medial and lateral heads of the gastrocnemius muscle (MG and LG, respectively). Despite being agonists, these muscles differ in many respects in humans. For example, the gastrocnemius is a biarticular muscle with a considerable proportion (⬃52%) of
Address for reprint requests and other correspondence: J.-S. Blouin, 2106081 Univ. Boulevard, Vancouver, BC V6T 1Z1 Canada (e-mail: jsblouin @interchange.ubc.ca). 140
fast-twitch muscle fibers, whereas SOL is a monoarticular muscle with a very high proportion (⬃88%) of slow-twitch muscle fibers (1, 13). Opposite electromyographic responses (i.e., facilitation vs. inhibition) have also been observed between MG and SOL and between gastrocnemius heads during reflex activation (8, 31, 40) and functional tasks such as weight shifting, walking, running, and hopping (7, 23, 31, 32). These anatomical, neurophysiological, and functional differences support the notion that despite sharing a common distal tendon, the muscles of the human triceps surae appear to have distinct functional roles. Whether these functional differences are associated with distinct motor unit (MU) discharge behavior between muscles and tasks has not been systematically examined. The distal portion of the triceps surae terminates into the Achilles tendon to transmit muscle force to the posterior aspect of the calcaneous (heel) bone and generate plantar flexion. Given this common tendon, torque measures cannot be used to distinguish MG, LG, and SOL contributions to overall plantar flexion. Although surface electromyography can be used to measure triceps surae neuromuscular activity during standing balance (14), its interpretation can be difficult because it is a filtered cumulative sum of nearby MU action potentials (9). This makes the association between surface electromyography and the physiological processes that generate muscle activity difficult to pinpoint and prone to erroneous conclusions (9, 10, 26). On the other hand, the measurement of action potentials from individual MUs provides a direct measure of motor output from the spinal cord because of the one-to-one association between the firing of ␣-motoneurons and their associated muscle fibers (25). This allows key features of MU behavior such as recruitment thresholds, mean discharge rates, and discharge rate variability to be measured. The ability to simultaneously record MU activity from several agonist muscles allows ␣-motoneuron firing behavior to be compared between muscles during the same task, and to investigate whether firing behavior is modified in tasks with different neural drive. In human standing balance, SOL MUs discharge continuously, and modulate their firing rate with changes in ankle plantar-flexor torque (29). In contrast, MG MU activity is intermittent, with minimal rate coding; relying instead on MU recruitment to produce additional torque (37). It is unknown whether this continuous vs. intermittent MU activity is associated with differences in recruitment threshold or spike-tospike variability between these two muscles. Equally important is whether these differences are intrinsic to the muscles involved or whether they result from the type of neural drive that
8750-7587/14 Copyright © 2014 the American Physiological Society
http://www.jappl.org
Triceps Surae Motor Unit Activity During Standing Balance
controls standing balance. For example, Mochizuki et al. (27– 29) reported greater levels of common drive to pairs of SOL MUs in standing compared with an isometric task, but similar levels of MU synchronization. Whether differences in neural drive affect the discharge behavior of individual MUs in the human triceps surae is unknown. Even less is known about LG MU behavior. As the lateral counterpart of the MG, LG has the same muscle fiber composition and distal tendon, and is also biarticular. Although it is generally assumed that MUs from the lateral portion of the gastrocnemius muscle behave similarly to MG MUs, this has not been investigated in humans (38). Here we examine the behavior of SOL, MG, and LG MUs in two tasks that involve different neural drive to the motoneurons of these muscles. Standing balance and voluntary ramp-and-hold contractions can be performed with similar levels of plantarflexor torque, but the neural drive to triceps surae motoneurons differs substantially between these tasks: standing balance is believed to involve subcortical structures with vestibular contributions, whereas muscle activation during a nonbalancing task is believed to be largely cortically driven, with little known vestibular contribution (21, 22). We recorded indwelling MU activity from SOL, MG, and LG in these two tasks to investigate several key questions about the neural control and functional role of the human triceps surae. Specifically, whether the phasic muscle activity observed in MG is also present in LG; the extent to which MU discharge rates and discharge variability differ between muscles of the triceps surae; whether muscle-specific MU behavior is dependent on the type of neural drive (balance vs. voluntary drive); and whether MU recruitment and rate coding both contribute to modulate torque output by the triceps surae in standing balance. METHODS
•
Héroux ME et al.
141
Fig. 1. Experimental setup and procedures. A: a total of 14 fine wire electrodes were inserted with ultrasound guidance in the triceps surae. Black dots indicate approximate site for each insertion. B: triceps surae motor unit (MU) activity was recorded during two 3-min standing balance trials. Subjects stood with their right foot on a forceplate from which vertical force, anterior-posterior (A-P) torque, and medio-lateral (M-L) torque were recorded. C: subjects were braced to a rigid board at the level of the chest and pelvis. MU activity was recorded during a series of ramp-and-hold isometric plantar flexion contractions at three different contraction intensities.
Testing Protocol
Subjects Seven healthy subjects (1 woman; age range 23– 48 years) with no known history of neurological disease or injury participated in the study. The experimental protocol was explained and written informed consent was obtained prior to the study. All study procedures conformed to the standards of the Declaration of Helsinki and were approved by the University of British Columbia’s Clinical Research Ethics Board. Single MU Recordings Single MU activity was recorded using custom-made, sterilized, fine-wire indwelling electrodes. The electrodes were made using two 0.05-mm insulated stainless steel wire strands wound together (California Fine Wire, Grover Beach, CA). The two wires were cut at the recording end, exposing the noninsulated cross-section of the wires (i.e., 0.05 mm). These tips were folded back to create two 1-mm barbs to anchor the wires in the muscle. Insertion of the wire electrodes into the right MG (six wires), LG (four wires), and SOL (four wires) was performed under ultrasound guidance (SonoSite Micro Maxx, Bothell, WA) (Fig. 1A). After cleaning the skin, a surface ground electrode was placed over the right lateral malleolus. MU activity was recorded using a high impedance GRASS 15CT amplifier (Astromed, West Warwick, RI), filtered (band-pass 30 – 6,000 Hz), amplified (⫻5,000), and digitized at 15,152 Hz using a 16-bit CED DAQ-card and Spike2 software (Cambridge Electronics Design, Cambridge, UK).
Participants stood barefoot with their right foot on a forceplate (Bertec 4060 – 80; Bertec, Columbus, OH) and their left foot on a large stable surface of equal height (Fig. 1B); the medial malleoli were 10 –15 cm apart. The plantar/dorsi-flexion axis of rotation of the right foot was aligned with the medio-lateral axis of the forceplate at a known distance from the midline of the forceplate. Vertical force, anterior-posterior (A-P) torque, and medio-lateral (M-L) torque were measured for the right foot. Subjects were instructed to stand with their eyes open and their hands by their side. MU activity and forceplate data were recorded during two 3-min trials. In four subjects, verbal cueing to stand with a slightly forward lean was required to elicit clear gastrocnemius activity. Subjects were provided with a 2-min rest period between trials. Subjects were then braced to a rigid board (Fig. 1C) and asked to perform a series of isometric ramp (15 s) and hold (45 s) plantar flexion contractions against the forceplate. The position of the feet and the mean ankle angle remained the same as during the balance trials. This allowed us, in most cases, to measure the same MUs during standing balance and voluntary ramp-and-hold trials and determine the effects of differing neural drive on MU discharge behavior. The ramp phase of the contractions was used to determine MU recruitment thresholds, and the hold phase was used to compare MU discharge behavior between balance and voluntary conditions. Because we tracked several MUs simultaneously, it was not feasible to have subjects produce all the contraction intensities required to have the same mean firing rate between balance and voluntary conditions for each MU. On the basis of
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
142
Triceps Surae Motor Unit Activity During Standing Balance
pilot experiments, target intensities that corresponded to the mean A-P torque and 35% of the mean A-P torque produced in balance trials were selected as target torques because they generated similar mean discharge rates between balance and voluntary trials in a majority of MUs. A third target intensity of 90% of maximal A-P torque produced in balance trials was chosen to identify the recruitment threshold and maintain a similar mean discharge rate to that produced during standing balance for higher threshold MUs. For each trial, subjects were provided with visual feedback of the target ramp-and-hold A-P torque profile and their own A-P torque production. Each contraction intensity was performed three times, with the order of presentation randomized. A 2-min rest period separated each trial.
•
Héroux ME et al.
active MUs across subjects and standing balance kinetic variables. For each SOL and MG MU we determined whether they were active (ISI ⬍250 ms ⫽ 1) or inactive (ISI ⬎250 ms ⫽ 0) for each sample of kinetic data. Normalized A-P torque, M-L torque, or vertical force values (range produced in standing balance trials corresponded to 0 –100%) were then separated into 10% bins and the mean for each bin’s binary data (i.e., 0 ⫽ not active, 1 ⫽ active) was calculated to determine whether a MU was, on average, active (mean ⬎0.5) or silent (mean ⱕ0.5). For each bin the total number of active SOL and MG MUs across all subjects was determined; this analysis provided an overall count of active MUs for normalized A-P torque, M-L torque, or vertical force in standing balance. Statistical Analysis
Data Analysis To have a representative sample of active MUs, our goal for standing balance trials was to obtain, across all subjects, a total of 5 MUs for each of the 14 insertion sites; the ramp-and-hold contractions for these MUs were also analyzed. MU action potentials were identified and extracted using Spike2 software (Cambridge Electronic Design, Cambridge, UK). The algorithm identified MUs on the basis of size, shape, and timing. The identified MUs were then manually reviewed and sorted to include unmatched action potentials and to solve for instances in which two MU discharges were superimposed. Once MUs were sorted, discharge times were exported to Matlab (Mathworks, Natick, MA) for further analysis. MU discharge behavior and recruitment threshold. For standing balance and voluntary ramp-and-hold trials the number of active MUs, activation ratios, recruitment thresholds, mean interspike intervals (ISI), and ISI variability were calculated. The activation ratio quantifies the percentage of time a MU had ISI ⬍250 ms during standing balance trials (11, 34), with 0% corresponding to no activity and 100% representing continuous activity. For each MU, mean ISI and ISI variability were computed for periods when ISI ⬍250 ms during standing balance (two trials) and voluntary ramp-and-hold conditions (45 s constant torque portion; three trials). MU discharge variability was calculated as the coefficient of variation of ISI. The recruitment thresholds of MUs active in standing balance were determined from ramp-and-hold trials. The A-P torque at which each identified MU was first recruited (three consecutive MU action potentials with ISI ⬎250 ms) was identified and normalized to the maximum A-P torque produced in standing balance trials (36). Because LG showed little to no MU activity in standing balance despite clear MU activity in high-intensity ramp-and-hold trials, the first three MUs recruited for all LG, MG, and SOL ramp-and-hold trials were identified and their recruitment thresholds determined. This allowed us to compare the lowest threshold MUs for LG (n ⫽ 54), MG (n ⫽ 83), and SOL (n ⫽ 51). Relationship between sway kinetics and MU discharge rate and recruitment. Whereas the primary action of the triceps surae is ankle plantar flexion (i.e., A-P torque), MG activity is also related to lateral body motion and the abduction moment about the ankle in standing balance (37). To investigate the relationship between sway kinetics and MU discharge rates in standing balance (i.e., rate coding), we determined A-P torque, M-L torque, and vertical force for each discharge of each MU. If an MU discharged 2,000 times over the course of a 3-min balance trial, there would be 2,000 values for A-P torque, M-L torque, and vertical force. Stepwise linear regression was used to determine the relationship between the discharge rate of each MU and the measured kinetic variables associated with standing balance in MUs that fired ⬎250 times in the two standing balance trials. For each MU we retained the number of times a significant regression model was found, the R2 value of the model, and which kinetic variables were retained as part of the model. To investigate the role of MU recruitment in the generation of vertical force and M-L and A-P torques produced in standing balance trials, we determined the linear relationship between the number of
Nonparametric statistics were used for data that were non-normally distributed (Kolmogorov-Smirnov test) or had unequal variance between muscle groups (homogeneity of variance test). The KruskalWallis test was used to determine whether the recruitment thresholds of the first three activated MUs from each recorded signal differed between muscles. When a significant main effect was found, Dunn’s test was used for post hoc comparisons. The Mann-Whitney rank-sum test was used to compare activation ratios, recruitment thresholds, and stepwise linear regression R2 values between muscles for standing balance trials. The Rank Transform test was used to compare mean ISI and ISI variability across muscles and neural drive conditions (balance and voluntary ramp-and-hold). This test is a nonparametric equivalent to a two-factor ANOVA and calculates the F statistic on the basis of rank (12, 35). When a significant main effect or interaction was found, the Tukey test was used for post hoc comparisons. Linear regression was used to determine the relationship between the overall number of active MUs and A-P torque, M-L torque, or vertical force produced in standing balance trials, and an unpaired t-test was used to determine whether slope values were significantly different between muscles. The threshold for significance for all tests was ␣ ⫽ 0.05. Boxplots are used in figures; on each box, the central mark is the median, the edges of the box are the 25th and 75th percentiles, the whiskers are 1.5 times the length of the box, and data points outside this range are plotted individually. Values reported in the text are median and interquartile ranges. RESULTS
Across all subjects, a total of 39 MG MUs and 21 SOL MUs were decomposed and analyzed for standing balance and rampand-hold trials, with 5– 6 MUs at each of the six MG and four SOL sites (Fig. 1A). Only one LG MU was consistently active in standing balance trials, whereas another three MUs were active 8.7%, 4.1%, and 2.3% of the time. LG Activity and MU Recruitment Thresholds A consistent observation across subjects was that a forward lean on the verge of eliciting a step response was required to elicit LG MU activity. In line with these observations, there was little to no LG MU activity in standing balance trials. This lack of LG MU activity can be seen in Fig. 2A, which shows 1 min of standing balance data from five MG, four LG, and three SOL recordings from a typical subject. Although there is clear activity from one or more MUs on all MG and SOL recordings, there is no MU activity on any of the LG recordings; this was the case for the entire 6 min of balance data for this subject. Because LG MUs required greater plantar-flexor torque to be activated, we investigated the possibility that they had elevated recruitment thresholds. Figure 2B presents 30 s of
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
Triceps Surae Motor Unit Activity During Standing Balance
•
Héroux ME et al.
143
Fig. 3. Recruitment thresholds of earliest recruited MUs in the triceps surae. The recruitment threshold of the first three MUs recorded on each recording with good quality signals were identified for MG, LG, and SOL. Recruitment thresholds for MG MUs were significantly higher than for SOL MUs. More dramatically, LG MUs had much higher recruitment thresholds than both MG and SOL MUs.
MG and SOL MU Characteristics and Discharge Behavior
Fig. 2. Example of motor unit activity from all three muscles during standing balance and ramp-and-hold trials. A: 60 s of standing balance data from five recordings from medial gastrocnemius (MG) and four from lateral gastrocnemius (LG), and three recordings from soleus (SOL) of one subject (top); A-P torque, M-L torque, and vertical force (bottom). Although there is clear MU on all MG and SOL recordings, there was absolutely no LG activity. B: data from the same subjects for a ramp-and-hold 90% maximum A-P torque trial. Although numerous MUs are recruited in MG and SOL in the early part of the ramped plantar-flexion contraction, LG MUs are recruited considerably later.
ramp-and-hold data from a 90% maximum A-P torque trial for the same subject and the same recording sites as those presented in Fig. 2A. Although MG and SOL MUs were recruited at the start of the ramp contraction and new MUs were recruited throughout the ramp phase, all but one LG MU were recruited in the hold phase of the contraction. When considering the recruitment threshold of the first three activated MUs from each recording with good signal quality, there was a significant difference between muscles (Kruskal-Wallis, P ⬍ 0.001; Fig. 3). Post hoc comparisons revealed LG MU recruitment thresholds [LG 67.3% (43.5–75.9%)] were ⬃20 times greater than MG MU recruitment thresholds [3.4% (1.5– 8.9%), P ⬍ 0.001] and ⬃35 times greater than SOL MU recruitment thresholds [2.0% (0.9 – 6.0%), P ⬍ 0.001]. Comparatively, MG MU thresholds were only 1.7 times greater than those of SOL MUs (P ⫽ 0.001).
In standing balance trials, MG MU activity was intermittent compared with that of the more continuously active SOL MUs (Fig. 4). When considering the entire sample of MG and SOL MUs, activation ratios were significantly lower in MG MUs [73.6% (51.8 –95.2%)] compared with SOL [100% (94.4 – 100%)] (P ⬍ 0.001; Fig. 5A). In line with activation ratio
Fig. 4. Example of MU activity during standing balance. Top: data from two recordings from MG and two recordings from SOL from one subject. Above each recording in the instantaneous discharge frequency of one decomposed MU for each recording. The superimposed templates of for each MU are plotted to the right of each recording. The horizontal line in each instantaneous discharge plot indicates 4 Hz. Bottom: A-P torque, M-L torque, and vertical force recorded during this 30 s period.
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
144
Triceps Surae Motor Unit Activity During Standing Balance
•
Héroux ME et al.
tary ramp-and-hold trials (muscle main effect, P ⬍ 0.001; muscle ⫻ neural drive interaction, P ⫽ 0.185). These results indicate that ISI variability is approximately twice as large in MG MUs. Relationship Between Sway Kinetics and MU Discharge Rate and Recruitment The possible relationship between instantaneous discharge rates and A-P torque, M-L torque, and vertical force was first plotted for visual inspection. Figure 6, A and B, shows examples of these plots for two MG MUs and one SOL MU, with vertical force or M-L torque along the x-axis and A-P torque along the y-axis. The figure shows that the range of discharge rates was lower in SOL MUs, and that the increase in discharge rate with greater A-P torque was gradual for this MU. The two MG MUs also modulated their firing rate with changes in A-P torque, but this modulation was less uniform. Stepwise regression analysis of MG and SOL MUs revealed a significant regression model in all but one MG MU. In significant models, A-P torque was retained in 70% of cases, vertical force in 46%, and M-L torque in 41%. Although there was no trend for any kinetic variables to be retained more often by MG or SOL MUs, the strength of the relationship between MU discharge rate and kinetic variables, as reflected by R2 values, was ⬃30% less in MG MUs (Fig. 6C, P ⫽ 0.039). Examples of the association between standing balance kinetic variables and MG MU recruitment are visible in Fig. 2A
Fig. 5. MG and SOL MU behavior during standing balance and ramp-and-hold trials. A: activation ratio results indicate that MUs from SOL were active much more consistently than those from MG during standing balance trials (sway). B: these results are in line with the higher recruited thresholds found for MG MUs during ramp-and-hold trials (braced). C: mean interspike intervals (ISI) were comparable between MG and SOL during standing balance trials. This was also the case for the ramp-and-hold trials. Overall, mean ISI was slightly higher during standing balance trials compared with the ramp-and-hold trials. D: the ISI coefficient of variation was slightly but significantly higher during standing balance trials compared with ramp-and-hold trials. The difference in ISI variability between MG and SOL MUs was marked and this for both standing balance and ramp-and-hold trials. *P ⬍ 0.05; **P ⬍ 0.001.
results and the difference between MG and SOL MUs recruitment thresholds for the earliest recruited MUs, the recruitment threshold of MUs studied in standing balance trials was higher in MG than in SOL (P ⫽ 0.022; Fig. 5B). Although mean ISI was ⬃5% lower in voluntary ramp-andhold trials compared with balance trials (Fig. 5C; neural drive main effect, P ⫽ 0.029), there was no difference between SOL and MG mean ISI (muscles main effect, P ⫽ 0.799; muscles ⫻ neural drive interaction, P ⫽ 0.547). In terms of spike-to-spike variability, ISI variability was ⬃8% higher in standing balance trials compared with voluntary ramp-and-hold trials (neural drive main effect, P ⫽ 0.026; Fig. 5D). However, more noticeable was the greater variability in MG MUs compared with SOL MUs (see Fig. 4, compare last 5 s of MG and SOL data). Overall, coefficient of variation values were 0.41 (0.24 – 0.48) in MG MUs compared with 0.19 (0.14 – 0.22) in SOL MUs in standing balance trials, and 0.37 (0.28 – 0.45) in MG MUs compared with 0.17 (0.14 – 0.23) in SOL MUs in volun-
Fig. 6. Relationship between MU discharge rates and kinetic variables during standing balance trials. A: forceplate data from both standing balance trials with A-P torque along the y-axis and vertical force along the x-axis (left); 2-dimensional heat maps for two MG and one SOL MU (three right panels). Motor unit discharge rates were separated into bins and averaged to create these plots. Discharge rates were much lower for the SOL MU, and all three MUs increased their discharge rate with increasing plantar flexor torque. The two MG MUs also increased their discharge rate with increasing vertical force. B: data from the same three MUs are replotted with M-L torque along the x-axis. C: results from the stepwise linear regression between MU discharge rate and A-P torque, M-L torque, and vertical force. Overall, A-P torque was retained in 70% of the regression models compared with 41% for M-L torque and 46% for vertical force, and R2 values were significantly greater for SOL MU compared with MG MU. *P ⬍ 0.05.
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
Triceps Surae Motor Unit Activity During Standing Balance
and Fig. 4, with MU recruitment and derecruitment that coincide with increases or decreases in A-P or M-L torque. Linear regression was used to quantify the strength of the relationship between the overall number of active MUs across subjects and kinetic variables recorded in standing balance (Fig. 7). There was a strong linear relationship between the number of active MG and SOL MUs and A-P and M-L torque production; vertical force measures were not significantly related to the number of active MUs in either muscle. When considering the slopes of these relationships, the total number of active MG MUs increased at a similar rate as in the SOL (P ⬎ 0.201). DISCUSSION
The present study investigated the behavior of motor units from all three parts of the human triceps surae in two conditions that involve different neural drive: standing balance and voluntary isometric contractions. Whereas previous work focused on surface electromyography or indwelling recordings from a single muscle, our experimental approach allowed us to make these key observations on the behavior of MUs from the three portions of the triceps surae. 1) Lateral gastrocnemius MUs showed little to no activity in standing balance, and this appears to be due to recruitment thresholds being 20 –35 times higher than MG and SOL MUs. 2) Medial gastrocnemius MU discharge rate variability was twice that of SOL MUs in both standing balance and voluntary contractions. 3) These MU properties were not a consequence of the type of neural drive to the motoneuron pool. 4) Higher A-P and M-L torque values in standing balance are associated with increased MU firing rates and a greater number of active MUs for both SOL and MG. The 20-fold difference in MU recruitment threshold between two heads of a same muscle was unexpected and provides evidence of an important physiological difference between MG and LG. Similarly, a twofold difference in MU discharge variability between two agonists (SOL and MG) is not typical in humans (4, 39) and may reflect an adaptation to the type of contraction most often accomplished by each muscle. These differences were present in low to moderate contractions in both standing balance and isometric contractions, which points
•
Héroux ME et al.
145
to local (i.e., spinal) differences in the motoneuron pools that innervate the triceps surae rather than differences in descending drive. In the following sections we interpret our results in the context of previous research findings and discuss how these new insights provide evidence for functional optimization of the three muscles of the triceps surae. Lack of LG Activity During Standing Balance The recruitment threshold of the first active MG MUs in a ramped contraction were ⬃2 times greater than SOL MUs. Similarly, MG MUs active in standing balance trials had recruitment thresholds ⬃3 times greater than their SOL counterparts. Although certainly not negligible, these results are in stark contrast to the 20-fold difference in recruitment threshold that was noted between the medial and lateral aspects of the gastrocnemius muscle. Several other human muscles have distinct compartments or muscular heads. However, we were unable to locate any examples of large differences in recruitment thresholds between distinct compartments of a single human muscle. For example, the flexor digitorum superficialis has four components that correspond to the four digits upon which it acts, and the earliest recruited MUs from each muscle component have similar recruitment thresholds (3). Similarly, Kamo (15) found no difference in recruitment threshold between the first recruited MUs in three heads of the quadriceps femoris muscle: the vastus medialis, vastus lateralis, and rectus femoris. Thus the gastrocnemius seems somewhat unique with its two muscular heads that have MUs with dramatically different recruitment thresholds. The LG and MG are the two heads of a single muscle, and an important question is whether possessing a distinct population of higher-threshold MUs in the anatomically smaller LG would have a functional or physiological benefit. If there was a truly distinct population of higher-threshold MUs with larger motoneurons in LG, it might be possible for descending neural drive to selectively activate this motoneuron pool. Similarly, global descending drive to the plantar flexor motoneuron pool would lead to MU recruitment on the basis of Henneman’s size principle (9, 10, 25), and thus later recruitment of LG MUs. However, it is not clear why this would be more advantageous
Fig. 7. Relationship between MU recruitment and kinetic variables during standing balance trials. Results of the linear regression between the overall number of active MG and SOL MUs across subjects and normalized to vertical force (A), A-P torque (B), and M-L torque (C). Kinetic variables were normalized to the minimum and maximum values produced during the standing balance trials and the count of active MU was determined in 10% bins. There was a moderate to strong linear relationship between the number of active MUs and A-P and M-L torque for both MG and SOL, whereas no relationship was present for vertical force. J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
146
Triceps Surae Motor Unit Activity During Standing Balance
than a uniform distribution of MU sizes across both heads of the gastrocnemius. An alternative explanation that we favor is that our results reflect an anatomical difference between the MG and LG rather than a purely neurophysiological difference. There are examples in humans in which a portion of a motoneuron pool is preferentially activated because the muscle fibers they innervate have a mechanical advantage for a specific task or posture. For example, the intercostal muscles [see Butler (2) for a review] show a rostro-caudal and dorso-ventral gradient of activation that closely parallels their calculated mechanical advantage, and these patterns of activation are preserved in animal models when all afferent feedback is removed (5). In terms of the triceps surae and standing balance, low-level isometric plantar flexion (⬍20% maximum voluntary contraction) with the ankle joint in a neutral position is associated with muscle fiber lengths and pennation angles that are similar for MG and SOL (⬃40 mm and ⬃20°), but different from those of LG muscle fibers (⬃80 mm and ⬃10°) (16, 24). This anatomical difference and the possible mechanical disadvantage it provides to LG may be associated with a reduction in motoneuron excitability, which could contribute to higher recruitment thresholds and minimal LG MU activity in standing balance. More broadly, we propose that this anatomical arrangement may reflect an optimization via which MG and LG are preferentially activated within relatively distinct knee and ankle configurations, which would increase the functional capabilities of the triceps surae. The greater muscle volume of the MG may reflect that as bipeds, humans generate plantar flexor torques within a somewhat limited range of motion that is optimal for MG and its muscle fascicles. Overall, additional experiments that closely control both knee and ankle angle, movement velocity, torque generation, and task requirements are needed to confirm these hypotheses for the gastrocnemius. SOL and MG MU Activity in Standing Balance Several reports have indicated that the SOL is continuously active in standing balance, whereas the gastrocnemius tends to be more intermittently active (14, 29, 30, 37). This has led the proposal that ongoing SOL activity in standing balance provides a background plantar-flexion torque to counteract the static gravitational body load, whereas intermittent gastrocnemius activity generates dynamic ankle plantar-flexor torque to stabilize the moving body (40). In the present study, these patterns of global activation were accompanied by a twofold difference in the amount of discharge rate variability between the SOL and MG. This difference in variability was independent of the descending drive to the muscles, suggesting functional adaptation of the motoneuron pool innervating the muscles of the triceps surae. The greater variability in MG does not appear to be due to the phasic behavior of its MUs in standing balance because it was still present in isometric contractions when these MUs were continuously active. In humans, it is unusual for MUs from agonist muscles or different compartments of a single muscle to have large differences in discharge rate variability. For example, discharge rate variability is similar between the biceps brachii and the brachialis (4), and between the various muscular heads of the quadriceps femoris (39). Thus this leads to the question of whether this large difference in discharge rate variability is of
•
Héroux ME et al.
functional importance. For SOL, low discharge rate variability would be beneficial in its role as the primary plantar flexor in standing balance: it would result in relatively stable MU firing rates that would experience gradual increases and decreases with changes in A-P torque. This type of MU activity would generate relatively stable muscle tension, which seems appropriate for a muscle that is continually active in standing balance. Conversely, the MG is regarded as a powerful muscle that is less suited for sustained contractions. This is in line with the higher recruitment thresholds in MG MUs and the intermittent nature of MG activity in standing balance required to stabilize the moving body. In the present experiment, we also noted that MG MUs had greater discharge rate variability compared with SOL MUs. Greater discharge rate variability is associated with larger force fluctuations in low-level contractions (6, 31), which would not be optimal for standing balance when contraction levels are relatively low. However, greater discharge rate variability is also associated with greater force production in higher-threshold MUs (17), which would be beneficial in a muscle such as MG, which is often required to generate high levels of torque. Thus, differences in SOL and MG MU discharge behavior appear to reflect a physiological optimization for the types of contractions these muscles most often produce. Relationship Between Sway Kinetics and MU Discharge Rate and Recruitment In the present experiment, stepwise linear regression was used to investigate the relationship between MU discharge rate and standing balance kinetic variables. In line with the key role the triceps surae plays in plantar flexion, A-P torque was retained in 70% of the regression models compared with 40 – 45% for M-L torque and vertical force measures. These findings agree with recent reports of an association between center of pressure A-P movements and MU firing rates (29, 31). When we consider the strength of the relationship between MU firing rate and A-P sway, Vieira et al. (37) reported an r value of 0.07 for MG MUs, whereas a similar analysis by Mochizuki et al. (29) yielded r values of 0.20 – 0.28 in SOL MUs, which support our current findings. The strength of these relationships indicate that the discharge rate of triceps surae MUs, and consequently the excitability of the associated ␣-motoneurons in the spinal cord, is influenced by factors not included in the regression analysis. One such factor is the intrinsic stiffness of the ankle joint. When we stand, small changes in muscle activity are not the only source of plantar flexion torque. There is a passive contribution (i.e., intrinsic stiffness) from in-series and in-parallel tissues (e.g., muscles, tendon, ligaments, fascia) that corresponds to approximately 65–90% of the normalized load stiffness of the human body (18, 19). Thus ankle moments and forces produced in standing balance are only partially due to active muscle tension from the triceps surae, and this likely explains the low to moderate strength of the regression models. The other mechanism by which muscle tension can be increased is the recruitment of additional, higher-threshold MUs. In the present study, there was a positive linear relationship between A-P and M-L torque in standing balance, and the overall number of active MG and SOL MUs across subjects. The strength of this relationship and the associated slope values
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
Triceps Surae Motor Unit Activity During Standing Balance
were similar between MG and SOL. This is in line with the results reported by Vieira et al. (37), who showed a linear increase in the number of recruited MG MUs with more forward body positions. Overall, both rate coding and MU recruitment contribute to the generation of SOL and MG muscle tension in standing balance, although the relative importance of these mechanisms may vary between muscles. Conclusion The most surprising result from the present study was the relative absence of LG MU activity in standing balance, which was accompanied by dramatically higher recruitment thresholds compared with both SOL and MG MUs. At present, the most plausible explanation for this finding is that the distinct orientation of LG muscle fascicles compared with those of the SOL and MG result in a difference in mechanical advantage that influences MU excitability. This leads to the hypothesis that the medial and lateral portions of the gastrocnemius do not share the same optimal range of knee and ankle positions for force production, and that this type of configuration would improve the function of the ankle without the need for an additional muscle. Confirming previous reports, SOL activity was tonic during standing balance, whereas MG activity was phasic, with MU recruitment and rate coding both contributing to the generation of active muscle tension in standing balance. This was accompanied by much greater ISI variability in MG MUs in both standing balance and voluntary trials. Although preliminary, it appears that the amount of MU discharge variability may be tailored to the type of contraction most often performed by a given muscle. Overall, these findings provide evidence at the MU level that the muscles of the triceps surae have distinct functional roles that are not a mere consequence of the neural drive associated with the performance of different tasks. GRANTS Support for this study was provided by the Natural Sciences and Engineering Research Council of Canada. J.-S. Blouin received salary support from the Canadian Institutes of Health Research, Canadian Chiropractic Research Foundation, and Michael Smith Foundation for Health Research. M. E. Héroux received postdoctoral funding from the Canadian Institutes of Health Research and the Michael Smith Foundation for Health Research. M.E. Héroux and B.L. Luu received salary support from the Natural Sciences and Engineering Research Council of Canada Discovery Accelerator Supplement awarded to J.T. Inglis DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: M.E.H., C.J.D., B.L.L., J.T.I., and J.-S.B. conception and design of research; M.E.H., C.J.D., B.L.L., and J.-S.B. performed experiments; M.E.H., C.J.D., and B.L.L. analyzed data; M.E.H. and J.-S.B. interpreted results of experiments; M.E.H. prepared figures; M.E.H. drafted manuscript; M.E.H., C.J.D., B.L.L., J.T.I., and J.-S.B. edited and revised manuscript; M.E.H., C.J.D., B.L.L., J.T.I., and J.-S.B. approved final version of manuscript. REFERENCES 1. Buchthal F, Schmalbruch H. Motor unit of mammalian muscle. Physiol Rev 60: 90 –142, 1980. 2. Butler JE. Drive to the human respiratory muscles. Respir Physiol Neurobiol 159: 115–126, 2007. 3. Butler TJ, Kilbreath SL, Gorman RB, Gandevia SC. Selective recruitment of single motor units in human flexor digitorum superficialis muscle during flexion of individual fingers. J Physiol 567: 301–309, 2005.
•
Héroux ME et al.
147
4. Dartnall TJ, Rogasch NC, Nordstrom MA, Semmler JG. Eccentric muscle damage has variable effects on motor unit recruitment thresholds and discharge patterns in elbow flexor muscles. J Neurophysiol 102: 413–423, 2009. 5. De Troyer A, Kirkwood PA, Wilson TA. Respiratory action of the intercostal muscles. Physiol Rev 85: 717–756, 2005. 6. Dideriksen JL, Negro F, Enoka RM, Farina D. Motor unit recruitment strategies and muscle properties determine the influence of synaptic noise on force steadiness. J Neurophysiol 107: 3357–3369, 2012. 7. Duysens J, Tax AA, van der Doelen B, Trippel M, Dietz V. Selective activation of human soleus or gastrocnemius in reflex responses during walking and running. Exp Brain Res 87: 193–204, 1991. 8. Duysens J, van Wezel BM, Prokop T, Berger W. Medial gastrocnemius is more activated than lateral gastrocnemius in sural nerve induced reflexes during human gait. Brain Res 727: 230 –232, 1996. 9. Farina D, Holobar A, Merletti R, Enoka RM. Decoding the neural drive to muscles from the surface electromyogram. Clin Neurophysiol 121: 1616 –1623, 2010. 10. Farina D, Merletti R, Enoka RM. The extraction of neural strategies from the surface EMG. J Appl Physiol 96: 1486 –1495, 2004. 11. Farina D, Holobar A, Gazzoni M, Zazula D, Merletti R, Enoka RM. Adjustments differ among low-threshold motor units during intermittent, isometric contractions. J Neurophysiol 101: 350 –359, 2009. 12. Iman RL. A power study of a rank transform for the two-way classification model when interaction may be present. Can J Stat 2: 227–239, 1974. 13. Johnson MA, Polgar D, Weightman D, Appleton D. Data on the distribution of fibre types in thirty-six human muscles: an autopsy study. J Neurol Sci 18: 111–129, 1973. 14. Joseph J, Nightingale A. Electromyography of muscles of posture: leg muscles in males. J Physiol 117: 484 –491, 1952. 15. Kamo M. Discharge behavior of motor units in knee extensors during the initial stage of constant-force isometric contraction at low force level. Eur J Appl Physiol 86: 375–381, 2002. 16. Kawakami Y, Ichinose Y, Fukunaga T. Architectural and functional features of human triceps surae muscles during contraction. J Appl Physiol 85: 398 –404, 1998. 17. Krutki P, Pogrzebna M, Drzymała H, Raikova R, Celichowski J. Force generated by fast motor units of the rat medial gastrocnemius muscle during stimulation with pulses at variable intervals. J Physiol Pharmacol 59: 85–100, 2008. 18. Loram ID, Lakie M. Direct measurement of human ankle stiffness during quiet standing: the intrinsic mechanical stiffness is insufficient for stability. J Physiol 545: 1041–1053, 2002. 19. Loram ID, Kelly SM, Lakie M. Human balancing of an inverted pendulum: is sway size controlled by ankle impedance? J Physiol 532: 879 –891, 2001. 20. Loram ID, Maganaris CN, Lakie M. Human postural sway results from frequent, ballistic bias impulses by soleus and gastrocnemius. J Physiol 564: 295–311, 2005. 21. Luu BL. Perception, perfusion and posture (doctoral thesis). Sydney, New South Wales, Australia: Univ. of New South Wales, 2010. 22. Luu BL, Inglis JT, Huryn TP, Van der Loos HF, Croft EA, Blouin JS. Human standing is modified by an unconscious integration of congruent sensory and motor signals. J Physiol 590: 5783–5794, 2012. 23. McLean L, Goudy N. Neuromuscular response to sustained low-level muscle activation: within- and between-synergist substitution in the triceps surae muscles. Eur J Appl Physiol 91: 204 –216, 2004. 24. Maganaris CN, Baltzopoulos V, Sargeant AJ. In vivo measurements of the triceps surae complex architecture in man: implications for muscle function. J Physiol 512: 603–614, 1998. 25. Merletti R, Farina D. Analysis of intramuscular electromyogram signals. Philos Trans A Math Phys Eng Sci 367, 357–368, 2009. 26. Merletti R, Holobar A, Farina D. Analysis of motor units with highdensity surface electromyography. J Electromyogr Kinesiol 18: 879 –890, 2008. 27. Mochizuki G, Ivanova TD, Garland SJ. Synchronization of motor units in human soleus muscle during standing postural task. J Neurophysiol 94: 62–69, 2005. 28. Mochizuki G, Ivanova TD, Garland SJ. Factors affecting the common modulation of bilateral motor unit discharge in human soleus muscles. J Neurophysiol 97: 3917–3925, 2007. 29. Mochizuki G, Semmler JG, Ivanova TD, Garland SJ. Low-frequency common modulation of soleus motor unit discharge is enhanced during postural control in humans. Exp Brain Res 175: 584 –595, 2006.
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
148
Triceps Surae Motor Unit Activity During Standing Balance
30. Mori S. Discharge patterns of soleus motor units with associated changes in force exerted by foot during quiet stance in man. J Neurophysiol 36: 458 –471, 1973. 31. Moritz CT, Barry BK, Pascoe MA, Enoka RM. Discharge rate variability influences the variation in force fluctuations across the working range of a hand muscle. J Neurophysiol 93: 2449 –2459, 2005. 32. Mouchnino L, Aurenty R, Massion J, Pedotti A. Coordination between equilibrium and head-trunk orientation during leg movement: a new strategy build up by training. J Neurophysiol 67: 1587–1598, 1992. 33. Nardone A, Corra` T, Schieppati M. Different activations of the soleus and gastrocnemii muscles in response to various types of stance perturbation in man. Exp Brain Res 80: 323–332, 1990. 34. Negro F, Holobar A, Farina D. Fluctuations in isometric muscle force can be described by one linear projection of low-frequency components of motor unit discharge rates. J Physiol 587: 5925–5938, 2009. 35. Toothaker LE, Newman D. Nonparametric competitors to the two-way ANOVA. J Educ Behav Stat 19: 237–273, 1994.
•
Héroux ME et al.
36. Vieira TM, Loram ID, Muceli S, Merletti R, Farina D. Postural activation of the human medial gastrocnemius muscle: are the muscle units spatially localised? J Physiol 589: 431–443, 2011. 37. Vieira TM, Loram ID, Muceli S, Merletti R, Farina D. Recruitment of motor units in the medial gastrocnemius muscle during human quiet standing: is recruitment intermittent? What triggers recruitment? J Neurophysiol 107: 666 –676, 2012. 38. Vieira TM, Windhorst U, Merletti R. Is the stabilization of quiet upright stance in humans driven by synchronized modulations of the activity of medial and lateral gastrocnemius muscles? J Appl Physiol 108: 85–97, 2010. 39. Welsh SJ, Dinenno DV, Tracy BL. Variability of quadriceps femoris motor neuron discharge and muscle force in human aging. Exp Brain Res 179: 219 –233, 2007. 40. Woollacott MH, Bonnet M, Yabe K. Preparatory process for anticipatory postural adjustments: modulation of leg muscles reflex pathways during preparation for arm movements in standing man. Exp Brain Res 55: 263–271, 1984.
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
Absence of lateral gastrocnemius activity and differential motor unit behavior in soleus and medial gastrocnemius during standing balance Martin E. Héroux,1 Christopher J. Dakin,1 Billy L. Luu,1 John Timothy Inglis,1,2,3 and Jean-Sébastien Blouin1,2,4 1
School of Kinesiology, University of British Columbia, Vancouver, British Columbia, Canada; 2Brain Research Center, University of British Columbia, Vancouver, British Columbia, Canada; 3International Collaboration for Repair Discoveries, University of British Columbia, Vancouver, British Columbia, Canada; and 4The Institute for Computing, Information and Cognitive Systems, University of British Columbia, Vancouver, British Columbia, Canada Submitted 6 August 2013; accepted in final form 2 December 2013
Héroux ME, Dakin CJ, Luu BL, Inglis JT, Blouin J-S. Absence of lateral gastrocnemius activity and differential motor unit behavior in soleus and medial gastrocnemius during standing balance. J Appl Physiol 116: 140 –148, 2014. First published December 5, 2013; doi:10.1152/japplphysiol.00906.2013.—In a standing position, the vertical projection of the center of mass passes in front of the ankle, which requires active plantar-flexor torque from the triceps surae to maintain balance. We recorded motor unit (MU) activity in the medial (MG) and lateral (LG) gastrocnemius muscle and the soleus (SOL) in standing balance and voluntary isometric contractions to understand the effect of functional requirements and descending drive from different neural sources on motoneuron behavior. Single MU activity was recorded in seven subjects with wire electrodes in the triceps surae. Two 3-min standing balance trials and several ramp-and-hold contractions were performed. Lateral gastrocnemius MU activity was rarely observed in standing. The lowest thresholds for LG MUs in ramp contractions were 20 –35 times higher than SOL and MG MUs (P ⬍ 0.001). Compared with MUs from the SOL, MG MUs were intermittently active (P ⬍ 0.001), had higher recruitment thresholds (P ⫽ 0.022), and greater firing rate variability (P ⬍ 0.001); this difference in firing rate variability was present in standing balance and isometric contractions. In SOL and MG MUs, both recruitment of new MUs (R2 ⫽ 0.59 – 0.79, P ⬍ 0.01) and MU firing rates (R2 ⫽ 0.05– 0.40, P ⬍ 0.05) were associated with anterior-posterior and medio-lateral torque in standing. Our results suggest that the two heads of the gastrocnemius may operate in different ankle ranges with the larger MG being of primary importance when standing, likely due to its fascicle orientation. These differences in MU discharge behavior were independent of the type of descending neural drive, which points to a muscle-specific optimization of triceps surae motoneurons. triceps surae; motor units; standing balance; human AS A PHYSICAL LOAD, THE STANDING
human body is an inverted pendulum and is thus inherently unstable. With the vertical projection of the body’s center of mass usually passing in front of the ankle joint and insufficient passive stiffness of the Achilles tendon and other tissues (18), active plantar-flexor torque is required to maintain standing balance (18, 20). In humans, plantar flexion is primarily accomplished by the triceps surae: the soleus muscle (SOL) and the medial and lateral heads of the gastrocnemius muscle (MG and LG, respectively). Despite being agonists, these muscles differ in many respects in humans. For example, the gastrocnemius is a biarticular muscle with a considerable proportion (⬃52%) of
Address for reprint requests and other correspondence: J.-S. Blouin, 2106081 Univ. Boulevard, Vancouver, BC V6T 1Z1 Canada (e-mail: jsblouin @interchange.ubc.ca). 140
fast-twitch muscle fibers, whereas SOL is a monoarticular muscle with a very high proportion (⬃88%) of slow-twitch muscle fibers (1, 13). Opposite electromyographic responses (i.e., facilitation vs. inhibition) have also been observed between MG and SOL and between gastrocnemius heads during reflex activation (8, 31, 40) and functional tasks such as weight shifting, walking, running, and hopping (7, 23, 31, 32). These anatomical, neurophysiological, and functional differences support the notion that despite sharing a common distal tendon, the muscles of the human triceps surae appear to have distinct functional roles. Whether these functional differences are associated with distinct motor unit (MU) discharge behavior between muscles and tasks has not been systematically examined. The distal portion of the triceps surae terminates into the Achilles tendon to transmit muscle force to the posterior aspect of the calcaneous (heel) bone and generate plantar flexion. Given this common tendon, torque measures cannot be used to distinguish MG, LG, and SOL contributions to overall plantar flexion. Although surface electromyography can be used to measure triceps surae neuromuscular activity during standing balance (14), its interpretation can be difficult because it is a filtered cumulative sum of nearby MU action potentials (9). This makes the association between surface electromyography and the physiological processes that generate muscle activity difficult to pinpoint and prone to erroneous conclusions (9, 10, 26). On the other hand, the measurement of action potentials from individual MUs provides a direct measure of motor output from the spinal cord because of the one-to-one association between the firing of ␣-motoneurons and their associated muscle fibers (25). This allows key features of MU behavior such as recruitment thresholds, mean discharge rates, and discharge rate variability to be measured. The ability to simultaneously record MU activity from several agonist muscles allows ␣-motoneuron firing behavior to be compared between muscles during the same task, and to investigate whether firing behavior is modified in tasks with different neural drive. In human standing balance, SOL MUs discharge continuously, and modulate their firing rate with changes in ankle plantar-flexor torque (29). In contrast, MG MU activity is intermittent, with minimal rate coding; relying instead on MU recruitment to produce additional torque (37). It is unknown whether this continuous vs. intermittent MU activity is associated with differences in recruitment threshold or spike-tospike variability between these two muscles. Equally important is whether these differences are intrinsic to the muscles involved or whether they result from the type of neural drive that
8750-7587/14 Copyright © 2014 the American Physiological Society
http://www.jappl.org
Triceps Surae Motor Unit Activity During Standing Balance
controls standing balance. For example, Mochizuki et al. (27– 29) reported greater levels of common drive to pairs of SOL MUs in standing compared with an isometric task, but similar levels of MU synchronization. Whether differences in neural drive affect the discharge behavior of individual MUs in the human triceps surae is unknown. Even less is known about LG MU behavior. As the lateral counterpart of the MG, LG has the same muscle fiber composition and distal tendon, and is also biarticular. Although it is generally assumed that MUs from the lateral portion of the gastrocnemius muscle behave similarly to MG MUs, this has not been investigated in humans (38). Here we examine the behavior of SOL, MG, and LG MUs in two tasks that involve different neural drive to the motoneurons of these muscles. Standing balance and voluntary ramp-and-hold contractions can be performed with similar levels of plantarflexor torque, but the neural drive to triceps surae motoneurons differs substantially between these tasks: standing balance is believed to involve subcortical structures with vestibular contributions, whereas muscle activation during a nonbalancing task is believed to be largely cortically driven, with little known vestibular contribution (21, 22). We recorded indwelling MU activity from SOL, MG, and LG in these two tasks to investigate several key questions about the neural control and functional role of the human triceps surae. Specifically, whether the phasic muscle activity observed in MG is also present in LG; the extent to which MU discharge rates and discharge variability differ between muscles of the triceps surae; whether muscle-specific MU behavior is dependent on the type of neural drive (balance vs. voluntary drive); and whether MU recruitment and rate coding both contribute to modulate torque output by the triceps surae in standing balance. METHODS
•
Héroux ME et al.
141
Fig. 1. Experimental setup and procedures. A: a total of 14 fine wire electrodes were inserted with ultrasound guidance in the triceps surae. Black dots indicate approximate site for each insertion. B: triceps surae motor unit (MU) activity was recorded during two 3-min standing balance trials. Subjects stood with their right foot on a forceplate from which vertical force, anterior-posterior (A-P) torque, and medio-lateral (M-L) torque were recorded. C: subjects were braced to a rigid board at the level of the chest and pelvis. MU activity was recorded during a series of ramp-and-hold isometric plantar flexion contractions at three different contraction intensities.
Testing Protocol
Subjects Seven healthy subjects (1 woman; age range 23– 48 years) with no known history of neurological disease or injury participated in the study. The experimental protocol was explained and written informed consent was obtained prior to the study. All study procedures conformed to the standards of the Declaration of Helsinki and were approved by the University of British Columbia’s Clinical Research Ethics Board. Single MU Recordings Single MU activity was recorded using custom-made, sterilized, fine-wire indwelling electrodes. The electrodes were made using two 0.05-mm insulated stainless steel wire strands wound together (California Fine Wire, Grover Beach, CA). The two wires were cut at the recording end, exposing the noninsulated cross-section of the wires (i.e., 0.05 mm). These tips were folded back to create two 1-mm barbs to anchor the wires in the muscle. Insertion of the wire electrodes into the right MG (six wires), LG (four wires), and SOL (four wires) was performed under ultrasound guidance (SonoSite Micro Maxx, Bothell, WA) (Fig. 1A). After cleaning the skin, a surface ground electrode was placed over the right lateral malleolus. MU activity was recorded using a high impedance GRASS 15CT amplifier (Astromed, West Warwick, RI), filtered (band-pass 30 – 6,000 Hz), amplified (⫻5,000), and digitized at 15,152 Hz using a 16-bit CED DAQ-card and Spike2 software (Cambridge Electronics Design, Cambridge, UK).
Participants stood barefoot with their right foot on a forceplate (Bertec 4060 – 80; Bertec, Columbus, OH) and their left foot on a large stable surface of equal height (Fig. 1B); the medial malleoli were 10 –15 cm apart. The plantar/dorsi-flexion axis of rotation of the right foot was aligned with the medio-lateral axis of the forceplate at a known distance from the midline of the forceplate. Vertical force, anterior-posterior (A-P) torque, and medio-lateral (M-L) torque were measured for the right foot. Subjects were instructed to stand with their eyes open and their hands by their side. MU activity and forceplate data were recorded during two 3-min trials. In four subjects, verbal cueing to stand with a slightly forward lean was required to elicit clear gastrocnemius activity. Subjects were provided with a 2-min rest period between trials. Subjects were then braced to a rigid board (Fig. 1C) and asked to perform a series of isometric ramp (15 s) and hold (45 s) plantar flexion contractions against the forceplate. The position of the feet and the mean ankle angle remained the same as during the balance trials. This allowed us, in most cases, to measure the same MUs during standing balance and voluntary ramp-and-hold trials and determine the effects of differing neural drive on MU discharge behavior. The ramp phase of the contractions was used to determine MU recruitment thresholds, and the hold phase was used to compare MU discharge behavior between balance and voluntary conditions. Because we tracked several MUs simultaneously, it was not feasible to have subjects produce all the contraction intensities required to have the same mean firing rate between balance and voluntary conditions for each MU. On the basis of
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
142
Triceps Surae Motor Unit Activity During Standing Balance
pilot experiments, target intensities that corresponded to the mean A-P torque and 35% of the mean A-P torque produced in balance trials were selected as target torques because they generated similar mean discharge rates between balance and voluntary trials in a majority of MUs. A third target intensity of 90% of maximal A-P torque produced in balance trials was chosen to identify the recruitment threshold and maintain a similar mean discharge rate to that produced during standing balance for higher threshold MUs. For each trial, subjects were provided with visual feedback of the target ramp-and-hold A-P torque profile and their own A-P torque production. Each contraction intensity was performed three times, with the order of presentation randomized. A 2-min rest period separated each trial.
•
Héroux ME et al.
active MUs across subjects and standing balance kinetic variables. For each SOL and MG MU we determined whether they were active (ISI ⬍250 ms ⫽ 1) or inactive (ISI ⬎250 ms ⫽ 0) for each sample of kinetic data. Normalized A-P torque, M-L torque, or vertical force values (range produced in standing balance trials corresponded to 0 –100%) were then separated into 10% bins and the mean for each bin’s binary data (i.e., 0 ⫽ not active, 1 ⫽ active) was calculated to determine whether a MU was, on average, active (mean ⬎0.5) or silent (mean ⱕ0.5). For each bin the total number of active SOL and MG MUs across all subjects was determined; this analysis provided an overall count of active MUs for normalized A-P torque, M-L torque, or vertical force in standing balance. Statistical Analysis
Data Analysis To have a representative sample of active MUs, our goal for standing balance trials was to obtain, across all subjects, a total of 5 MUs for each of the 14 insertion sites; the ramp-and-hold contractions for these MUs were also analyzed. MU action potentials were identified and extracted using Spike2 software (Cambridge Electronic Design, Cambridge, UK). The algorithm identified MUs on the basis of size, shape, and timing. The identified MUs were then manually reviewed and sorted to include unmatched action potentials and to solve for instances in which two MU discharges were superimposed. Once MUs were sorted, discharge times were exported to Matlab (Mathworks, Natick, MA) for further analysis. MU discharge behavior and recruitment threshold. For standing balance and voluntary ramp-and-hold trials the number of active MUs, activation ratios, recruitment thresholds, mean interspike intervals (ISI), and ISI variability were calculated. The activation ratio quantifies the percentage of time a MU had ISI ⬍250 ms during standing balance trials (11, 34), with 0% corresponding to no activity and 100% representing continuous activity. For each MU, mean ISI and ISI variability were computed for periods when ISI ⬍250 ms during standing balance (two trials) and voluntary ramp-and-hold conditions (45 s constant torque portion; three trials). MU discharge variability was calculated as the coefficient of variation of ISI. The recruitment thresholds of MUs active in standing balance were determined from ramp-and-hold trials. The A-P torque at which each identified MU was first recruited (three consecutive MU action potentials with ISI ⬎250 ms) was identified and normalized to the maximum A-P torque produced in standing balance trials (36). Because LG showed little to no MU activity in standing balance despite clear MU activity in high-intensity ramp-and-hold trials, the first three MUs recruited for all LG, MG, and SOL ramp-and-hold trials were identified and their recruitment thresholds determined. This allowed us to compare the lowest threshold MUs for LG (n ⫽ 54), MG (n ⫽ 83), and SOL (n ⫽ 51). Relationship between sway kinetics and MU discharge rate and recruitment. Whereas the primary action of the triceps surae is ankle plantar flexion (i.e., A-P torque), MG activity is also related to lateral body motion and the abduction moment about the ankle in standing balance (37). To investigate the relationship between sway kinetics and MU discharge rates in standing balance (i.e., rate coding), we determined A-P torque, M-L torque, and vertical force for each discharge of each MU. If an MU discharged 2,000 times over the course of a 3-min balance trial, there would be 2,000 values for A-P torque, M-L torque, and vertical force. Stepwise linear regression was used to determine the relationship between the discharge rate of each MU and the measured kinetic variables associated with standing balance in MUs that fired ⬎250 times in the two standing balance trials. For each MU we retained the number of times a significant regression model was found, the R2 value of the model, and which kinetic variables were retained as part of the model. To investigate the role of MU recruitment in the generation of vertical force and M-L and A-P torques produced in standing balance trials, we determined the linear relationship between the number of
Nonparametric statistics were used for data that were non-normally distributed (Kolmogorov-Smirnov test) or had unequal variance between muscle groups (homogeneity of variance test). The KruskalWallis test was used to determine whether the recruitment thresholds of the first three activated MUs from each recorded signal differed between muscles. When a significant main effect was found, Dunn’s test was used for post hoc comparisons. The Mann-Whitney rank-sum test was used to compare activation ratios, recruitment thresholds, and stepwise linear regression R2 values between muscles for standing balance trials. The Rank Transform test was used to compare mean ISI and ISI variability across muscles and neural drive conditions (balance and voluntary ramp-and-hold). This test is a nonparametric equivalent to a two-factor ANOVA and calculates the F statistic on the basis of rank (12, 35). When a significant main effect or interaction was found, the Tukey test was used for post hoc comparisons. Linear regression was used to determine the relationship between the overall number of active MUs and A-P torque, M-L torque, or vertical force produced in standing balance trials, and an unpaired t-test was used to determine whether slope values were significantly different between muscles. The threshold for significance for all tests was ␣ ⫽ 0.05. Boxplots are used in figures; on each box, the central mark is the median, the edges of the box are the 25th and 75th percentiles, the whiskers are 1.5 times the length of the box, and data points outside this range are plotted individually. Values reported in the text are median and interquartile ranges. RESULTS
Across all subjects, a total of 39 MG MUs and 21 SOL MUs were decomposed and analyzed for standing balance and rampand-hold trials, with 5– 6 MUs at each of the six MG and four SOL sites (Fig. 1A). Only one LG MU was consistently active in standing balance trials, whereas another three MUs were active 8.7%, 4.1%, and 2.3% of the time. LG Activity and MU Recruitment Thresholds A consistent observation across subjects was that a forward lean on the verge of eliciting a step response was required to elicit LG MU activity. In line with these observations, there was little to no LG MU activity in standing balance trials. This lack of LG MU activity can be seen in Fig. 2A, which shows 1 min of standing balance data from five MG, four LG, and three SOL recordings from a typical subject. Although there is clear activity from one or more MUs on all MG and SOL recordings, there is no MU activity on any of the LG recordings; this was the case for the entire 6 min of balance data for this subject. Because LG MUs required greater plantar-flexor torque to be activated, we investigated the possibility that they had elevated recruitment thresholds. Figure 2B presents 30 s of
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
Triceps Surae Motor Unit Activity During Standing Balance
•
Héroux ME et al.
143
Fig. 3. Recruitment thresholds of earliest recruited MUs in the triceps surae. The recruitment threshold of the first three MUs recorded on each recording with good quality signals were identified for MG, LG, and SOL. Recruitment thresholds for MG MUs were significantly higher than for SOL MUs. More dramatically, LG MUs had much higher recruitment thresholds than both MG and SOL MUs.
MG and SOL MU Characteristics and Discharge Behavior
Fig. 2. Example of motor unit activity from all three muscles during standing balance and ramp-and-hold trials. A: 60 s of standing balance data from five recordings from medial gastrocnemius (MG) and four from lateral gastrocnemius (LG), and three recordings from soleus (SOL) of one subject (top); A-P torque, M-L torque, and vertical force (bottom). Although there is clear MU on all MG and SOL recordings, there was absolutely no LG activity. B: data from the same subjects for a ramp-and-hold 90% maximum A-P torque trial. Although numerous MUs are recruited in MG and SOL in the early part of the ramped plantar-flexion contraction, LG MUs are recruited considerably later.
ramp-and-hold data from a 90% maximum A-P torque trial for the same subject and the same recording sites as those presented in Fig. 2A. Although MG and SOL MUs were recruited at the start of the ramp contraction and new MUs were recruited throughout the ramp phase, all but one LG MU were recruited in the hold phase of the contraction. When considering the recruitment threshold of the first three activated MUs from each recording with good signal quality, there was a significant difference between muscles (Kruskal-Wallis, P ⬍ 0.001; Fig. 3). Post hoc comparisons revealed LG MU recruitment thresholds [LG 67.3% (43.5–75.9%)] were ⬃20 times greater than MG MU recruitment thresholds [3.4% (1.5– 8.9%), P ⬍ 0.001] and ⬃35 times greater than SOL MU recruitment thresholds [2.0% (0.9 – 6.0%), P ⬍ 0.001]. Comparatively, MG MU thresholds were only 1.7 times greater than those of SOL MUs (P ⫽ 0.001).
In standing balance trials, MG MU activity was intermittent compared with that of the more continuously active SOL MUs (Fig. 4). When considering the entire sample of MG and SOL MUs, activation ratios were significantly lower in MG MUs [73.6% (51.8 –95.2%)] compared with SOL [100% (94.4 – 100%)] (P ⬍ 0.001; Fig. 5A). In line with activation ratio
Fig. 4. Example of MU activity during standing balance. Top: data from two recordings from MG and two recordings from SOL from one subject. Above each recording in the instantaneous discharge frequency of one decomposed MU for each recording. The superimposed templates of for each MU are plotted to the right of each recording. The horizontal line in each instantaneous discharge plot indicates 4 Hz. Bottom: A-P torque, M-L torque, and vertical force recorded during this 30 s period.
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
144
Triceps Surae Motor Unit Activity During Standing Balance
•
Héroux ME et al.
tary ramp-and-hold trials (muscle main effect, P ⬍ 0.001; muscle ⫻ neural drive interaction, P ⫽ 0.185). These results indicate that ISI variability is approximately twice as large in MG MUs. Relationship Between Sway Kinetics and MU Discharge Rate and Recruitment The possible relationship between instantaneous discharge rates and A-P torque, M-L torque, and vertical force was first plotted for visual inspection. Figure 6, A and B, shows examples of these plots for two MG MUs and one SOL MU, with vertical force or M-L torque along the x-axis and A-P torque along the y-axis. The figure shows that the range of discharge rates was lower in SOL MUs, and that the increase in discharge rate with greater A-P torque was gradual for this MU. The two MG MUs also modulated their firing rate with changes in A-P torque, but this modulation was less uniform. Stepwise regression analysis of MG and SOL MUs revealed a significant regression model in all but one MG MU. In significant models, A-P torque was retained in 70% of cases, vertical force in 46%, and M-L torque in 41%. Although there was no trend for any kinetic variables to be retained more often by MG or SOL MUs, the strength of the relationship between MU discharge rate and kinetic variables, as reflected by R2 values, was ⬃30% less in MG MUs (Fig. 6C, P ⫽ 0.039). Examples of the association between standing balance kinetic variables and MG MU recruitment are visible in Fig. 2A
Fig. 5. MG and SOL MU behavior during standing balance and ramp-and-hold trials. A: activation ratio results indicate that MUs from SOL were active much more consistently than those from MG during standing balance trials (sway). B: these results are in line with the higher recruited thresholds found for MG MUs during ramp-and-hold trials (braced). C: mean interspike intervals (ISI) were comparable between MG and SOL during standing balance trials. This was also the case for the ramp-and-hold trials. Overall, mean ISI was slightly higher during standing balance trials compared with the ramp-and-hold trials. D: the ISI coefficient of variation was slightly but significantly higher during standing balance trials compared with ramp-and-hold trials. The difference in ISI variability between MG and SOL MUs was marked and this for both standing balance and ramp-and-hold trials. *P ⬍ 0.05; **P ⬍ 0.001.
results and the difference between MG and SOL MUs recruitment thresholds for the earliest recruited MUs, the recruitment threshold of MUs studied in standing balance trials was higher in MG than in SOL (P ⫽ 0.022; Fig. 5B). Although mean ISI was ⬃5% lower in voluntary ramp-andhold trials compared with balance trials (Fig. 5C; neural drive main effect, P ⫽ 0.029), there was no difference between SOL and MG mean ISI (muscles main effect, P ⫽ 0.799; muscles ⫻ neural drive interaction, P ⫽ 0.547). In terms of spike-to-spike variability, ISI variability was ⬃8% higher in standing balance trials compared with voluntary ramp-and-hold trials (neural drive main effect, P ⫽ 0.026; Fig. 5D). However, more noticeable was the greater variability in MG MUs compared with SOL MUs (see Fig. 4, compare last 5 s of MG and SOL data). Overall, coefficient of variation values were 0.41 (0.24 – 0.48) in MG MUs compared with 0.19 (0.14 – 0.22) in SOL MUs in standing balance trials, and 0.37 (0.28 – 0.45) in MG MUs compared with 0.17 (0.14 – 0.23) in SOL MUs in volun-
Fig. 6. Relationship between MU discharge rates and kinetic variables during standing balance trials. A: forceplate data from both standing balance trials with A-P torque along the y-axis and vertical force along the x-axis (left); 2-dimensional heat maps for two MG and one SOL MU (three right panels). Motor unit discharge rates were separated into bins and averaged to create these plots. Discharge rates were much lower for the SOL MU, and all three MUs increased their discharge rate with increasing plantar flexor torque. The two MG MUs also increased their discharge rate with increasing vertical force. B: data from the same three MUs are replotted with M-L torque along the x-axis. C: results from the stepwise linear regression between MU discharge rate and A-P torque, M-L torque, and vertical force. Overall, A-P torque was retained in 70% of the regression models compared with 41% for M-L torque and 46% for vertical force, and R2 values were significantly greater for SOL MU compared with MG MU. *P ⬍ 0.05.
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
Triceps Surae Motor Unit Activity During Standing Balance
and Fig. 4, with MU recruitment and derecruitment that coincide with increases or decreases in A-P or M-L torque. Linear regression was used to quantify the strength of the relationship between the overall number of active MUs across subjects and kinetic variables recorded in standing balance (Fig. 7). There was a strong linear relationship between the number of active MG and SOL MUs and A-P and M-L torque production; vertical force measures were not significantly related to the number of active MUs in either muscle. When considering the slopes of these relationships, the total number of active MG MUs increased at a similar rate as in the SOL (P ⬎ 0.201). DISCUSSION
The present study investigated the behavior of motor units from all three parts of the human triceps surae in two conditions that involve different neural drive: standing balance and voluntary isometric contractions. Whereas previous work focused on surface electromyography or indwelling recordings from a single muscle, our experimental approach allowed us to make these key observations on the behavior of MUs from the three portions of the triceps surae. 1) Lateral gastrocnemius MUs showed little to no activity in standing balance, and this appears to be due to recruitment thresholds being 20 –35 times higher than MG and SOL MUs. 2) Medial gastrocnemius MU discharge rate variability was twice that of SOL MUs in both standing balance and voluntary contractions. 3) These MU properties were not a consequence of the type of neural drive to the motoneuron pool. 4) Higher A-P and M-L torque values in standing balance are associated with increased MU firing rates and a greater number of active MUs for both SOL and MG. The 20-fold difference in MU recruitment threshold between two heads of a same muscle was unexpected and provides evidence of an important physiological difference between MG and LG. Similarly, a twofold difference in MU discharge variability between two agonists (SOL and MG) is not typical in humans (4, 39) and may reflect an adaptation to the type of contraction most often accomplished by each muscle. These differences were present in low to moderate contractions in both standing balance and isometric contractions, which points
•
Héroux ME et al.
145
to local (i.e., spinal) differences in the motoneuron pools that innervate the triceps surae rather than differences in descending drive. In the following sections we interpret our results in the context of previous research findings and discuss how these new insights provide evidence for functional optimization of the three muscles of the triceps surae. Lack of LG Activity During Standing Balance The recruitment threshold of the first active MG MUs in a ramped contraction were ⬃2 times greater than SOL MUs. Similarly, MG MUs active in standing balance trials had recruitment thresholds ⬃3 times greater than their SOL counterparts. Although certainly not negligible, these results are in stark contrast to the 20-fold difference in recruitment threshold that was noted between the medial and lateral aspects of the gastrocnemius muscle. Several other human muscles have distinct compartments or muscular heads. However, we were unable to locate any examples of large differences in recruitment thresholds between distinct compartments of a single human muscle. For example, the flexor digitorum superficialis has four components that correspond to the four digits upon which it acts, and the earliest recruited MUs from each muscle component have similar recruitment thresholds (3). Similarly, Kamo (15) found no difference in recruitment threshold between the first recruited MUs in three heads of the quadriceps femoris muscle: the vastus medialis, vastus lateralis, and rectus femoris. Thus the gastrocnemius seems somewhat unique with its two muscular heads that have MUs with dramatically different recruitment thresholds. The LG and MG are the two heads of a single muscle, and an important question is whether possessing a distinct population of higher-threshold MUs in the anatomically smaller LG would have a functional or physiological benefit. If there was a truly distinct population of higher-threshold MUs with larger motoneurons in LG, it might be possible for descending neural drive to selectively activate this motoneuron pool. Similarly, global descending drive to the plantar flexor motoneuron pool would lead to MU recruitment on the basis of Henneman’s size principle (9, 10, 25), and thus later recruitment of LG MUs. However, it is not clear why this would be more advantageous
Fig. 7. Relationship between MU recruitment and kinetic variables during standing balance trials. Results of the linear regression between the overall number of active MG and SOL MUs across subjects and normalized to vertical force (A), A-P torque (B), and M-L torque (C). Kinetic variables were normalized to the minimum and maximum values produced during the standing balance trials and the count of active MU was determined in 10% bins. There was a moderate to strong linear relationship between the number of active MUs and A-P and M-L torque for both MG and SOL, whereas no relationship was present for vertical force. J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
146
Triceps Surae Motor Unit Activity During Standing Balance
than a uniform distribution of MU sizes across both heads of the gastrocnemius. An alternative explanation that we favor is that our results reflect an anatomical difference between the MG and LG rather than a purely neurophysiological difference. There are examples in humans in which a portion of a motoneuron pool is preferentially activated because the muscle fibers they innervate have a mechanical advantage for a specific task or posture. For example, the intercostal muscles [see Butler (2) for a review] show a rostro-caudal and dorso-ventral gradient of activation that closely parallels their calculated mechanical advantage, and these patterns of activation are preserved in animal models when all afferent feedback is removed (5). In terms of the triceps surae and standing balance, low-level isometric plantar flexion (⬍20% maximum voluntary contraction) with the ankle joint in a neutral position is associated with muscle fiber lengths and pennation angles that are similar for MG and SOL (⬃40 mm and ⬃20°), but different from those of LG muscle fibers (⬃80 mm and ⬃10°) (16, 24). This anatomical difference and the possible mechanical disadvantage it provides to LG may be associated with a reduction in motoneuron excitability, which could contribute to higher recruitment thresholds and minimal LG MU activity in standing balance. More broadly, we propose that this anatomical arrangement may reflect an optimization via which MG and LG are preferentially activated within relatively distinct knee and ankle configurations, which would increase the functional capabilities of the triceps surae. The greater muscle volume of the MG may reflect that as bipeds, humans generate plantar flexor torques within a somewhat limited range of motion that is optimal for MG and its muscle fascicles. Overall, additional experiments that closely control both knee and ankle angle, movement velocity, torque generation, and task requirements are needed to confirm these hypotheses for the gastrocnemius. SOL and MG MU Activity in Standing Balance Several reports have indicated that the SOL is continuously active in standing balance, whereas the gastrocnemius tends to be more intermittently active (14, 29, 30, 37). This has led the proposal that ongoing SOL activity in standing balance provides a background plantar-flexion torque to counteract the static gravitational body load, whereas intermittent gastrocnemius activity generates dynamic ankle plantar-flexor torque to stabilize the moving body (40). In the present study, these patterns of global activation were accompanied by a twofold difference in the amount of discharge rate variability between the SOL and MG. This difference in variability was independent of the descending drive to the muscles, suggesting functional adaptation of the motoneuron pool innervating the muscles of the triceps surae. The greater variability in MG does not appear to be due to the phasic behavior of its MUs in standing balance because it was still present in isometric contractions when these MUs were continuously active. In humans, it is unusual for MUs from agonist muscles or different compartments of a single muscle to have large differences in discharge rate variability. For example, discharge rate variability is similar between the biceps brachii and the brachialis (4), and between the various muscular heads of the quadriceps femoris (39). Thus this leads to the question of whether this large difference in discharge rate variability is of
•
Héroux ME et al.
functional importance. For SOL, low discharge rate variability would be beneficial in its role as the primary plantar flexor in standing balance: it would result in relatively stable MU firing rates that would experience gradual increases and decreases with changes in A-P torque. This type of MU activity would generate relatively stable muscle tension, which seems appropriate for a muscle that is continually active in standing balance. Conversely, the MG is regarded as a powerful muscle that is less suited for sustained contractions. This is in line with the higher recruitment thresholds in MG MUs and the intermittent nature of MG activity in standing balance required to stabilize the moving body. In the present experiment, we also noted that MG MUs had greater discharge rate variability compared with SOL MUs. Greater discharge rate variability is associated with larger force fluctuations in low-level contractions (6, 31), which would not be optimal for standing balance when contraction levels are relatively low. However, greater discharge rate variability is also associated with greater force production in higher-threshold MUs (17), which would be beneficial in a muscle such as MG, which is often required to generate high levels of torque. Thus, differences in SOL and MG MU discharge behavior appear to reflect a physiological optimization for the types of contractions these muscles most often produce. Relationship Between Sway Kinetics and MU Discharge Rate and Recruitment In the present experiment, stepwise linear regression was used to investigate the relationship between MU discharge rate and standing balance kinetic variables. In line with the key role the triceps surae plays in plantar flexion, A-P torque was retained in 70% of the regression models compared with 40 – 45% for M-L torque and vertical force measures. These findings agree with recent reports of an association between center of pressure A-P movements and MU firing rates (29, 31). When we consider the strength of the relationship between MU firing rate and A-P sway, Vieira et al. (37) reported an r value of 0.07 for MG MUs, whereas a similar analysis by Mochizuki et al. (29) yielded r values of 0.20 – 0.28 in SOL MUs, which support our current findings. The strength of these relationships indicate that the discharge rate of triceps surae MUs, and consequently the excitability of the associated ␣-motoneurons in the spinal cord, is influenced by factors not included in the regression analysis. One such factor is the intrinsic stiffness of the ankle joint. When we stand, small changes in muscle activity are not the only source of plantar flexion torque. There is a passive contribution (i.e., intrinsic stiffness) from in-series and in-parallel tissues (e.g., muscles, tendon, ligaments, fascia) that corresponds to approximately 65–90% of the normalized load stiffness of the human body (18, 19). Thus ankle moments and forces produced in standing balance are only partially due to active muscle tension from the triceps surae, and this likely explains the low to moderate strength of the regression models. The other mechanism by which muscle tension can be increased is the recruitment of additional, higher-threshold MUs. In the present study, there was a positive linear relationship between A-P and M-L torque in standing balance, and the overall number of active MG and SOL MUs across subjects. The strength of this relationship and the associated slope values
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
Triceps Surae Motor Unit Activity During Standing Balance
were similar between MG and SOL. This is in line with the results reported by Vieira et al. (37), who showed a linear increase in the number of recruited MG MUs with more forward body positions. Overall, both rate coding and MU recruitment contribute to the generation of SOL and MG muscle tension in standing balance, although the relative importance of these mechanisms may vary between muscles. Conclusion The most surprising result from the present study was the relative absence of LG MU activity in standing balance, which was accompanied by dramatically higher recruitment thresholds compared with both SOL and MG MUs. At present, the most plausible explanation for this finding is that the distinct orientation of LG muscle fascicles compared with those of the SOL and MG result in a difference in mechanical advantage that influences MU excitability. This leads to the hypothesis that the medial and lateral portions of the gastrocnemius do not share the same optimal range of knee and ankle positions for force production, and that this type of configuration would improve the function of the ankle without the need for an additional muscle. Confirming previous reports, SOL activity was tonic during standing balance, whereas MG activity was phasic, with MU recruitment and rate coding both contributing to the generation of active muscle tension in standing balance. This was accompanied by much greater ISI variability in MG MUs in both standing balance and voluntary trials. Although preliminary, it appears that the amount of MU discharge variability may be tailored to the type of contraction most often performed by a given muscle. Overall, these findings provide evidence at the MU level that the muscles of the triceps surae have distinct functional roles that are not a mere consequence of the neural drive associated with the performance of different tasks. GRANTS Support for this study was provided by the Natural Sciences and Engineering Research Council of Canada. J.-S. Blouin received salary support from the Canadian Institutes of Health Research, Canadian Chiropractic Research Foundation, and Michael Smith Foundation for Health Research. M. E. Héroux received postdoctoral funding from the Canadian Institutes of Health Research and the Michael Smith Foundation for Health Research. M.E. Héroux and B.L. Luu received salary support from the Natural Sciences and Engineering Research Council of Canada Discovery Accelerator Supplement awarded to J.T. Inglis DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: M.E.H., C.J.D., B.L.L., J.T.I., and J.-S.B. conception and design of research; M.E.H., C.J.D., B.L.L., and J.-S.B. performed experiments; M.E.H., C.J.D., and B.L.L. analyzed data; M.E.H. and J.-S.B. interpreted results of experiments; M.E.H. prepared figures; M.E.H. drafted manuscript; M.E.H., C.J.D., B.L.L., J.T.I., and J.-S.B. edited and revised manuscript; M.E.H., C.J.D., B.L.L., J.T.I., and J.-S.B. approved final version of manuscript. REFERENCES 1. Buchthal F, Schmalbruch H. Motor unit of mammalian muscle. Physiol Rev 60: 90 –142, 1980. 2. Butler JE. Drive to the human respiratory muscles. Respir Physiol Neurobiol 159: 115–126, 2007. 3. Butler TJ, Kilbreath SL, Gorman RB, Gandevia SC. Selective recruitment of single motor units in human flexor digitorum superficialis muscle during flexion of individual fingers. J Physiol 567: 301–309, 2005.
•
Héroux ME et al.
147
4. Dartnall TJ, Rogasch NC, Nordstrom MA, Semmler JG. Eccentric muscle damage has variable effects on motor unit recruitment thresholds and discharge patterns in elbow flexor muscles. J Neurophysiol 102: 413–423, 2009. 5. De Troyer A, Kirkwood PA, Wilson TA. Respiratory action of the intercostal muscles. Physiol Rev 85: 717–756, 2005. 6. Dideriksen JL, Negro F, Enoka RM, Farina D. Motor unit recruitment strategies and muscle properties determine the influence of synaptic noise on force steadiness. J Neurophysiol 107: 3357–3369, 2012. 7. Duysens J, Tax AA, van der Doelen B, Trippel M, Dietz V. Selective activation of human soleus or gastrocnemius in reflex responses during walking and running. Exp Brain Res 87: 193–204, 1991. 8. Duysens J, van Wezel BM, Prokop T, Berger W. Medial gastrocnemius is more activated than lateral gastrocnemius in sural nerve induced reflexes during human gait. Brain Res 727: 230 –232, 1996. 9. Farina D, Holobar A, Merletti R, Enoka RM. Decoding the neural drive to muscles from the surface electromyogram. Clin Neurophysiol 121: 1616 –1623, 2010. 10. Farina D, Merletti R, Enoka RM. The extraction of neural strategies from the surface EMG. J Appl Physiol 96: 1486 –1495, 2004. 11. Farina D, Holobar A, Gazzoni M, Zazula D, Merletti R, Enoka RM. Adjustments differ among low-threshold motor units during intermittent, isometric contractions. J Neurophysiol 101: 350 –359, 2009. 12. Iman RL. A power study of a rank transform for the two-way classification model when interaction may be present. Can J Stat 2: 227–239, 1974. 13. Johnson MA, Polgar D, Weightman D, Appleton D. Data on the distribution of fibre types in thirty-six human muscles: an autopsy study. J Neurol Sci 18: 111–129, 1973. 14. Joseph J, Nightingale A. Electromyography of muscles of posture: leg muscles in males. J Physiol 117: 484 –491, 1952. 15. Kamo M. Discharge behavior of motor units in knee extensors during the initial stage of constant-force isometric contraction at low force level. Eur J Appl Physiol 86: 375–381, 2002. 16. Kawakami Y, Ichinose Y, Fukunaga T. Architectural and functional features of human triceps surae muscles during contraction. J Appl Physiol 85: 398 –404, 1998. 17. Krutki P, Pogrzebna M, Drzymała H, Raikova R, Celichowski J. Force generated by fast motor units of the rat medial gastrocnemius muscle during stimulation with pulses at variable intervals. J Physiol Pharmacol 59: 85–100, 2008. 18. Loram ID, Lakie M. Direct measurement of human ankle stiffness during quiet standing: the intrinsic mechanical stiffness is insufficient for stability. J Physiol 545: 1041–1053, 2002. 19. Loram ID, Kelly SM, Lakie M. Human balancing of an inverted pendulum: is sway size controlled by ankle impedance? J Physiol 532: 879 –891, 2001. 20. Loram ID, Maganaris CN, Lakie M. Human postural sway results from frequent, ballistic bias impulses by soleus and gastrocnemius. J Physiol 564: 295–311, 2005. 21. Luu BL. Perception, perfusion and posture (doctoral thesis). Sydney, New South Wales, Australia: Univ. of New South Wales, 2010. 22. Luu BL, Inglis JT, Huryn TP, Van der Loos HF, Croft EA, Blouin JS. Human standing is modified by an unconscious integration of congruent sensory and motor signals. J Physiol 590: 5783–5794, 2012. 23. McLean L, Goudy N. Neuromuscular response to sustained low-level muscle activation: within- and between-synergist substitution in the triceps surae muscles. Eur J Appl Physiol 91: 204 –216, 2004. 24. Maganaris CN, Baltzopoulos V, Sargeant AJ. In vivo measurements of the triceps surae complex architecture in man: implications for muscle function. J Physiol 512: 603–614, 1998. 25. Merletti R, Farina D. Analysis of intramuscular electromyogram signals. Philos Trans A Math Phys Eng Sci 367, 357–368, 2009. 26. Merletti R, Holobar A, Farina D. Analysis of motor units with highdensity surface electromyography. J Electromyogr Kinesiol 18: 879 –890, 2008. 27. Mochizuki G, Ivanova TD, Garland SJ. Synchronization of motor units in human soleus muscle during standing postural task. J Neurophysiol 94: 62–69, 2005. 28. Mochizuki G, Ivanova TD, Garland SJ. Factors affecting the common modulation of bilateral motor unit discharge in human soleus muscles. J Neurophysiol 97: 3917–3925, 2007. 29. Mochizuki G, Semmler JG, Ivanova TD, Garland SJ. Low-frequency common modulation of soleus motor unit discharge is enhanced during postural control in humans. Exp Brain Res 175: 584 –595, 2006.
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
148
Triceps Surae Motor Unit Activity During Standing Balance
30. Mori S. Discharge patterns of soleus motor units with associated changes in force exerted by foot during quiet stance in man. J Neurophysiol 36: 458 –471, 1973. 31. Moritz CT, Barry BK, Pascoe MA, Enoka RM. Discharge rate variability influences the variation in force fluctuations across the working range of a hand muscle. J Neurophysiol 93: 2449 –2459, 2005. 32. Mouchnino L, Aurenty R, Massion J, Pedotti A. Coordination between equilibrium and head-trunk orientation during leg movement: a new strategy build up by training. J Neurophysiol 67: 1587–1598, 1992. 33. Nardone A, Corra` T, Schieppati M. Different activations of the soleus and gastrocnemii muscles in response to various types of stance perturbation in man. Exp Brain Res 80: 323–332, 1990. 34. Negro F, Holobar A, Farina D. Fluctuations in isometric muscle force can be described by one linear projection of low-frequency components of motor unit discharge rates. J Physiol 587: 5925–5938, 2009. 35. Toothaker LE, Newman D. Nonparametric competitors to the two-way ANOVA. J Educ Behav Stat 19: 237–273, 1994.
•
Héroux ME et al.
36. Vieira TM, Loram ID, Muceli S, Merletti R, Farina D. Postural activation of the human medial gastrocnemius muscle: are the muscle units spatially localised? J Physiol 589: 431–443, 2011. 37. Vieira TM, Loram ID, Muceli S, Merletti R, Farina D. Recruitment of motor units in the medial gastrocnemius muscle during human quiet standing: is recruitment intermittent? What triggers recruitment? J Neurophysiol 107: 666 –676, 2012. 38. Vieira TM, Windhorst U, Merletti R. Is the stabilization of quiet upright stance in humans driven by synchronized modulations of the activity of medial and lateral gastrocnemius muscles? J Appl Physiol 108: 85–97, 2010. 39. Welsh SJ, Dinenno DV, Tracy BL. Variability of quadriceps femoris motor neuron discharge and muscle force in human aging. Exp Brain Res 179: 219 –233, 2007. 40. Woollacott MH, Bonnet M, Yabe K. Preparatory process for anticipatory postural adjustments: modulation of leg muscles reflex pathways during preparation for arm movements in standing man. Exp Brain Res 55: 263–271, 1984.
J Appl Physiol • doi:10.1152/japplphysiol.00906.2013 • www.jappl.org
