Download Vastus lateralis single motor unit EMG at the same absolute torque

Articles in PresS. J Appl Physiol (May 21, 2009). doi:10.1152/japplphysiol.90322.2008
1
Vastus lateralis single motor unit EMG at the same absolute torque production at
2
different knee angles
3
4
TM Altenburg1,2, A de Haan1,2, PWL Verdijk2, W van Mechelen3 and CJ de Ruiter2
5
6
1
7
Metropolitan University, Cheshire, UK
8
2
9
Amsterdam, The Netherlands
Institute for Biomedical Research into Human Movement and Health, Manchester
Research Institute MOVE, Faculty of Human Movement Sciences, VU University
3
11
Centre, Amsterdam, The Netherlands
EMGO Institute and Department of Public and Occupational Health, VU University Medical
12
13
Running head: Motor unit EMG at different knee angles
14
15
Correspondence to:
16
C.J. de Ruiter, Research Institute MOVE, Faculty of Human Movement Sciences, VU
17
University Amsterdam, Van der Boechorststraat 9, 1081 BT, Amsterdam, The Netherlands.
18
Tel.: (+31)(20)5988578
19
Fax: (+31)(20)5988529
20
Email: [email protected]
21
22
23
Copyright © 2009 by the American Physiological Society.
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
10
Abstract
25
Single motor unit electromyographic (EMG) activity of the knee extensors was investigated at
26
different knee angles with subjects (n=10) exerting the same absolute submaximal isometric
27
torque at each angle. Measurements were made over a 20° range around the optimum angle for
28
torque production (AngleTmax) and, where feasible, over a wider range (50°). Forty six vastus
29
lateralis (VL) motor units were recorded at 20.7±17.9%MVC together with the rectified surface
30
EMG (rsEMG) of the superficial VL muscle. Due to the lower maximal torque capacity at
31
positions more flexed and extended than AngleTmax, single motor unit recruitment thresholds
32
were expected to decrease and discharge rates were expected to increase at angles above and
33
below AngleTmax. Unexpectedly, the recruitment threshold was higher (P<0.05) at knee
34
angles 10° more extended (43.7±22.2Nm) and not different (P>0.05) at knee angles 10° more
35
flexed (35.2±17.9Nm), compared to recruitment threshold at AngleTmax (41.8±21.4Nm).
36
Also, unexpectedly the discharge rates were similar (P>0.05) at the three angles: 11.6±2.2,
37
11.6±2.1 and 12.3±2.1Hz. Similar angle independent discharge rates were also found for 12
38
units (n=5; 7.4±5.4%MVC) studied over the wider (50°) range, while recruitment threshold
39
only decreased at more flexed angles. In conclusion, the similar recruitment threshold and
40
discharge behavior of vastus lateralis motor units during submaximal isometric torque
41
production suggests that net motor unit activation did not change very much along the
42
ascending limb of the knee-angle torque relationship. Several factors such as length-dependent
43
twitch potentiation, which may contribute to this unexpected aspect of motor control, are
44
discussed.
45
Key words: recruitment threshold, discharge rate, muscle length, quadriceps muscle
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
24
2
46
47
Introduction
The maximal isometric force that skeletal muscles can exert varies with the length of
the contractile part of the muscle – tendon complex, resulting in the well described force –
49
length relationship. Force declines when a muscle is at a shorter or a longer length compared to
50
the length at which overlap between actin and myosin filaments is optimal (Lo; 17, 30).
51
Consequently, when the same absolute force has to be exerted during submaximal voluntary
52
contractions at different lengths, it might be expected that the muscle would need to be
53
activated to a greater extent at muscle lengths either longer or shorter than Lo. Thus, even
54
during relatively simple tasks it seems likely that motor control is complicated by the
55
adaptations that have to be made for the length (joint angle) dependent changes in force
56
generating capacity of the muscles.
57
There have been a number of investigations of the effect of muscle length on motor unit
58
recruitment and discharge rate (7, 12, 33). In the study of Vander Linden et al. (33) higher
59
discharge rates were found per change in torque at short compared to long muscle lengths. In
60
contrast, Bigland-Ritchie et al. (6) and Del Valle & Thomas (12) reported no differences in
61
discharge rate at short compared to long muscle lengths. However, in these studies different
62
motor unit populations were investigated at the different muscle lengths. It could not be
63
excluded, therefore, that at short muscle lengths the same force (or rather torque) was generated
64
by the recruitment of additional or larger motor units, without the necessity of increasing the
65
discharge rate. Alternatively, discharge rates could have been higher at short compared to long
66
muscle lengths, but due to pooling of all motor unit data, these potentially higher rates could
67
have been masked by newly recruited motor units which discharged at low rates. Moreover, by
68
studying different motor unit populations (6, 12), length dependent changes in recruitment and
69
derecruitment thresholds can not be investigated.
3
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
48
70
To our knowledge there are only two studies which have investigated the same motor
units at different muscle lengths. The outcomes of these studies can not be compared directly,
72
because force levels, muscles and muscle lengths (relative to Lo) were different. Christova et
73
al. (7) found higher discharge rates for the majority of the biceps brachii motor units studied at
74
shorter compared to longer muscle lengths (60° joint angle range), when torque was delivered
75
at the same percentage of the maximal voluntary contraction torque (MVC) obtained at each
76
elbow angle. Pasquet et al. (27) found no changes in discharge rate between the different
77
muscle lengths (20° joint angle range) of tibialis anterior muscle. However, in that study
78
discharge rate was obtained just above recruitment threshold, thus at different absolute and
79
relative (% MVC) torques at each joint angle (muscle length). The study of Pasquet et al. (27)
80
is the only study that investigated recruitment thresholds of the same motor units at different
81
muscle lengths in addition to motor unit discharge rate. A lower threshold was found at the
82
shorter compared to the longer muscle length indicating that additional motor units were
83
recruited. However, the authors stated (27) that both the short and the long muscle lengths were
84
probably on the ascending limb of the force – length relationship. Thus, the findings of Pasquet
85
et al. (27) indicate that at muscle lengths shorter than Lo, motor unit recruitment threshold was
86
decreased and discharge rate remained unchanged at contraction intensities just above
87
recruitment threshold at each muscle length. It remains to be seen whether motor unit
88
recruitment threshold decreases and discharge rates remain unchanged when the same absolute
89
torque has to be generated at muscle lengths shorter or longer than Lo. Moreover, the
90
derecruitment threshold of the same motor units at different muscle lengths has, to the best of
91
our knowledge, not been studied before.
92
The present study investigated motor unit (de)recruitment thresholds and discharge rates
93
at muscle lengths shorter and longer than Lo, in the quadriceps femoris muscle group,
94
specifically in the vastus lateralis. We could not measure Lo of the vastus lateralis separately,
4
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
71
95
but we determined the knee angle at which knee extensor torque production was maximal
96
(AngleTmax). In vivo it is unclear whether quadriceps muscle fibers are (on average) at optimal
97
length for force production at the joint angle at which maximal torque is exerted (AngleTmax),
98
which is ~60° for the quadriceps muscle (4). Because considerations such as the quadriceps
99
moment arm, Ca2+ sensitivity, potentiation and slackness of the series elastic components,
differ with muscle length, the knee angle at which the muscle fibers of the individual
101
component muscles are at optimum length may be different from AngleTmax. However, this is
102
a complication that all in vivo studies have to deal with and the net effect of these factors is
103
unknown. Consequently, in the present study we assumed that the optimal length for force
104
production of the quadriceps muscle fibers was at AngleTmax (see also discussion).
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
100
105
The purpose of the present study was to investigate the (de)recruitment thresholds and
106
the discharge rates of the same vastus lateralis motor units when the same absolute torque had
107
to be exerted at knee angles both more extended and flexed than AngleTmax. In addition,
108
surface EMG of the three superficial quadriceps femoris muscles was also investigated at the
109
different knee angles. Because single motor unit measurements are technically difficult, we
110
studied a (limited) 20° joint range, similar to Pasquet et al. (27). However, when feasible the
111
firing behavior of the same motor unit was investigated over a wider joint range (up to 50°).
112
Maximal isometric knee extensor torque capacity is lower at knee angles more extended and
113
flexed than AngleTmax. Thus, when the same absolute torque has to be generated this is
114
relatively more difficult at angles other than AngleTmax and it would necessitate the
115
recruitment of additional motor units and/or more intense motor unit activity, to compensate for
116
the relatively low force capacity of the muscle fibers at these angles. Therefore, it was
117
hypothesized that, when the same torque had to be produced at knee angles more extended and
118
flexed than AngleTmax, motor unit recruitment and de-recruitment thresholds would be lower
119
and their discharge rates would be higher compared to those observed at AngleTmax.
120
5
121
Methods
122
Subjects
Ten healthy subjects (5 men and 5 women) with a mean ± SD age of 26.7 ± 6.1 yr,
124
height 178.2 ± 9.0 cm and mass 70.3 ± 8.1 kg voluntarily participated in two experimental
125
sessions. The study was performed with the approval of the ethics committee of the VU
126
University Medical Center in Amsterdam, The Netherlands, and in accordance with the
127
Declaration of Helsinki. After written and verbal explanations of the objectives and procedures
128
of the experiment, the subjects signed informed consent forms. All subjects refrained from
129
heavy exercise 48 hours prior to the experimental sessions.
130
131
132
Study design
The present study was divided into two sessions. In the first session, the effect of knee
133
angle on quadriceps femoris surface EMG during constant torque contractions was studied.
134
This was done at the same absolute contraction torques at each angle, representing 10 and 50 %
135
of MVC at AngleTmax. In the second session vastus lateralis motor unit (de)recruitment
136
thresholds, discharge rate and surface EMG were studied during constant torque contractions at
137
different knee angles. In this case the torque was set approximately 20 % above the recruitment
138
threshold of the identified motor units. Subsequently the same absolute torque had to be
139
produced at all knee angles. These two sessions (referred to as the first and the second session)
140
were preceded by a habituation session in which subjects practiced delivering constant torque
141
as well as smooth ramp up and ramp down changes in torque at different knee angles.
142
143
144
145
Torque recordings
Isometric knee extension contractions were performed on a custom made dynamometer,
which recorded the exerted torque at its axis of rotation (4, 5, 15). Subjects sat in an upright
6
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
123
position with a hip angle of 85° (0° = full extension) while straps restrained hip and shoulders.
147
A cuff was fastened around the lower leg and to the lever of the dynamometer. A shin guard
148
ensured subjects could exert maximal forces without discomfort at the shin. The axis of
149
rotation of the dynamometer was aligned with the lateral femoral condyle, while subjects
150
delivered a submaximal torque (about 20 % MVC) with the lever arm of the dynamometer set
151
in a position that corresponded with 60°. Torques signals were AD-converted and stored on
152
disk for off-line analysis. At each knee angle torque signals were automatically corrected for
153
gravity. Fifty ms before the start of each measurement, and with the subject sitting relaxed in
154
the dynamometer, the average torque applied by the weight of the limb and/or possibly due to
155
any passive joint and muscle forces, was set at zero by the computer program (11).
156
Consequently the measured motor recruitment thresholds represent active net knee torque.
157
It is well known that during isometric contractions when using a fixed dynamometer
158
lever arm, the knee angle decreases with increasing knee extension torque (1, 31). In the
159
present study the submaximal contractions were performed at different torque levels, and
160
therefore probably at slightly different knee angles. However, for practical reasons, we changed
161
the dynamometer lever arm angle, rather than of the active knee angle, in steps of 10° during
162
the experimental sessions. Afterwards, for each subject, the active knee angles were measured
163
at the different dynamometer angles and torques produced during the experiments over which
164
single motor unit EMG was successfully recorded. Knee angles were measured with a
165
modified handheld goniometer with extendable arms (model G300, Whitehall Manufacturing)
166
using the greater trochanter of the femur, the lateral epicondyle and the lateral malleolus as
167
anatomical landmarks, while the subjects performed isometric contractions at torque levels
168
between 5 and 100 % MVC for the range of dynamometer angles. When knee angles were
169
measured several times with other contractions in between, the repeated knee angle
170
measurements differed only 1°, indicating that these knee angle measurements were reliable.
7
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
146
171
172
Surface electromyography
Surface EMG data of 5 muscles, vastus lateralis (VL), vastus medialis, rectus femoris,
174
biceps femoris and semitendinosis, was obtained during the first session. Electrical activity of
175
the biceps femoris and semitendinosis were recorded to determine co-contraction. For the VL
176
muscle two electrode pairs were placed on the muscle. The first pair was placed on the part of
177
the muscle from which motor unit EMG would be obtained in the second session (referred to as
178
mid VL position). The second electrode pair was placed more distally on the VL muscle
179
(referred to as distal VL position). This was to determine if possible effects of knee angle on
180
surface EMG were similar in the mid VL position compared to the more distal VL position
181
from which the surface EMG is commonly recorded. During the second session surface EMG
182
was only obtained from the mid position of the VL muscle and the electrodes were placed on
183
both sides of the intramuscular wire electrodes (see Intramuscular electromyography). During
184
the second session only surface EMG from the biceps femoris was obtained to monitor co-
185
contraction.
186
EMG was recorded using surface electrodes (Blue Sensor, Ambu, Ølstykke, Denmark,
187
lead-off area: 1.0 cm2). After shaving and cleaning the skin with 70 % ethanol two electrodes
188
were placed on the belly of each muscle in a bipolar configuration, in line with the muscle fiber
189
direction and with an inter-electrode distance of 25 mm (centre to centre). For the mid VL
190
position the inter-electrode distance was 5 mm greater (30 mm) as they were placed on both
191
sides of the intramuscular wire electrodes needed for motor unit recording in the second
192
session. Reference electrodes were placed on the right patella and on the lateral epicondyle of
193
the femur of the right leg. The location of each electrode was accurately marked with a
194
waterproof felt tip pen for precise electrode re-application in the subsequent session. Surface
195
EMG signals were amplified (x1000) with a biosignal amplifier (g.tec, Austria, 10-500 Hz,
8
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
173
196
input impedance 110 MOhm), AD-converted with a Simultaneous-Sampling AtoD board (PCI-
197
6143, National Instruments, Austin, Texas, USA), digitized (10 kHz), band-pass filtered (10-
198
400 Hz) and stored, together with the torque signal, on computer disk.
199
In pilot experiments supra-maximal twitch stimulation was applied to the femoral nerve
200
and the M-wave shapes were studied to confirm that the EMG electrodes remained distal to the
201
motor end plate zone following a change in knee angle. M-wave amplitude was found to be
202
constant for the range of knee angles used in the present study and there was no change in the
203
sequence of the positive and negative phases of the M-wave.
205
206
Intramuscular electromyography
The method used to record and analyze single motor unit action potentials has been
207
described previously (9, 10). In brief, custom-made, fine-wire electrodes were constructed from
208
four polytereftalate-butylene-coated wires (0.044 mm diameter, stainless steel core, Capable,
209
The Netherlands) to record motor unit action potentials. The four wires with a length of 10 cm
210
were glued together at the tip and cut transversely, passed through a hypodermic needle (0.75 x
211
40 mm) and bent backwards over the needle tip for the last 2 mm. The needles containing the
212
electrodes were sterilized for 30 min at 150° C. A new needle-electrode was used for each
213
measurement. Before insertion, the skin at the electrode site was shaved and cleaned with 70 %
214
ethanol. The needle-electrode-combination was inserted to a depth of 40 mm in the middle of
215
the vastus lateralis muscle belly and the needle then withdrawn leaving the quadrifillar tip of
216
the electrode in the muscle. After insertion the subjects were asked to perform a few maximal
217
(brisk) voluntary contractions to stabilize the position of the electrode in the muscle. Six
218
bipolar EMG signals were led off from the four free ends of the electrode which were
219
connected to 50-cm-long isolated cables. The signals were amplified (x1000) with a biosignal
220
amplifier (g.tec, Austria, 0.01-10 kHz, input impedance 110 MOhm), AD-converted with a
9
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
204
221
Simultaneous-Sampling AtoD board (PCI-6143, National Instruments, America), digitized
222
(44.1 kHz), played through loud-speakers online and saved on computer disk. A reference
223
electrode was placed on the patella of the right knee. Following band-pass filtering (1-10 kHz)
224
the six bipolar signals were displayed immediately on-screen.
225
Intramuscular EMG was considered stable and selective, when one or two motor units
had a distinctive amplitude and shape on at least one of the six EMG channels during a few
227
submaximal isometric contractions. If no unit could be selectively recorded after insertion of
228
the wire electrode, the electrode was slightly re-positioned by gentle pulling until selective
229
recording became possible.
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
226
230
231
232
Experimental protocol
At the beginning of the first session, AngleTmax was determined for each subject
233
individually with subjects performing extension MVCs at different dynamometer angles
234
between 30 and 80° with 5° steps, in random order and with 2 min rest between contractions.
235
AngleTmax was determined by fitting a second order polynomial (R2 = 0.98 ± 0.05).
236
(AngleTmax as determined for each subject in the first session was also used in the second
237
session.) Subsequently, subjects were asked to track a trapezoidal torque trajectory that was
238
displayed on the computer monitor (Figure 1). A slow isometric ramp up (5 s duration) was
239
performed to a constant absolute torque, followed by a torque plateau of 10 s and a ramp down
240
of 5 s. The torque trajectory was performed at AngleTmax and at 10° more extended and flexed
241
angles both at 10 and 50 % of the MVC at AngleTmax. Each contraction was performed twice
242
and in random order. To determine whether fatigue had occurred at the end of the first session,
243
subjects were asked to perform extension MVCs at the three knee angles and a flexion MVC at
244
AngleTmax.
10
245
In the second session, in which motor unit EMG was recorded, a torque trajectory, at
about 120% recruitment threshold of the motor unit of interest, was performed at AngleTmax
247
and at 10° more extended and flexed angles. Additionally, discharge behavior of twelve motor
248
units in five subjects could be recorded over a wider range of dynamometer angles from 30°
249
more extended up to 20° more flexed. Torque level was set approximately at 120 % recruitment
250
threshold and ranged from 4 – 76 % MVC. In a similar way to the first session, contractions
251
were performed twice and in random knee angle order. At the end of the second session,
252
subjects performed extension MVCs at those knee angles where motor unit activity had been
253
studied.
254
255
256
EMG analysis
During maximal isometric contractions, rectified surface EMG (rsEMG) was calculated
257
for a 500 ms segment before maximum torque was reached. During the submaximal isometric
258
contractions at the different dynamometer angles, rsEMG and motor unit discharge rates (the
259
inverse of the inter spike discharge intervals) were calculated over the first 3 s of the 10 s
260
torque plateau (Figure 1: t = 8 – 11 s). In addition, to study possible knee angle dependent
261
changes in rsEMG and discharge rate during the 10 s torque plateau, rsEMG and discharge rate
262
were also obtained at the end of the isometric plateau (Figure 1: t = 15 – 18 s). Recruitment and
263
de-recruitment thresholds of motor units were defined as the torque at the moment of,
264
respectively, the first (during ramp up) and the last (during ramp down) motor unit action
265
potential. RsEMG, discharge rates and recruitment thresholds of the two attempts at each knee
266
angle were averaged. Parameters obtained at each knee angle were normalized to their values at
267
AngleTmax before statistical analysis.
268
269
Motor unit action potentials were identified on the basis of their amplitude and shape
using custom-written Matlab software (9, 10). Usually, small changes in action potential
11
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
246
270
amplitude and/or shape occurred gradually during the different contractions (Figure 1), due to
271
small and unavoidable changes in electrode position. Therefore, in the analysis, motor unit
272
templates made for unit identification were updated within and also between contractions.
273
It is important to note that the changes in knee angle occurred at slow speeds (10°/s) and
274
with the subjects exerting constant torque, thus the motor units of interest were also
275
discharging during the changes in knee angle. We were therefore confident that the same motor
276
units were studied at the different knee angles.
277
Statistics
279
Data were presented as mean ± SD. For the normalized recruitment torque, discharge
280
rate and rsEMG of the submaximal contractions of the second session, the averaged value of
281
the different motor units studied for each subject (range: 1 – 10 units per subject) were
282
calculated and used in the statistical analysis (n=10). For both the first and the second
283
experimental session, repeated measures analysis of variance (ANOVA, SPSS version 12.0)
284
was used to analyze the effect of knee angle on torque and rsEMG during MVCs, and on
285
normalized (de)recruitment thresholds, discharge rates and rsEMG during submaximal
286
contractions. Additionally, this analysis was also performed with each motor unit (instead of
287
each subject) considered as an independent measure (n=46 and n=12, respectively for the small
288
and the wider knee angle range). Paired-samples t-tests were used to compare recruitment vs.
289
de-recruitment thresholds, and discharge rate and rsEMG during the first vs. the last 3 s of
290
submaximal torque production. If significant main effects were observed, Bonferroni tests were
291
performed for post hoc analysis. When the sphericity assumption was violated, a Greenhouse-
292
Geisser correction was applied. The level of significance of all statistical analyses were set at P
293
< 0.05.
294
12
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
278
295
Results
296
Knee angle
297
Table 1 shows, for both sessions, the active knee angles obtained during the
contractions at the different torque levels. At all dynamometer angles, the knee angles
299
decreased significantly with increasing torque level up to 50 % MVC (F4,16 = 179.6, P < 0.001).
300
This torque dependent decrease in knee angle was slightly greater at the more extended knee
301
positions (F20,80 = 8.1, P < 0.04). When torque increased from 5 to 50 % MVC, knee angle
302
decreased (average values across subjects) from 3 to 11° at a dynamometer angle of 30° and
303
from 2 to 11° at a dynamometer angle of 80°. All data below are presented as a function of the
304
active knee angles of the individual subjects.
305
306
307
Verification of the data
There were no significant differences between the normalized rsEMG of the three
308
measured component quadriceps muscles (vastus medialis, VL and rectus femoris) during the
309
submaximal contractions at the three knee angles studied in the first session (F2,18 = 1.9, P =
310
0.19). Furthermore, the behavior of rsEMG of the two measured hamstring muscles (biceps
311
femoris and semitendinosis) was similar at the three knee angles studied (F2,18 = 1.5; P = 0.24).
312
The rsEMG at the mid VL position was not significantly different from the distal VL position
313
at the different knee angles (F1,9 = 2.4, P > 0.16). Since in the second session motor unit EMG
314
was sampled from the mid VL, only the results of rsEMG from the VL mid position will be
315
presented. Furthermore, since in the second session rsEMG was only obtained from the biceps
316
femoris, only the rsEMG of the biceps femoris will be presented as an indication for co-
317
contraction.
318
319
Maximal isometric torque and surface EMG
13
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
298
320
With the measurements made at the beginning of the first session, there was a
significant main effect of knee angle on the variation of extension MVC (F2,18 = 11.8, P =
322
0.001). Extension MVCs were significantly lower at angles 10° more extended (258.1 ± 69.2
323
Nm) and 10° more flexed (259.5 ± 62.9 Nm) compared to values at AngleTmax (271.9 ± 74.8
324
Nm). AngleTmax corresponded with an active knee angle of 55 ± 4°. Maximal extension
325
torques at the end of the first session did not differ from those at the beginning (F1,9 = 0.02, P =
326
0.87), indicating that no significant fatigue had developed and/or that no shift in the angle for
327
maximal torque production occurred during the session. There were no differences in maximal
328
rsEMG between the three knee angles (F2,18 = 0.35, P = 0.71) for any of the quadriceps
329
muscles.
330
Normalized (% AngleTmax) maximal extension torques are shown in figure 2A at the
331
different active knee angles during the second session. Similar to the first session, there was a
332
significant main effect of knee angle on the variation of extension MVC (F2,18 = 15.9, P <
333
0.001); normalized maximal torque was significantly lower at angles 10° more extended and
334
flexed compared to AngleTmax (n = 10 subjects). The normalized maximal extension torques
335
of the 5 subjects from whom motor units could also be recorded over a wider joint angle range,
336
are shown in figure 3A. AngleTmax for these 5 subjects corresponded with an active knee
337
angle of 56 ± 2°. Again, there was a significant main effect of knee angle on variation of MVC
338
(F5,20 = 12.6, P < 0.001); normalized maximal torque was significantly lower at angles 30°, 20°
339
and 10° more extended and 20° more flexed compared to AngleTmax. Maximal VL rsEMG
340
was not significantly different between knee angles during both the first (F2,18 = 3.3, P = 0.06)
341
or the second session (F5,20 = 2.0, P = 0.12) although the highest values were recorded at
342
AngleTmax (Table 2).
343
344
Motor unit EMG during submaximal isometric contractions
14
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
321
345
A total of 46 VL motor units were followed twice during submaximal isometric
346
contractions at the three knee angles around AngleTmax in the second session (n = 10
347
subjects). Twelve of these units could also be followed over a wider range of joint angles (n = 5
348
subjects). Relative torque levels (percentage of the maximal torque at AngleTmax) during the
349
submaximal contractions varied from 4 – 76 % MVC and were, on average, 20.7 ± 17.9 %
350
MVC for the 10 subjects and 7.4 ± 5.4 % MVC when motor units were studied in the subset of
351
5 subjects.
352
Recruitment thresholds were 43.7 ± 22.2, 41.8 ± 21.4 and 35.2 ± 17.9 Nm, and
discharge rates (first 3 s torque plateau) 11.6 ± 2.2, 11.6 ± 2.1 and 12.3 ± 2.1 Hz at,
354
respectively, angles 10° more extended, AngleTmax and 10° more flexed (n = 10 subjects).
355
Normalized (% AngleTmax) recruitment and de-recruitment thresholds and discharge rates at
356
the different angles for 10 subjects are shown in figure 2B. There was a significant main effect
357
of angle on variation in normalized recruitment threshold (F2,18 = 7.6, P < 0.004); normalized
358
recruitment threshold was significantly higher at 10° more extended compared to AngleTmax,
359
but was not different between AngleTmax and the angle 10° more flexed. However, note that
360
the submaximal contractions took place with active knee angles of ~50, 60 and 70° (Figure
361
2B), thus at knee angles that were about 5° less extended than during maximal torque
362
production (Figure 2A). This indicates that these submaximal contractions (Figure.2B) were
363
performed somewhat more towards the descending limb of the maximal torque – knee angle
364
relationship (Figure 2A). Normalized de-recruitment was not significantly different between
365
knee angles (F2,18 = 2.2, P = 0.14). Furthermore, and also unexpected, there were no differences
366
in discharge rates between these three angles (F2,18 = 2.7, P = 0.13). In addition, discharge rates
367
during the first 3 s of the 10 s torque plateau were not significantly different from the last 3 s
368
(data not presented; t = -0.8, P = 0.45).
15
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
353
369
For the 12 VL motor units (n = 5 subjects) which could also be studied over the wider
range of angles (30°, 20° and 10° more extended and 10° and 20° more flexed than
371
AngleTmax) recruitment thresholds were, respectively, 13.0 ± 9.9, 15.0 ± 12.4, 15.6 ± 12.9,
372
13.5 ± 11.7, 11.5 ± 9.2 and 9.8 ± 7.3 Nm, and discharge rates (first 3 s torque plateau) were,
373
respectively, 11.5 ± 1.1, 10.9 ± 1.6, 11.1 ± 1.7, 11.4 ± 1.9, 11.9 ± 1.4 and 11.7 ± 1.5 Hz. In
374
figure 3, normalized (% AngleTmax) recruitment thresholds (B), de-recruitment thresholds (C)
375
and discharge rates (D) are shown for these 12 single motor units at the different knee angles.
376
There was a main effect of knee angle on recruitment threshold (F5,20 = 6.3, P = 0.018) and a
377
trend of knee angle on derecruitment threshold (F5,20 = 4.0, P = 0.07), but without post hoc
378
differences being evident. Unexpectedly, (de)recruitment thresholds did not decrease at the
379
more extended knee angles. Again, there were no differences in discharge rates between knee
380
angles (F5,20 = 1.8, P = 0.20). Also, within this subset of subjects and motor units, the discharge
381
rates of the first and the last 3 s of the 10 s torque plateau were not significantly different (data
382
not presented; t = 2.2, P = 0.09).
383
When each motor unit was considered as an independent measure for the ANOVA
384
analyses, similar results were found for recruitment and de-recruitment thresholds. However, a
385
main effect of angle for discharge rate became evident, both for the units studied over the small
386
(F2,90 = 4.4, P = 0.016, n = 46 motor units) and over the wider knee angle range (F5,55 = 2.7, P =
387
0.03, n = 12 motor units). Although there were no post hoc differences, these findings were
388
indicative of a slight increase in discharge rates at the more flexed knee angles.
389
In figure 4 the ratio of recruitment to de-recruitment threshold is shown at different
390
active knee angles. There were no changes in the recruitment to de-recruitment ratio among
391
knee angles, both for the small (F2,18 = 0.4, P = 0.65, n = 10 subjects) and the wider (F5,20 = 1.0,
392
P = 0.44, n = 5 subjects) knee angle range. However, at each knee angle the ratios were
393
significantly higher than 1.0 for the small knee angle range (t = 5.1, P = 0.001) and tended to be
16
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
370
394
higher than 1.0 for the wider range (t = 4.0, P = 0.09), indicating that the torque at which the
395
motor units started discharging was higher than the torque at which the units stopped
396
discharging.
397
398
399
Surface EMG during submaximal isometric contractions
During the submaximal contractions at 10 and 50 % of the maximal torque in the first
session, there was a significant main effect of knee angle on normalized VL rsEMG (%
401
AngleTmax, first 3 s of the torque plateau; F2,18 = 10.0, P = 0.001). Normalized VL rsEMG
402
was, as expected, significantly higher at angles 10° more flexed (105.9 ± 9.6 and 107.5 ± 6.8 %
403
for contractions at 10 and 50 % respectively) compared to the rsEMG at the same torque at
404
AngleTmax. Normalized rsEMG at 10° more extended (93.4 ± 8.2 and 102.3 ± 9.6 %
405
respectively at 10 and 50 % MVC) was, however, not significantly higher than the rsEMG
406
obtained at AngleTmax. These findings were very similar to the results obtained during the
407
submaximal contractions at the three knee angles in the second session (F2,18 = 19.2, P < 0.001,
408
n = 10 subjects), where contraction intensities were 20.7 ± 17.9 % MVC (Figure 5). For the
409
wider range of knee angles (7.4 ± 5.4 % MVC) there was a main effect of knee angle on
410
rsEMG (F5,20 = 6.3, P = 0.001, n = 5 subjects). The lowest normalized VL rsEMG was found at
411
an averaged knee angle of 53° (Figure 5, open circles), which was very near the knee angle for
412
maximal torque production in these five subjects (56°, Table 1, Figure 3A). However, probably
413
due to the limited number of subjects, the only significant post hoc effect was found between
414
the 83 ± 3° and 53 ± 3° knee angles.
415
Both in the first and second session, VL rsEMG of the last 3 s of the torque plateau was
416
significantly (4 – 13 %) higher compared to the first 3 s (t = -10.4, P < 0.001), indicating that
417
rsEMG increased slightly during the 10 s torque plateau (Table 3). These increases were
418
similar at all knee angles (Table 3).
17
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
400
419
In table 4 the normalized (% maximum) biceps femoris rsEMG (first 3 s of the torque
420
plateau) for both the first and the second session are shown at the different knee angles. There
421
were no significant differences in normalized biceps femoris rsEMG among knee angles for
422
both sessions (P > 0.10). Moreover, and similar to what was found for the knee extensors,
423
rsEMG of biceps femoris muscle was slightly, but significantly, higher during the last 3 s (4.0
424
± 2.9 %) compared to the first 3 s (3.6 ± 2.6 %) of the torque plateau (P < 0.05).
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
18
425
Discussion
The present study is the first in which (de)recruitment thresholds and discharge rates of
427
the same single VL motor units were investigated during submaximal quadriceps contractions
428
at the same absolute torque at knee angles more extended and more flexed than the optimum
429
angle for maximal torque production. The main finding of the present study is that the expected
430
decrease in (de)recruitment thresholds and increase in discharge rates during contractions at
431
knee angles different from AngleTmax, were only found for the more flexed knee angles
432
(longer muscle lengths).
433
Consistency of the measurements
434
Although only 12 units in 5 subjects were studied over the wider knee angle range,
435
these units behaved in a very similar fashion compared to the other 34 units that were studied
436
over the smaller knee angle range. Moreover, the discharge behavior was very similar and
437
consistent among units and between repeated contractions. At all knee angles, motor units were
438
studied during two contractions, which were not consecutive but randomly ordered between
439
knee angles. In addition, when the knee angle was changed, constant torque was produced by
440
the subjects such that the units investigated continued discharging. Therefore, we are confident
441
that the same motor units were studied at all the different knee angles. However, since these
442
type of measurements are technically challenging a limited number of VL motor units were
443
studied.
444
Torque angle relationship and length force relationship
445
It is well known that during ‘isometric contractions’, the knee extends with increasing knee
446
extension torque from rest to full contraction (about 7° in the present study) (1, 32). In the
447
present study, the actual (active) knee angles during the submaximal contractions (Figure 2B)
448
at dynamometer angles 10° more extended and flexed were on average respectively about 50°
449
and 70 while AngleTmax determined during maximal contractions was 55⁰ (Figure 2A).
19
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
426
Consequently, for these submaximal contractions, it is reasonable to assume that motor unit
451
EMG was studied predominantly on the plateau and on the descending limb of the torque –
452
angle relationship. Twelve units in five subjects could be studied over a wider knee angle range
453
of 20-80⁰ (Figure 3) indicating that these units were studied both on the ascending and the
454
descending limb of the torque – angle relationship of the quadriceps femoris muscle group.
455
This is in line with the observation that the lowest values for normalized rsEMG during
456
submaxial torque generation were found around AngleTmax at knee angles of about 53°
457
(Figure 5). However, surface EMG may not be a very good measure of muscle activation
458
especially when knee angle is changed (14). Moreover, surface EMG was only significantly
459
higher at the most flexed angles (Figure 5). Nevertheless, we are confident that measurements
460
were made on both limbs of the knee-angle torque relationship. It is however more difficult to
461
state where exactly on the force – length relationship of the quadriceps muscle fibers the
462
measurements were made. Several factors change with muscle length. The maximal internal
463
moment arm of the quadriceps is reached at ~30° (3, 26), therefore the knee angle at which the
464
muscle fibers are at optimal length for maximal force generation may be shifted to angles more
465
flexed than AngleTmax. In addition, due to an increase in Ca2+-sensitivity at longer muscle
466
lengths (2) a shift of the optimum towards more flexed knee angles may also be expected
467
during submaximal contractions (18, 29) because AngleTmax was determined during maximal
468
effort. In contrast, during repeated submaximal contractions the knee angle at which muscle
469
fiber force production is optimal may be shifted to angles more extended than AngleTmax,
470
because potentiation of submaximal force is more pronounced at shorter muscle lengths (18,
471
25). Furthermore, the series elastic components are slacker with more extended knees
472
potentially allowing the contractile elements to shorten more during isometric force
473
development at short length (22). In addition, optimal fiber length may vary for the different
474
quadriceps muscle groups. However, the net effect of these factors is unknown. Therefore, we
20
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
450
475
assumed that the optimal length for force production of the vastus lateralis muscle fibers
476
probably was near AngleTmax.
477
The descending limb of the torque – angle relationship
During the submaximal contractions at more flexed knee angles, thus on the descending
479
limb of the torque – angle relationship, VL rsEMG was, as expected, higher than at AngleTmax
480
(Figure 5). VL (de)recruitment thresholds decreased slightly with more flexed knees, which
481
was also in line with our expectations (Figures 2 and 3). Moreover, when each motor unit was
482
considered as an independent measure (thereby providing more power to the analysis),
483
discharge rates slightly increased with more flexed knees. These results suggest that in order to
484
produce a certain submaximal torque on the descending limb of the torque – angle relationship,
485
additional VL motor units have to be recruited and already recruited motor units have to
486
discharge at higher rates.
487
The ascending limb of the torque – angle relationship
488
At a knee angle of 30° maximal torque is about 60% of the maximum obtained at a knee angle
489
55° (Figure 3A). Due to the higher quadriceps internal moment arm at 30° compared to 55° (3,
490
26), maximal muscle force at 30° is probably even less than 60% of the force at 55°.
491
Therefore, exerting the same absolute submaximal torque at 30° was expected to be relatively
492
more demanding than at AngleTmax, not only in terms of torque but certainly in terms of
493
muscle force. However, on the ascending limb of the torque – angle relationship VL
494
(de)recruitment thresholds were not lower and motor unit discharge rates were not higher
495
compared to AngleTmax (Figure 3). These knee angle independent discharge rates at the
496
shorter muscle lengths are in line with the findings of Pasquet et al. (27) with the tibialis
497
anterior muscle. However, in the study of Pasquet et al. (27) submaximal contractions were
498
performed at lower absolute and relative torques at the shorter compared to the longer muscle
499
lengths, which makes their findings difficult to compare with ours. Christova et al. (7) found
21
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
478
higher discharge rates at the shorter muscle lengths in the biceps brachii muscle, despite the
501
fact that submaximal contractions at the different joint angles were performed at the same
502
relative (% MVC) torques. If we would have studied contractions at the same relative
503
intensities at each knee angle, we probably would have found decreases in discharge rate at
504
more extended knees. In the present study, the expected decrease in recruitment thresholds did
505
not occur at the shorter muscle lengths, which is not in line with the study of Pasquet et al. (27)
506
in the tibialis anterior muscle. The finding of higher recruitment thresholds than de-recruitment
507
thresholds, confirms earlier observations (13, 31). It has been suggested that such hysteresis of
508
motor unit discharge could be a manifestation of persistent inward currents increasing
509
motoneuron excitability during a muscle contraction (16, 19).
510
The most interesting finding of the present study was that torque production on the ascending
511
limb of the torque – angle relationship (shorter muscle lengths), probably neither necessitated
512
recruitment of additional VL motor units (recruitment thresholds did not change) nor increased
513
activation of already recruited units (discharge rates did not increase). Noteworthy, in this
514
respect, is that recent studies have shown that when the same relative torque was produced
515
during isometric contractions at different knee angles oxygen consumption of all quadriceps
516
components was found to be lower (8) and resistance to fatigue higher at a knee flexion angle
517
of 30° than at 60° and 90° (24, 28). The present findings are in line with these earlier data and
518
suggest that at smaller knee flexion angles the lower maximal torque capacity of the muscle
519
fibers is somehow compensated for during sub maximal in vivo torque generation. There are
520
several factors that have to be discussed in relation to this surprising finding.
521
Co-contraction and contribution of other quadriceps muscles
522
One factor is that co-contraction could differ among knee angles. The similar normalized
523
biceps femoris rsEMG during submaximal knee extensions at different knee angles (~2%,
22
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
500
Table 4) imply that the effects of co-contraction on knee extension contraction intensity were
525
similar at the different knee angles.
526
However, the internal moment arms of the hamstring muscles change with knee angle.
527
Consequently it cannot be excluded that the opposing hamstring torque was somewhat greater
528
during knee extensions with more extended knees. However, this would necessitate additional
529
motor unit activation making the present findings even more surprising.
530
It can also not be excluded that the relative force contribution of the vastus medialis, rectus
531
femoris and/or vastus intermedius was larger than that of the VL on the ascending limb of the
532
torque – angle relationship. Moreover, additional motor units may have been recruited in the
533
other knee extensors when the same torque had to be produced at more extended knee angles,
534
but this remains to be investigated.
535
536
Motor unit excitation
537
Although the exerted torque was the same at all knee angles, VL muscle force was
538
probably not the same and therefore afferent activity from Golgi tendon organs may have
539
varied with knee angle. In addition, afferent activity from muscle spindles in both quadriceps
540
and hamstring muscles varies with muscle length and may therefore influence motor unit
541
activity. Reciprocal inhibition of the knee extensors may have been more pronounced during
542
contractions with more extended knees when hamstring muscles were relatively more
543
stretched. However, clearly any inhibition was overcome since the same torque could be
544
generated with more extended knees. However, muscle spindle activity may also influence
545
persistent inward currents of alpha motoneurons. Persistent inward currents can amplify
546
motoneuron excitability (19), and have shown to be joint angle dependent (21). Hyngstrom et
547
al. (21) found that persistent inward currents and consequently motoneuron excitability in the
548
agonist muscle were diminished when the antagonist muscles were at longer lengths in
23
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
524
decerebrated cats, probably due to Ia reciprocal inhibition from muscle spindles. If this also
550
occurred in the present study, this could explain why recruitment thresholds were relatively
551
high at the more extended knee angles and did not show the expected decrease. However, the
552
ratio of recruitment to de-recruitment threshold was found to be independent of knee angle
553
(Figure 4) and motor unit discharge rates at recruitment and de-recruitment were also similar
554
between knee angles (data not presented), these findings do not seem to support the idea of a
555
knee angle dependent modification of persistent inward currents on motoneuron excitation.
556
Notwithstanding all possible influences either of a central or more peripheral origin in the
557
nervous system, the final net effect appears to be that VL motor unit activity does not need to
558
increase to compensate for the lower torque generating capacity of the knee extensors during
559
constant torque generation with more extended knees.
560
Twitch potentiation
561
Length dependent twitch potentiation may have contributed to the present findings.
562
Place et al. (28) found potentiation of twitch force following a submaximal contraction with
563
extended knees (35° angle) with the muscle relatively short, but not at 75°, at longer muscle
564
lengths. The present study involved many submaximal trials to sample as many motor units as
565
possible. This did not fatigue the muscle (maximal torque production did not decrease at any of
566
the knee angles during the experiments), but, in a similar way to the findings of Place et al.
567
(28), the repeated contractions may have lead to a more pronounced twitch potentiation at more
568
extended knees (short muscle length) than with more flexed knees. Klein et al. (23) found a
569
significant correlation between the decline in motor unit discharge rate and the amount of
570
twitch potentiation in triceps brachii muscle, indicating that twitch potentiation may help
571
maintain a constant force output despite the fall in motor unit discharge rates. It is difficult to
572
quantitatively compare their results with the present findings. Klein et al. (23) found decreases
573
in motor unit discharge rate up to 20% during sub maximal (10-30 % MVC) contractions
24
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
549
performed 10 s after a 5 s ‘conditioning’ contraction at 75 % MVC, causing substantial
575
increases ( about 45 %) in twitch force. The ‘conditioning’ contraction was more forceful than
576
the average force level during the repeated sub maximal contractions in the present study. This
577
may have mitigated twitch potentiation and the consequent effects on motor unit discharge rate
578
in the present study. However, Klein et al. (23) only investigated these effects at one
579
intermediate muscle length and twitch potentiation probably is more pronounced at short
580
lengths (28). Therefore, and although we did not measure changes in twitch force at the
581
different muscle lengths in the present study, we suggest that muscle length dependent twitch
582
potentiation could reduce the expected need for an increase in motor unit activity during
583
submaximal torque production with more extended knees.
584
Thus length dependent twitch potentiation may contribute to a simplification of motor control,
585
particularly because the knee extensors generally operate predominantly on the ascending limb
586
during daily life activities. Twitch potentiation may help to maintain a constant torque output
587
during sub maximal activities without the necessity of substantially increasing motor unit
588
activity at the shorter muscle lengths, as would be expected from the torque knee angle
589
relationship established during maximal contractions.
590
591
In conclusion, when generating a certain absolute submaximal torque on the descending
592
limb of the torque – angle relationship (thus with relatively long muscles), the lowest
593
recruitment thresholds and the highest discharge rates were found, suggesting that additional
594
motor units were recruited and that already activated motor units discharged at higher rates to
595
compensate for the lower torque generating capacity of the muscle. In contrast, on the
596
ascending limb of the torque – angle relationship (thus with relatively short muscles), motor
597
unit recruitment thresholds did not decrease and motor unit discharge rates did not increase.
598
This indicates that no additional motor units were recruited and discharge rates of already
25
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
574
599
recruited motor units were unchanged, despite the fact that the maximal torque capacity at
600
these knee angles was about 40% lower than at the optimal knee angle for maximal torque
601
production. The presence of length-dependent twitch potentiation may explain the lack of
602
increase in motor unit activity down the ascending limb of the torque – angle relationship.
603
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
26
References
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
1.
Arampatzis A, Karamanidis K, De Monte G, Stafilidis S, Morey-Klapsing G, and
Bruggemann GP. Differences between measured and resultant joint moments during
voluntary and artificially elicited isometric knee extension contractions. Clin Biomech (Bristol,
Avon) 19: 277-283, 2004.
2.
Balnave CD and Allen DG. The effect of muscle length on intracellular calcium and
force in single fibers from mouse skeletal muscle. J Physiol 492: 705-713, 1996.
3.
Baltzopoulos V. A videofluoroscopy method for optical distortion correction and
measurement of knee-joint kinematics. Clin Biomech 10: 85-92, 1995.
4.
Beltman JG, Sargeant AJ, van Mechelen W, and de Haan A. Voluntary activation
level and muscle fiber recruitment of human quadriceps during lengthening contractions. J
Appl Physiol 97: 619-626, 2004.
5.
Beltman JG, van der Vliet MR, Sargeant AJ, and de Haan A. Metabolic cost of
lengthening, isometric and shortening contractions in maximally stimulated rat skeletal muscle.
Acta Physiol Scand 182: 179-187, 2004.
6.
Bigland-Ritchie B, Furbush FH, Gandevia SC, and Thomas CK. Voluntary
discharge frequencies of human motoneurons at different muscle lengths. Muscle Nerve 15:
130-137, 1992.
7.
Christova P, Kossev A, and Radicheva N. Discharge rate of selected motor units in
human biceps brachii at different muscle lengths. J Electromyogr Kinesiol 8: 287-294, 1998.
8.
de Ruiter CJ, de Boer MD, Spanjaard M, and de Haan A. Knee angle-dependent
oxygen consumption during isometric contractions of the knee extensors determined with nearinfrared spectroscopy. J Appl Physiol 99: 579-586, 2005.
9.
de Ruiter CJ, Elzinga MJ, Verdijk PW, van Mechelen W, and de Haan A. Changes
in force, surface and motor unit EMG during post-exercise development of low frequency
fatigue in vastus lateralis muscle. Eur J Appl Physiol 94: 659-669, 2005.
10.
de Ruiter CJ, Elzinga MJH, Verdijk PWL, van Mechelen W, and de Haan A.
Voluntary drive-dependent changes in vastus lateralis motor unit firing rates during a sustained
isometric contraction at 50% of maximum knee extension force. Pflugers Arch - Eur J Physiol
447: 436-444, 2004.
11.
de Ruiter CJ, Kooistra RD, Paalman MI, and de Haan A. Initial phase of maximal
voluntary and electrically stimulated knee extension torque development at different knee
angles. J Appl Physiol 97: 1693-1701, 2004.
12.
Del Valle A and Thomas CK. Motor unit firing rates during isometric voluntary
contractions performed at different muscle lengths. Can J Physiol Pharmacol 82: 769-776,
2004.
13.
Denier van der Gon JJ, ter Haar Romeny BM, and van Zuylen EJ. Behaviour of
motor units of human arm muscles: differences between slow isometric contraction and
relaxation. J Physiol 359: 107-118, 1985.
14.
Farina D, Merletti R, Nazzaro M, and Caruso I. Effect of joint angle on EMG
variables in leg and thigh muscles. IEEE Engineering in Medicine and Biology 20: 62-71,
2001.
15.
Gerrits KH, Maganaris CN, Reeves ND, Sargeant AJ, Jones DA, and de Haan A.
Influence of knee joint angle on muscle properties of paralyzed and nonparalyzed human knee
extensors. Muscle Nerve 32: 73-80, 2005.
16.
Gorassini M, Yang JF, Siu M, and Bennett DJ. Intrinsic activation of human
motoneurons: possible contribution to motor unit excitation. J Neurophysiol 87: 1850-1858,
2002.
27
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
604
17.
Gordon AM, Huxley AF, and Julian FJ. The variation in isometric tension with
sarcomere length in vertebrate muscle fibres. J Physiol 184: 170-192, 1966.
18.
Hansen EA, Lee HD, Barrett K, and Herzog W. The shape of the force-elbow angle
relationship for maximal voluntary contractions and sub-maximal electrically induced
contractions in human elbow flexors. J Biomech 36: 1713-1718, 2003.
19.
Heckman CJ, Johnson M, Mottram C, and Schuster J. Persistent inward currents in
spinal motoneurons and their influence on human motoneuron firing patterns. Neuroscientist
14: 264-275, 2008.
20.
Herzog W and Read LJ. Lines of action and moment arms of the major force-carrying
structures crossing the human knee joint. J Anat 182 (Pt 2): 213-230, 1993.
21.
Hyngstrom AS, Johnson MD, Miller JF, and Heckman CJ. Intrinsic electrical
properties of spinal motoneurons vary with joint angle. Nat Neurosci 10: 363-369, 2007.
22.
Ichinose Y, Kawakami Y, Ito M, and Fukunaga T. Estimation of active force-length
characteristics of human vastus lateralis muscle. Acta Anat (Basel) 159: 78-83, 1997.
23.
Klein CS, Ivanova TD, Rice CL, and Garland SJ. Motor unit discharge rate
following twitch potentiation in human triceps brachii muscle. Neurosci Lett 316: 153-156,
2001.
24.
Kooistra RD, Blaauboer ME, Born JR, de Ruiter CJ, and de Haan A. Knee
extensor muscle oxygen consumption in relation to muscle activation. Eur J Appl Physiol,
2006.
25.
Levine RJ, Kensler RW, Yang Z, Stull JT, and Sweeney HL. Myosin light chain
phosphorylation affects the structure of rabbit skeletal muscle thick filaments. Biophys J 71:
898-907, 1996.
26.
Ng AV, Agre JC, Hanson P, Harrington MS, and Nagle FJ. Influence of muscle
length and force on endurance and pressor responses to isometric exercise. J Appl Physiol 76:
2561-2569, 1994.
27.
Pasquet B, Carpentier A, and Duchateau J. Change in muscle fascicle length
influences the recruitment and discharge rate of motor units during isometric contractions. J
Neurophysiol 94: 3126-3133, 2005.
28.
Place N, Maffiuletti NA, Ballay Y, and Lepers R. Twitch potentiation is greater after
a fatiguing submaximal isometric contraction performed at short vs. long quadriceps muscle
length. J Appl Physiol 98: 429-436, 2005.
29.
Rack PMH and Westbury DR. The effects of length and stimulus rate on tension in
the isometric cat soleus muscle. J Physiol 204: 443-460, 1969.
30.
Rassier DE, MacIntosh BR, and Herzog W. Length dependence of active force
production in skeletal muscle. J Appl Physiol 86: 1445-1457, 1999.
31.
Romaiguere P, Vedel JP, and Pagni S. Comparison of fluctuations of motor unit
recruitment and de-recruitment thresholds in man. Exp Brain Res 95: 517-522, 1993.
32.
Tsaopoulos DE, Baltzopoulos V, Richards PJ, and Maganaris CN. In vivo changes
in the human patellar tendon moment arm length with different modes and intensities of muscle
contraction. J Biomech 40: 3325-3332, 2007.
33.
Vander Linden DW, Kukulka CG, and Soderberg GL. The effect of muscle length
on motor unit discharge characteristics in human tibialis anterior muscle. Exp Brain Res 84:
210-218, 1991.
28
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
Tables
700
701
Table 1. Active knee angles (mean ± SD) during maximal and submaximal torque production
702
(% MVC at AngleTmax) at different dynamometer angles with more extended (top) and more
703
flexed (bottom) knees.
First experimental session
Second experimental session
Dynamometer
Torque (n = 10 subjects)
Torque (n = 10 subjects)
Torque (n = 5 subjects)
angle
10%
20.7 ± 17.9%
7.4 ± 5.4%
100%
AngleTmax -30°
33 ± 3° *
26 ± 2°
AngleTmax -20°
43 ± 3° *
36 ± 2°
50%
100%
100%
51 ± 4° *
45 ± 4°
44 ± 4°
50 ± 4° *
44 ± 4°
53 ± 3° *
46 ± 2°
AngleTmax
61 ± 4° *
56 ± 4°
55 ± 4°
60 ± 4° *
55 ± 4°
63 ± 3° *
56 ± 2°
AngleTmax +10°
71 ± 4° *
66 ± 4°
66 ± 3°
70 ± 4° *
66 ± 3°
73 ± 3° *
67 ± 2°
83 ± 3° *
77 ± 2°
AngleTmax +20°
704
* Denotes significantly less extended than at 100% torque.
705
706
Table 2. Absolute rsEMG (mean ± SD, μV) of the VL muscle during maximal isomeric
707
contractions at different active knee angles during the second session, for 10 and for the subset
708
of 5 subjects. For the SDs of the different knee angles see table 1.
Knee
rsEMG
Knee
rsEMG
angle
N = 10
angle
N=5
26°
339.1 ± 94.7
36°
354.7 ± 119.1
44°
329.2 ± 157.0
46°
385.3 ± 149.2
55°
366.4 ± 167.3
56°
403.5 ± 138.3
66°
327.9 ± 158.7
67°
380.2 ± 125.8
77°
343.7 ± 110.6
709
Maximal VL rsEMG was not different between knee angles, both for n = 10 and n = 5 subjects
710
(P > 0.05).
711
712
713
714
29
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
AngleTmax - 10°
715
Table 3. Percentage increase (last 3 s to first 3 s) in rsEMG (mean ± SD) of the VL and the
716
biceps femoris (BF) muscles during the 10 s torque plateau at different active knee angles, both
717
for the first session (10 and 50 % of the MVC obtained at AngleTmax) and the second session
718
(20.7 ± 17.9 and 7.4 ± 5.4 % MVC respectively for n = 10 and n = 5 subjects). For the SDs of
719
the different knee angles see table 1.
First session
Second session
10 % MVC
50 % MVC
20.7 ± 17.9 % MVC
7.4 ± 5.4 % MVC
Angle Δ%
Δ%
Angle Δ%
Δ%
Angle Δ%
Δ%
Angle Δ%
Δ%
VL
BF
VL
BF
VL
BF
VL
BF
33°
8±2
8±12
43°
4±16
6±7
7±4
9±10
45°
12±3
7±11 50°
11±4
12±5 53°
10±4
13±7
61°
7±5
9±12
56°
12±4
6±8
60°
12±5
11±5 63°
10±4
10±8
71°
11±8
13±17
66°
13±4
12±7 70°
11±5
11±6 73°
11±4
7±6
83°
12±5
10±3
720
There were significant increases (P < 0.05) of VL and biceps femoris rsEMG during the 10 s
721
torque plateau. However, these increases were not different among knee angles (P > 0.05).
722
723
Table 4. Normalized (% maximum) rsEMG (mean ± SD) of the biceps femoris muscle during
724
the first 3 s of the 10 s torque plateau at different active knee angles, both for the first (10 and
725
50 % of the MVC obtained at AngleTmax) and the second session (20.7 ± 17.9 and 7.4 ± 5.4 %
726
MVC respectively for n = 10 and n = 5 subjects). For the SDs of the different knee angles see
727
table 1.
First session
728
Second session
10 % MVC
50 % MVC
20.7 ± 17.9 % MVC
7.4 ± 5.4 % MVC
Angle
Angle
Angle
Angle
%EMG
33°
1.6 ± 1.1
43°
1.0 ± 0.4
%EMG
%EMG
%EMG
51°
2.0 ± 1.1
45°
6.8 ± 2.4
50°
3.9 ± 2.8
53°
1.0 ± 0.3
61°
2.1 ± 0.8
56°
6.9 ± 2.9
60°
3.8 ± 2.8
63°
1.0 ± 0.4
71°
2.0 ± 1.0
66°
7.5 ± 3.6
704°
4.3 ± 3.3
73°
1.2 ± 1.7
83°
1.3 ± 0.8
There were no significant differences in biceps femoris rsEMG among knee angles (P > 0.05).
30
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
51°
729
730
Figure legends
731
Figure 1. Submaximal torque (upper traces) and action potentials of the same single motor unit
732
(lower traces) as a function of time for 1 subject at active knee angles of 37° (A), 67° (B) and
733
87° (C). In (D) the consecutive action potentials are shown of the same single motor unit for 1 s
734
at knee angles of 37, 47, 57, 67, 77, 87°. The motor unit started discharging at a higher torque
735
level at 37° (11.7 Nm) compared to 77 and 87° (10.4 and 8.3 Nm respectively). Note the
736
gradual change in the shape of the motor unit action potentials across the different active knee
737
angles.
739
Figure 2. Normalized (to values obtained at AngleTmax) maximal extension torque (A),
740
recruitment (B, closed circles) and derecruitment thresholds (B, open circles) and discharge
741
rate during submaximal contractions of the first 3 s of the 10 s torque plateau (B, closed
742
squares) as a function of active knee angle for 10 subjects (mean + SD). * Indicates
743
significantly lower at 44 ± 4° and 66 ± 3° compared to 55 ± 4°; # indicates significantly higher
744
recruitment threshold at 50 ± 4° compared to 60 ± 4° (P < 0.05). Note the horizontal SDs (see
745
Table 1).
746
747
Figure 3. Normalized (to values obtained at AngleTmax) maximal torque (A), recruitment (B)
748
and derecruitment threshold (C) and discharge rate (D) as a function of active knee angle for 5
749
subjects (indicated by different symbols) and for 12 motor units (indicated by different shaded
750
of grey). * Indicates significantly lower at 26 ± 2°, 36 ± 2°, 46 ± 2° and 77 ± 2° compared to 56
751
± 2°; # indicates significant main effect of knee angle.
752
753
31
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
738
754
Figure 4. Recruitment to derecruitment ratios at different active knee angles for 10 (closed
755
circles) and 5 (open circles) subjects (mean + or - SD). The recruitment to derecruitment ratio
756
was not significantly different among knee angles, however at each knee angle, the recruitment
757
threshold was significantly higher than the derecruitment threshold. Note the horizontal SDs
758
(see Table 1).
759
Figure 5. Normalized (to values obtained at AngleTmax) rsEMG of the VL muscle of the first
761
3 s of the 10 s torque plateau as a function of active knee angle for 10 (closed circles) and 5
762
(open circles) subjects (mean + or - SD). * Indicates significantly higher at 70 ± 4° compared to
763
60 ± 4°; # indicates significantly higher at 83 ± 3° compared to 53 ± 3°. Note that the small
764
shift of the relation to greater knee angles for n = 5 compared to n = 10 is because on average
765
the produced submaximal torques were lower for n = 5 (7.4 ± 5.4 % MVC at AngleTmax) than
766
for n = 10 (20.7 ± 17.9 % MVC at AngleTmax). Note the horizontal SDs (see Table 1).
767
32
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
760
A
37°
10 Nm
0.5 mV
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
time (s)
B
67°
0.5 mV
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
time (s)
C
87°
10 Nm
0.5 mV
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
time (s)
D
37°
0.5 mV
1 ms
47°
57°
67°
77°
87°
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
10 Nm
A
100
*
*
95
90
85
80
Normalized recruitment torque and discharge rate (%)
40
130
45
50
55
60
65
70
75
55
60
65
70
75
B
120
#
110
100
90
80
70
40
45
extension
50
Knee angle (°)
flexion
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
Normalized torque (%)
105
Maximal torque (%)
110
A
*
100
*
*
90
80
*
70
60
50
40
Derecruitment threshold (%)
B
25
30
35
40
45
50
55
60
65
70
75
80
85
#
150
100
50
0
20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95
250
C
200
150
100
50
0
20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95
200
Discharge rate (%)
20
D
150
100
50
0
20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95
extension
Knee angle (°)
flexion
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
Recruitment threshold (%)
15
200
1.8
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
Recruitment / Derecruitment threshold
2.0
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
25
30
35
extension
40
45
50
55
60
65
Knee angle (°)
70
75
80
85
flexion
90
Downloaded from http://jap.physiology.org/ by 10.220.33.6 on June 17, 2017
Normalized VL rsEMG (%)
130
*
120
110
#
100
90
80
25
30
35
extension
40
45
50
55
60
65
Knee angle (°)
70
75
80
flexion
85
90