Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
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