- Email: [email protected]
Relationship between motor unit short-term synchronization and common drive in human first dorsal interosseous muscle
Brain Research 767 Ž1997. 314–320
Research report
Relationship between motor unit short-term synchronization and common drive in human first dorsal interosseous muscle John G. Semmler, Michael A. Nordstrom ) , Christopher J. Wallace Department of Physiology, UniÕersity of Adelaide, Adelaide, S.A. 5005, Australia Accepted 30 April 1997
Abstract We assessed the strength of motor unit ŽMU. short-term synchronization and common fluctuations in mean firing rate Žcommon drive. in the same pairs of MUs in order to evaluate whether these features of voluntary MU discharge arise from a common mechanism. Shared, branched-axon inputs, with the most important being widely divergent monosynaptic projections to motoneurons from motor cortical cells, are regarded as the principal determinants of MU short-term synchronization. It is not known to what extent these synaptic inputs are responsible for common drive behaviour of MUs. MU spike trains from 77 pairs of concurrently active MUs in first dorsal interosseous muscle of 17 subjects were discriminated with the high reliability needed for common drive analysis. For each MU pair, the data used for comparison of the two analyses of correlated MU discharge came from a single trial Ž1–5 min duration. of isometric abduction of the index finger. Linear regression revealed a weak, significant positive correlation between the strength of MU short-term synchronization and the strength of common drive in the MU pairs Ž r 2 s 0.06, P - 0.05, n s 77., which was slightly stronger when MU pairs with broad synchronous peaks Ž) 20 ms. were excluded Ž r 2 s 0.09, P - 0.05, n s 63.. These data suggest that less than 10% of the variation in the strength of common drive exhibited by pairs of MUs could be accounted for by differences in the strength of MU short-term synchronization. These two phenomena are therefore likely to arise predominantly from separate mechanisms. At least under these task conditions, the widely divergent, branched-axon inputs from single corticospinal neurons which are important in the generation of MU short-term synchronization play only a minor role in the production of common drive of MU discharge rates. q 1997 Elsevier Science B.V. Keywords: Short-term synchronization; Corticospinal tract
1. Introduction The discharge of voluntarily activated human motor units ŽMUs. is not completely independent. Two examples of correlated discharge patterns that can be demonstrated in pairs of concurrently active MUs in many muscles are short-term synchronization Žthe greater than chance tendency for concurrently active MUs to discharge within a few milliseconds of each other. and common drive Žsimultaneous fluctuations in mean discharge rate.. MU shortterm synchronization is believed to arise from the joint occurrence at the motoneurons of excitatory post-synaptic potentials from branched axons of common pre-synaptic neurons w6,29x. These highly correlated post-synaptic potentials generated by a proportion of the inputs to the tonically active motoneurons slightly increases the probability that they will discharge at the same time. A number )
Corresponding author. Fax: q61 Ž8. 83033356.
0006-8993r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 0 6 2 1 - 5
of neuronal classes with widely divergent projections within the motoneuron pool may potentially contribute to MU short-term synchronization. In recent years, evidence has accumulated that inputs to motoneurons from the contralateral motor cortex are important for the generation of MU short-term synchronization w5,11–13x. The corticomotoneuronal ŽCM. cells, which have widely divergent monosynaptic excitatory connections with motoneurons Žreviewed in w26x., seem likely to be the most important source of inputs responsible for MU short-term synchronization w13x. Concurrently active MUs exhibit an in-phase, 1–2 Hz common modulation of mean discharge frequency which has been termed ‘common drive’ w8x. This phenomenon, which is revealed by smoothing the time-varying firing rate over several successive discharges, is distinct from short-term synchronization which is the result of simultaneous shifts in discharge times in both MUs on a spikeby-spike basis. The mechanisms that may give rise to
J.G. Semmler et al.r Brain Research 767 (1997) 314–320
common drive have not been elucidated experimentally. It presumably reflects a rhythmic 1–2 Hz modulation of activity in a population of last-order neurons. Widely divergent inputs to motoneurons from last-order neurons Ži.e., the same mechanism that is believed to produce short-term synchronization. could be responsible for common drive. However, in the genesis of common drive, the simultaneous arrival of unitary post-synaptic potentials at the motoneurons which is believed to promote synchronous discharge is less important than the effectiveness of widely divergent inputs in transmitting slow fluctuations in the net excitatory drive simultaneously to a large proportion of the motoneuron pool. In theory, single last-order neurons need not have widely divergent inputs to the motoneuron pool to produce common drive; firing rate modulation that is highly correlated in a population of last-order neurons might be sufficient to produce common drive in the mean discharge rates of active MUs even if the motoneurons share few inputs from single last-order neurons. In the present study we have examined the relative importance of the branched-axon inputs to motoneurons, which are recognized as important for the generation of short-term synchronization, for the genesis of common drive fluctuations in mean firing rate. We studied this by comparing, for a number of MU pairs in different subjects, the strength of MU short-term synchronization and the strength of the common drive fluctuations in their mean firing rates. The two analyses were performed on MU spike-train data from first dorsal interosseous ŽFDI. muscle obtained during a single trial Ž1–5 min duration. of weak isometric abduction of the index finger. It was expected that if widely divergent, branched-axon inputs to motoneurons were important mediators of the common drive phenomenon, there would be a strong positive correlation between the strength of MU short-term synchronization and the common drive cross-correlation coefficient in MU pairs. As CM cell activity is believed to be important for MU synchrony, this approach also gives information about the importance of CM cell activity for the generation of common drive. Preliminary results have been presented in abstract form w33x.
2. Materials and methods The experiments were approved by the Committee for the Ethics of Human Experimentation at the University of Adelaide. Data are reported from 77 pairs of concurrently active MUs recorded in FDI muscle of 17 neurologically normal subjects Žages 18–47 years. who volunteered to participate in the study. The data were obtained as part of a larger study of the effects of training on MU discharge patterns w30,32x, and this subset comprises data from those pairs of MUs that could be discriminated with close to 100% accuracy Žnecessary for common drive analysis. and
315
for long Ž1–5 min. epochs of tonic discharge Žnecessary for reliable short-term synchronization analysis.. Three subjects Ž14 MU pairs. were musicians and four others Ž21 MU pairs. regularly lifted weights, while the remainder Ž5 left-handed ŽLH. subjects, 28 MU pairs; 5 right-handed ŽRH. subjects, 14 MU pairs. reported no special use of their hands. FDI MU pairs were obtained from dominant Ž34 pairs. and non-dominant Ž43 pairs. hands, with the degree of hand dominance determined by the Edinburgh handedness inventory w25x. MU activity was recorded with two separate fine-wire electrodes inserted percutaneously into the FDI. Each electrode consisted of three Teflon-insulated fine wires Ž45 mm core diameter. threaded through the lumen of a 25-gauge disposable needle. The surface electromyogram ŽEMG. of the left and right FDI was recorded with bipolar Ag–AgCl electrodes. Myoelectric signals were amplified Ž1000 = ., filtered Žbandwidth 2 Hz–10 kHz. and recorded on FM tape ŽVetter model 400D, Rebersburg, PA, USA, 22 kHzrch. for off-line analysis. The index finger abduction force signal Žbandwidth 0–5 kHz. was also recorded on tape. Subjects were seated with their arm and hand secured in a manipulandum, the details of which have been described previously w31x. The distal interphalangeal joint of the index finger was aligned with a load cell which measured the force of abduction. The manipulandum was designed so that abduction of FDI against the load cell involved only the index finger. To begin the experiment, subjects performed a steady, low-force, isometric abduction of the index finger. A single MU was chosen by the experimenters for the subject to control at a comfortable discharge rate Žtermed the feedback unit.. Subjects were provided with audio and visual feedback of MU discharge on an oscilloscope screen. The subject’s task was to control the mean firing rate of the feedback unit at a constant level for 1–5 min. The activity of additional MUs was monitored during the trial to confirm that discriminable MU potentials were present in each channel for off-line cross-correlation. The procedure was repeated following repositioning of both electrodes, in order to sample from as many different MUs as possible. All analyses were performed off-line from the taped records. Single MUs were discriminated using a computer-based template-matching algorithm ŽSPS 8701; Signal Processing Systems, Malvern, S.A., Australia.. Action potentials belonging to a particular MU were identified on the basis of waveform shape, and great care was taken to confirm the discrimination accuracy. Interspike intervals ŽISIs. of identified MUs were measured Ž"250 ms resolution. using an in-built function of the SPS 8701 and stored on computer. With the aid of ISI histograms, which were constructed from the discharge times of each MU, ISI records were scrutinized for every trial and each discriminated MU to assess discrimination accuracy. Abnormally short and long ISIs that were clearly the result of
316
J.G. Semmler et al.r Brain Research 767 (1997) 314–320
discrimination error Žcf. w21x. were noted, and files with discrimination errors greater than 1% of total discharges were not analysed further for the present study. Files satisfying this criterion for discrimination accuracy that could be paired with another containing ISIs from a concurrently active MU Žrecorded from a separate electrode. were reanalysed on a spike-by-spike basis with the operator manually identifying unclassified spikes Žusually superimpositions with other active MUs. using the off-line spike-sorting facilities of the SPS 8701. The resulting file of MU discharge times contained close to zero incorrect ISI values due to discrimination error. An example of ISI vs. time plots from files used for the analyses is shown in Fig. 1A,B. MUs detected with separate electrodes in the same trial were paired for cross-correlation. Trial durations ranged from 53 to 265 s. The cross-correlation histogram of the individual discharge times of each MU was used to determine the degree of MU short-term synchronization w6,21,29x. All cross-correlation histograms had 1-ms bin widths and spanned a period 100 ms before and after the discharge of the reference unit. An example is shown in the lower trace of Fig. 1C for the MU pair whose ISI vs. time plots appear in Fig. 1A,B. Cross-correlation histograms were restricted to periods in which both units of the pair were tonically active. Histograms with a mean bin-count - 4 were not analysed further. The position and
duration of the synchronous peak was judged visually through the use of the cumulative sum procedure ŽCUSUM; w10x; dotted vertical lines in Fig. 1C.. The significance of synchronous peaks in the cross-correlogram was assessed using the method described by Wiegner and Wierzbicka w35x. If no significant peak was identified from the crosscorrelogram using this test, then a standard peak width of 11 ms, centred at time zero, was used for quantification of the strength of synchrony in that MU pair. The strength of MU short-term synchronization was quantified as the number of synchronous action potentials in the MU pair in excess of chance Ždark area in Fig. 1C. divided by the duration of the trial; i.e., the frequency of extra synchronous discharges in the MU pair. This measure of synchronization was termed an index of common input strength ŽCIS. by Nordstrom et al. w21x, who argued that its magnitude was directly related to the number of shared, branched-axon inputs, their discharge frequency and synaptic efficacy. The synchronization index CIS is independent of the discharge rate of the MUs contributing to the cross-correlogram w21x, an important advantage over other commonly used indices of MU synchronization. Common drive analysis for the MU pairs was performed using spike-train data obtained from the same 1–5 min trial of isometric index finger abduction used for the synchronization analysis. To obtain an estimate of common drive that was representative of the entire trial we
Fig. 1. Quantification of motor unit ŽMU. synchronization and common drive in the same MU pairs. A, B: interspike interval ŽISI. vs. time plots for two concurrently active MUs. Mean ISI was 92.5 ms for the MU shown in A, and 83.2 ms for the MU in B. The duration of the trial was 89.7 s. Note the absence of abnormally short or long ISIs, indicating very accurate discrimination of MU action potentials. C: lower trace shows the cross-correlation histogram of the individual discharge times of MUs shown in A and B. The position and duration of the synchronous peak Žvertical dotted lines. was judged visually from the cumulative sum ŽCUSUM, upper trace.. The width of the peak was 15 ms centred at t s y4 ms. The mean bin count of off-peak bins in the cross-correlation histogram was 11.1 Ždashed horizontal line.. This value served to distinguish the counts expected due to chance Žlight-shaded area. from those counts in excess of chance Ždark-shaded area. in the region of the synchronous peak. The synchronization index CIS for this MU pair was 0.98 extra counts sy1 . D: a 5-s epoch of the time-varying smoothed firing rates Žusing a 400-ms symmetric Hanning window digital filter. of the MUs shown in A Žsolid line. and B Ždotted line.. E: the high-pass filtered Žfilter characteristics; H Ž f . s 1 y Žsin p f .rp f with a low frequency cut-off of 0.75 Hz. version of the smoothed firing rate data shown in D. F: cross-correlation function of the data shown in E, revealing the extent of any underlying common variation in mean firing rates for lags of "0.5 s during the 5-s epoch. The common drive coefficient r for this MU pair was 0.69 at t s 12.5 ms.
J.G. Semmler et al.r Brain Research 767 (1997) 314–320
randomly selected four 5-s epochs from each trial that contained periods of relatively stable firing rate and no evidence of discrimination errors in the raw ISI vs. time plots. The method of common drive analysis was that described by De Luca et al. w8x and was implemented on a Macintosh computer. The time-varying instantaneous discharge frequency of the MU was smoothed Že.g. Fig. 1D. using a 400-ms symmetric Hanning window digital filter of the form hŽt . s 2.50w1 q cosŽptr200.x for
317
3. Results Forty-nine MU pairs Ž64%. had statistically significant peaks near time zero in the cross-correlation histogram. Mean width of the synchronous peak in the cross-correlograms for these pairs was 17 ms Žrange 9–37 ms.. Mean synchronization index CIS for the 77 MU pairs was 0.65 Žrange y0.16–2.97.. Mean common drive coefficient r was 0.44 Žrange 0.03–0.74.. Both measures of correlated MU discharge varied over a large range in different MU pairs. There was no significant difference between the dominant Ž n s 34 MU pairs. and non-dominant hand Ž n s 43 MU pairs. for the mean Ž"S.E.M.. MU synchronization index CIS Ždominant vs. non-dominant; 0.70 " 0.10 vs. 0.61 " 0.10. or the common drive coefficient Ždominant vs. non-dominant; 0.44 " 0.03 vs. 0.44 " 0.02.. For all MU pairs, linear regression analysis revealed a weak but statistically significant positive correlation between the MU synchronization index CIS and the common drive coefficient Žfitted line in Fig. 2A, r 2 s 0.06, P 0.05.. The linear regression correlation coefficient ŽFig. 2A. suggests that only about 6% of the variation in the strength of MU synchronization is associated with changes in the extent of common drive of firing rates. Removal of the 14 MU pairs with a synchronous peak width greater than 20 ms slightly improved the correlation ŽFig. 2B; r 2 s 0.09, P - 0.05.. When the comparison was restricted to the 49 MU pairs with statistically significant synchronous peaks in the cross-correlation histogram, there was no statistically significant correlation between the MU synchronization index CIS and the common drive coefficient r Ž r 2 s 0.11, P ) 0.05.. The weak interdependence of MU synchrony and common drive in the same MU pairs suggests that these two features of voluntary MU discharge arise by relatively independent sources.
Fig. 2. Relationship between the strength of MU synchronization and common drive in the same MU pairs. A: data from 77 MU pairs showing the synchronization strength ŽCIS. plotted against the common drive coefficient r for each MU pair. Linear regression Žfitted line shown. revealed a weak positive correlation between these variables Ž r 2 s 0.06, P - 0.05.. B: data as in A, showing the 63 MU pairs with a synchronous peak less than 20 ms in width. Removal of the MU pairs with broad synchronous peaks slightly improved the relationship between synchronization strength and common drive Žfitted line, r 2 s 0.09, P - 0.05..
318
J.G. Semmler et al.r Brain Research 767 (1997) 314–320
4. Discussion The strength of MU synchrony in FDI can vary considerably in different MU pairs in a single muscle, and there are consistent differences between individuals w3,21x. MU synchrony is also task dependent in the same individual w4x. These differences may be related to anatomical factors Že.g. the degree of divergence of last-order inputs. andror activity of the last-order neurons responsible. We have made use of the wide range in strength of MU synchrony in different MU pairs to examine whether this property of MU discharge is linked with common drive of MU discharge rates. The size and width of the central peaks in cross-correlograms of MU discharge found in the present study Žmean 17 ms. are in agreement with the features of short-term synchronization reported previously in FDI w3,6,21x. Narrow peaks Ž- 20 ms. are consistent with the hypothesis that short-term synchronization arises from the joint generation in the motoneurons of excitatory post-synaptic potentials from branched axons of common pre-synaptic neurons w3,6,29x, which slightly increases the probability that the motoneurons will discharge within a few milliseconds of each other. At least for intrinsic hand muscles, there is a body of evidence suggesting that CM cells from the contralateral motor cortex play a major role in the generation of MU short-term synchrony. In primates the motor cortex CM cells have widely divergent monosynaptic excitatory connections with motoneurons in their target muscles Žreviewed in w26x.. Supporting evidence for a corticospinal origin of short-term synchronization in man includes a gradient of MU synchrony in different muscles which matches the effectiveness of corticospinal inputs w5,11x, and loss of short-term synchrony following lesions of the corticospinal pathway w5,13x. Normal MU synchrony in a deafferented patient w2x supports a central origin. We have preliminary evidence using magnetic transcutaneous cortical stimulation ŽTCS. that differences in FDI MU shortterm synchronization between hands are associated with hemispheric differences in activity of corticospinal neurons during the task w23,24x. Probably the most convincing evidence, although pathophysiological, is the finding of strong MU short-term synchrony in concurrently active MUs in FDI muscles from opposite hands in a patient with mirror movements w12x. This is never seen in normal individuals. Using magnetic TCS, this patient was shown to have bilateral corticospinal projections to both FDI motoneuron pools from the contralateral motor cortex. Common drive is the simultaneous modulation of mean firing rates of the motoneurons during a voluntary isometric contraction which has a predominant frequency in the 1–2 Hz range w8x. This low-frequency modulation must be a feature of the net excitatory drive to the motoneurons, and could arise from the discharge patterns of last-order neurons with inhibitory or excitatory influences on motoneurons. If the common modulation in firing rate is
sufficiently strong in the population of last-order neurons, then the fluctuations in net excitatory drive will be effectively transmitted to the motoneuron pool even without wide divergence of axons from single last-order neurons. Although not essential, a high degree of divergence in the inputs to motoneurons carrying the common modulation in excitatory drive would tend to accentuate the common low-frequency modulation in firing rates of the motoneuron pool. The source of the inputs producing common drive of MU firing rates has not been established. The lowfrequency modulation could be a feature of the descending command signal from supraspinal centres, or arise from the operation of segmental interneuronal circuits, or peripheral afferents. Several lines of evidence suggest a suprasegmental component in the generation of common drive. The nature of the task being performed under voluntary control influences the pattern of common drive seen following cross-correlation of firing rate fluctuations in MUs from antagonist muscles controlling flexionrextension of the interphalangeal joint of the thumb w9x. When the antagonists were coactivated so as to stiffen the joint, fluctuations in mean discharge rate were positively correlated at zero lag for MUs in the two muscles. In contrast, when a force tracking task was performed, fluctuations in mean discharge rate in MUs of the opposing muscles were negatively correlated at zero lag. The report that common drive is higher in FDI MUs from the dominant hand w15x suggests a central origin, which may reflect lateral differences in supraspinal drive or operation of spinal interneuronal circuits. However, in the present study there was no tendency for common drive to be influenced by hand preference. The present study may not be directly comparable with Kamen et al. w15x on this issue, as it included subjects who had trained their muscles for skill or strength, as well as untrained individuals. Similarly, the inclusion of skill- and strength-trained subjects, who have no consistent differences in FDI MU synchrony between hands w30,32x, explains why hand preference and MU synchrony were not related in the present study. In untrained RH subjects, MU synchrony in FDI w31x and extensor carpi radialis w28x differs between sides Žalthough in opposite directions in the hand and wrist muscles.. Motor cortex CM cells, which have widely divergent monosynaptic projections to hand muscle motoneurons, and exert a powerful excitatory influence on them, are an obvious candidate for the generation of common drive. CM cells innervate multiple synergist and even antagonist muscles Žreviewed in w26x., so they could potentially mediate the De Luca and Mambrito w9x findings. A degree of synchronization is seen in the discharge of motor cortex neurons w19x, which include putative w1x and physiologically identified w34x CM cells. Synchrony of motor cortex neurons raises the possibility that correlated discharge in the population of CM cells active during a task is reflected in a modulation of the net excitatory drive to the
J.G. Semmler et al.r Brain Research 767 (1997) 314–320
motoneuron pool. Transient synchronous oscillatory activity has been seen in motor cortex neurons in the 20–40 Hz frequency band w20x, but low-frequency Ž1–2 Hz. oscillations that might produce common drive have not been noted. In the present study there was a weak positive relationship between the extent of MU synchrony and common drive in the same MU pairs in FDI ŽFig. 2.. The correlation was slightly stronger when MU pairs with synchronous peaks wider than 20 ms were excluded from the comparison ŽFig. 2B.. Synchronization of pre-synaptic neurons contributes to broader synchronous peaks in the cross-correlation histogram w16,17x, whereas narrow peaks are thought to predominantly arise from shared, branchedaxon inputs to motoneurons w3,6x. As motor cortex CM cell activity is important for MU short-term synchronization, the relative independence of MU synchrony and common drive suggests that CM cells are not responsible for common drive of MU firing rates. A caveat is that both MU synchronization w4x and common drive w9x have been shown to be task-dependent. It is possible that performance of another task requiring a different pattern of corticospinal neuron involvement might reveal a stronger relationship between MU synchronization and common drive. The findings of the present study are in agreement with the results of Farmer et al. w11x who used both time- and frequency-domain analyses to investigate correlated MU discharge in FDI. These authors found coherence between MU discharge rates in the 1–12 Hz and 16–32 Hz range. Coherence in the 1–3 Hz range was present in 25% of cases in which short-term synchronization was absent. Voluntary common modulation of MU firing rates at low frequencies Ž- 1 Hz. produced high coherence in the low frequency band without modifying short-term synchronization. Farmer et al. w11x concluded that MU coherence in the 16–32 Hz range was produced by the rhythmic discharge of the same inputs producing MU short-term synchronization, which are likely to be of corticospinal origin. MU coherence in the 1–12 Hz range was weakly associated with MU short-term synchronization, and likely to arise from activity in a separate pathway. The low-frequency Ž1–2 Hz. common drive modulation of MU firing rates must therefore arise from oscillatory activity in an indirect descending pathway, or segmental action of afferents or interneuronal circuits. Several possibilities acting at a segmental level do not seem to be important for common drive. The widely divergent, monosynaptic excitatory projections to motoneurons from muscle spindle Ia afferents are not necessary for common drive, as a muscle lacking muscle spindles Žorbicularis oris inferior of the lip. has common drive of MU firing rates that is similar to that found in other muscles that contain spindles w14x. Renshaw cell recurrent inhibition mediated by motoneuron axon collaterals is not essential for common drive, as we have observed the common drive phenomenon in masseter MUs w22x, and the trigeminal motor
319
system lacks recurrent inhibition w18x. Recurrent inhibition is also thought to be weak in distal muscles of the upper limb w27x. In summary, we have shown that short-term synchronization and common drive of firing rates are relatively independent discharge properties of FDI MU pairs voluntarily activated during isometric index finger abduction. These two phenomena are therefore likely to arise from separate mechanisms. At least under these task conditions, the widely divergent, branched-axon inputs from single corticospinal neurons which are important in the generation of MU short-term synchronization play only a minor role in the production of common drive of MU discharge rates.
Acknowledgements The encouragement and support of Dr. T.S. Miles is gratefully acknowledged. This work forms part of the Ph.D. studies of J.G.S., who was supported by a University of Adelaide Postgraduate Research Scholarship. M.A.N. is an R.D. Wright Fellow of the NH and MRC of Australia. C.J.W. was supported by an Australian Postgraduate Research Award.
References w1x J.H. Allum, M.C. Hepp-Reymond, R. Gysin, Cross-correlation analysis of interneuronal connectivity in the motor cortex of the monkey, Brain Res. 231 Ž1982. 325–334. w2x J.R. Baker, F.D. Bremner, J.D. Cole, J.A. Stephens, Short-term synchronization of intrinsic hand muscle motor units in a ‘deafferented’ man, J. Physiol. ŽLond.. 396 Ž1988. 155P. w3x F.D. Bremner, J.R. Baker, J.A. Stephens, Correlation between the discharges of motor units recorded from the same and from different finger muscles in man, J. Physiol. 432 Ž1991. 355–380. w4x F.D. Bremner, J.R. Baker, J.A. Stephens, Effect of task on the degree of synchronization of intrinsic hand muscle motor units in man, J. Neurophysiol. 66 Ž1991. 2072–2083. w5x A.K. Datta, S.F. Farmer, J.A. Stephens, Central nervous pathways underlying synchronization of human motor unit firing studied during voluntary contractions, J. Physiol. ŽLond.. 432 Ž1991. 401–425. w6x A.K. Datta, J.A. Stephens, Synchronization of motor unit activity during voluntary contraction in man, J. Physiol. 422 Ž1990. 397–419. w7x C.J. De Luca, Z. Erim, Common drive of motor units in regulation of muscle force, Trends Neurosci. 17 Ž1994. 299–305. w8x C.J. De Luca, R.S. LeFever, M.P. McCue, A.P. Xenakis, Control scheme governing concurrently active human motor units during voluntary contractions, J. Physiol. ŽLond.. 329 Ž1982. 129–142. w9x C.J. De Luca, B. Mambrito, Voluntary control of motor units in human antagonist muscles: coactivation and reciprocal activation, J. Neurophysiol. 58 Ž1987. 525–542. w10x P.H. Ellaway, Cumulative sum technique and its application to the analysis of peristimulus time histograms, Electroencephalogr. Clin. Neurophysiol. 45 Ž1978. 302–304. w11x S.F. Farmer, F.D. Bremner, D.M. Halliday, J.R. Rosenberg, J.A. Stephens, The frequency content of common synaptic inputs to motoneurones studied during voluntary isometric contraction in man, J. Physiol. 470 Ž1993. 127–155.
320
J.G. Semmler et al.r Brain Research 767 (1997) 314–320
w12x S.F. Farmer, D.A. Ingram, J.A. Stephens, Mirror movements studied in a patient with Klippel-Feil syndrome, J. Physiol. 428 Ž1990. 467–484. w13x S.F. Farmer, M. Swash, D.A. Ingram, J.A. Stephens, Changes in motor unit synchronization following central nervous lesions in man, J. Physiol. 463 Ž1993. 83–105. w14x G. Kamen, C.J. De Luca, Firing rate interactions among human orbicularis oris motor units, Int. J. Neurosci. 64 Ž1992. 167–175. w15x G. Kamen, S.S. Greenstein, C.J. De Luca, Lateral dominance and motor unit firing behavior, Brain Res. 576 Ž1992. 165–167. w16x P.A. Kirkwood, T.A. Sears, D.L. Tuck, R.H. Westgaard, Variations in the time course of the synchronization of intercostal motoneurones in the cat, J. Physiol. 327 Ž1982. 105–135. w17x P.A. Kirkwood, T.A. Sears, R.H. Westgaard, Restoration of function in external intercostal motoneurones of the cat following partial central deafferentation, J. Physiol. 350 Ž1984. 225–251. w18x E. Luschei, L.J. Goldberg, Neural mechanisms of mandibular control: mastication and voluntary biting, in: V.B. Brooks ŽEd.., Handbook of Physiology, sect. 1, The Nervous System, vol. II, Motor Control, part 1, American Physiological Society, Bethesda, MD, 1981, pp. 1237–1274. w19x J.T. Murphy, H.C. Kwan, Y.C. Wong, Cross correlation studies in primate motor cortex: event related correlation, Can. J. Neurol. Sci. 12 Ž1985. 24–30. w20x V.N. Murthy, E.E. Fetz, Oscillatory activity in sensorimotor cortex of awake monkeys: synchronization of local field potentials and relation to behaviour, J. Neurophysiol. 76 Ž1996. 3949–3967. w21x M.A. Nordstrom, A.J. Fuglevand, R.M. Enoka, Estimating the strength of common input to human motoneurons from the crosscorrelogram, J. Physiol. 453 Ž1992. 547–574. w22x M.A. Nordstrom, T.S. Miles, K.S. Turker, Evidence for common ¨ drive in motor units of human masseter, Proc. XXX Int. Congr. Physiol. Sci. XVI Ž1986. 5P. w23x M.A. Nordstrom, J.G. Semmler, Hemispheric differences in motor cortex excitability during simple index finger abduction performed with dominant and non-dominant hands, Proc. Aust. Neurosci. Soc. 7 Ž1996. 50.
w24x M.A. Nordstrom, J.G. Semmler, Motor cortex excitability during task performance is related to hand preference in man, Soc. Neurosci. Abstr. 22 Ž1996. 658. w25x R.C. Oldfield, The assessment and analysis of handedness: the Edinburgh inventory, Neuropsychologia 9 Ž1971. 97–113. w26x R. Porter, R.N. Lemon, Corticospinal Function and Voluntary Movement, Oxford University Press, New York, 1993. w27x A. Rossi, R. Mazzocchio, Renshaw recurrent inhibition to motoneurones innervating proximal and distal muscles of the human upper and lower limb, in: L. Jami, E. Pierrot-Deseilligny, D. Zytnicki ŽEds.., Muscle Afferents and Spinal Control of Movement, Pergamon Press, Oxford, England, 1992, pp. 313–319. w28x A. Schmied, J.P. Vedel, S. Pagni, Human spinal lateralization assessed from motoneurone synchronization: dependence on handedness and motor unit type, J. Physiol. 480 Ž1994. 369–387. w29x T.A. Sears, D. Stagg, Short-term synchronization of intercostal motoneurone activity, J. Physiol. 263 Ž1976. 357–381. w30x J.G. Semmler, M.A. Nordstrom, Motor unit synchronization and force tremor are reduced in skill-compared to strength-trained subjects, Soc. Neurosci. Abstr. 21 Ž1995. 1433. w31x J.G. Semmler, M.A. Nordstrom, Influence of handedness on motor unit discharge properties and force tremor, Exp. Brain Res. 104 Ž1995. 115–125. w32x J.G. Semmler, M.A. Nordstrom, Influence of muscle usage patterns on motor unit synchronization and force tremor in skill- and strength-trained subjects, Proc. Aust. Neurosci. Soc. 7 Ž1996. 49. w33x J.G. Semmler, C.J. Wallace, M.A. Nordstrom, Short-term synchronization and common drive of firing rates are relatively independent properties of voluntary motor unit discharge, Proc. Aust. Neurosci. Soc. 7 Ž1996. 196. w34x W.S. Smith, E.E. Fetz, Effects of synchrony between primate corticomotoneuronal cells on post-spike facilitation of muscles and motor units, Neurosci. Lett. 96 Ž1989. 76–81. w35x A.W. Wiegner, M.M. Wierzbicka, A method for assessing significance of peaks in cross-correlation histograms, J. Neurosci. Methods 22 Ž1987. 125–131.
Research report
Relationship between motor unit short-term synchronization and common drive in human first dorsal interosseous muscle John G. Semmler, Michael A. Nordstrom ) , Christopher J. Wallace Department of Physiology, UniÕersity of Adelaide, Adelaide, S.A. 5005, Australia Accepted 30 April 1997
Abstract We assessed the strength of motor unit ŽMU. short-term synchronization and common fluctuations in mean firing rate Žcommon drive. in the same pairs of MUs in order to evaluate whether these features of voluntary MU discharge arise from a common mechanism. Shared, branched-axon inputs, with the most important being widely divergent monosynaptic projections to motoneurons from motor cortical cells, are regarded as the principal determinants of MU short-term synchronization. It is not known to what extent these synaptic inputs are responsible for common drive behaviour of MUs. MU spike trains from 77 pairs of concurrently active MUs in first dorsal interosseous muscle of 17 subjects were discriminated with the high reliability needed for common drive analysis. For each MU pair, the data used for comparison of the two analyses of correlated MU discharge came from a single trial Ž1–5 min duration. of isometric abduction of the index finger. Linear regression revealed a weak, significant positive correlation between the strength of MU short-term synchronization and the strength of common drive in the MU pairs Ž r 2 s 0.06, P - 0.05, n s 77., which was slightly stronger when MU pairs with broad synchronous peaks Ž) 20 ms. were excluded Ž r 2 s 0.09, P - 0.05, n s 63.. These data suggest that less than 10% of the variation in the strength of common drive exhibited by pairs of MUs could be accounted for by differences in the strength of MU short-term synchronization. These two phenomena are therefore likely to arise predominantly from separate mechanisms. At least under these task conditions, the widely divergent, branched-axon inputs from single corticospinal neurons which are important in the generation of MU short-term synchronization play only a minor role in the production of common drive of MU discharge rates. q 1997 Elsevier Science B.V. Keywords: Short-term synchronization; Corticospinal tract
1. Introduction The discharge of voluntarily activated human motor units ŽMUs. is not completely independent. Two examples of correlated discharge patterns that can be demonstrated in pairs of concurrently active MUs in many muscles are short-term synchronization Žthe greater than chance tendency for concurrently active MUs to discharge within a few milliseconds of each other. and common drive Žsimultaneous fluctuations in mean discharge rate.. MU shortterm synchronization is believed to arise from the joint occurrence at the motoneurons of excitatory post-synaptic potentials from branched axons of common pre-synaptic neurons w6,29x. These highly correlated post-synaptic potentials generated by a proportion of the inputs to the tonically active motoneurons slightly increases the probability that they will discharge at the same time. A number )
Corresponding author. Fax: q61 Ž8. 83033356.
0006-8993r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 0 6 2 1 - 5
of neuronal classes with widely divergent projections within the motoneuron pool may potentially contribute to MU short-term synchronization. In recent years, evidence has accumulated that inputs to motoneurons from the contralateral motor cortex are important for the generation of MU short-term synchronization w5,11–13x. The corticomotoneuronal ŽCM. cells, which have widely divergent monosynaptic excitatory connections with motoneurons Žreviewed in w26x., seem likely to be the most important source of inputs responsible for MU short-term synchronization w13x. Concurrently active MUs exhibit an in-phase, 1–2 Hz common modulation of mean discharge frequency which has been termed ‘common drive’ w8x. This phenomenon, which is revealed by smoothing the time-varying firing rate over several successive discharges, is distinct from short-term synchronization which is the result of simultaneous shifts in discharge times in both MUs on a spikeby-spike basis. The mechanisms that may give rise to
J.G. Semmler et al.r Brain Research 767 (1997) 314–320
common drive have not been elucidated experimentally. It presumably reflects a rhythmic 1–2 Hz modulation of activity in a population of last-order neurons. Widely divergent inputs to motoneurons from last-order neurons Ži.e., the same mechanism that is believed to produce short-term synchronization. could be responsible for common drive. However, in the genesis of common drive, the simultaneous arrival of unitary post-synaptic potentials at the motoneurons which is believed to promote synchronous discharge is less important than the effectiveness of widely divergent inputs in transmitting slow fluctuations in the net excitatory drive simultaneously to a large proportion of the motoneuron pool. In theory, single last-order neurons need not have widely divergent inputs to the motoneuron pool to produce common drive; firing rate modulation that is highly correlated in a population of last-order neurons might be sufficient to produce common drive in the mean discharge rates of active MUs even if the motoneurons share few inputs from single last-order neurons. In the present study we have examined the relative importance of the branched-axon inputs to motoneurons, which are recognized as important for the generation of short-term synchronization, for the genesis of common drive fluctuations in mean firing rate. We studied this by comparing, for a number of MU pairs in different subjects, the strength of MU short-term synchronization and the strength of the common drive fluctuations in their mean firing rates. The two analyses were performed on MU spike-train data from first dorsal interosseous ŽFDI. muscle obtained during a single trial Ž1–5 min duration. of weak isometric abduction of the index finger. It was expected that if widely divergent, branched-axon inputs to motoneurons were important mediators of the common drive phenomenon, there would be a strong positive correlation between the strength of MU short-term synchronization and the common drive cross-correlation coefficient in MU pairs. As CM cell activity is believed to be important for MU synchrony, this approach also gives information about the importance of CM cell activity for the generation of common drive. Preliminary results have been presented in abstract form w33x.
2. Materials and methods The experiments were approved by the Committee for the Ethics of Human Experimentation at the University of Adelaide. Data are reported from 77 pairs of concurrently active MUs recorded in FDI muscle of 17 neurologically normal subjects Žages 18–47 years. who volunteered to participate in the study. The data were obtained as part of a larger study of the effects of training on MU discharge patterns w30,32x, and this subset comprises data from those pairs of MUs that could be discriminated with close to 100% accuracy Žnecessary for common drive analysis. and
315
for long Ž1–5 min. epochs of tonic discharge Žnecessary for reliable short-term synchronization analysis.. Three subjects Ž14 MU pairs. were musicians and four others Ž21 MU pairs. regularly lifted weights, while the remainder Ž5 left-handed ŽLH. subjects, 28 MU pairs; 5 right-handed ŽRH. subjects, 14 MU pairs. reported no special use of their hands. FDI MU pairs were obtained from dominant Ž34 pairs. and non-dominant Ž43 pairs. hands, with the degree of hand dominance determined by the Edinburgh handedness inventory w25x. MU activity was recorded with two separate fine-wire electrodes inserted percutaneously into the FDI. Each electrode consisted of three Teflon-insulated fine wires Ž45 mm core diameter. threaded through the lumen of a 25-gauge disposable needle. The surface electromyogram ŽEMG. of the left and right FDI was recorded with bipolar Ag–AgCl electrodes. Myoelectric signals were amplified Ž1000 = ., filtered Žbandwidth 2 Hz–10 kHz. and recorded on FM tape ŽVetter model 400D, Rebersburg, PA, USA, 22 kHzrch. for off-line analysis. The index finger abduction force signal Žbandwidth 0–5 kHz. was also recorded on tape. Subjects were seated with their arm and hand secured in a manipulandum, the details of which have been described previously w31x. The distal interphalangeal joint of the index finger was aligned with a load cell which measured the force of abduction. The manipulandum was designed so that abduction of FDI against the load cell involved only the index finger. To begin the experiment, subjects performed a steady, low-force, isometric abduction of the index finger. A single MU was chosen by the experimenters for the subject to control at a comfortable discharge rate Žtermed the feedback unit.. Subjects were provided with audio and visual feedback of MU discharge on an oscilloscope screen. The subject’s task was to control the mean firing rate of the feedback unit at a constant level for 1–5 min. The activity of additional MUs was monitored during the trial to confirm that discriminable MU potentials were present in each channel for off-line cross-correlation. The procedure was repeated following repositioning of both electrodes, in order to sample from as many different MUs as possible. All analyses were performed off-line from the taped records. Single MUs were discriminated using a computer-based template-matching algorithm ŽSPS 8701; Signal Processing Systems, Malvern, S.A., Australia.. Action potentials belonging to a particular MU were identified on the basis of waveform shape, and great care was taken to confirm the discrimination accuracy. Interspike intervals ŽISIs. of identified MUs were measured Ž"250 ms resolution. using an in-built function of the SPS 8701 and stored on computer. With the aid of ISI histograms, which were constructed from the discharge times of each MU, ISI records were scrutinized for every trial and each discriminated MU to assess discrimination accuracy. Abnormally short and long ISIs that were clearly the result of
316
J.G. Semmler et al.r Brain Research 767 (1997) 314–320
discrimination error Žcf. w21x. were noted, and files with discrimination errors greater than 1% of total discharges were not analysed further for the present study. Files satisfying this criterion for discrimination accuracy that could be paired with another containing ISIs from a concurrently active MU Žrecorded from a separate electrode. were reanalysed on a spike-by-spike basis with the operator manually identifying unclassified spikes Žusually superimpositions with other active MUs. using the off-line spike-sorting facilities of the SPS 8701. The resulting file of MU discharge times contained close to zero incorrect ISI values due to discrimination error. An example of ISI vs. time plots from files used for the analyses is shown in Fig. 1A,B. MUs detected with separate electrodes in the same trial were paired for cross-correlation. Trial durations ranged from 53 to 265 s. The cross-correlation histogram of the individual discharge times of each MU was used to determine the degree of MU short-term synchronization w6,21,29x. All cross-correlation histograms had 1-ms bin widths and spanned a period 100 ms before and after the discharge of the reference unit. An example is shown in the lower trace of Fig. 1C for the MU pair whose ISI vs. time plots appear in Fig. 1A,B. Cross-correlation histograms were restricted to periods in which both units of the pair were tonically active. Histograms with a mean bin-count - 4 were not analysed further. The position and
duration of the synchronous peak was judged visually through the use of the cumulative sum procedure ŽCUSUM; w10x; dotted vertical lines in Fig. 1C.. The significance of synchronous peaks in the cross-correlogram was assessed using the method described by Wiegner and Wierzbicka w35x. If no significant peak was identified from the crosscorrelogram using this test, then a standard peak width of 11 ms, centred at time zero, was used for quantification of the strength of synchrony in that MU pair. The strength of MU short-term synchronization was quantified as the number of synchronous action potentials in the MU pair in excess of chance Ždark area in Fig. 1C. divided by the duration of the trial; i.e., the frequency of extra synchronous discharges in the MU pair. This measure of synchronization was termed an index of common input strength ŽCIS. by Nordstrom et al. w21x, who argued that its magnitude was directly related to the number of shared, branched-axon inputs, their discharge frequency and synaptic efficacy. The synchronization index CIS is independent of the discharge rate of the MUs contributing to the cross-correlogram w21x, an important advantage over other commonly used indices of MU synchronization. Common drive analysis for the MU pairs was performed using spike-train data obtained from the same 1–5 min trial of isometric index finger abduction used for the synchronization analysis. To obtain an estimate of common drive that was representative of the entire trial we
Fig. 1. Quantification of motor unit ŽMU. synchronization and common drive in the same MU pairs. A, B: interspike interval ŽISI. vs. time plots for two concurrently active MUs. Mean ISI was 92.5 ms for the MU shown in A, and 83.2 ms for the MU in B. The duration of the trial was 89.7 s. Note the absence of abnormally short or long ISIs, indicating very accurate discrimination of MU action potentials. C: lower trace shows the cross-correlation histogram of the individual discharge times of MUs shown in A and B. The position and duration of the synchronous peak Žvertical dotted lines. was judged visually from the cumulative sum ŽCUSUM, upper trace.. The width of the peak was 15 ms centred at t s y4 ms. The mean bin count of off-peak bins in the cross-correlation histogram was 11.1 Ždashed horizontal line.. This value served to distinguish the counts expected due to chance Žlight-shaded area. from those counts in excess of chance Ždark-shaded area. in the region of the synchronous peak. The synchronization index CIS for this MU pair was 0.98 extra counts sy1 . D: a 5-s epoch of the time-varying smoothed firing rates Žusing a 400-ms symmetric Hanning window digital filter. of the MUs shown in A Žsolid line. and B Ždotted line.. E: the high-pass filtered Žfilter characteristics; H Ž f . s 1 y Žsin p f .rp f with a low frequency cut-off of 0.75 Hz. version of the smoothed firing rate data shown in D. F: cross-correlation function of the data shown in E, revealing the extent of any underlying common variation in mean firing rates for lags of "0.5 s during the 5-s epoch. The common drive coefficient r for this MU pair was 0.69 at t s 12.5 ms.
J.G. Semmler et al.r Brain Research 767 (1997) 314–320
randomly selected four 5-s epochs from each trial that contained periods of relatively stable firing rate and no evidence of discrimination errors in the raw ISI vs. time plots. The method of common drive analysis was that described by De Luca et al. w8x and was implemented on a Macintosh computer. The time-varying instantaneous discharge frequency of the MU was smoothed Že.g. Fig. 1D. using a 400-ms symmetric Hanning window digital filter of the form hŽt . s 2.50w1 q cosŽptr200.x for
317
3. Results Forty-nine MU pairs Ž64%. had statistically significant peaks near time zero in the cross-correlation histogram. Mean width of the synchronous peak in the cross-correlograms for these pairs was 17 ms Žrange 9–37 ms.. Mean synchronization index CIS for the 77 MU pairs was 0.65 Žrange y0.16–2.97.. Mean common drive coefficient r was 0.44 Žrange 0.03–0.74.. Both measures of correlated MU discharge varied over a large range in different MU pairs. There was no significant difference between the dominant Ž n s 34 MU pairs. and non-dominant hand Ž n s 43 MU pairs. for the mean Ž"S.E.M.. MU synchronization index CIS Ždominant vs. non-dominant; 0.70 " 0.10 vs. 0.61 " 0.10. or the common drive coefficient Ždominant vs. non-dominant; 0.44 " 0.03 vs. 0.44 " 0.02.. For all MU pairs, linear regression analysis revealed a weak but statistically significant positive correlation between the MU synchronization index CIS and the common drive coefficient Žfitted line in Fig. 2A, r 2 s 0.06, P 0.05.. The linear regression correlation coefficient ŽFig. 2A. suggests that only about 6% of the variation in the strength of MU synchronization is associated with changes in the extent of common drive of firing rates. Removal of the 14 MU pairs with a synchronous peak width greater than 20 ms slightly improved the correlation ŽFig. 2B; r 2 s 0.09, P - 0.05.. When the comparison was restricted to the 49 MU pairs with statistically significant synchronous peaks in the cross-correlation histogram, there was no statistically significant correlation between the MU synchronization index CIS and the common drive coefficient r Ž r 2 s 0.11, P ) 0.05.. The weak interdependence of MU synchrony and common drive in the same MU pairs suggests that these two features of voluntary MU discharge arise by relatively independent sources.
Fig. 2. Relationship between the strength of MU synchronization and common drive in the same MU pairs. A: data from 77 MU pairs showing the synchronization strength ŽCIS. plotted against the common drive coefficient r for each MU pair. Linear regression Žfitted line shown. revealed a weak positive correlation between these variables Ž r 2 s 0.06, P - 0.05.. B: data as in A, showing the 63 MU pairs with a synchronous peak less than 20 ms in width. Removal of the MU pairs with broad synchronous peaks slightly improved the relationship between synchronization strength and common drive Žfitted line, r 2 s 0.09, P - 0.05..
318
J.G. Semmler et al.r Brain Research 767 (1997) 314–320
4. Discussion The strength of MU synchrony in FDI can vary considerably in different MU pairs in a single muscle, and there are consistent differences between individuals w3,21x. MU synchrony is also task dependent in the same individual w4x. These differences may be related to anatomical factors Že.g. the degree of divergence of last-order inputs. andror activity of the last-order neurons responsible. We have made use of the wide range in strength of MU synchrony in different MU pairs to examine whether this property of MU discharge is linked with common drive of MU discharge rates. The size and width of the central peaks in cross-correlograms of MU discharge found in the present study Žmean 17 ms. are in agreement with the features of short-term synchronization reported previously in FDI w3,6,21x. Narrow peaks Ž- 20 ms. are consistent with the hypothesis that short-term synchronization arises from the joint generation in the motoneurons of excitatory post-synaptic potentials from branched axons of common pre-synaptic neurons w3,6,29x, which slightly increases the probability that the motoneurons will discharge within a few milliseconds of each other. At least for intrinsic hand muscles, there is a body of evidence suggesting that CM cells from the contralateral motor cortex play a major role in the generation of MU short-term synchrony. In primates the motor cortex CM cells have widely divergent monosynaptic excitatory connections with motoneurons in their target muscles Žreviewed in w26x.. Supporting evidence for a corticospinal origin of short-term synchronization in man includes a gradient of MU synchrony in different muscles which matches the effectiveness of corticospinal inputs w5,11x, and loss of short-term synchrony following lesions of the corticospinal pathway w5,13x. Normal MU synchrony in a deafferented patient w2x supports a central origin. We have preliminary evidence using magnetic transcutaneous cortical stimulation ŽTCS. that differences in FDI MU shortterm synchronization between hands are associated with hemispheric differences in activity of corticospinal neurons during the task w23,24x. Probably the most convincing evidence, although pathophysiological, is the finding of strong MU short-term synchrony in concurrently active MUs in FDI muscles from opposite hands in a patient with mirror movements w12x. This is never seen in normal individuals. Using magnetic TCS, this patient was shown to have bilateral corticospinal projections to both FDI motoneuron pools from the contralateral motor cortex. Common drive is the simultaneous modulation of mean firing rates of the motoneurons during a voluntary isometric contraction which has a predominant frequency in the 1–2 Hz range w8x. This low-frequency modulation must be a feature of the net excitatory drive to the motoneurons, and could arise from the discharge patterns of last-order neurons with inhibitory or excitatory influences on motoneurons. If the common modulation in firing rate is
sufficiently strong in the population of last-order neurons, then the fluctuations in net excitatory drive will be effectively transmitted to the motoneuron pool even without wide divergence of axons from single last-order neurons. Although not essential, a high degree of divergence in the inputs to motoneurons carrying the common modulation in excitatory drive would tend to accentuate the common low-frequency modulation in firing rates of the motoneuron pool. The source of the inputs producing common drive of MU firing rates has not been established. The lowfrequency modulation could be a feature of the descending command signal from supraspinal centres, or arise from the operation of segmental interneuronal circuits, or peripheral afferents. Several lines of evidence suggest a suprasegmental component in the generation of common drive. The nature of the task being performed under voluntary control influences the pattern of common drive seen following cross-correlation of firing rate fluctuations in MUs from antagonist muscles controlling flexionrextension of the interphalangeal joint of the thumb w9x. When the antagonists were coactivated so as to stiffen the joint, fluctuations in mean discharge rate were positively correlated at zero lag for MUs in the two muscles. In contrast, when a force tracking task was performed, fluctuations in mean discharge rate in MUs of the opposing muscles were negatively correlated at zero lag. The report that common drive is higher in FDI MUs from the dominant hand w15x suggests a central origin, which may reflect lateral differences in supraspinal drive or operation of spinal interneuronal circuits. However, in the present study there was no tendency for common drive to be influenced by hand preference. The present study may not be directly comparable with Kamen et al. w15x on this issue, as it included subjects who had trained their muscles for skill or strength, as well as untrained individuals. Similarly, the inclusion of skill- and strength-trained subjects, who have no consistent differences in FDI MU synchrony between hands w30,32x, explains why hand preference and MU synchrony were not related in the present study. In untrained RH subjects, MU synchrony in FDI w31x and extensor carpi radialis w28x differs between sides Žalthough in opposite directions in the hand and wrist muscles.. Motor cortex CM cells, which have widely divergent monosynaptic projections to hand muscle motoneurons, and exert a powerful excitatory influence on them, are an obvious candidate for the generation of common drive. CM cells innervate multiple synergist and even antagonist muscles Žreviewed in w26x., so they could potentially mediate the De Luca and Mambrito w9x findings. A degree of synchronization is seen in the discharge of motor cortex neurons w19x, which include putative w1x and physiologically identified w34x CM cells. Synchrony of motor cortex neurons raises the possibility that correlated discharge in the population of CM cells active during a task is reflected in a modulation of the net excitatory drive to the
J.G. Semmler et al.r Brain Research 767 (1997) 314–320
motoneuron pool. Transient synchronous oscillatory activity has been seen in motor cortex neurons in the 20–40 Hz frequency band w20x, but low-frequency Ž1–2 Hz. oscillations that might produce common drive have not been noted. In the present study there was a weak positive relationship between the extent of MU synchrony and common drive in the same MU pairs in FDI ŽFig. 2.. The correlation was slightly stronger when MU pairs with synchronous peaks wider than 20 ms were excluded from the comparison ŽFig. 2B.. Synchronization of pre-synaptic neurons contributes to broader synchronous peaks in the cross-correlation histogram w16,17x, whereas narrow peaks are thought to predominantly arise from shared, branchedaxon inputs to motoneurons w3,6x. As motor cortex CM cell activity is important for MU short-term synchronization, the relative independence of MU synchrony and common drive suggests that CM cells are not responsible for common drive of MU firing rates. A caveat is that both MU synchronization w4x and common drive w9x have been shown to be task-dependent. It is possible that performance of another task requiring a different pattern of corticospinal neuron involvement might reveal a stronger relationship between MU synchronization and common drive. The findings of the present study are in agreement with the results of Farmer et al. w11x who used both time- and frequency-domain analyses to investigate correlated MU discharge in FDI. These authors found coherence between MU discharge rates in the 1–12 Hz and 16–32 Hz range. Coherence in the 1–3 Hz range was present in 25% of cases in which short-term synchronization was absent. Voluntary common modulation of MU firing rates at low frequencies Ž- 1 Hz. produced high coherence in the low frequency band without modifying short-term synchronization. Farmer et al. w11x concluded that MU coherence in the 16–32 Hz range was produced by the rhythmic discharge of the same inputs producing MU short-term synchronization, which are likely to be of corticospinal origin. MU coherence in the 1–12 Hz range was weakly associated with MU short-term synchronization, and likely to arise from activity in a separate pathway. The low-frequency Ž1–2 Hz. common drive modulation of MU firing rates must therefore arise from oscillatory activity in an indirect descending pathway, or segmental action of afferents or interneuronal circuits. Several possibilities acting at a segmental level do not seem to be important for common drive. The widely divergent, monosynaptic excitatory projections to motoneurons from muscle spindle Ia afferents are not necessary for common drive, as a muscle lacking muscle spindles Žorbicularis oris inferior of the lip. has common drive of MU firing rates that is similar to that found in other muscles that contain spindles w14x. Renshaw cell recurrent inhibition mediated by motoneuron axon collaterals is not essential for common drive, as we have observed the common drive phenomenon in masseter MUs w22x, and the trigeminal motor
319
system lacks recurrent inhibition w18x. Recurrent inhibition is also thought to be weak in distal muscles of the upper limb w27x. In summary, we have shown that short-term synchronization and common drive of firing rates are relatively independent discharge properties of FDI MU pairs voluntarily activated during isometric index finger abduction. These two phenomena are therefore likely to arise from separate mechanisms. At least under these task conditions, the widely divergent, branched-axon inputs from single corticospinal neurons which are important in the generation of MU short-term synchronization play only a minor role in the production of common drive of MU discharge rates.
Acknowledgements The encouragement and support of Dr. T.S. Miles is gratefully acknowledged. This work forms part of the Ph.D. studies of J.G.S., who was supported by a University of Adelaide Postgraduate Research Scholarship. M.A.N. is an R.D. Wright Fellow of the NH and MRC of Australia. C.J.W. was supported by an Australian Postgraduate Research Award.
References w1x J.H. Allum, M.C. Hepp-Reymond, R. Gysin, Cross-correlation analysis of interneuronal connectivity in the motor cortex of the monkey, Brain Res. 231 Ž1982. 325–334. w2x J.R. Baker, F.D. Bremner, J.D. Cole, J.A. Stephens, Short-term synchronization of intrinsic hand muscle motor units in a ‘deafferented’ man, J. Physiol. ŽLond.. 396 Ž1988. 155P. w3x F.D. Bremner, J.R. Baker, J.A. Stephens, Correlation between the discharges of motor units recorded from the same and from different finger muscles in man, J. Physiol. 432 Ž1991. 355–380. w4x F.D. Bremner, J.R. Baker, J.A. Stephens, Effect of task on the degree of synchronization of intrinsic hand muscle motor units in man, J. Neurophysiol. 66 Ž1991. 2072–2083. w5x A.K. Datta, S.F. Farmer, J.A. Stephens, Central nervous pathways underlying synchronization of human motor unit firing studied during voluntary contractions, J. Physiol. ŽLond.. 432 Ž1991. 401–425. w6x A.K. Datta, J.A. Stephens, Synchronization of motor unit activity during voluntary contraction in man, J. Physiol. 422 Ž1990. 397–419. w7x C.J. De Luca, Z. Erim, Common drive of motor units in regulation of muscle force, Trends Neurosci. 17 Ž1994. 299–305. w8x C.J. De Luca, R.S. LeFever, M.P. McCue, A.P. Xenakis, Control scheme governing concurrently active human motor units during voluntary contractions, J. Physiol. ŽLond.. 329 Ž1982. 129–142. w9x C.J. De Luca, B. Mambrito, Voluntary control of motor units in human antagonist muscles: coactivation and reciprocal activation, J. Neurophysiol. 58 Ž1987. 525–542. w10x P.H. Ellaway, Cumulative sum technique and its application to the analysis of peristimulus time histograms, Electroencephalogr. Clin. Neurophysiol. 45 Ž1978. 302–304. w11x S.F. Farmer, F.D. Bremner, D.M. Halliday, J.R. Rosenberg, J.A. Stephens, The frequency content of common synaptic inputs to motoneurones studied during voluntary isometric contraction in man, J. Physiol. 470 Ž1993. 127–155.
320
J.G. Semmler et al.r Brain Research 767 (1997) 314–320
w12x S.F. Farmer, D.A. Ingram, J.A. Stephens, Mirror movements studied in a patient with Klippel-Feil syndrome, J. Physiol. 428 Ž1990. 467–484. w13x S.F. Farmer, M. Swash, D.A. Ingram, J.A. Stephens, Changes in motor unit synchronization following central nervous lesions in man, J. Physiol. 463 Ž1993. 83–105. w14x G. Kamen, C.J. De Luca, Firing rate interactions among human orbicularis oris motor units, Int. J. Neurosci. 64 Ž1992. 167–175. w15x G. Kamen, S.S. Greenstein, C.J. De Luca, Lateral dominance and motor unit firing behavior, Brain Res. 576 Ž1992. 165–167. w16x P.A. Kirkwood, T.A. Sears, D.L. Tuck, R.H. Westgaard, Variations in the time course of the synchronization of intercostal motoneurones in the cat, J. Physiol. 327 Ž1982. 105–135. w17x P.A. Kirkwood, T.A. Sears, R.H. Westgaard, Restoration of function in external intercostal motoneurones of the cat following partial central deafferentation, J. Physiol. 350 Ž1984. 225–251. w18x E. Luschei, L.J. Goldberg, Neural mechanisms of mandibular control: mastication and voluntary biting, in: V.B. Brooks ŽEd.., Handbook of Physiology, sect. 1, The Nervous System, vol. II, Motor Control, part 1, American Physiological Society, Bethesda, MD, 1981, pp. 1237–1274. w19x J.T. Murphy, H.C. Kwan, Y.C. Wong, Cross correlation studies in primate motor cortex: event related correlation, Can. J. Neurol. Sci. 12 Ž1985. 24–30. w20x V.N. Murthy, E.E. Fetz, Oscillatory activity in sensorimotor cortex of awake monkeys: synchronization of local field potentials and relation to behaviour, J. Neurophysiol. 76 Ž1996. 3949–3967. w21x M.A. Nordstrom, A.J. Fuglevand, R.M. Enoka, Estimating the strength of common input to human motoneurons from the crosscorrelogram, J. Physiol. 453 Ž1992. 547–574. w22x M.A. Nordstrom, T.S. Miles, K.S. Turker, Evidence for common ¨ drive in motor units of human masseter, Proc. XXX Int. Congr. Physiol. Sci. XVI Ž1986. 5P. w23x M.A. Nordstrom, J.G. Semmler, Hemispheric differences in motor cortex excitability during simple index finger abduction performed with dominant and non-dominant hands, Proc. Aust. Neurosci. Soc. 7 Ž1996. 50.
w24x M.A. Nordstrom, J.G. Semmler, Motor cortex excitability during task performance is related to hand preference in man, Soc. Neurosci. Abstr. 22 Ž1996. 658. w25x R.C. Oldfield, The assessment and analysis of handedness: the Edinburgh inventory, Neuropsychologia 9 Ž1971. 97–113. w26x R. Porter, R.N. Lemon, Corticospinal Function and Voluntary Movement, Oxford University Press, New York, 1993. w27x A. Rossi, R. Mazzocchio, Renshaw recurrent inhibition to motoneurones innervating proximal and distal muscles of the human upper and lower limb, in: L. Jami, E. Pierrot-Deseilligny, D. Zytnicki ŽEds.., Muscle Afferents and Spinal Control of Movement, Pergamon Press, Oxford, England, 1992, pp. 313–319. w28x A. Schmied, J.P. Vedel, S. Pagni, Human spinal lateralization assessed from motoneurone synchronization: dependence on handedness and motor unit type, J. Physiol. 480 Ž1994. 369–387. w29x T.A. Sears, D. Stagg, Short-term synchronization of intercostal motoneurone activity, J. Physiol. 263 Ž1976. 357–381. w30x J.G. Semmler, M.A. Nordstrom, Motor unit synchronization and force tremor are reduced in skill-compared to strength-trained subjects, Soc. Neurosci. Abstr. 21 Ž1995. 1433. w31x J.G. Semmler, M.A. Nordstrom, Influence of handedness on motor unit discharge properties and force tremor, Exp. Brain Res. 104 Ž1995. 115–125. w32x J.G. Semmler, M.A. Nordstrom, Influence of muscle usage patterns on motor unit synchronization and force tremor in skill- and strength-trained subjects, Proc. Aust. Neurosci. Soc. 7 Ž1996. 49. w33x J.G. Semmler, C.J. Wallace, M.A. Nordstrom, Short-term synchronization and common drive of firing rates are relatively independent properties of voluntary motor unit discharge, Proc. Aust. Neurosci. Soc. 7 Ž1996. 196. w34x W.S. Smith, E.E. Fetz, Effects of synchrony between primate corticomotoneuronal cells on post-spike facilitation of muscles and motor units, Neurosci. Lett. 96 Ž1989. 76–81. w35x A.W. Wiegner, M.M. Wierzbicka, A method for assessing significance of peaks in cross-correlation histograms, J. Neurosci. Methods 22 Ž1987. 125–131.