Ne~r~seie~e Vol. 36, No. 3,
pp. 779-784,
0~522/~
1990
$3.00+ 0.00 plc Q 1990IBRO
Pergamon Pms
Printed in Great Britain
REFLEX RESPONSES OF FUSIMOTONEUR~NS SINUSOIDAL MUSCLE STRETCHING
TO
R. ANASTASIJEVIC,*I(. JOVANOVICand J. VuCo Institute for Medical Research, 11001 Beograd, Bulevar JNA 10, Yugoslavia Abstract-Reflex response of fusimotone~ons to sinusoidal muscle stretching were investigated in decerebrated cats, Nerve impulses of single fusimotoneurons were recorded from thin filaments dissected from otherwise intact nerves to triceps surae muscles. Amplitude of the sinusoidal stretching of these muscles was 3 mm peak-to-peak and the frequencies 0.1-10 Hz. Electric muscle activity was also recorded in some experiments. Fusimotor responses were similar to those of the skeletomotoneurons in that both were advanced in phase with respect to muscle length changes, while their amplitude increased with increase in stretching frequency. Modulation at the frequency corresponding to the second harmonic of the input signal was predominant in fusimotor responses. It is supposed to appear mainly due to the convergence to fusimotoneurons of agerent impulses from different muscle receptors arriving after different delays. Its functional role is discussed.
carried out under halothane anaesthesia. All the nerves of electrical muscle activity*‘*” as well inne~ating the right hind limb except those for the triceps as of skeletomotoneuron responses to sinusoidal surae muscles were severed. Impulses of single fusimotomuscle stretchingti*24*2s,37*4’ indicated that both the neurons were recorded from fine filaments isolated from amplitude and velocity of the input signal are otherwise intact nerves to triceps surae muscles. They were identified as fusimotor if their conduction velocity, deterreflected in these cells output. These findings were confirmed by analysis of skeletomotor responses to mined by spike triggered averaging technique26s’swas within the range 10-45 m/s. Sinusoidal muscle stretching of the afferent discharges induced by sinusoidally modutriceps surae muscles at an amplitude of 3 mm peak-to-peak lated eiectrical stimulationz7 as well as to sinusoidally and frequencies 0.1-10.0 Hz, was produced by the means of modulated’2,zs and ramp-like” transmembrane a servo-controlled Pye Ling electromagnetic puller. The current stimulation. Input-output characteristics of muscles were prestretched by 6 mm. Muscle length changes were recorded simultaneously with fusimotor spikes and fusimotoneurons have not been investigated except stored on magnetic tape. In some experiments electric for their responses to long-lasting depolarizing muscle activity (EMG) of either gastrocemius medialis or Modulations of fusimotor current injections.” gastrocnemius lateralis muscle was also recorded. Cycle histograms of fusimotor spikes were computed discharge rate during the rising phase of vibrationfrom 128 sinusoidal muscle stretches (64 for 0.1 Hz stretchinduced reflex muscle contraction’ as well as during ing frequency). Bin widths of the histograms were l-40 ms, small oscillatory changes in maintained muscle depending on the stretching frequency (i.e. cycle duration). tension levelm indicated that fusimotoneurons might Sine and cosine Fourier coefficients (u, b) of the first six also be responsive to both amplitude and velocity harmonics were computed off-line on Hewlett Packard 2817 computer and amplitudes of these harmonics (,,/m), of the input signal. In order to get more information reflecting modulation in fusimotor discharge rate, calcuon fusimotoneuron input-output characteristics, lated. In some cases power spectra were also computed, changes in discharge rate of these cells were investiusing Fast Fourier Transform, from either the cycle gated in this work during sinusoidal muscle stretchhistograms, or the records of fusimotor discharge during a ing. Stretching frequencies and amplitude are chosen sequence of sinusolidal muscle stretches of an available length, or else from poststimulus time histograms (PSTHs) so as to correspond to those applied in most of the above cited works on skeletomotoneurons.4*5,24~zg,30*4’computed during 128 IO-cycle sequences of sinusoidal stretching. The same analyses were performed on the The data obtained are also intended to serve as a basis records of spontaneous discharges of a corresponding for further studying of fusimotoneuron response length. Since the system was obviously non-linear and the characteristics using an approach appropriate for changes in fusimotor discharge rate substantiaBy distorted by higher harmonics, phase and amplitudes of fusimotor non-linear system analysis.3’ Some preliminary results have been presented.2,3 responses shown in Fig. 2 were computed from the cycle
Phase relations
EXPERIMENTAL PROCEDURES The experiments were performed on 12 adult decerebrate cats. The operative procedure before decerebration was -__.. *To whom correspondence should be addressed. Abbreviations: EMG, electric muscle activity; PSTH, poststimulus time histogram.
histograms: phase advance was calculated from the distance between either the peak discharge rate or the starting point of its increase and the maximal muscle length; response amplitude was expressed as either the difference between the peak discharge rate during the muscle stretching and the mean spontaneous discharge rate or as the ratio of this difference to the mean discharge rate. The discharge rates were calculated from the histograms by dividing the number of spikes per bin with bin width and number of cycles and expressed in imp/s.
779
780
R.
ANASTASIJEV~ef u!.
RESULTS
Changes in discharge rate of 1.5 spontaneously firing fusimotoneurons from 12 experiments during sinusoidat muscle stretching were investigated. Spontaneous discharge rate of these cells ranged from 5 to 80 imp/s and the conduction velocity of their axons IO-22 m/s. Changes in discharge rate induced by sinusolidal muscle stretching were usually barely visible and could be ascertained only after cycle histograms had been computed. The largest ones encountered are shown in Fig. 1, which represents a general picture of the recordings. Amplitudes und phases qffusimotor responses. Corn parison with skeletomotor responses
Phases of the onset and the maximum of the reflex response (increase in discharge rate) of nine fusimotoneurons (mean k SD.) with respect to the muscle length changes at different stretching frequencies are
shown in Fig. 2A. Both the onset and the maximum of the fusimotor reflex responses appeared in advance of the maximal muscle length, by about 65’ and 45’, respectively at all the stretching frequencies applied. Phase lagging of the reflex response occurred in only two fusimotoneurons during the first stretch in the sequence at 10 Hz stretching frequency and also in one of them at 5 Hz. Phase advance of EMG, measured from its onset, although rather variable from experiment to experiment, covered the same range of degrees as that of fusimotor responses. Amplitudes of the reflex responses (peak increase in discharge rate) of the same nine fusimotoneurons (mean + S.B.) to muscle stretching at different frequencies are shown in Fig. 2B. Both the absolute and relative (normalized with respect to the spontaneous firing rate) response ampIitude increased with increase in stretching frequency. Frequency of spikes in EMG activity increased while its duration decreased with muscle stretching frequency
A
8
Fig. I. Changes in discharge rate of a f~imotoneuron during sinusoidal muscle stretching. (A) Spontaneous discharge. (3) During muscle stretching. Upper traces: PSTHs computed during 128 sequences of sinusoidal muscle stretching (B) and the corresponding period of spontaneous firing (A); middle traces: spike discharges of the cell; lower traces: muscle length changes. Sinusoidal stretching 10 Hz, 3mm. Bin width 4ms. Fusimotor discharge recorded from a gastrocnemius medialis nerve fascicle, conduction velocity 14 m/s. Lower-most trace in B: EMG of the gastrocnemius medialis muscle.
781
Fusimotor responses to sinusoidal stretching
I:
0:
~~~~~~~FR~~~
5.0 STRETCHING
FREQUENCY
Fig. 2. Phase advance (A) and amplitude (B) of the fusimotor reflex response to sinusoidal muscle stretching at different frequencies (mean + S.D. for nine fusimotoneurons). (A) Dashed and continuous line: phase advance of the onset and peak increase in discharge rate, respectively. Dotted line: phase advance of EMG onset (mean f S.D. for eight cats). Abscissa: stretching frequency; ordinate: angle measured in advance from the maximum muscle length. (B) Abscissa: stretching frequency; left and right ordinates: relative and absolute increase in fusimotor discharge rate (Fm maximal discharge rate: Fs, mean spontaneous discharge rate). Inset: EMG records at 0.1, 1 and 10 Hz stretching frequency.
until at 5 and 10 Hz it turned to a short-lasting synchronous burst (Fig. 2B, insets). No attempts were made to estimate these changes in a quantitative way. Analysis of changes Harmonic distortion
in fusimotor
discharge
rate.
Most prominent changes in fusimotor discharge rate revealed by cycle histograms appeared at a frequency corresponding to the second harmonic of sinusoidal muscle stretching frequency. The highest peaks in power spectra appeared at the frequency corresponding to the second harmonic and the amplitude of this harmonic, calculated from the Fourier coefficients, was the largest. While the amplitudes of all the six harmonics increased with muscle stretching frequency, the increase in the second one was steeper than that of the others. These findings are summarized in Fig. 3. Histograms of a fusimotoneuron discharge at three different muscle stretching frequencies are shown in Fig. 3A. Thin lines drawn through the cycle histograms at 0.1 and 1 Hz are obtained by summing up the first four harmonics. At 5 Hz stretching frequency a fairly good fit to a cycle histogram was obtained in this cell by summing the first and the second harmonic. PSTH is shown in Fig. 3 instead, to make the time relations of changes in fusimotor discharge rate to those of muscle length more clearly visible. Amplitudes of the first to sixth harmonics in the response of the same fusimotoneuron at these frequencies of muscle stretching are shown in Fig. 3B
and power spectra in Fig. 3C. In some cells, the fourth harmonic was also prominent and its amplitude attained the values of the first one. Modulations appearing in spontaneous discharge rate (not shown) were smaller in all the investigated cells by an order of magnitude and showed no prominent harmonic, or else the prominent one was not the same as that during muscle stretching. The phases computed from the Fourier coefficients should not be relied on in a non-linear system. It seems obvious, however, from Fig. 3 that at least the second harmonic contributed both to the phase advance of the whole fusimotor response as well as to the steepness of the initial increase in firing rate. DISCUSSION
Phase advance and response amplitude The reflex response of fusimotoneurons to sinusoidal muscle stretching showed the expected similarities to the response of skeletomotoneurons. Phase advance of the fusimotor reflex response with respect to the muscle length changes corresponded to the phase advance of skeletomotor spikes4-6.39 and membrane potential changes24~37~4’as well as of EMG (Lippold et af.29130and these experiments). The increase in phase advance at higher stretching frequencies, present in skeletomotor reflex discharges, was not found in these experiments. Its absence may be a consequence of much longer delays in the reflex path to fusimotoneurons and the lower conduction
782
I
I
I
1
1
1
I
1
p-lOOtu--
I
1.0
0.25
0.M
.5
0.75
L 32 256
i HZ
1.0s
FREQUENCY
Fig. 3. Harmonics in fusimotor responses. (A) Cycle histograms of the discharge of a fusimotor neuron (to gastrocnemius medialis muscle, conduction velocity 14 m/s) at 0. I and 1.OHz stretching frequencies as indicated (bin width 40 and 4 ms, 64 and 128 trials, respectively). At 5.0 Hz PSTH computed during lOcycle sequence of sinusoidal stretching (bin width 10 ms, 128 trials). Thin lines, sums of the first four (at 0. I and 1 Hz) and first two (at 5 Hz) harmonics; thick lines, muscle length changes; dashed lines, mean spontaneous discharge rate. (B) Amplitude (ordinates, imp/s) of the first to sixth harmonics (abscissa) in the response of the same fusimotoneuron. Stretching frequencies as in A. (C) Power spectra of the fusimotor discharge.
velocity of their axons.* Amplitude responses responses
of the fusimotor is not directly comparable with the of skeletomotoneurons since the latter are
not spontaneously active and the number of their spikes per stretching cycle decreases with muscle stretching frequency, i.e. with cycle durationk6 It showed, however, some similarities to the modulations of skeletomotor discharge rate induced by intracellular injection of sinusoidal currents:‘*.*’ in both cases the response amplitude increased with increasing input frequency, with a change in slope at a frequency between 1 and 5 Hz. Both the phase *Central delay for excitation of fusimotoneurons by 50 pm triangular muscle stretches is shown to be 5.1-13.7 ms (1-2 ms less for larger stretches) and for skeletomotoneurons 0.7-3 ms.lh Since fusimotor spikes were recorded from muscle-nerve filaments, their conduction from the spinal cord to the recording site would take another 3.5-14 ms (for conduction velocity range 1040 m/s and the distance of 140 mm). This would make a delay of 6.6-25.7 ms. Such a delay would mean approximately 0.2”VI.9” (or less than one PSTH bin) at 0.1 Hz stretching frequency, but 23.8”-92.5” (six to 25 bins) at 10 Hz. For skeletomotor spikes recorded from ventral root filaments (or their membrane potential changes recorded intracellularly) this delay would be only about 2.5~10.8” at 10 Hz stretching frequency.
advance and the increase in the response amplitude with muscle stretching frequency indicate that the velocity of the input signal is reflected in the output of fusimotoneurons. Phase advance of skeletomotor reflex responses to sinusoidal muscle stretching compensates for the phase lag of muscle tension changes with respect to skeletomotor spikes improving, in this way, frequency characteristics of the stretch reflex.37.39Due to the phase advance of fusimotor responses, a rather synchronous alpha-gamma activation would be preserved (see Windhorst and Koehler43) and the fusimotor impulses would arrive in time to exert their effects on the muscle spindle before the maximal muscle length is attained. Harmonics in fisimotor
discharge rate
Predominant changes in fusimotor discharge rate at a frequency double that of the sinusoidal muscle stretching were not expected. Due to the large stretching amplitude applied in this work muscle spindle behaviour would be non-linear32.36 and might also be affected by its unloading by the reflex muscle contraction. This, however, could not account for the appearance of the second harmonic in changes of fusimotor discharge rate since a pause rather than a
783
Fusimotor responses to sinusoidal stretching burst of impulses would appear in muscle spindle afferent discharges during shortening of the muscle. The second harmonic appears in membrane potential changes of skeletomotoneurons at a comparable range of stretching amplitudes24 but not nearly so large as the one in fusimotor responses. Prominent second harmonic (in some cases also the fourth one) may appear if the sinusoidal input signal is fully rectified (the so-called square-law responses).2i This process, if present, could hardly be ascribed to fusimotoneurons themselves: if the properties of fusimotoneurons are similar to those of skeletomotoneurons,42 half-rectification could be expected at best.*’ Therefore, the source of the second harmonic in fusimotor responses must be looked for in reflex pathways in the spinal cord. Modulations in response amplitude of vestibular nuclear cells at a frequency double that of the input sinusoidal signal4 are supposed to appear as a consequence of afferent inputs of opposite sign converging upon them. Many feedback loops with different delays may also induce different periodic and aperiodic oscillations in responses of the target neurons.*’ Afferent impulses both of opposite sign and with different delays might converge upon fusimotoneurons in our experiments. Muscle receptors other than muscle spindle primary endings respond to sinusoidal muscle stretching.24,32,39 Discharges in some group III and IV afferent fibres might be induced by muscle stretching and/or contraction.33 Afferent impulses from muscle receptors, activated either by electrical stimulation of muscle nerves’ 9 or by muscle stretch and/or contraction’.‘E’7.34 induce both excitation and inhibition of fusimotoneurons. Latencies of the fusimotor responses are different, due to the differences in both conduction velocity of the afferent fibres involved as well as in central delays indicating anything from di- to polysynaptic transmission. Timing of arrival to fusimotoneurons of afferent impulses from different receptors, if evoked during sinusoidal muscle stretching, could hardly be predicted with enough certainty to decide whether they have induced the modulations found in fusimotor discharge rate. Recurrent inhibi-
tionlO,l4
may also induce a transient decrease in fusimotor discharge rate. It has been shown3’ that burst-like discharges appear in responses of Renshaw cells to sinusoidal muscle stretches when stretching velocity is at or above 0.5 mm/s. Since stretching velocity in our experiments would range from 0.6 to 6Omm/s the decrease in fusimotor discharge rate coincident with the peak muscle length could appear to be due to their recurrent inhibition as is supposed to be the case with the discharge of skeletomotoneurons.3s.4’ However induced, second harmonic in fusimotor discharge rate makes these cells respond as movement detectors. This information can be transmitted through muscle spindle afferents to the central nervous system. Second harmonic also contributes to the steeper rise in fusimotor discharge rate as well as to the phase advance of their responses during muscle lengthening. In this way the ensuing contraction of intrafusal muscle fibres would appear in parallel with the whole muscle reflex contraction. It might also be enhanced if the intrafusal muscle fibres share the “catch properties”‘-’ with extrafusal muscle fibres. Second increase in fusimotor discharge rate, coincident with muscle shortening, might help to keep the muscle spindles from slackening. During sinusoidal changes in muscle length occurring in natural movements such as locomotion, this would prepare extensor muscle spindles for the support phase of the step cycle. Peaks in discharge rate of fusimotoneurons, enhanced or induced by the harmonics in their responses to sinusoida muscle stretching, followed by either further increase or decrease in muscle length, may exert different after-effects on the responsiveness of muscle spindle sensory endings.l8,19.22.23 Both magnitude** and latency23 of the receptor responses to subsequent changes in muscle length and/or affected.
fusimotor
discharge
rate
could
be
Acknowledgements-This work was supported by a grant from the Serbian Medical Research Foundation. We thank M. JociC for help with computer programs.
REFERENCE 1. Anastasijevii: R. and VuEo J. (1984) Changes in fusimotor outflow during vibration-induced contraction of triceps surae muscles in decerebrate cats. E.x@ Neuroi. 85, 523-532. 2. Anastasijevib R. and Vui-o J. (1985) Parallel modulation of skeletomotor and fusimotor activity induced by muscle tension changes. Iugoslav. Physiol. Pharmac. Acta (Suppl.) 21, l-2. 3. AnastasijeviC R. and VuEo J. (1986) Fusimotor response to sinusoidal muscle stretching in decerebrated cats. Neurosci. Lert. Suppl. 26, S.567. 4. AnastasijeviC R., AnojGE M., TodoroviC. B. and
5. 6. 7. 8.
VuEo J. (1969) Effect of fusimotor stimulation on the reflex response of spinal alpha motoneurones to sinusoidal stretching of the muscle. Expl Neural. 25, 559-570. AnojGi: M., PaSiCM., TodoroviC B. and VuEo J. (1967) The reflex activity of spinal alpha motoneurones under dynamic conditions of stretching the tendon to triceps surae muscle in cat. lugoslao. Physioi. Pharmac. Acta 3, 109-121. AnojCic M., Pa&C M., Todorovii: B. and VuEo J. (1968) The reflex activity of spinal motoneurones under dynamic conditions of stretching the tendon to triceps. J. Physiot. 194, 30-33P. Appelberg B., Huiliger M., Johansson H. and Sojka P. (1983) Action on gamma motoneurones elicited by electrical stimulation of group I muscle afferent fibres in the hind limb of the cat. J. Physioi. 335, 237-253. Appelberg B., Hulliger M., Johansson H. and Sojka P. (1983) Action on gamma motone~ones elicited by electrical stimulation of group II muscle afferent fibres in the hind limb of the cat. J. Physiul. 335, 255273.
784
R. ANASTA~IJEVI~ et a/
9. Appelberg B., Hulliger M., Johansson H. and Sojka P. (1983) Action on gamma motoneurones elicited by electrical stimulation of group III muscle afferent fibres in the hind limb of the cat. J. Phvsiol. 335. 275-292. IO. Appelberg B., Hulliger M., Johansson H. and Sojka P. (1983) Recurrent actions on gamma motoneurones mediated via large and small ventral root fibres in the cat. J. Physiol. 335, 2933305. II. Baldissera F., Campadelli P. and Piccinelli L. (I(982) Neural encoding of input transients investigated by intracellular injection of ramp currents in cat a-motoneurones. J. Physiot. 3uI, 73-86.
12. Bafdissera F., Campadelli P. and Piccinelli L. (1984) The dynamic response of cat a-motoneurones inv~tjgated by intracellular injection of sinusoidal currents. Expl Brain Res. 54, 2755282. 13. Burke R. E., Rudomin P. and Zajac F. E. (1970) Catch property in single mammalian motor units. Science 168, 122-124. 14. Ellaway P. H. (1971) Recurrent inhibition of fusimotor neurones exhibiting background discharge in the decerebrate and the spinal cat. J. Physiol. 216, 419439. 15. Ellaway P. H. and Murphy P. R. (1980) Autogenetic effects of muscle contraction on extensor gamma motoneurones in the cat. Expl Brain Res. 38, 305-312. 16. Eflaway P. H. and Trott J. R. (1978) Autogenetic reflex action on the gamma motoneurones by stretch of triceps surae in the decerebrated cats. /. Physiol. 276, 49-66. 17. Ellaway P. H., Murphy P. R. and Tripathi A. (1982) Closely coupled excitation of gamma motoneurones by group III muscle afferents with low mechanical threshold in the cat. J. Physiol. 331, 481489. 18. Emonet-Denand F., Hunt C. C. and Laporte Y. (1985) Fusimotor after-effects on response of primary endings to test dynamic stimuli in cat muscle spindles. J. Physiol. 360, 187-200. 19. Emonet-Denand F.. Hunt C. C. and Laporte Y. (1985) Effects of stretch on dynamic fusimotor after-effects in cat muscle spindle. J. Physiol. 360, 201-213. 20. Glass L., Beuter A. and Laraque D. (1988) Time delays, oscillations and chaos in physiological control systems. Math. Biosci. 90, 11l-125. 21. Graham D. and McRuer D. (1961) Analysis of Nonlinear Control Systems. John Wiley, New York. 22. Gregory J. E., Morgan D. L. and Proske U. (1986) After-effects in the responses of cat muscle spindles. J. Neurophysiol. 56, 451461.
23. Gregory J. E., Morgan D. L. and Proske U. (1988) Responses of muscle spindles depend on their history of activation and movement. In Progress in Brain Research (eds Haman W. and Iggo A.), Vol. 74>pp. 85-90, Elsevier, Amsterdam, 24. JociC M., Anastasijevic R. and V&o J. (1988) Membrane potential changes of skeletomotor neurones durina sinusoidal muscle stretching. Frequency characteristics. Iugosla. Physial. Pharmac. Acta 2.4, 213-222. 2.5. JociC M., Anastasijevic R. and Vuco J. (1989) Membrane potential changes of skeletomotor neurones during sinusoidal muscle stretching. Mathematical modeling. Iugoslav. Physiol. Pharmac. Acta 25, 89994. 26. Kanda K., Burke R. and Walmsley B. (1977) Differential control of fast and slow twitch motor units in the dccerebrate - .__ _--. cat. Cxpi Brain Kes. ZY, 57-74. 27. Kostyukov A. I. (1982) Transformation of the frequency-modulated afferent activity by cat spinal motoneurons. ~e~ro~zio~ogiya 14, 198-200. 28. Kostyukov A. I. and Kry~anovsky M. V. (1982) Impulse activity evoked in the cat spinal neurons by sinusoidal depolarizing currents. Neiro~zio~og~ya 14, 3542. 29. Lippold 0. C. J., Redfearn J. W. T. and VuEo J. (1957) The rhythmical activity of groups of motor units in the voluntary contraction of muscle. J. Physiol. 137, 473-487.
30. Lippold 0. C. J., Redfearn J. W. T. and Vuco J. (1958) The effect of sinusoidal stretching upon the activity of stretch receptors in voluntary muscle and their reflex responses. J. Physiol. 144, 373-386. 31. Marmarelis P. Z. and Marmarelis V. Z. (1978) Analysis of Physiological Systems. The White Noise Approach. Plenum Press, New York. 32. Matthews P. 3. C. and Stein R. B. (1969) The sensitivity of muscle spindle afferents to small sinusoidal changes in length. J. Physic& 200, 723-743. 33. Mense S. and Meyer H. (1985) Different types of slowly conducting aRerent units in cat skeletal muscle and tendon. J. Physiol. 363, 403-417. 34. Noth J. and Thilmann A. (1980) Autogenetic excitation of extensor gamma motoneurones in the cat. Neurosci. Lett. 17, 23-26.
by group I1 muscle afferents
35. Pompeiano O., Wand P. and Sontag K. H. (1975) The sensitivity of Renshaw cells to velocity of sinusoidal stretches of the triceps surae muscle. Archs ital. Biol. 113, 280-294. 36. Poppcle R. E. and Bowman R. J. (1970) Quantitative description of linear behavior of mammalian muscle spindles. J. Neurophysiol. 33, 59-72.
37. Poppele -R: E. and Terzuolo C. A. (1968) Myotatic reflex: its input-output relation. Science 159, 743-745. 38. Post E. M.. Rymer W. Z. and Hasan Z. (1980) Relation between intrafusal and extrafusal activity in triceps surae muscles of the decerebrate cat: evidence for b action. J. Neurophysiol. 44, 383-404. 39. Rosenthal N. P., McKean T. A., Roberts W. J. and Terzuolo C. A. (1970) Frequency analysis of stretch reflex and its main subsystems in triceps surae muscles of the cat. J. ~~rophys~o~. 33, 713-749. 40. Vuco J. and Anastasijevic R. (1985) Modulations of fusimotor discharge induced by tension changes during reflex muscle contraction. In The Muscle Spindle (eds Boyd I. A. and Gladden M. H.), pp. 2855289. Macmillan, Basingstoke. 41. Westbury D. R. (1971) The response of a-motoneurones of the cat to sinusoidal movements of the muscle they innervate. Brain Res. 25, 75-86. 42. Westbury D. R. (1981) Electrophysiological characteristics of spinal gamma motoneurons in the cat. In Muscle Receptors and Movement (eds Taylor A. and Prochazka A.), pp. 87-96. Macmillan, London. 43. Windhorst U. and Koehler W. (1986) The dynamic effects of inputs to spinal motoneurones of different type upon the outputs of ~-motoneurones mediated via recurrent inhibition. Bioi. Cybern. 54, 1677177. 44. Xerri C., Barth~l~my J., Harlay F., Bore1 L. and Lacour M. (1988) Neural coding of linear motion in the vestibular nuclei of the alert cat. Expi Brain Res. 65, 5699581. (Accepted 15 November 1989)
pp. 779-784,
0~522/~
1990
$3.00+ 0.00 plc Q 1990IBRO
Pergamon Pms
Printed in Great Britain
REFLEX RESPONSES OF FUSIMOTONEUR~NS SINUSOIDAL MUSCLE STRETCHING
TO
R. ANASTASIJEVIC,*I(. JOVANOVICand J. VuCo Institute for Medical Research, 11001 Beograd, Bulevar JNA 10, Yugoslavia Abstract-Reflex response of fusimotone~ons to sinusoidal muscle stretching were investigated in decerebrated cats, Nerve impulses of single fusimotoneurons were recorded from thin filaments dissected from otherwise intact nerves to triceps surae muscles. Amplitude of the sinusoidal stretching of these muscles was 3 mm peak-to-peak and the frequencies 0.1-10 Hz. Electric muscle activity was also recorded in some experiments. Fusimotor responses were similar to those of the skeletomotoneurons in that both were advanced in phase with respect to muscle length changes, while their amplitude increased with increase in stretching frequency. Modulation at the frequency corresponding to the second harmonic of the input signal was predominant in fusimotor responses. It is supposed to appear mainly due to the convergence to fusimotoneurons of agerent impulses from different muscle receptors arriving after different delays. Its functional role is discussed.
carried out under halothane anaesthesia. All the nerves of electrical muscle activity*‘*” as well inne~ating the right hind limb except those for the triceps as of skeletomotoneuron responses to sinusoidal surae muscles were severed. Impulses of single fusimotomuscle stretchingti*24*2s,37*4’ indicated that both the neurons were recorded from fine filaments isolated from amplitude and velocity of the input signal are otherwise intact nerves to triceps surae muscles. They were identified as fusimotor if their conduction velocity, deterreflected in these cells output. These findings were confirmed by analysis of skeletomotor responses to mined by spike triggered averaging technique26s’swas within the range 10-45 m/s. Sinusoidal muscle stretching of the afferent discharges induced by sinusoidally modutriceps surae muscles at an amplitude of 3 mm peak-to-peak lated eiectrical stimulationz7 as well as to sinusoidally and frequencies 0.1-10.0 Hz, was produced by the means of modulated’2,zs and ramp-like” transmembrane a servo-controlled Pye Ling electromagnetic puller. The current stimulation. Input-output characteristics of muscles were prestretched by 6 mm. Muscle length changes were recorded simultaneously with fusimotor spikes and fusimotoneurons have not been investigated except stored on magnetic tape. In some experiments electric for their responses to long-lasting depolarizing muscle activity (EMG) of either gastrocemius medialis or Modulations of fusimotor current injections.” gastrocnemius lateralis muscle was also recorded. Cycle histograms of fusimotor spikes were computed discharge rate during the rising phase of vibrationfrom 128 sinusoidal muscle stretches (64 for 0.1 Hz stretchinduced reflex muscle contraction’ as well as during ing frequency). Bin widths of the histograms were l-40 ms, small oscillatory changes in maintained muscle depending on the stretching frequency (i.e. cycle duration). tension levelm indicated that fusimotoneurons might Sine and cosine Fourier coefficients (u, b) of the first six also be responsive to both amplitude and velocity harmonics were computed off-line on Hewlett Packard 2817 computer and amplitudes of these harmonics (,,/m), of the input signal. In order to get more information reflecting modulation in fusimotor discharge rate, calcuon fusimotoneuron input-output characteristics, lated. In some cases power spectra were also computed, changes in discharge rate of these cells were investiusing Fast Fourier Transform, from either the cycle gated in this work during sinusoidal muscle stretchhistograms, or the records of fusimotor discharge during a ing. Stretching frequencies and amplitude are chosen sequence of sinusolidal muscle stretches of an available length, or else from poststimulus time histograms (PSTHs) so as to correspond to those applied in most of the above cited works on skeletomotoneurons.4*5,24~zg,30*4’computed during 128 IO-cycle sequences of sinusoidal stretching. The same analyses were performed on the The data obtained are also intended to serve as a basis records of spontaneous discharges of a corresponding for further studying of fusimotoneuron response length. Since the system was obviously non-linear and the characteristics using an approach appropriate for changes in fusimotor discharge rate substantiaBy distorted by higher harmonics, phase and amplitudes of fusimotor non-linear system analysis.3’ Some preliminary results have been presented.2,3 responses shown in Fig. 2 were computed from the cycle
Phase relations
EXPERIMENTAL PROCEDURES The experiments were performed on 12 adult decerebrate cats. The operative procedure before decerebration was -__.. *To whom correspondence should be addressed. Abbreviations: EMG, electric muscle activity; PSTH, poststimulus time histogram.
histograms: phase advance was calculated from the distance between either the peak discharge rate or the starting point of its increase and the maximal muscle length; response amplitude was expressed as either the difference between the peak discharge rate during the muscle stretching and the mean spontaneous discharge rate or as the ratio of this difference to the mean discharge rate. The discharge rates were calculated from the histograms by dividing the number of spikes per bin with bin width and number of cycles and expressed in imp/s.
779
780
R.
ANASTASIJEV~ef u!.
RESULTS
Changes in discharge rate of 1.5 spontaneously firing fusimotoneurons from 12 experiments during sinusoidat muscle stretching were investigated. Spontaneous discharge rate of these cells ranged from 5 to 80 imp/s and the conduction velocity of their axons IO-22 m/s. Changes in discharge rate induced by sinusolidal muscle stretching were usually barely visible and could be ascertained only after cycle histograms had been computed. The largest ones encountered are shown in Fig. 1, which represents a general picture of the recordings. Amplitudes und phases qffusimotor responses. Corn parison with skeletomotor responses
Phases of the onset and the maximum of the reflex response (increase in discharge rate) of nine fusimotoneurons (mean k SD.) with respect to the muscle length changes at different stretching frequencies are
shown in Fig. 2A. Both the onset and the maximum of the fusimotor reflex responses appeared in advance of the maximal muscle length, by about 65’ and 45’, respectively at all the stretching frequencies applied. Phase lagging of the reflex response occurred in only two fusimotoneurons during the first stretch in the sequence at 10 Hz stretching frequency and also in one of them at 5 Hz. Phase advance of EMG, measured from its onset, although rather variable from experiment to experiment, covered the same range of degrees as that of fusimotor responses. Amplitudes of the reflex responses (peak increase in discharge rate) of the same nine fusimotoneurons (mean + S.B.) to muscle stretching at different frequencies are shown in Fig. 2B. Both the absolute and relative (normalized with respect to the spontaneous firing rate) response ampIitude increased with increase in stretching frequency. Frequency of spikes in EMG activity increased while its duration decreased with muscle stretching frequency
A
8
Fig. I. Changes in discharge rate of a f~imotoneuron during sinusoidal muscle stretching. (A) Spontaneous discharge. (3) During muscle stretching. Upper traces: PSTHs computed during 128 sequences of sinusoidal muscle stretching (B) and the corresponding period of spontaneous firing (A); middle traces: spike discharges of the cell; lower traces: muscle length changes. Sinusoidal stretching 10 Hz, 3mm. Bin width 4ms. Fusimotor discharge recorded from a gastrocnemius medialis nerve fascicle, conduction velocity 14 m/s. Lower-most trace in B: EMG of the gastrocnemius medialis muscle.
781
Fusimotor responses to sinusoidal stretching
I:
0:
~~~~~~~FR~~~
5.0 STRETCHING
FREQUENCY
Fig. 2. Phase advance (A) and amplitude (B) of the fusimotor reflex response to sinusoidal muscle stretching at different frequencies (mean + S.D. for nine fusimotoneurons). (A) Dashed and continuous line: phase advance of the onset and peak increase in discharge rate, respectively. Dotted line: phase advance of EMG onset (mean f S.D. for eight cats). Abscissa: stretching frequency; ordinate: angle measured in advance from the maximum muscle length. (B) Abscissa: stretching frequency; left and right ordinates: relative and absolute increase in fusimotor discharge rate (Fm maximal discharge rate: Fs, mean spontaneous discharge rate). Inset: EMG records at 0.1, 1 and 10 Hz stretching frequency.
until at 5 and 10 Hz it turned to a short-lasting synchronous burst (Fig. 2B, insets). No attempts were made to estimate these changes in a quantitative way. Analysis of changes Harmonic distortion
in fusimotor
discharge
rate.
Most prominent changes in fusimotor discharge rate revealed by cycle histograms appeared at a frequency corresponding to the second harmonic of sinusoidal muscle stretching frequency. The highest peaks in power spectra appeared at the frequency corresponding to the second harmonic and the amplitude of this harmonic, calculated from the Fourier coefficients, was the largest. While the amplitudes of all the six harmonics increased with muscle stretching frequency, the increase in the second one was steeper than that of the others. These findings are summarized in Fig. 3. Histograms of a fusimotoneuron discharge at three different muscle stretching frequencies are shown in Fig. 3A. Thin lines drawn through the cycle histograms at 0.1 and 1 Hz are obtained by summing up the first four harmonics. At 5 Hz stretching frequency a fairly good fit to a cycle histogram was obtained in this cell by summing the first and the second harmonic. PSTH is shown in Fig. 3 instead, to make the time relations of changes in fusimotor discharge rate to those of muscle length more clearly visible. Amplitudes of the first to sixth harmonics in the response of the same fusimotoneuron at these frequencies of muscle stretching are shown in Fig. 3B
and power spectra in Fig. 3C. In some cells, the fourth harmonic was also prominent and its amplitude attained the values of the first one. Modulations appearing in spontaneous discharge rate (not shown) were smaller in all the investigated cells by an order of magnitude and showed no prominent harmonic, or else the prominent one was not the same as that during muscle stretching. The phases computed from the Fourier coefficients should not be relied on in a non-linear system. It seems obvious, however, from Fig. 3 that at least the second harmonic contributed both to the phase advance of the whole fusimotor response as well as to the steepness of the initial increase in firing rate. DISCUSSION
Phase advance and response amplitude The reflex response of fusimotoneurons to sinusoidal muscle stretching showed the expected similarities to the response of skeletomotoneurons. Phase advance of the fusimotor reflex response with respect to the muscle length changes corresponded to the phase advance of skeletomotor spikes4-6.39 and membrane potential changes24~37~4’as well as of EMG (Lippold et af.29130and these experiments). The increase in phase advance at higher stretching frequencies, present in skeletomotor reflex discharges, was not found in these experiments. Its absence may be a consequence of much longer delays in the reflex path to fusimotoneurons and the lower conduction
782
I
I
I
1
1
1
I
1
p-lOOtu--
I
1.0
0.25
0.M
.5
0.75
L 32 256
i HZ
1.0s
FREQUENCY
Fig. 3. Harmonics in fusimotor responses. (A) Cycle histograms of the discharge of a fusimotor neuron (to gastrocnemius medialis muscle, conduction velocity 14 m/s) at 0. I and 1.OHz stretching frequencies as indicated (bin width 40 and 4 ms, 64 and 128 trials, respectively). At 5.0 Hz PSTH computed during lOcycle sequence of sinusoidal stretching (bin width 10 ms, 128 trials). Thin lines, sums of the first four (at 0. I and 1 Hz) and first two (at 5 Hz) harmonics; thick lines, muscle length changes; dashed lines, mean spontaneous discharge rate. (B) Amplitude (ordinates, imp/s) of the first to sixth harmonics (abscissa) in the response of the same fusimotoneuron. Stretching frequencies as in A. (C) Power spectra of the fusimotor discharge.
velocity of their axons.* Amplitude responses responses
of the fusimotor is not directly comparable with the of skeletomotoneurons since the latter are
not spontaneously active and the number of their spikes per stretching cycle decreases with muscle stretching frequency, i.e. with cycle durationk6 It showed, however, some similarities to the modulations of skeletomotor discharge rate induced by intracellular injection of sinusoidal currents:‘*.*’ in both cases the response amplitude increased with increasing input frequency, with a change in slope at a frequency between 1 and 5 Hz. Both the phase *Central delay for excitation of fusimotoneurons by 50 pm triangular muscle stretches is shown to be 5.1-13.7 ms (1-2 ms less for larger stretches) and for skeletomotoneurons 0.7-3 ms.lh Since fusimotor spikes were recorded from muscle-nerve filaments, their conduction from the spinal cord to the recording site would take another 3.5-14 ms (for conduction velocity range 1040 m/s and the distance of 140 mm). This would make a delay of 6.6-25.7 ms. Such a delay would mean approximately 0.2”VI.9” (or less than one PSTH bin) at 0.1 Hz stretching frequency, but 23.8”-92.5” (six to 25 bins) at 10 Hz. For skeletomotor spikes recorded from ventral root filaments (or their membrane potential changes recorded intracellularly) this delay would be only about 2.5~10.8” at 10 Hz stretching frequency.
advance and the increase in the response amplitude with muscle stretching frequency indicate that the velocity of the input signal is reflected in the output of fusimotoneurons. Phase advance of skeletomotor reflex responses to sinusoidal muscle stretching compensates for the phase lag of muscle tension changes with respect to skeletomotor spikes improving, in this way, frequency characteristics of the stretch reflex.37.39Due to the phase advance of fusimotor responses, a rather synchronous alpha-gamma activation would be preserved (see Windhorst and Koehler43) and the fusimotor impulses would arrive in time to exert their effects on the muscle spindle before the maximal muscle length is attained. Harmonics in fisimotor
discharge rate
Predominant changes in fusimotor discharge rate at a frequency double that of the sinusoidal muscle stretching were not expected. Due to the large stretching amplitude applied in this work muscle spindle behaviour would be non-linear32.36 and might also be affected by its unloading by the reflex muscle contraction. This, however, could not account for the appearance of the second harmonic in changes of fusimotor discharge rate since a pause rather than a
783
Fusimotor responses to sinusoidal stretching burst of impulses would appear in muscle spindle afferent discharges during shortening of the muscle. The second harmonic appears in membrane potential changes of skeletomotoneurons at a comparable range of stretching amplitudes24 but not nearly so large as the one in fusimotor responses. Prominent second harmonic (in some cases also the fourth one) may appear if the sinusoidal input signal is fully rectified (the so-called square-law responses).2i This process, if present, could hardly be ascribed to fusimotoneurons themselves: if the properties of fusimotoneurons are similar to those of skeletomotoneurons,42 half-rectification could be expected at best.*’ Therefore, the source of the second harmonic in fusimotor responses must be looked for in reflex pathways in the spinal cord. Modulations in response amplitude of vestibular nuclear cells at a frequency double that of the input sinusoidal signal4 are supposed to appear as a consequence of afferent inputs of opposite sign converging upon them. Many feedback loops with different delays may also induce different periodic and aperiodic oscillations in responses of the target neurons.*’ Afferent impulses both of opposite sign and with different delays might converge upon fusimotoneurons in our experiments. Muscle receptors other than muscle spindle primary endings respond to sinusoidal muscle stretching.24,32,39 Discharges in some group III and IV afferent fibres might be induced by muscle stretching and/or contraction.33 Afferent impulses from muscle receptors, activated either by electrical stimulation of muscle nerves’ 9 or by muscle stretch and/or contraction’.‘E’7.34 induce both excitation and inhibition of fusimotoneurons. Latencies of the fusimotor responses are different, due to the differences in both conduction velocity of the afferent fibres involved as well as in central delays indicating anything from di- to polysynaptic transmission. Timing of arrival to fusimotoneurons of afferent impulses from different receptors, if evoked during sinusoidal muscle stretching, could hardly be predicted with enough certainty to decide whether they have induced the modulations found in fusimotor discharge rate. Recurrent inhibi-
tionlO,l4
may also induce a transient decrease in fusimotor discharge rate. It has been shown3’ that burst-like discharges appear in responses of Renshaw cells to sinusoidal muscle stretches when stretching velocity is at or above 0.5 mm/s. Since stretching velocity in our experiments would range from 0.6 to 6Omm/s the decrease in fusimotor discharge rate coincident with the peak muscle length could appear to be due to their recurrent inhibition as is supposed to be the case with the discharge of skeletomotoneurons.3s.4’ However induced, second harmonic in fusimotor discharge rate makes these cells respond as movement detectors. This information can be transmitted through muscle spindle afferents to the central nervous system. Second harmonic also contributes to the steeper rise in fusimotor discharge rate as well as to the phase advance of their responses during muscle lengthening. In this way the ensuing contraction of intrafusal muscle fibres would appear in parallel with the whole muscle reflex contraction. It might also be enhanced if the intrafusal muscle fibres share the “catch properties”‘-’ with extrafusal muscle fibres. Second increase in fusimotor discharge rate, coincident with muscle shortening, might help to keep the muscle spindles from slackening. During sinusoidal changes in muscle length occurring in natural movements such as locomotion, this would prepare extensor muscle spindles for the support phase of the step cycle. Peaks in discharge rate of fusimotoneurons, enhanced or induced by the harmonics in their responses to sinusoida muscle stretching, followed by either further increase or decrease in muscle length, may exert different after-effects on the responsiveness of muscle spindle sensory endings.l8,19.22.23 Both magnitude** and latency23 of the receptor responses to subsequent changes in muscle length and/or affected.
fusimotor
discharge
rate
could
be
Acknowledgements-This work was supported by a grant from the Serbian Medical Research Foundation. We thank M. JociC for help with computer programs.
REFERENCE 1. Anastasijevii: R. and VuEo J. (1984) Changes in fusimotor outflow during vibration-induced contraction of triceps surae muscles in decerebrate cats. E.x@ Neuroi. 85, 523-532. 2. Anastasijevib R. and Vui-o J. (1985) Parallel modulation of skeletomotor and fusimotor activity induced by muscle tension changes. Iugoslav. Physiol. Pharmac. Acta (Suppl.) 21, l-2. 3. AnastasijeviC R. and VuEo J. (1986) Fusimotor response to sinusoidal muscle stretching in decerebrated cats. Neurosci. Lert. Suppl. 26, S.567. 4. AnastasijeviC R., AnojGE M., TodoroviC. B. and
5. 6. 7. 8.
VuEo J. (1969) Effect of fusimotor stimulation on the reflex response of spinal alpha motoneurones to sinusoidal stretching of the muscle. Expl Neural. 25, 559-570. AnojGi: M., PaSiCM., TodoroviC B. and VuEo J. (1967) The reflex activity of spinal alpha motoneurones under dynamic conditions of stretching the tendon to triceps surae muscle in cat. lugoslao. Physioi. Pharmac. Acta 3, 109-121. AnojCic M., Pa&C M., Todorovii: B. and VuEo J. (1968) The reflex activity of spinal motoneurones under dynamic conditions of stretching the tendon to triceps. J. Physiot. 194, 30-33P. Appelberg B., Huiliger M., Johansson H. and Sojka P. (1983) Action on gamma motoneurones elicited by electrical stimulation of group I muscle afferent fibres in the hind limb of the cat. J. Physioi. 335, 237-253. Appelberg B., Hulliger M., Johansson H. and Sojka P. (1983) Action on gamma motone~ones elicited by electrical stimulation of group II muscle afferent fibres in the hind limb of the cat. J. Physiul. 335, 255273.
784
R. ANASTA~IJEVI~ et a/
9. Appelberg B., Hulliger M., Johansson H. and Sojka P. (1983) Action on gamma motoneurones elicited by electrical stimulation of group III muscle afferent fibres in the hind limb of the cat. J. Phvsiol. 335. 275-292. IO. Appelberg B., Hulliger M., Johansson H. and Sojka P. (1983) Recurrent actions on gamma motoneurones mediated via large and small ventral root fibres in the cat. J. Physiol. 335, 2933305. II. Baldissera F., Campadelli P. and Piccinelli L. (I(982) Neural encoding of input transients investigated by intracellular injection of ramp currents in cat a-motoneurones. J. Physiot. 3uI, 73-86.
12. Bafdissera F., Campadelli P. and Piccinelli L. (1984) The dynamic response of cat a-motoneurones inv~tjgated by intracellular injection of sinusoidal currents. Expl Brain Res. 54, 2755282. 13. Burke R. E., Rudomin P. and Zajac F. E. (1970) Catch property in single mammalian motor units. Science 168, 122-124. 14. Ellaway P. H. (1971) Recurrent inhibition of fusimotor neurones exhibiting background discharge in the decerebrate and the spinal cat. J. Physiol. 216, 419439. 15. Ellaway P. H. and Murphy P. R. (1980) Autogenetic effects of muscle contraction on extensor gamma motoneurones in the cat. Expl Brain Res. 38, 305-312. 16. Eflaway P. H. and Trott J. R. (1978) Autogenetic reflex action on the gamma motoneurones by stretch of triceps surae in the decerebrated cats. /. Physiol. 276, 49-66. 17. Ellaway P. H., Murphy P. R. and Tripathi A. (1982) Closely coupled excitation of gamma motoneurones by group III muscle afferents with low mechanical threshold in the cat. J. Physiol. 331, 481489. 18. Emonet-Denand F., Hunt C. C. and Laporte Y. (1985) Fusimotor after-effects on response of primary endings to test dynamic stimuli in cat muscle spindles. J. Physiol. 360, 187-200. 19. Emonet-Denand F.. Hunt C. C. and Laporte Y. (1985) Effects of stretch on dynamic fusimotor after-effects in cat muscle spindle. J. Physiol. 360, 201-213. 20. Glass L., Beuter A. and Laraque D. (1988) Time delays, oscillations and chaos in physiological control systems. Math. Biosci. 90, 11l-125. 21. Graham D. and McRuer D. (1961) Analysis of Nonlinear Control Systems. John Wiley, New York. 22. Gregory J. E., Morgan D. L. and Proske U. (1986) After-effects in the responses of cat muscle spindles. J. Neurophysiol. 56, 451461.
23. Gregory J. E., Morgan D. L. and Proske U. (1988) Responses of muscle spindles depend on their history of activation and movement. In Progress in Brain Research (eds Haman W. and Iggo A.), Vol. 74>pp. 85-90, Elsevier, Amsterdam, 24. JociC M., Anastasijevic R. and V&o J. (1988) Membrane potential changes of skeletomotor neurones durina sinusoidal muscle stretching. Frequency characteristics. Iugosla. Physial. Pharmac. Acta 2.4, 213-222. 2.5. JociC M., Anastasijevic R. and Vuco J. (1989) Membrane potential changes of skeletomotor neurones during sinusoidal muscle stretching. Mathematical modeling. Iugoslav. Physiol. Pharmac. Acta 25, 89994. 26. Kanda K., Burke R. and Walmsley B. (1977) Differential control of fast and slow twitch motor units in the dccerebrate - .__ _--. cat. Cxpi Brain Kes. ZY, 57-74. 27. Kostyukov A. I. (1982) Transformation of the frequency-modulated afferent activity by cat spinal motoneurons. ~e~ro~zio~ogiya 14, 198-200. 28. Kostyukov A. I. and Kry~anovsky M. V. (1982) Impulse activity evoked in the cat spinal neurons by sinusoidal depolarizing currents. Neiro~zio~og~ya 14, 3542. 29. Lippold 0. C. J., Redfearn J. W. T. and VuEo J. (1957) The rhythmical activity of groups of motor units in the voluntary contraction of muscle. J. Physiol. 137, 473-487.
30. Lippold 0. C. J., Redfearn J. W. T. and Vuco J. (1958) The effect of sinusoidal stretching upon the activity of stretch receptors in voluntary muscle and their reflex responses. J. Physiol. 144, 373-386. 31. Marmarelis P. Z. and Marmarelis V. Z. (1978) Analysis of Physiological Systems. The White Noise Approach. Plenum Press, New York. 32. Matthews P. 3. C. and Stein R. B. (1969) The sensitivity of muscle spindle afferents to small sinusoidal changes in length. J. Physic& 200, 723-743. 33. Mense S. and Meyer H. (1985) Different types of slowly conducting aRerent units in cat skeletal muscle and tendon. J. Physiol. 363, 403-417. 34. Noth J. and Thilmann A. (1980) Autogenetic excitation of extensor gamma motoneurones in the cat. Neurosci. Lett. 17, 23-26.
by group I1 muscle afferents
35. Pompeiano O., Wand P. and Sontag K. H. (1975) The sensitivity of Renshaw cells to velocity of sinusoidal stretches of the triceps surae muscle. Archs ital. Biol. 113, 280-294. 36. Poppcle R. E. and Bowman R. J. (1970) Quantitative description of linear behavior of mammalian muscle spindles. J. Neurophysiol. 33, 59-72.
37. Poppele -R: E. and Terzuolo C. A. (1968) Myotatic reflex: its input-output relation. Science 159, 743-745. 38. Post E. M.. Rymer W. Z. and Hasan Z. (1980) Relation between intrafusal and extrafusal activity in triceps surae muscles of the decerebrate cat: evidence for b action. J. Neurophysiol. 44, 383-404. 39. Rosenthal N. P., McKean T. A., Roberts W. J. and Terzuolo C. A. (1970) Frequency analysis of stretch reflex and its main subsystems in triceps surae muscles of the cat. J. ~~rophys~o~. 33, 713-749. 40. Vuco J. and Anastasijevic R. (1985) Modulations of fusimotor discharge induced by tension changes during reflex muscle contraction. In The Muscle Spindle (eds Boyd I. A. and Gladden M. H.), pp. 2855289. Macmillan, Basingstoke. 41. Westbury D. R. (1971) The response of a-motoneurones of the cat to sinusoidal movements of the muscle they innervate. Brain Res. 25, 75-86. 42. Westbury D. R. (1981) Electrophysiological characteristics of spinal gamma motoneurons in the cat. In Muscle Receptors and Movement (eds Taylor A. and Prochazka A.), pp. 87-96. Macmillan, London. 43. Windhorst U. and Koehler W. (1986) The dynamic effects of inputs to spinal motoneurones of different type upon the outputs of ~-motoneurones mediated via recurrent inhibition. Bioi. Cybern. 54, 1677177. 44. Xerri C., Barth~l~my J., Harlay F., Bore1 L. and Lacour M. (1988) Neural coding of linear motion in the vestibular nuclei of the alert cat. Expi Brain Res. 65, 5699581. (Accepted 15 November 1989)
