Effects of FM Vibration on Muscle Spindles in the Cat MUNEAKI MIZOTE Department of Physiology, School of Medicine, Chiba University, Chiba (Japan)
INTRODUCTION Constant frequency vibratory stimulation may allow us to distinguish the response of a primary ending from that of a secondary ending in muscle spindles. But the responses of nuclear bag fibers have never been divided functionally by mechanical stimulations from those of nuclear chain fibers in muscle spindles. It has been suggested that nuclear bag fibers are more viscous and nuclear chain fibers purely elastic in muscle spindles of the cat (Boyd, 1971). The mechanical properties obtained by ramp stretches of muscle spindles led t o a model of intrafusal muscle fibers (Matthews, 1964;Houk, 1966;Crowe, 1968). It is possible, therefore, that the different visco-elastic properties of the two kinds of intrafusal muscle fiber may allow us t o distinguish responses of nuclear bag fibers from those of nuclear chain fibers by using various modes of stretch. It is believed that frequency modulated ( F M ) vibration influences particularly the velocity sensitivity of intrafusal muscle fibers. The present paper shows that responses of primary endings can be divided into categories which correspond to the two types of intrafusal muscle fibers. METHODS The experiments were carried out on cats weighing 2.0-3.5 kg, which were anesthetized intraperitoneally and muscles dissected free of surrounding tissue, while keeping intact as much as possible of the blood supply. All other nerves in that hindlimb were cut except the gastrocnemius and soleus nerves. A laminectomy was performed to expose the dorsal and ventral roots between L5 and S1. They were cut at their entry into the spinal cord.'The dorsal root L7 was split into fine filaments for the isolation of single primary afferent fibers, which were identified as spindle endings by their behavior during twitch contractions of the muscle elicited by stimulating its nerve (Matthews, 1933). They were classified as primary or secondary endings on the basis of the conduction velocity of their afferent fibers. All primary endings studied had afferent fibers which conducted at over 80 m/sec. The peripheral ends of the cut L7 ventral root were subdivided into about 20 approximately equal filaments
134 which were classified as exerting a static or dynamic type of fusimotor action by the effect of stimulation of the filament at 100 Hz on the response of a spindle to different kinds of stretch. Filaments which on stimulation caused a marked increase of the dynamic index to a ramp stretch and a silent period on the releasing phase of 3 Hz sinusoidal stretch (the amplitude is less than 500 pm) were classified as having a dynamic fusimotor action, while those which on stimulation caused a decrease of the dynamic index in response t o 3 Hz sinusoidal stretch and a driving effect were classified as static fusimotor (Crowe and Matthews, 1964a, b). A primary ending was always excited strongly by gamma fusimotor fibers, in spite of the concomitant contraction of the extrafusal fibers due to excitation of alpha motor fibers on stimulating a ventral rootlet. Mechanical sinusoidal vibration was applied to the tendon of the muscle through a steel hook. Vibration of 0.5 sec duration was repeated 5 times or more every 2.0 sec (Homma et al., 1972). The initial muscle length was determined by the single shock stimulus of ventral root. Primary endings discharging spontaneously at the shortest muscle length were then neglected. The threshold of a primary ending to vibratory stimulation was obtained as the smallest amplitude of vibration which elicited only a single Ia spike and discharged a Ia spike of more than 80% on repeated stimulation periods. Two types of vibration were used, characterized by their different frequency components. One was a constant frequency of vibration (upper trace in Fig. 1)and in the other the frequency was modulated by continuously increasing frequency from one value to another (lower trace in Fig. 1).In this paper, they are termed CF vibration and FM vibration respectively. Frequencies of vibration were varied from 1 0 to 100 Hz. The amplitude excursion of vibration was detected electronically by a difference transformer in the vibrator and could be compensated for as the frequency was changed by a feedback system, to an accuracy of the order of 5% in the frequency band used. Amplitude values were indicated on an electronic meter. The amplified action potentials of the single Ia fiber and the displacement due to vibration were recorded simultaneously on separate channels of the magnetic tape recorder.
constant frequency vibration I
upward FM vibration
Fig. 1. Vibration of 0.5 sec duration was repeated 5 times every 2.0 sec. Two types of vibration (CF and FM) were used characterized by their different frequency components. In this figure the action potential of the single Ia fiber and the displacement due t o vibration are shown in the upper and lower trace respectively.
135 RESULTS Fig. 2 shows the threshold amplitude at which a constant frequency of vibration elicits a Ia spike on each occasion of steps of 10 Hz from 10 t o 100 Hz. The abscissa shows the frequency of vibration and the ordinate shows the smallest amplitude of vibration which elicits only the one spike. These frequency characteristics could be divided into two types by CF vibration over this frequency band. One shows a basically hyperbolic curve which decreases monotonically as the frequency increases. The other shows a parabolic curve which opens upward and has the lowest threshold in the middle of the band of applied frequency. In this paper, the former type of curve is labeled B and the latter labeled K. In Fig. 2, 5 primary endings are illustrated for both types of curve. The effects of FM vibration on primary endings showing curves of both type B and K were investigated in Fig. 3. An arrow shows the threshold t o FM vibration of primary endings. The direction of an arrow indicates an increasing frequency for FM vibration. In type B all the arrows cross the curve of a constant frequency vibration. On the other hand, arrows of type K never cross the curve. This means that for type K receptors tested below about 50 Hz the same value of threshold is given by the two methods of determination. In contrast, type B
Fig. 2. The abscissa shows t h e frequency of vibration and the ordinate shows the smallest amplitude of vibration which elicits only t h e one spike. These frequency characteristics were classified by CF vibration and could be divided into tw o types (type B and type K ) over this frequency band.
Fig. 3. An arrow shows the threshold to FM vibration of primary endings. The direction of a n arrow indicates an increasing frequency for FM vibration. All t h e arrows cross t h e curve of a constant frequency vibration. O n t h e other hand, in t y p e K, arrows never cross the curve.
receptors are shown as having a lower threshold when tested by CF rather than by FM stimulation. The significance of this difference between types B and K seems of potential interest. For frequencies above 50 Hz the FM method can give no measure of the rising phase of the curve determined by CF and simply indicates the lowest threshold occurring in the “trough” at about 50 Hz. After primary endings were classified as either type B or type K by CF vibration and FM vibration respectively, gamma fusimotor fibers were stimulated electrically at 100 Hz in Fig. 4.
Fig. 4. After stimulating static fusimotor fibers the response pattern of a primary ending which showed type B properties (solid line) changes into o n e showing type K properties (dotted line). In contrast, as shown on t h e right figure, type K receptors are not changed by stimulation of static gamma fusimotor fibers.
Fig. 5. After eliciting dynamic gamma fusirnotor fiber excitation, type K receptors (solid line) change into type B (dotted line). In contrast, as shown on the right figure, type B receptors remain unchanged by dynamic fusimotor fiber stimulation.
After producing static fusimotor excitation the response pattern of a primmy ending which showed type B properties changes into one showing type K properties, and thresholds are always significantly lower. In contrast, type K receptors are not changed by stimulation of static gamma fusimotor fibers, in spite of thresholds having been lowered. In Fig. 5, after eliciting dynamic gamma fusimotor fiber excitation, type K
0 1 50 'DOH z Fig. 6. This figure shows the response properties of the separate secondary endings to CF vibration and FM vibration. These response patterns are very similar to those of the type K receptors already described for primary endings. '
138 receptors change into type B and thresholds are also lowered. In contrast, type B receptors remain unchanged by dynamic fusimotor activation. Fig. 6 shows the response property of the separate secondary endings to CF vibration and FM vibration. The response patterns are very similar t o those of the type K receptors already described for primary endings. DISCUSSION The frequency-response curves of primary endings were not always divided simply into two types. The various complex types in which types B and K were mixed were observed about 30%. But in this paper two typical types are described. The threshold of type B is lower on CF vibration than on FM vibration and is lowest in the highest frequency band. Type B receptors respond effectively to the velocity of vibration. In contrast, in type K the threshold t o CF vibration almost coincides with that to FM and is lowest in the middle of the frequency band. Type K receptors respond effectively to the displacement of vibration. These results indicate that the type B receptors show viscous properties and the type K receptors elastic properties respectively. It has been reported that the nuclear bag fibers are viscous and the nuclear chain fibers are purely elastic (Boyd, 1971). Boyd and Ward (1975) reported that repetitive stimulation of a fusimotor axon produced visible contraction in the bundle of nuclear chain fibers or contraction in nuclear bag fibers, but not in both. On the gamma fusimotor fiber stimulation, it is suggested that changes of type are relative t o those of contractile intrafusal muscle fibers in their study, In the present paper, two kinds of frequency-response curve of primary endings were obtained on sinusoidal stretching (type B and K). On the other hand, Goodwin and Matthews (1971) have obtained the frequency-response curve of primary endings which increased monotonically as frequency increased from 0.1 t o 100 Hz. These differences depend upon the experimental arrangement, namely, they used a very small amplitude of stretch and a very large initial muscle length. In contrast, the author used a large amplitude and a very small muscle length. As a result, type B shows the frequency-response of the nuclear bag fibers and type K of the nuclear chain fibers. SUMMARY (1)Longitudinal C F and FM vibrations were applied t o the de-efferented gastrocnemius or soleus muscles of anesthetized cats while recording the discharge of single afferent fibers from the proprioceptors within the muscle. (2) Frequencies of vibration of 10-100 Hz were used. The maximum amplitude of vibration was 500 pm (peak t o peak) at 20 Hz. (3) The frequency-response of the primary endings t o CF and F M vibration was divided into two types, type B and type K. Type B receptors are very sensitive t o CF vibration but not so t o FM vibration, while, in contrast, type K receptors are very sensitive t o FM vibration.
139 ( 4 ) Type B properties changed into those showing type K properties after static gamma fusimotor fibers were stimulated electrically. On the other hand, type K were changed into type B by the stimulation of dynamic gamma fusimotor fibers. (5) Type B receptors show the response properties of nuclear bag fibers and type K receptors those of nuclear chain fibers.
I would like to thank Dr. P.B.C. Matthews and Dr. P.H. Ellaway for valuable advice during the preparation of this manuscript. REFERENCES Boyd, I.A. ( 1 9 7 1 ) Specific fusimotor control of nuclear bag and nuclear chain fibers in cat muscle spindles. J. Physial. (Lond.), 214: 30-31P. Boyd, I.A. and Ward, J. (1975) Motor control of nuclear bag and nuclear chain intrafusal fibers in isolated living muscle spindles from the cat. J. Physiol. (Lond.), 244: 83-112. Crowe, A . ( 1 9 6 8 ) A mechanical model of the mammalian muscle spindles. J. theoret. Biol., 2 1 : 21-41. Crowe, A and Matthews, P.B.C. (1964a) The effects of stimulation of static and dynamic f u s h o t o r fibers o n the response t o stretching of the primary endings of muscle spindles. J. Physiol. (Lond.), 1 7 4 : 109-131. Crowe, A . and Matthews, P.B.C. (1964b) Further studies of static and dynamic fusimotor fibers. J. Physiol. (Lond.), 1 7 4 : 132-151. Goodwin. G.M. and Matthews, P.B.C. ( 1 9 7 1 ) Effects of fusimotor stimulation o n the sensitivity of muscle spindle endings t o small-amplitude sinusoidal stretching. J. Physiol. (Lond.), 218: 56-58P. Homma, S., Mizote, M., Nakajima, Y. and Watanabe, S. ( 1 9 7 2 ) Muscle afferent discharges during vibratory stimulation of muscles and gamma fusimotor activities. Agressologie, 13: 45-53. Houk, J.C.( 1 9 6 6 ) A model adaptation in amphibian spindle receptors. J. theoret. Biol., 1 2 : 196-215. Matthews, B.H.C. (1933) Nerve endings in mammalian muscle. J. Physiol. (Lond.), 7 8 : 1-53, Matthews, P.B.C. ( 1 9 6 4 ) Muscle spindles and their motor control. Physiol. Rev., 4 4 : 219288.
DISCUSSION MATTHEWS: Here we have had the second paper o n sinusoidal stretching today, and you may perhaps feel that it conflicts somewhat with what I have already said. But I should emphasize that we are doing different things in different ways. Dr. Mizote is classifying his primary ehdings into t w o distinct groups o n sinusoidal stretching. I have my primary endings in one group with sinusoidal stretching. The first difference between our experimental arrangements, which I have had the opportunity of learning by being in Dr. Mizote’s laboratory, is that we are using different sizes of stretch. He is using what I call a large stretch, 400 pm, but this is an equally acceptable stretch, and so we are working in different parts of the range. That, however, is a doubtful difference. The interesting difference is that we have been working o n nearly the full length of the muscle with everything stretched tight. Dr. Mizote is working with t h e muscle very very slack. In Dr. Mizotc’s experiments, the inside
of t h e muscle spindle must look like the pictures tha t Dr. Gladden has shown us before she put the acetylcholine on, and it is quite possible that when the spindle is made so slack, then t h e branches o n one kind of intrafusal fiber are working and those o n another kind of intrafusal fiber are inactive. So it is possible that Dr. Mizote is showing differences in behavior which one seesin the very slack muscle, but which one does not see when the muscle is made tight.