Brain Research, 401 (1987) 152-154 Elsevier
The reinnervation of cat muscle spindles by skeletofusimotor axons J.J.A.
Department of Zoology, University of Durham, Science Laboratories, Durham (U.K.) (Accepted 26 August 1986) Key words: Nerve regeneration; Muscle spindle; Skeletofusimotor axon
Tests were made to ascertain the numbers of skeletofusimotor axons reinnervating muscle spindles following crush or section of the nerve to peroneus tertius in adult cats. After short periods of recovery there was no change in the proportion of skeletofusimotor axons compared with normal animals. The mammalian muscle spindle receives efferent innervation from fusimotor (V) and skeletofusimotor (/3) axons which are classified as static or dynamic according to their effects on the afferents' responses to muscle stretch. Following nerve injury both the V and fl efferents regenerate and reinnervate the spindle forming functional motor endings 2'3'6. Barker and Boddy 1 observed the presence of additional, aberrant motor connections in spindles reinnervated after nerve section and they suggested that these constituted abortive attempts by regenerating a axons to form endplates on the intrafusal fibres. Brown and Butler: also reported that there was an increase in functional fl innervation following nerve section compared with normal or after nervecrush injury. The numbers of efferents isolated, however, only represented a small sample of the total populations of the muscles. Further, if the data from the two muscles they studied (tenuissimus and peroneus longus) are separated, the figures are contradictory with peroneus longus showing an increase whereas tenuissimus showed a decrease in fl innervation (cf. their Table I). In order to examine this question in greater detail a quantitative analysis was attempted of the fl innervation of muscle spindles in peroneus tertius (PT) muscle in the cat hindlimb following nerve crush or section. Peroneus tertius was chosen because it has relatively few spindles (a mean of 12 spindle capsules
containing all three intrafusal fibre types - - bag 1, bag2 and chain 7) and the fl innervation has been examined in detail 4. Jami et al. 4 found that 31% of the a motor axons innervating extrafusal muscle fibres also had fusimotor actions and that static fl axons ~s) occurred twice as frequently as dynamic (fld) axons. The experiments were performed on 10 adult cats. Three were used as unoperated controls. In a further 3 the nerve to PT was crushed close to the point of muscle entry for 1 min between the tips of a pair of watchmaker's forceps 1. In the other four animals the nerve was cut at muscle entry with no attempt being made to reconnect the nerve ends in accordance with the procedure of Brown and Butler 2. After 6 weeks in the case of the nerve crush and 7 weeks for the nerve section animals acute experiments were performed to assess the proportion offl innervation. The hindlimb was extensively denervated except for the nerve to PT which was dissected free for a distance of 10 mm or more and passed over a recording electrode central to the injury site. The dorsal and ventral L 7 and $1 spinal roots were exposed by lumbar laminectomy. The muscle was dissected free at its distal end and attached directly to a tension transducer (Kulite) in series with an electromagnetic puller (Ling Dynamic Systems, V201). The leg was fixed rigidly at the knee and the ankle. All procedures were performed under sodium pentobarbitone anaesthesia (Nembutal, 40 mg/kg i.p., supplemented
Correspondence: J.J.A. Scott, Department of Zoology, Science Site, South Road, Durham DH1 3LE, U.K. 0006-8993/87/$03.50 © 1987 Elsevier Science Publishers B.V. (Biomedical Division)
153 as required) and, during the acute experiments, the core temperature of the animal and that of the mineral pools over the muscle and the cord were maintained at 37 °C. The L 7 and S 1 dorsal roots were subdivided to isolate single spindle afferents that gave primary-like responses to ramp-and-hold stretch of the muscles (although afferent conduction velocity could be employed to identify the afferents as Ia or II in the normal and crush-injured animals, it could not be used following nerve section since the disruption at the injury site could have led, for example, to a primary afferent innervating a secondary region or vice versa). Such afferents will be referred to as primary afferents since it is assumed that they innervated the primary regions of the spindles although they were not necessarily Ia axons. In each experiment as many as possible (up to 10) of these afferents were isolated and placed individually onto a multi-polar electrode in order to have a large proportion of PT's primary afferent population available for testing. The ventral roots were also subdivided to isolate single a motor axons. Each filament was stimulated and the orthodromic action potential was recorded from the electrode on the muscle nerve which also recorded the occurrence of any muscle action potential. Where necessary, the filament was further divided until a single all-or-nothing potential was recorded from the nerve. Special care was taken that no ~, axons were present in the filament which was stimulated supramaximally to ensure no further recruitment of efferent axons, y Fusimotor axons were characterized by their slow conduction velocities (less than 55 m/s) combined with their inability to generate a muscle action potential. Following nerve section, some of the a efferents (11%, n = 102) failed to elicit a muscle action potential and generated no contractile tension. These efferents presumably had failed to make functional connections with any extrafusal muscle fibres but they were still tested for their effects on the spindle afferents though only one was found to have any action. Overall, therefore, 89% of the sectioned a axons were successful in reinnervating extrafusal muscle fibres though the proportions varied between 76% (n = 25) and 93% (n = 29) in the individual animals. Functional ? efferents were found in all the preparations but no such axons were observed to ge-
nerate any tension. Each a axon was tested against each spindle afferent in order to ascertain the occurrence of any intrafusal activation. The procedure for identifying the/3 axons was similar to that employed by Jami et al. 4. The efferents were stimulated for a period of 1 s, initially at 150/s for 500 ms and then at 250/s for the subsequent 500 ms. A n y excitation of the afferent, especially when the stimulus frequency was stepped up, was noted. Where such excitation occurred, further tests were carried out to confirm that the efferent was a/3. The first of these was to test for the persistence of afferent excitation following the fatigue of the extrafusal component of the efferent by repeated stimulation cycles of 2 s duration at 100/s. If this test also proved positive then the muscle was stretched with a ramp-and-hold stretch and the afferent response recorded in the presence and absence of efferent stimulation at 150/s. This last procedure also enabled the identification of the fl as static or dynamic according to its effect on the afferent response s . As many as possible (up to 32) a axons were isolated and tested in each experiment (PT has an average of 35 a axons (G. Horcholle-Bossavit, L. Jami and D. Zytnicki, personal communication)). In the three normal animals 65 efferents were tested against 26 primary-ending afferents. Twenty of these (31%) were identified as having both a fusimotor and a skeletomotor action and were therefore/3 axons (Table I). This finding is identical to the results of Jami et al. 4. Of these efferents, 13 were/3s and 7 were/3d which is also in close agreement with the 2:1 ratio reported by Jami et al. 4. After nerve crush 26% of the efferents (21 of 81) were identified as/3 axons and the ratio of static to dynamic was 2:1. Following nerve section, only 23% (22 of 96) were/3 axons of which 17 were static and 5 dynamic. In all the experiments, where a/3 axon was found to have an effect on
The proportions of fl axons in normal animals (N) and after nerve crush (C) and section (S)
N C S
No, of ct's
65 81 96
31 26 23
1.8:1 2:1 3.4:1
571 820 730
5,2 4.75 5.2
154 m o r e than one afferent, the static or dynamic nature of its action was consistent for all the afferents activated. D u e to the reduction in the numbers of afferents isolated after nerve section (an average of 7.5 p e r experiment) c o m p a r e d with the o t h e r experiments (averages of 8,7 and 9.5 p e r e x p e r i m e n t in n o r m a l and nerve-crushed animals, respectively) there would have been a reduced probability of a given a having an excitatory effect on one of the afferents. To compensate for this effect the results were also expressed in terms of the p r o p o r t i o n of the actions of a skeletom o t o r axons on primary-ending afferents which were identified as being fl effects. In this case, 5.2% of the interactions in n o r m a l animals constituted fl effects c o m p a r e d with 4.75% and 5.2% following nerve crush and nerve section, respectively (Table I). In the normal animals 77% of the p r i m a r y afterents were activated by one or m o r e fl axons compared with 73% and 67% after nerve crush and section, respectively (Table If). A small p r o p o r t i o n of afferents was activated by m o r e than two fl efferents but only in one case, following nerve section, was a spindle innervated by m o r e than two of one type of ft. In this case the spindle received three fls and two fld axons. These results indicate that, even after relatively short recovery periods, there does not a p p e a r to be any increase in functional skeletofusimotor innerva-
1 Barker, D. and Boddy, A., Reinnervation of stretch receptors in cat muscle after nerve crush. In J. Taxi (Ed.), Ontogenesis and Functional Mechanisms of Peripheral Synapses, Elsevier, Amsterdam, 1980, pp. 251-263. 2 Brown, M.C. and Butler, R.G., Regeneration of afferent and efferent fibres to muscle spindles after nerve injury in adult cats, J. Physiol. (London), 260 (1976) 253-266. 3 Hyde, D. and Scott, J.J.A., Responses of cat peroneus brevis muscle-spindle afferents during recovery from nervecrush injury, J. Neurophysiol., 50 (1983) 344-357. 4 Jami, L., Murthy, K. and Petit, J., A quantitative study of
skeletofusimotor innervation in the cat peroneus tertius muscle, J. Physiol. (London), 325 (1982) 125-144. 5 Matthews, P.B.C., Mammalian Muscle Receptors and their CentralActions, Arnold, London, 1972. 6 Scott, J.J.A., Regeneration of y-fusimotor axons after nerve-freeze injury in the cat, Brain Research, 348 (1985) 159-162. 7 Scott, J.J.A. and Young, H., The number and distribution of muscle spindles and tendon organs in the peroneal muscles of the cat, J. Anat., in press.
The proportions of prima~-ending afferents activated by fl axons
N C S
No. of primaries
% Primaries activated by fl axons 0 1 2 3 4 5
No. of ct's/Expt.
26 30 30
23 27 33
23 (18-27) 27 (25-29) 25.5 (16-32)
50 33 30
19 27 27
4 10 7
4 3 0
0 0 3
tion and therefore the often bizarre connections formed by the "a invaders q after nerve section are unlikely to evoke contraction of the intrafusal muscle fibres. Following nerve section, however, there was a relative increase in the p r o p o r t i o n of fls efferents at the expense of the flds (Table I). The p r o p o r t i o n of n o r m a l spindles receiving fl innervation (77%) was higher than the 49% r e p o r t e d by Jami et al. 4. In their experiments, though, they only isolated an average of 11.4 ( 9 - 1 5 ) a axons for testing in each experiment c o m p a r e d with 23 (18-27) in the present study. In both series, however, the proportion of a axons having fusimotor actions was found to be 31%. S u p p o r t e d by the D u r h a m University R e s e a r c h Foundation. My thanks to Mrs. H. Young and Mrs. M. E d g e for technical assistance and to M m e L. Jami and Prof. D. B a r k e r for their helpful criticisms.