Acta physiol. scand. 1978. 103. 81-91 From the Institute of Physiology, University of Oslo, Norway
Reinnervation of partially denervated rat soleus muscle BY
WESLEY THOMPSON Received 25 November 1977
-
Abstract
THOMPSON, W. Reinnervation of partially denervated rat soleus muscle. Acta physiol. scand. 1978. 103. 81-91. The reinnervation of partially denervated rat soleus muscles by their interrupted, regenerating motor axons has been examined in adult white rats. If reinnervation occurred after the remaining, intact motor axons had sprouted to their full, maximal extent, then the regenerating axons formed synapses preferentially with denervated muscle fibers and not with fibers innervated by sprouts. The sprouted motor units retained their size as if no reinnervation had occurred. On the other hand, if reinnervation occurred early during the sprouting process, the sprouting motor units were never able to attain their maximalsize. Further, some muscle fibers became innervated by both sprouted and regenerating axons. These “hyperinnervated” fibers lost their dual innervation within a few weeks. The sprouted axons seemed to be the nerve fibers preferentially eliminated from these hyperinnervated fibers, since during the loss of hyperinnervation the sprouted motor units decreased in size while the motor units formed by the regenerating axons did not change in size. It is proposed that the occurrence of hyperinnervation is influenced by the amount of time sprouting axons have to consolidate their synapses with muscle fibers. Further, it is proposed that on muscle fibers which can become hyperinnervated, the sprouted motor neurons are at a disadvantage in the competition for maintenance because of their larger unit sizes.
The number of synapses which a motor neuron forms within its target muscle is limited. In the soleus muscle of the adult rat this limitation is set by a constant or nearly constant number of muscle fibers, a narrowly defined number of motor neurons, and a requirement that each muscle fiber receive innervation from only one neuron. A result of these limitations and an incompletely understood post-natal development (Redfern 1970, Brown el al. 1976) is that the soleus motor neurons divide the soleus muscle fibers among themselves so that each neuron innervates about the same number of muscle fibers, i.e. the motor units are of roughly uniform size. Adult soleus motor neurons can, however, be induced to increase the number of muscle fibers they innervate if the muscle is partially denervated, i.e. if the axons of a portion of the population of the motor neurons are interrupted (Guth 1962, Thompson and Jansen 1977). This expansion in motor unit size constitutes the well-known phenomenon of collateral sprouting which has been shown to occur in many central and peripheral neurons in response to partial denervation (Edds 1953, Raisman and Field 1973). In the case of the adult soleus 6 - 785875
81
82
WESLEY THOMPSON
muscle, partial denervation causes motor units to increase their size to, on the average, about 3-4 times their normal size (Thompson and Jansen 1977). This increase in size seems to persist indefinitely. at least as long as the axons interrupted upon partial denervation d o not reinnervate the muscle. Guth (1962) has reported that when interrupted axons return to partially denervated rat soleus or plantaris muscles many of the muscle fibers already innervated by the sprouted motor neurons also came to be innervated (“hyperinnervated”) by the regenerating axons. This observation is interesting sinc: it means that muscle fibers which are innervated and active accept additional innervation. Normally such muscle fibers are refractory to hyperinnervation (Elsberg 1917). I have reexamined the problem of reinnervation of partially denervated rat soleus muscle and confirm that hyperinnervation does occur, although to a limited extent. Furthermore, the hyperinnervation is transient. Many of the muscle fibers lose their innervation from the sprouted nerve and remain innervated only by the regenerating nerve. Independently, Brown and lronton (1977) and Betz (1977) have obtained similar results from other muscles.
Materials and Methods Animals used in these experiments were a special strain of albino rats ( A 0 strain, M R C Cellular Immunology Unit. Oxford) whose soleus muscles possess an unusual pattern of innervation (Thompson and Jansen 1977). I n addition to the normal soleus nerve which is a branch of the tibial nerve, the soleus muscles of these rats frequently receive innervation via a second, smaller nerve, the so-called “aberrant nerve”, which arises as a branch of the plantar nerve. The earlier experiments have established that this aberrant nerve possesses a variable number of the normal complement of soleus motor axons which reach the muscle via this unusual path instead of through the soleus nerve. Soleus muscles in these animals were partially denervated by either resecting, cutting, or crushing the lateral gastrocnemius-soleus nerve at the point where it branches from the tibial nerve in the thigh. These operations were conducted under ether anesthesia. The separate methods of partially denervating the muscles were selected because they provide for different nerve regeneration times (Frank rt a / . 1975). In the first method, the nerbe was cut with a pair of dissecting scissors. In the second method, the nerve was crushed seceral times between the smooth tips of Dumont number 5 watchmaker’s forceps. The third method consisted of resecting a 1 mm piece of the nerve and ligating the proximal stump with 9-0 surgical thread. Wounds were sutured closed and a local anesthetic and penicillin ( I S 0 000 unitsikg. i m . ) administered. For the acute experiments, the muscles and their nerves were dissected under a microscope in oxygenated Ringer’s solution (composition in m M : NaCI. 137; KCI, 5 ; CaCI,. 10: MgCI,, I ; glucose 10; TrisCI, 10, buffered to pH 7.4). After dissection the muscle/nerve preparation was removed to a chamber through which oxygenated Ringer‘s was perfused (rate ca. I mlimin). Isometric tensions were measured by attaching the distal tendon by 6-0 silk suture to a sensitive strain gauge (0.15 V/g). The length of the muscle was adjusted to obtain the maximum twitch tension. Nerves were stimulated through suction electrodes. Muscles were directly stimulated by I msec duration pulses delivered through silver electrodes o n each side of each musclc. Evoked synaptic potentials were measured in curarized muscles (2-6 lo-‘ giml D-tubocurarine) by intraccllular micropipettes (20-50 M R . potassium acetate filled). After physiological examination, most of the partially denervated muscles were prepared for histology by fixation in 2 ” , paraf0rmaldehyde-2.5~,, glutaraldehyde and post-fixation in 1 96 osmium tetroxide. The muscles were embedded in plastic and 3 p thick cross-sections were made through the central region and stained with toluidine blue. The number of muscle fibers in the muscle cross-sections was counted o n photographic montages. -2
Results The result of the reinnervation of partially denervated soleus muscles differed according to the time of arrival at the muscle of the regenerating motor axons.
83
REINNERVATION OF MUSCLE
Fig. 1. Summation of twitches from the sprouted nerve and the regenerated nerve in a muscle partially denervated by soleus nerve cut 133 days earlier. Superimposed traces of the twitch to stimulation of the soleus nerve alone (sol.), to the aberrant nerve alone (ah.) and to stimulation of both nerves simultaneously (ah. t sol.). The timing of the stimuli was adjusted so as to give the minimal twitch. The aberrant nerve to this muscle contained 3 motor units.
-
&- [ ab. +sol.
59
100ms
Muscles partially denervated by nerve cuts
When partial denervation was achieved by cutting the soleus nerve, reinnervation of the soleus muscle was first observed at day 12 after partial denervation. Six rats were examined 16 to 133 days after such partial denervation in order to determine the extent of hyperinnervation and the sizes of motor units. The degree of hyperinnervation in these muscles was determined by two procedures. First, the summation of twitch tensions to the aberrant and the regenerated soleus nerve was measured. A typical result of this type of measurement is illustrated in Fig. 1. Here it can be seen that the tension generated by the two nerves stimulated together is approximately equal to the sum of the tensions when they were stimulated separately. Such a result was the case for all 6 muscles examined (Table I), and this result suggests that the aberrant nerve and the soleus nerve innervate different populations of muscle fibers in the muscle. A more accurate way to estimate the degree of hyperinnervation in these muscles is to sample the innervation of individual muscle fibers by microelectrode recording. This method also has the advantage that it will reveal the existence of any innervation which might be too weak to elicit contraction in the muscle. The end-plate potentials (epp’s) from the sprouting and regenerating axons were found only near the center of each muscle fiber, at the site of the old denervated end-plates. With a few exceptions, muscle fibers gave epp’s only to one of the nerves (cf. Table I). Only 2 of 92 fibers in 5 muscles had epp’s to stimulation of both the aberrant and the soleus nerves. Hence, both tension measurements and intracellular recordings indicate that the regenerated and the sprouted nerves innervated, with rare excepTABLE I. Hyperinnervation of muscles partially denervated by nerve cut. Surviva1 age (days)
16 22 32 31 49 133
Tension measurements Aberrant N.
Soleus N.
Tension ( 9)
Number of units
Tension (g)
1.1 7.6 16.6 8.1 18.2 7.5
ca. 10 5 I1 3 6 3
1.1 0.6 1.3 1.2 2.1 2.1
Epp recordings
-
Ah N.+ Sol. N.
% convergence
No. of fibers with sol.n.epp.
No. of these also with ab.n.epp.
o,,
7.7 1.9 11.4 9.2 19.1 10.0
6.5 3.8 2.9 1.6 6.3 2.0
30 18 12 12
1 1 0
3.3 5.6 0
0
0
-
-
-
20
0
0
Number of units 10
3 3 4
> I 10
convergence
84
WESLEY THOMPSON
A.
Sprouted Nerve Units
B.
Regenerated Nerve Units
n
I Fig. 3. Tuitches of sprouted (aberrant) nerve motor units (part A) and regenerated (soleus) nerve motor units (part B). The muscle was partially denervated by crush of the soleus nerve 71 days earlier. The aberrant and the soleus nerve were in turn stimulated with shocks of carefully graded intensity while measuring the isometric tension generated by the muscle. Each all-or-nothing increment in the twitch tension produced by the activation of a motor axon in the nerve was recorded on the face of a storage oscilloscope. The stimulus intensity was adjusted several times to ensure reproducibility of the responses. Part A shows the 4 units in the aherrant nerve. Part B shows the 5 units of lowest threshold in the soleus nerve.
tions. separate fibers in these muscles. It should also be noted that the two doubly innervated fibers were found in the animals with the shortest survival times. To determine whether reinnervation changed the sizes of the sprouted motor units, we examined the motor unit sizes of aberrant motor axons in these partially denervated muscles with soleus reinnervation and compared them with the aberrant motor unit sizes in partially denervated muscles without soleus reinnervation. Motor unit sizes were determined using the procedure described previously (Thompson and Jansen 1977). Fig. 2 provides a n illustration of this procedure for a n adult muscle partially denervated 71 days previously. Fig. 2 A shows the tension increments for the 4 units present in the aberrant nerve; 2 B shows the tension increments for the 5 units of lowest threshold in the soleus nerve which had regenerated to
z
-2 0
A.
With Soleus Nerve Reinnervation 12 S.D. = 517t217 (n-23)
6.
Without Soleus Nerve Reinnervation x?S.D.=524 192 (n-29)
200
400
600
a00
'
1000
Number of Musclefibers Fig. 3. Sprouted nerve motor unit sizes with ( A ) and without (B) soleus nerve reinnervation. Part A shows 2.3 aberrant nerve motor units obtained from 6 muscles 16-133 days after partial denervation by soleus nerve cut. In each of these 6 muscles, there was extensike soleus nerve reinnervation. Part B shows 29 aberrant n e n e motor units obtained from 8 muscles 19-67 days after partial denervation. I n these muscles partial denervation was accomplished by resection and ligation of the soleus nerve which did not reinnervate t h e muscle.
85
REINNERVATION OF MUSCLE
TABLE 11. Hyperinnervation of muscles partially denervated by nerve crush. Sur-
Tension measurements
viva1 age (days)
Ab Nerve
Sol Nerve
Tension No. (g) of units
Tension No. (g) of units
5.9 4.3 2.6 4.2 8.2 1.4 4.6 12.1 8.3 6.4
1.9 3.4 5.8 2.4 12.1 13.2 17.9 10.7 12.5 11.4
9 9 10 10 37 38 40 48 70 71
9 5 2 7 6 1 2 8 6 4
Action potentials
>14 >I9 >I9 >16 >20 >26 >20 >I9 >2 0 >20
End-plate potentials
Ab + Sol % con- Sol N Sol & % con- Sol N. Sol b: contension vergAb vergN + verg(g) ence ence Ab.N. ence
6.7 6.6 7.1 5.4 17.5 13.9 20.0 18.6 16.8 15.4
16 17 18 22 16 5 12 23 24 16
-
-
-
20 32 11
3 6 2
I5 19 18
-
-
21 20
1 1
-
-
47 26
0 1
-
18 18 20 28 27
5 3 3 2 0
28 17 15 7 0
5
-
_
_
5
0
-
18 33
0 4
11
10
0 0
0 3 0 0
1
the muscle. In addition to the procedure shown in Fig. 2, the muscle was also stimulated directly by currents passed between two electrodes on each side of the muscle in order to determine the maximal twitch the muscle was capable of generating. The ratio of each increment in twitch tension to the maximal direct tension is taken to be a measure of the fraction of the muscle fibers in the muscle innervated by each motor axon. Each fraction was then converted into a number of muscle fibers by multiplying by the number of muscle fibers counted in a cross-section of the muscle. The results of these unit size measurements for the muscles partially denervated by lateral gastrocnemius-soleus nerve cut are shown in the histogram of Fig. 3. The units from the 6 muscles had an average unit size of 517 fibers. For comparison, the results are also shown from 8 muscles in which there was no soleus nerve reinnervation 19-67 days after partial denervation (reinnervation was prevented in these muscles by resecting a portion of the nerve and ligating the proximal stump). Clearly there is no discernible difference in the unit sizes. Hence, the sprouted aberrant motor units are not appreciably decreased in their expanded size upon reinnervation by the soleus nerve. The regenerating axons seem to have preferentially innervated muscle fibers which had not been claimed by the sprouting axons. Muscles partially deriervated by nerve crushes
In a second group of rats partial denervation of the soleus was achieved by crushing the lateral gastrocnemius-soleus nerve with forceps. This procedure allowed for more rapid reinnervation by the interrupted soleus nerve axons. Strong reinnervation was noted at 9 and 10 days post-partial denervation in 4 of these animals; weak reinnervation was seen in one rat examined at 7 days. Twitch tensions were measured in these muscles and once again the degree of convergence was only moderate. The values are given in Table 11, and they were between 5 and 25%. However, when action potentials and end-plate potentials were recorded in these muscles, a considerable hyperinnervation was detected at early times. In 4 curarized muscles examined
WESLEY THOMPSON
aberrant nerve soieus nerve
smv t----r
Fig. 4. lntracellular recording from a muscle fiber innervated by both the sprouted (aberrant) nerve and the regenerated (soleus) nerve. The fiber was in a muscle partially denervated by soleus nerve crush 9 days earlier. D-tubocurarine 2 lo-’ g/ml.
lOms
9 and 10 days after partial denervation, from 7 to 28”, of the muscle fibers which gave a n epp to the aberrant nerve also gave an epp to stimulation of the regenerated soleus nerve (Table 11). An example of one such recording can be seen in Fig. 4.In contrast only one such doubly innervated fiber was found in muscles examined after long survival times. Comparable results were obtained by examining the frequency of action potentials elicited by stimulation of the two nerves in non-curarized muscles (Table 11). Thus, it appears that at early times after partial denervation a moderate number of muscle fibers accept innervation from both nerles and that a later times most of this hyperinnervation disappears. There is not a similar trend for hyperinnervation as measured by muscle twitches (Table 11). This could have a number of explanations. The intracellular recording measurements give estimates of hyperinnervation only in that portion of each muscle reinnervated by the soleus motor neurons. This portion on the average was only of the total tension in the 9 and 10 day survival muscles (Table 11). Assuming a 20°, hyperinnervation in this portion of the muscle, only I !I0 or so of the total muscle would be hyperinnervated. This small degree of hyperinnervation may well be beyond our limits of detection in tension occlusion measurements, especially since we normally measure a small degree (up to 1406) of “apparent” convergence in normal adult muscles with aberrant and soleus nerves where in fact no convergence exists (Thompson and Jansen 1977). This inaccuracy in our measurement technique is apparently due to a lack of pure isometric conditions. In addition to determining whether muscle fibers became innervated by both aberrant and soleus nerve fibers. we also determined by microelectrode recording and grading the stimuli to each nerve independently whether multiple axons in each made synapses onto the individual muscle fibers. Of 216 muscle fibers giving a n epp to the aberrant nerve, none were polyneuronally innervated by the aberrant nerve. Hence, as previously reported (Thompson and Jansen 1977), the aberrant nerve fibers form single synaptic contacts upon sprouting. Of 330 muscle fibers giving an epp to the regenerating soleus nerve, only 14 fibers were polyneuronally innervated by the soleus nerve fibers. Even at the earliest times after soleus nerve regeneration, only 4 of 93 fibers examined were found to be polyneuronally innervated. Hence, in contrast to the reinnervation of neonatal soleus muscle (Brown et nl. 1976) and the reinnervation of an adult fast muscle of the rat ( McArdle 1975). polyneuronal reinnervation of the soleus muscle is rare. The sizes of the motor units in the muscles partially denervated by nerve crush were determined as described above and the results are presented in Fig. 5 . The material has been divided according to the survival time following partial denervation. The sprouted motor units after nerve crush (Fig. 5 A), average size 315 fibers, did not become as large as those in
87
REINNERVATION OF MUSCLE A.
R
Sprouted Nmrrc Motor Unih
rtsa-9835 ("-15)
ntS.D.-3Yl~111 (n-23) , "
RegasraledNwve M o h Unih
5
3
2 0 37-71danoftercruh
Number gf MurcleFiberr
2 I5l
37-71 donoftsr crush
I.lumb*i of hluscleFibers
Fig. 5. Sizes of sprouted and regenerated motor units in muscles partially denervated by crush of the soleus nerve. The data are divided into two histograms according to survival time after nerve crush: top histograms in both A & B give the motor unit sizes for 4 muscles examined 9-10 days after nerve crush; the lower histograms give the motor unit sizes for 6 muscles examined 37-71 days after nerve crush. Motor unit sizes for the data presented here were computed using 3 400 fibers for the normal number present in such muscles. In muscle fiber counts made of some of the muscles, an average of 3 340 was obtained.
the muscles where reinnervation by the soleus nerve was either prevented or delayed about one week longer (averages of 524 and 517 fibers, respectively; see Fig. 3). Apparently, if regenerating nerve fibers reach the muscle soon after partial denervation they are able to compete with the sprouting nerve fibers for innervation of the available, denervated muscle fibers. However, the regenerating soleus nerve motor units were never able to regain their normal size. The soleus motor units after regeneration through the crush (a +S.D. -94 +. 32; Fig. 5 B) are significantly smaller (Wilcoxon, P ;.O.OOl) than the normal soleus motor units X = 150, Thompson and Jansen 1977). Further, the measurements of the sizes of the sprouted motor units indicate that these units may be decreasing in size with time after partial denervation. The sprouted motor units are larger at 9-10 days after partial denervation than they are at later times (37-71 days). This difference is statistically significant (Wilcoxon, p
Reinnervation of partially denervated rat soleus muscle BY
WESLEY THOMPSON Received 25 November 1977
-
Abstract
THOMPSON, W. Reinnervation of partially denervated rat soleus muscle. Acta physiol. scand. 1978. 103. 81-91. The reinnervation of partially denervated rat soleus muscles by their interrupted, regenerating motor axons has been examined in adult white rats. If reinnervation occurred after the remaining, intact motor axons had sprouted to their full, maximal extent, then the regenerating axons formed synapses preferentially with denervated muscle fibers and not with fibers innervated by sprouts. The sprouted motor units retained their size as if no reinnervation had occurred. On the other hand, if reinnervation occurred early during the sprouting process, the sprouting motor units were never able to attain their maximalsize. Further, some muscle fibers became innervated by both sprouted and regenerating axons. These “hyperinnervated” fibers lost their dual innervation within a few weeks. The sprouted axons seemed to be the nerve fibers preferentially eliminated from these hyperinnervated fibers, since during the loss of hyperinnervation the sprouted motor units decreased in size while the motor units formed by the regenerating axons did not change in size. It is proposed that the occurrence of hyperinnervation is influenced by the amount of time sprouting axons have to consolidate their synapses with muscle fibers. Further, it is proposed that on muscle fibers which can become hyperinnervated, the sprouted motor neurons are at a disadvantage in the competition for maintenance because of their larger unit sizes.
The number of synapses which a motor neuron forms within its target muscle is limited. In the soleus muscle of the adult rat this limitation is set by a constant or nearly constant number of muscle fibers, a narrowly defined number of motor neurons, and a requirement that each muscle fiber receive innervation from only one neuron. A result of these limitations and an incompletely understood post-natal development (Redfern 1970, Brown el al. 1976) is that the soleus motor neurons divide the soleus muscle fibers among themselves so that each neuron innervates about the same number of muscle fibers, i.e. the motor units are of roughly uniform size. Adult soleus motor neurons can, however, be induced to increase the number of muscle fibers they innervate if the muscle is partially denervated, i.e. if the axons of a portion of the population of the motor neurons are interrupted (Guth 1962, Thompson and Jansen 1977). This expansion in motor unit size constitutes the well-known phenomenon of collateral sprouting which has been shown to occur in many central and peripheral neurons in response to partial denervation (Edds 1953, Raisman and Field 1973). In the case of the adult soleus 6 - 785875
81
82
WESLEY THOMPSON
muscle, partial denervation causes motor units to increase their size to, on the average, about 3-4 times their normal size (Thompson and Jansen 1977). This increase in size seems to persist indefinitely. at least as long as the axons interrupted upon partial denervation d o not reinnervate the muscle. Guth (1962) has reported that when interrupted axons return to partially denervated rat soleus or plantaris muscles many of the muscle fibers already innervated by the sprouted motor neurons also came to be innervated (“hyperinnervated”) by the regenerating axons. This observation is interesting sinc: it means that muscle fibers which are innervated and active accept additional innervation. Normally such muscle fibers are refractory to hyperinnervation (Elsberg 1917). I have reexamined the problem of reinnervation of partially denervated rat soleus muscle and confirm that hyperinnervation does occur, although to a limited extent. Furthermore, the hyperinnervation is transient. Many of the muscle fibers lose their innervation from the sprouted nerve and remain innervated only by the regenerating nerve. Independently, Brown and lronton (1977) and Betz (1977) have obtained similar results from other muscles.
Materials and Methods Animals used in these experiments were a special strain of albino rats ( A 0 strain, M R C Cellular Immunology Unit. Oxford) whose soleus muscles possess an unusual pattern of innervation (Thompson and Jansen 1977). I n addition to the normal soleus nerve which is a branch of the tibial nerve, the soleus muscles of these rats frequently receive innervation via a second, smaller nerve, the so-called “aberrant nerve”, which arises as a branch of the plantar nerve. The earlier experiments have established that this aberrant nerve possesses a variable number of the normal complement of soleus motor axons which reach the muscle via this unusual path instead of through the soleus nerve. Soleus muscles in these animals were partially denervated by either resecting, cutting, or crushing the lateral gastrocnemius-soleus nerve at the point where it branches from the tibial nerve in the thigh. These operations were conducted under ether anesthesia. The separate methods of partially denervating the muscles were selected because they provide for different nerve regeneration times (Frank rt a / . 1975). In the first method, the nerbe was cut with a pair of dissecting scissors. In the second method, the nerve was crushed seceral times between the smooth tips of Dumont number 5 watchmaker’s forceps. The third method consisted of resecting a 1 mm piece of the nerve and ligating the proximal stump with 9-0 surgical thread. Wounds were sutured closed and a local anesthetic and penicillin ( I S 0 000 unitsikg. i m . ) administered. For the acute experiments, the muscles and their nerves were dissected under a microscope in oxygenated Ringer’s solution (composition in m M : NaCI. 137; KCI, 5 ; CaCI,. 10: MgCI,, I ; glucose 10; TrisCI, 10, buffered to pH 7.4). After dissection the muscle/nerve preparation was removed to a chamber through which oxygenated Ringer‘s was perfused (rate ca. I mlimin). Isometric tensions were measured by attaching the distal tendon by 6-0 silk suture to a sensitive strain gauge (0.15 V/g). The length of the muscle was adjusted to obtain the maximum twitch tension. Nerves were stimulated through suction electrodes. Muscles were directly stimulated by I msec duration pulses delivered through silver electrodes o n each side of each musclc. Evoked synaptic potentials were measured in curarized muscles (2-6 lo-‘ giml D-tubocurarine) by intraccllular micropipettes (20-50 M R . potassium acetate filled). After physiological examination, most of the partially denervated muscles were prepared for histology by fixation in 2 ” , paraf0rmaldehyde-2.5~,, glutaraldehyde and post-fixation in 1 96 osmium tetroxide. The muscles were embedded in plastic and 3 p thick cross-sections were made through the central region and stained with toluidine blue. The number of muscle fibers in the muscle cross-sections was counted o n photographic montages. -2
Results The result of the reinnervation of partially denervated soleus muscles differed according to the time of arrival at the muscle of the regenerating motor axons.
83
REINNERVATION OF MUSCLE
Fig. 1. Summation of twitches from the sprouted nerve and the regenerated nerve in a muscle partially denervated by soleus nerve cut 133 days earlier. Superimposed traces of the twitch to stimulation of the soleus nerve alone (sol.), to the aberrant nerve alone (ah.) and to stimulation of both nerves simultaneously (ah. t sol.). The timing of the stimuli was adjusted so as to give the minimal twitch. The aberrant nerve to this muscle contained 3 motor units.
-
&- [ ab. +sol.
59
100ms
Muscles partially denervated by nerve cuts
When partial denervation was achieved by cutting the soleus nerve, reinnervation of the soleus muscle was first observed at day 12 after partial denervation. Six rats were examined 16 to 133 days after such partial denervation in order to determine the extent of hyperinnervation and the sizes of motor units. The degree of hyperinnervation in these muscles was determined by two procedures. First, the summation of twitch tensions to the aberrant and the regenerated soleus nerve was measured. A typical result of this type of measurement is illustrated in Fig. 1. Here it can be seen that the tension generated by the two nerves stimulated together is approximately equal to the sum of the tensions when they were stimulated separately. Such a result was the case for all 6 muscles examined (Table I), and this result suggests that the aberrant nerve and the soleus nerve innervate different populations of muscle fibers in the muscle. A more accurate way to estimate the degree of hyperinnervation in these muscles is to sample the innervation of individual muscle fibers by microelectrode recording. This method also has the advantage that it will reveal the existence of any innervation which might be too weak to elicit contraction in the muscle. The end-plate potentials (epp’s) from the sprouting and regenerating axons were found only near the center of each muscle fiber, at the site of the old denervated end-plates. With a few exceptions, muscle fibers gave epp’s only to one of the nerves (cf. Table I). Only 2 of 92 fibers in 5 muscles had epp’s to stimulation of both the aberrant and the soleus nerves. Hence, both tension measurements and intracellular recordings indicate that the regenerated and the sprouted nerves innervated, with rare excepTABLE I. Hyperinnervation of muscles partially denervated by nerve cut. Surviva1 age (days)
16 22 32 31 49 133
Tension measurements Aberrant N.
Soleus N.
Tension ( 9)
Number of units
Tension (g)
1.1 7.6 16.6 8.1 18.2 7.5
ca. 10 5 I1 3 6 3
1.1 0.6 1.3 1.2 2.1 2.1
Epp recordings
-
Ah N.+ Sol. N.
% convergence
No. of fibers with sol.n.epp.
No. of these also with ab.n.epp.
o,,
7.7 1.9 11.4 9.2 19.1 10.0
6.5 3.8 2.9 1.6 6.3 2.0
30 18 12 12
1 1 0
3.3 5.6 0
0
0
-
-
-
20
0
0
Number of units 10
3 3 4
> I 10
convergence
84
WESLEY THOMPSON
A.
Sprouted Nerve Units
B.
Regenerated Nerve Units
n
I Fig. 3. Tuitches of sprouted (aberrant) nerve motor units (part A) and regenerated (soleus) nerve motor units (part B). The muscle was partially denervated by crush of the soleus nerve 71 days earlier. The aberrant and the soleus nerve were in turn stimulated with shocks of carefully graded intensity while measuring the isometric tension generated by the muscle. Each all-or-nothing increment in the twitch tension produced by the activation of a motor axon in the nerve was recorded on the face of a storage oscilloscope. The stimulus intensity was adjusted several times to ensure reproducibility of the responses. Part A shows the 4 units in the aherrant nerve. Part B shows the 5 units of lowest threshold in the soleus nerve.
tions. separate fibers in these muscles. It should also be noted that the two doubly innervated fibers were found in the animals with the shortest survival times. To determine whether reinnervation changed the sizes of the sprouted motor units, we examined the motor unit sizes of aberrant motor axons in these partially denervated muscles with soleus reinnervation and compared them with the aberrant motor unit sizes in partially denervated muscles without soleus reinnervation. Motor unit sizes were determined using the procedure described previously (Thompson and Jansen 1977). Fig. 2 provides a n illustration of this procedure for a n adult muscle partially denervated 71 days previously. Fig. 2 A shows the tension increments for the 4 units present in the aberrant nerve; 2 B shows the tension increments for the 5 units of lowest threshold in the soleus nerve which had regenerated to
z
-2 0
A.
With Soleus Nerve Reinnervation 12 S.D. = 517t217 (n-23)
6.
Without Soleus Nerve Reinnervation x?S.D.=524 192 (n-29)
200
400
600
a00
'
1000
Number of Musclefibers Fig. 3. Sprouted nerve motor unit sizes with ( A ) and without (B) soleus nerve reinnervation. Part A shows 2.3 aberrant nerve motor units obtained from 6 muscles 16-133 days after partial denervation by soleus nerve cut. In each of these 6 muscles, there was extensike soleus nerve reinnervation. Part B shows 29 aberrant n e n e motor units obtained from 8 muscles 19-67 days after partial denervation. I n these muscles partial denervation was accomplished by resection and ligation of the soleus nerve which did not reinnervate t h e muscle.
85
REINNERVATION OF MUSCLE
TABLE 11. Hyperinnervation of muscles partially denervated by nerve crush. Sur-
Tension measurements
viva1 age (days)
Ab Nerve
Sol Nerve
Tension No. (g) of units
Tension No. (g) of units
5.9 4.3 2.6 4.2 8.2 1.4 4.6 12.1 8.3 6.4
1.9 3.4 5.8 2.4 12.1 13.2 17.9 10.7 12.5 11.4
9 9 10 10 37 38 40 48 70 71
9 5 2 7 6 1 2 8 6 4
Action potentials
>14 >I9 >I9 >16 >20 >26 >20 >I9 >2 0 >20
End-plate potentials
Ab + Sol % con- Sol N Sol & % con- Sol N. Sol b: contension vergAb vergN + verg(g) ence ence Ab.N. ence
6.7 6.6 7.1 5.4 17.5 13.9 20.0 18.6 16.8 15.4
16 17 18 22 16 5 12 23 24 16
-
-
-
20 32 11
3 6 2
I5 19 18
-
-
21 20
1 1
-
-
47 26
0 1
-
18 18 20 28 27
5 3 3 2 0
28 17 15 7 0
5
-
_
_
5
0
-
18 33
0 4
11
10
0 0
0 3 0 0
1
the muscle. In addition to the procedure shown in Fig. 2, the muscle was also stimulated directly by currents passed between two electrodes on each side of the muscle in order to determine the maximal twitch the muscle was capable of generating. The ratio of each increment in twitch tension to the maximal direct tension is taken to be a measure of the fraction of the muscle fibers in the muscle innervated by each motor axon. Each fraction was then converted into a number of muscle fibers by multiplying by the number of muscle fibers counted in a cross-section of the muscle. The results of these unit size measurements for the muscles partially denervated by lateral gastrocnemius-soleus nerve cut are shown in the histogram of Fig. 3. The units from the 6 muscles had an average unit size of 517 fibers. For comparison, the results are also shown from 8 muscles in which there was no soleus nerve reinnervation 19-67 days after partial denervation (reinnervation was prevented in these muscles by resecting a portion of the nerve and ligating the proximal stump). Clearly there is no discernible difference in the unit sizes. Hence, the sprouted aberrant motor units are not appreciably decreased in their expanded size upon reinnervation by the soleus nerve. The regenerating axons seem to have preferentially innervated muscle fibers which had not been claimed by the sprouting axons. Muscles partially deriervated by nerve crushes
In a second group of rats partial denervation of the soleus was achieved by crushing the lateral gastrocnemius-soleus nerve with forceps. This procedure allowed for more rapid reinnervation by the interrupted soleus nerve axons. Strong reinnervation was noted at 9 and 10 days post-partial denervation in 4 of these animals; weak reinnervation was seen in one rat examined at 7 days. Twitch tensions were measured in these muscles and once again the degree of convergence was only moderate. The values are given in Table 11, and they were between 5 and 25%. However, when action potentials and end-plate potentials were recorded in these muscles, a considerable hyperinnervation was detected at early times. In 4 curarized muscles examined
WESLEY THOMPSON
aberrant nerve soieus nerve
smv t----r
Fig. 4. lntracellular recording from a muscle fiber innervated by both the sprouted (aberrant) nerve and the regenerated (soleus) nerve. The fiber was in a muscle partially denervated by soleus nerve crush 9 days earlier. D-tubocurarine 2 lo-’ g/ml.
lOms
9 and 10 days after partial denervation, from 7 to 28”, of the muscle fibers which gave a n epp to the aberrant nerve also gave an epp to stimulation of the regenerated soleus nerve (Table 11). An example of one such recording can be seen in Fig. 4.In contrast only one such doubly innervated fiber was found in muscles examined after long survival times. Comparable results were obtained by examining the frequency of action potentials elicited by stimulation of the two nerves in non-curarized muscles (Table 11). Thus, it appears that at early times after partial denervation a moderate number of muscle fibers accept innervation from both nerles and that a later times most of this hyperinnervation disappears. There is not a similar trend for hyperinnervation as measured by muscle twitches (Table 11). This could have a number of explanations. The intracellular recording measurements give estimates of hyperinnervation only in that portion of each muscle reinnervated by the soleus motor neurons. This portion on the average was only of the total tension in the 9 and 10 day survival muscles (Table 11). Assuming a 20°, hyperinnervation in this portion of the muscle, only I !I0 or so of the total muscle would be hyperinnervated. This small degree of hyperinnervation may well be beyond our limits of detection in tension occlusion measurements, especially since we normally measure a small degree (up to 1406) of “apparent” convergence in normal adult muscles with aberrant and soleus nerves where in fact no convergence exists (Thompson and Jansen 1977). This inaccuracy in our measurement technique is apparently due to a lack of pure isometric conditions. In addition to determining whether muscle fibers became innervated by both aberrant and soleus nerve fibers. we also determined by microelectrode recording and grading the stimuli to each nerve independently whether multiple axons in each made synapses onto the individual muscle fibers. Of 216 muscle fibers giving a n epp to the aberrant nerve, none were polyneuronally innervated by the aberrant nerve. Hence, as previously reported (Thompson and Jansen 1977), the aberrant nerve fibers form single synaptic contacts upon sprouting. Of 330 muscle fibers giving an epp to the regenerating soleus nerve, only 14 fibers were polyneuronally innervated by the soleus nerve fibers. Even at the earliest times after soleus nerve regeneration, only 4 of 93 fibers examined were found to be polyneuronally innervated. Hence, in contrast to the reinnervation of neonatal soleus muscle (Brown et nl. 1976) and the reinnervation of an adult fast muscle of the rat ( McArdle 1975). polyneuronal reinnervation of the soleus muscle is rare. The sizes of the motor units in the muscles partially denervated by nerve crush were determined as described above and the results are presented in Fig. 5 . The material has been divided according to the survival time following partial denervation. The sprouted motor units after nerve crush (Fig. 5 A), average size 315 fibers, did not become as large as those in
87
REINNERVATION OF MUSCLE A.
R
Sprouted Nmrrc Motor Unih
rtsa-9835 ("-15)
ntS.D.-3Yl~111 (n-23) , "
RegasraledNwve M o h Unih
5
3
2 0 37-71danoftercruh
Number gf MurcleFiberr
2 I5l
37-71 donoftsr crush
I.lumb*i of hluscleFibers
Fig. 5. Sizes of sprouted and regenerated motor units in muscles partially denervated by crush of the soleus nerve. The data are divided into two histograms according to survival time after nerve crush: top histograms in both A & B give the motor unit sizes for 4 muscles examined 9-10 days after nerve crush; the lower histograms give the motor unit sizes for 6 muscles examined 37-71 days after nerve crush. Motor unit sizes for the data presented here were computed using 3 400 fibers for the normal number present in such muscles. In muscle fiber counts made of some of the muscles, an average of 3 340 was obtained.
the muscles where reinnervation by the soleus nerve was either prevented or delayed about one week longer (averages of 524 and 517 fibers, respectively; see Fig. 3). Apparently, if regenerating nerve fibers reach the muscle soon after partial denervation they are able to compete with the sprouting nerve fibers for innervation of the available, denervated muscle fibers. However, the regenerating soleus nerve motor units were never able to regain their normal size. The soleus motor units after regeneration through the crush (a +S.D. -94 +. 32; Fig. 5 B) are significantly smaller (Wilcoxon, P ;.O.OOl) than the normal soleus motor units X = 150, Thompson and Jansen 1977). Further, the measurements of the sizes of the sprouted motor units indicate that these units may be decreasing in size with time after partial denervation. The sprouted motor units are larger at 9-10 days after partial denervation than they are at later times (37-71 days). This difference is statistically significant (Wilcoxon, p
