030~322/~ $3.00 + 0.00 PergamonPressplc 0 1990 IBRO
Neura~e~~ceVol. 34, No. 2, pp. 525-532, 1990 Printedin Great Britain
THE EFFECT OF MUSCLE ACTIVITY ON MOTOR UNIT SIZE IN PARTIALLY DENERVATED RAT SOLEUS MUSCLES A. L.
CONNOLD
and G.
VRBOVA*
Department of Anatomy and Developmental Biology, Centre for Neuroscience, University College London, Gower Street, London WCIE 6BT, U.K. Ahstraet-4 has been shown previously that after section of L5 ventral ramus at 5 days the intact axons of L4 ventral ramus retain their large neonatal peripheral field in the rat soleus muscles. Soleus muscles of S-day-old rats were partially denervated by section of their major neural input, L5 ventral ramus, and in addition paralysed with a-bungarotoxin for 3-5 days. The motor unit size was examined 2 months later. The tension developed by individual motor units from muscles that were partially denervated and in addition temporally paralysed was much less than that after partial denervation Blone. This reduced tension output was not due to muscle atrophy but to a smaller number of muscle fibres supplied by individual axons. Thus, unlike after partial denervation only, motoneurons were unable to maintain their large neonatal territory when the muscle was temporarily paralysed and they were unable to reoccupy this territory after the muscles recovered from the paralysis. The possibility that arrested muscle maturation due to paralysis has a permanent effect on motor unit size is discussed.
The soleus muscle of the rat is innervated by axons from two lumbar spinal ventral roots, L4 and L5. In neonates, individual motor units are large and their axons contact up to five times as many muscle fibres as do the corresponding axons in adults3~“~” This large motor unit size is due to the fact that individual muscle fibres are contacted by several axons and motor unit territories overlap. A reduction of the number of synaptic contacts, on individual muscle fibres (elimination of polyneuronal innervation), occurs during the second and third week after birth in rats, and it is then that the adult size of the motor unit emerges.‘2*‘g~23 It is we11documented that the territory of a motor unit is not rigidly determined since, as well as having its size reduced during development,iJ an adult motoneuron has the ability to expand its peripheral field when a skeletal muscle becomes Moreover, the reduction partially denervated. 2~7~8~1’~‘2 in motor unit size during development is not a pre-programmed event since after partial denervation of the soleus in neonates it does not occur.8 It is known that neuromuscular activity during the period of rapid synapse elimination (postnatal days 9-12 in the rat) enhances the rate of elimination of polyneuronal innervationi while inactivity of the target at this time favours the persistence of neuromuscular contacts.s*22 Recent results show that while inactivity of the muscles during the time when synapse elimination is in progress (days g-16) does *To whom correspondence should be addressed. ACh, acetylcholine; AChE, acetylcholinesterase; a-BTX, a-bungarotoxin; EMG, electromyogram; SDH, succinate dehydrogenase; vr., ventral ramus.
Abbreuiurions:
preserve nerve-muscle contacts, it has the opposite effect during earlier stages of development (at birth), when paralysis of the muscle causes a dramatic decrease of neuromuscular contacts.” These results show that the target activity is im~rtant for the maintenance of immature synapses, and they suggest that the interaction between the nerve ending and the muscle is needed for further synapse development. In experiments where partial denervation of the soleus muscle is carried out by removing the L5 ventral ramus (vr.) the terminals of I.4 axons at the time of the operation are relatively ineficient but they develop with time and their quanta1 content becomes similar to that of terminals in a fully innervated soleus, even though their axons occupy a territory two to three times larger.12 It is likely that after partial denervation the overall activity of the muscle is reduced and this may allow the maintenance of neuromuscular contacts of L4 axons and hence the maintenance of a larger motor unit territory. Whether inactivity of the target and thus prevention of the interaction between nerve terminals and muscle fibres at a time when many of the L4 terminals are still weak will effect the territory of L4 axons after removal of L5 v.r. is studied in the present work. EXPERIMENTAL PROCEDURES Surgery
Section of the L5 v.r. was performed on one side in 5-day old Wistar Albino rats (University College London) under ether anaesthesia. Two days later a second operation was carried out on the same side, during which a silicone rubber strip containing a-bungarotoxin (20 ng a-BTX, Sigma or purified a-BTX, gift from Dr Dolly, and 0.3 mg NaCI/l mg of silicone) was implanted alongside the 525
526
A. L. CONNOLD and G. VRLIOVA
soleus muscle. The preparation of the strips is similar to that described previously.4 All the operations were carried out under sterile conditions In control experiments L5 v.r. section was followed by the implantation of a silicone rubber strip containing sodium chloride (0.3 mg NaCI/I mg of silicone). In some experiments the rats had only L5 v.r. section without any implant. The muscles were then tested 2 weeks or 2-10 months after the implantation of the strips. Tension recordings Two weeks or 2210 months after the operations, the rats were anaesthetized with chloralhydrate (4.5%, 1 ml/100 g body weight) and prepared for tension recording of their muscles. The soleus muscle was freed from the surrounding musculature and the Achilles tendon cut. The leg of the animal was rigidly fixed by pins. The tendon was then connected to a strain gauge via a silk thread. Soleus muscle contractions were elicited by stimulating its motor nerve with a pair of silver chloride electrodes using a pulse width of 0.0220.05 ms and the contractions were displayed on an oscilloscope screen (Tetronix R5113). The muscle length was adjusted to obtain maximum twitch tension at supramaximal stimulus intensity. The effect of the treatment was assessed by comparing the maximal tetanic tension developed by the operated muscle to that developed by its contralateral control and was expressed as a percentage in relation to the control, unoperated muscle. To test the possibility that denervated fibres were present at the time of the experiment, directly elicited tetanic contractions were also recorded. The two silver wire electrodes were placed either side of the soleus muscle and the muscle was stimulated by pulses of 2ms duration and supramaximal intensity. The maximum tetanic tension developed directly was compared to that obtained from
indirect stimulation, and expressed as a percentage of the directly elicited tension. Motor unit numbers and sizes To estimate the number of motor units in each muscle, the nerves to the operated and contralateral soleus muscles were stimulated by single pulses every 4 s. The stimulus strength was gradually increased to obtain stepwise increments of twitch tension, as individual motor axons become recruited. The number of stepwise increments was counted to give an estimate of the number of motor axons present in the nerve. Mean motor unit tension was obtained by dividing the maximum tetanic tension of each muscle by the number of motor units identified in it. The mean value obtained from an operated muscle was expressed as a percentage of the mean value obtained from its contralateral control muscle. After all tension measurements had been completed, the control and operated soleus muscles were dissected out from the rats, weighed, and frozen for subsequent histochemistry. Histochrmislry and morphomelry The dissected soleus muscles were stretched and secured along a pin at a length similar to that in the leg. The control and operated muscles were mounted side by side for each animal and frozen in isopentane cooled with liquid nitrogen. To determine the muscle fibre types present, fresh-frozen cross-sections (10 pm), cut from the middle of control and operated muscles, were collected on gelatinized (0.5%) glass slides and stained for succinate dehydrogenase” and alkalipreincubated myosin ATPasezO activities. The total number of muscle fibres and a random sample of their cross-sectional areas were also measured from these stained sections. The areas of 300&400 muscle fibres were obtained from a random sample using camera lucida drawings and a computer graphics programme. The sample was obtained by placing a grid of squares over each muscle section and including the muscle fibres in every sixth square.
Electromyogram recording To test whether the a-BTX applied in the manner described caused paralysis, electromyogram (EMG) recordings were carried out daily on treated and untreated soleus muscles as described by Navarette and Vrbov6.r4 In three previously unoperated rats that had electrodes implanted for EMG recording, a-BTX was applied to one soleus muscle on postnatal day 7.
RESULTS
The efect of temporary paralysis on partially denervated soleus muscles When partial denervation of rat soleus muscles is carried out by removing the L5 v.r. at 5 days the remaining L4 axons retain their relatively large territory so that 14 days later they still occupy 63% of the soleus muscle fibres. Without the removal of L5 v.r., L4 axons would have retracted and supplied only 20-30% of the muscles.‘2,24 To study the possibility that target activity is important for the maintenance of the enlarged territory of L4 axons after partial denervation, the muscles were paralysed by cr-BTX 2 days after section of the L5 v.r. and examined 2 weeks and 2210 months later. Muscle tensions and weights. Contractions of unoperated contralateral, and NaCI- and c(-BTXtreated partially denervated soleus muscles were elicited by stimulation of their motor nerves. The experiments were carried out either 16 days or 2 months after section of the L5 v.r. Figure 1 shows examples from two experiments examined 2 months after the operation. Traces from contralateral, untreated muscles are shown on the left, those from operated muscles on the right. The operated muscle at the top right was treated with NaCl, that at the bottom right with a-BTX. It is apparent that while both NaCl- and a-BTX-treated muscles are weaker than the unoperated contralateral, the reduction in force output is much greater after treatment with a-BTX. The results obtained in this manner are summarized in Table 1. In each animal the maximum tetanic tension of the NaCI- or a-BTX-treated muscle was expressed as a percentage of unoperated contralateral muscle. It is clear from Table 1 that the a-BTXtreated muscles developed a much smaller proportion of the tension of their unoperated contralateral muscle then either the untreated or the NaCl-treated muscles. This was so that no matter whether the muscles were tested 2 weeks or 2-10 months after a-BTX treatment. Thus the relative weakness of the treated muscles developed rapidly and was not temporary, but persisted. In some experiments, the ratio of directly to indirectly elicited maximum tetanic tension was also assessed in animals tested at 2-10 months. There was no difference between the tensions obtained by these two methods of stimulation, indicating that at 2-10 months after the rhizotomy there were no denervated fibres present in the operated muscles. These results
Muscle
activity
and motor
a
521
unit size
b
3OOm.s Fig. I. Records of single twitch and tetanic contractions (IO, 20, 40 and 80 Hz), from the soleus muscles of two rats elicited by stimulation of their motor nerves. Trace (b) is from an animal that had NaCl implanted in addition to L5 VI. section, while (a) represents the control unoperated muscle. Trace (d) is from an animal where a-BTX implant followed the L5 v.r. section, and (c) is a trace from the control soleus. could
not be due to the presence
strip
for,
as seen
in Table
of the silicone
1, if the
strip
rubber
contained
NaCl the tension produced by the partially denervated muscles was similar to that of rhizotomized muscles with no implants. Table I also shows that muscle weight of the a-BTX-treated muscles was reduced at 2-10 months, but not at 2 weeks. Motor unit number and sizes. The low tension output and weight of the paralysed partially denervated muscle is not due to a lower than usual number of motor axons in the L4 v.r. The number of motor units in the soleus muscles after LS v.r. section was assessed by counting the increments of twitch tension produced by stimulation of the soleus nerve by increasing current intensity. It varied between 5 and 13, with a mean of 8.7 f 1.4 S.E.M. (n = 6) in the partially denervated soleus muscles, and between 3 Table 1. Maximal tetanic tension and muscle weight of each operated muscle expressed as a percentage of its contralateral
unoperated
muscle
Treatment after L5 section
Tetanic tension (% op/con) 14 days 2 months
Muscle weight (% op/con) 14 days 2 months
None
69 + 1 (n = 10)
62+4 (n = 12)
92 + 4 (n = 12)
81 +3 (n = 8)
NaCl
65+ 11 (n = 3)
65 & 9 (n = 6)
88 + 5 (n = 3)
80 + 5 (n = 5)
a-BTX
31 f 6 (n = 11)
39 & 7 (n = 8)
13 +2
65 k 6 (n = 8)
(n = 6)
The operated muscles were from rats that had LS v.r. section at 5 days followed by either NaCl or a-BTX containing silicone implants. The muscles were examined 14 days or 2 months later. N = number of muscles in each group. Values are kS.E.M.
and 13, with a mean of 9.5 f 1.2 S.E.M. (n = 8), in the partially denervated a-BTX-treated muscles. These numbers of motor axons are not significantly different from each other. To determine changes of motor unit size in operated muscles, the mean motor unit tension was calculated for each muscle dividing the maximal tetanic tension by the number of motor units. The value obtained for the operated muscle was then expressed as a percentage of the mean motor unit tension of the contralateral control soleus. For normal muscles this value would be 100%. Table 2 shows
Table 2. Mean motor unit tension of operated muscle expressed as a percentage of the mean motor unit tension of the control unoperated muscle. Treatment after L5 v.r. section
Motor unit size (% op/con)
None NaCl a-BTX
225 & 13 (n = 8) 208 + 14 (n = 6) 121 + 23 (n = 8)
Mean motor unit tension calculated for each soleus muscle by dividing the maximal tetanic tension by the number of motor units present. The values obtained for operated muscle was then expressed as a percentage of the mean motor unit tension of the control unoperated soleus. First value represents results from animals with L5 rhizotomy at 5 days, values two and three represent L5 results where rhizotomy at 5 days was followed by a-BTX or NaCl treatment two days later. Experiments were carried out at 2 months. n-number of muscles in each group. Values are &S.E.M. con, control unoperated; op, operated.
528
A.L. CONNOLD and G. VRBOV/k Table 3. The numbers and areas of muscle fibres in control and operated muscles, from six rats aged 2 months, that had L5 ventral ramus section at 5 days followed by either NaCI or ~-BTX treatment 2 days later Treatment after L5 v.r. section
Muscle fibre number con op % op/con
Muscle fibre area ( ~ m 2) con op % op/con
NaC1
3118 2158 2679
2301 1699 2105
74 79 79
3228 4438 3839
3517 4708 4458
109 106 116
2246 2600 2484
1152 980 1367
51 38 55
4122 3590 3634
6339 3806 4912
154 106 135
-BTX
Values for individual muscles are given as well as calculated percentage of operated muscles in relation to their control. Con, control unoperated; op, operated.
that if L5 v.r. section is carried out at 5 days then 2 10 m o n t h s later this ratio is 225% + 13S.E.M., (n = 8). 8 W h e n L5 v.r. section was followed by NaCI t r e a t m e n t the value was 208% _+ 14 S.E.M. (n = 6), which is similar to that obtained from untreated rhizotomized muscles. If ~ - B T X is applied to the partially denervated muscles the m e a n m o t o r unit size, 2-10 m o n t h s later, is only 121% + 2 3 S.E.M. (n = 8), which is significantly less t h a n the m e a n m o t o r unit size of either partially denervated or N a C l - t r e a t e d partially denervated muscles. Muscle fibre development. The relatively smaller m o t o r unit sizes in the ~ - B T X - t r e a t e d muscles could be due to a p e r m a n e n t effect of the ~ - B T X t r e a t m e n t on the d e v e l o p m e n t and growth of muscle fibres. Muscle fibre n u m b e r s a n d areas were therefore measured in 2-10 m o n t h old rats. Table 3 summarizes the results obtained for all the muscles examined. Muscle fibre n u m b e r s for the operated muscles were expressed as a percentage of their contralateral controls. In three rats t h a t had L5 v.r. section a n d an N a C l implant, the values were 74, 79 a n d 79%. Thus some p e r m a n e n t loss of muscle fibres occurred due to the partial denervation. In three rats t h a t had L5 v.r. section followed by ~ - B T X implant, the n u m b e r of muscle fibres was 51, 38 and 55% only of the contralateral-control muscle, showing that by 2 m o n t h s the loss of muscle fibres is considerably greater in c¢-BTX-treated muscles t h a n in NaCItreated ones. The muscle fibre areas were also measured. Figure 2 c o m p a r e s the distribution of fibre areas in control soleus with those in (a) an N a C l - t r e a t e d and (b) an ~ - B T X - t r e a t e d partially denervated muscle. There is little difference between the distribution of muscle fibre area o f control and rhizotomized, NaC1treated soleus muscles (Fig. 2a), but after a - B T X t r e a t m e n t o f the partially denervated muscles the fibre area distribution was shifted to the right, due to the presence o f a greater n u m b e r of large fibres (Fig. 2b). This is consistent with results shown in Table 3, where the m e a n area o f N a C l - t r e a t e d muscles is only slightly increased above normal, whereas a more p r o n o u n c e d increase occurs in the ct-BTX-treated muscles.
These results, taken together, show that the difference in muscle tension between partially denervated muscles a n d partially denervated muscles t h a t had t e m p o r a r y paralysis is due to a p e r m a n e n t loss of muscle fibres and not to muscle fibre atrophy. Thus, a brief paralysis o f the developing muscle does not allow the m a i n t e n a n c e of the e x p a n d e d peripheral field of neonatal m o t o n e u r o n s that occurs in the absence of ct-BTX treatment.
Changes of contractile properties and histochemistry. W h e t h e r t r e a t m e n t with ~ - B T X affects the d e v e l o p m e n t of contractile characteristics was also
~ CONTROL ~2'~LS*NaCI
4(3
3(3
2(3
10
0
o~
b
D CONTROL ] LS+c~gTx
_~ 4c 3C 2C lC
5O00 Fibre area (j.im 2)
100O0
Fig. 2. Frequency distribution of muscle fibre areas. (a) Results from an animal that had L5 v.r. section at 5 days and NaC1 treatment 2 days later. (b) Results from an animal that had L5 v.r. section at 5 days and ~-BTX treatment 2 days later. In both (a) and (b) the empty areas represent the distribution of fibre areas from control muscles and the shaded areas represent those from operated muscles.
Muscle activity and motor unit size examined. Table 4 summarizes the results and shows that a-BTX has little effect on the time course of contraction. The effects of partial denervation and/or paralysis on muscle fibre type distribution were investigated by staining muscle sections for succinate dehydrogenase (SDH) and alkaline preincubated (fast) ATPase activities. In contralateral control soleus muscles stained for SDH, mainly large fibres interspersed with a few small, slightly darker stained fibres were seen (Fig. 3a). After LS v.r. section, these small dark fibres were conspicuously absent from the muscles. The absence of these small oxidative fibres was also seen in all the partially denervated muscles, no matter whether they had treatment with NaCl (Fig. 3b) or with a-BTX (Fig. 3d). The fast-myosin ATPase stain showed that the dark fast fibres present in the control soleus (Fig. 3e) were absent after L5 v.r. section and also after L5 v.r. section followed by NaCl or ol-BTX treatment (Fig. 3f and h). The eficts of treatment with u-bungarotoxin on the actiuity of soleus. To determine the degree and time course of the paralysis induced by the application of c(-BTX on the neonatal muscles, EMG recordings were carried out in freely moving animals with implanted electrodes. These records were taken from the control soleus and from soleus muscles that had had a-BTX applied to them. Three hours after the or-BTX implant was placed onto the muscle the control soleus showed considerable activity during locomotion (Fig. 4a) while almost no activity could be recorded from the operated muscle. Even 24 h later there was no EMG activity seen in the treated soleus (Fig. 4b) during spontaneous locomotion or when attempts were made to elicit activity by displacing the animal or moving its foot. These manipulations elicited vigorous activity in the control muscle, but none in the treated soleus. EMG activity was seen to gradually return, and 5 days after the operation the amounts of activity in the treated soleus were indistinguishable from those in the control muscle. Thus the paralysis recovered completely within 5 days. DISCUSSION
Following the removal of a major input to the rat soleus muscle the remaining motoneurons can successfully maintain a larger than normal peripheral field.*,‘* The present results show that this does not occur when the response of the target is temporarily prevented. Although the paralysis lasted for only 5 days the L4 axons were unable to maintain their enlarged neonatal territory. Neither were they able to reoccupy it later, when the muscle was no longer paralysed and the denervated fibres were still present. This is surprising, for these “vacated” muscle fibres were available for the recently disconnected terminals to establish contacts. The findings that (a) existing contacts are lost when the target is paralysed and (b) the denervated muscle fibres do not become
529
530
A. L. CONNOLV and G. VuaovA
Fig. 3. Examples of cross-sections of control and operated soleus muscles from 5-month-old rats. In ad the sections were stained for SDH activity and in e-h for alkaline preincubated ATPase activities. a, c, e and g were contralateral muscles to the soleus muscles that were treated in the following way: b and f, v.r. sectioned and NaCl implanted; and h, v.r. sectioned and a-BTX implanted.
Muscle activity and motor unit size Spontaneous activity
a Con
24h Fig. 4. EMG activity recorded from control and a-BTX treated soleus muscles in freely moving 7-day-old-rats. (a) The activity present in control and operated muscles 3 h after the application of GI-BTX. (b) The activity present in control and operated muscles 24 h after the application of a-BTX.
reinnervated, (see also Ref. 18), suggest that the reduction of the peripheral field after paralysis could be due to retarded development of those muscle fibres that at the time of paralysis are relatively immature, and lose their contact with the nerve terminal on
531
recovery from the paralysis. Development of muscle fibres is retarded as a result of the reduced input to the muscle, and additional paralysis. It could be that in muscle fibres where development is arrested the accumulation of a~tyIcholinesterase (AChE) does not take place. It has indeed been shown that paralysis of the embryonic muscies prevents the accumulation of AChE.9 In addition, the conversion of the slow neonatal acetylcholine (ACh) channels to the more mature form may also be halted,*’ for this transition too is activity-dependent,~ This could, during the period of recovery from the paralysis, lead to a situation seen in muscles treated with cr-BTX immediately after birth and examined 5 days later, where a very prolonged end-plate potential is seen.” In this case, too, immediately after recovery from the paralysis, a loss of many contacts occurs. A prolonged depola~zation causes disruption of neuromuscular connections, For exposure of soleus muscles to ACh or to inhibitors of AChE is known to bring about a reduciton of neuromuscular contacts.5.6.‘7 It is likely that in the present experiments in the paralysed muscles a number of muscle fibres have end-plates with low levels of AChE, and these endplates may then rapidly loose contact with their nerve terminals during the period of recovery from the paralysis. Moreover these fibres, for the same reasons, will then be unable to accept innervation during later stages of recovery. Such a mechanism could explain the ~rmanent nature of this short-term paralysis. Acknowledgements-The authors would like to thank the Vivian Smith Foundation, The National Fund for Research into Crippling Diseases and the Wellcome Trust for support. They are grateful to Dr Dolly for the gift of the purified a-bungarotoxin.
REFERENCES
1. Bagust J., Lewis D. M. and Westerman R. A. (1973) Polyneuronal innervation of kitten skeletal muscle. J. Physiol. 229, 241-255. 2. Brown M. C., Holland R. L. and Hopkins W. G. (1981) Restoration of focal multiple innervation in rat muscles by transmission block during a critical stage of development. J. Physiol., Land. 318, 355-364. 3. Brown M. C., Jansen K. K. S. and Van Essen D. (1976) Poiyneuronal innervation of skeletal muscle in new-born rats and its elimination during maturation. J. Physiof. 261, 387-422. 4. Connold A. L., Evers J. V. and VrbovB G. (1986) Effect of low calcium and protease inhibitor on synapse elimination during postnatal development in the rat soleus muscle. Devl Brain Res. 28, 99-107. 5. Duxson M. J. (1982) The effect of postsynaptic block on development of the neuromuscular junction in postnatal rats. J. Neurocytol. 11, 395-408. 6. Duxson M. J. and Vrbova G. (1984) Inhibition of acetylcholinesterase accelerates axon terminal withdrawal. 1. Neurocytol. 14, 337-363. 7. Edds M. V. and Small W. T. (1957) The behaviour of residual axons in partially denervated muscles of the monkey. J. exp. Med. 93, 207-215. 8. Fisher T. J., VrbovL G. and Widjetunge A. (1989) Partial denervation of the rat soleus muscle at two different development stages. Neuroscience 28, 755-763. 9. Gordon T., Tuffery A., Perry R. and Vrbovi G. (1974) Possible mechanisms determining synapse formation in developing skeletal muscles of the chick. Cell Tiss. Res. 155, 13-25. 10. Greensmith L., LeHouelleur J. and VrbovB G. (1988) The effect of early muscles paralysis on the distribution of inne~ation in the newborn rat. J. Physiol. 401, 52P. 11. Hoffman H. (1950) Local reinnervation of partially denervated muscle: a histophysiological study. Aust. J. exp. Biol. Med. Sci. 28, 383-397. 12. Lowrie M. B., O’Brien R. A. D. and Vrbovi G. (1985) the effect of altered peripheral field on motoneurone in developing rat soleus muscles. J. Physiol. 368, 513-524.
function
13. Nachlas M. M., Tson K. C., De Souza E., Chang C. H. and Se&man A. M. (1957) Cytochemical demonstration of succinate dehydrogenase by the use of a new p-nitrophenyl substituted ditetrazole. J. Hisfochem. Cytochem. S,420-436.
532
A. L. CONNOLD and G. VRBOV,~
14. Navarette R. and Vrbova G. (1983) Changes of activity patterns in slow and fast muscles during postnatal development. Devl Brain Res. 8, 11-19. 15. O’Brien R. A. D., &&erg A. J. C. and Vrbova G. (1978) Observations on the elimination of polyneuronal innervation in developing mammalian skeletal muscle. J. Physiol. 282, 571-582. 16. O’Brien R. A. D., &berg A. J. C. and Vrbova G. (1980) The effect of acetylcholine on the function and structure of the developing mammalian neuromuscular junction. Neuroscience 5, 1367-1379. 17. O’Brien R. A. D., &berg A. J. C. and Vrbova G. (1984) Protease inhibitors reduce the loss of nerve terminals induced by activity and calcium in developing rat soleus muscles in uitro. Neuroscience 12, 637-646. 18. Pestronk A. and Drachman D. B. (1985) Motor nerve terminal outgrowth and acetylcholine receptors: inhibition of terminal outgrowth by u-bungarotoxin and anti-acetylcholine receptor antibody. J. Neurosci. 5, 751-758. 19. Redfern P. A. (1970) Neuromuscular transmission in newborn rats. J. Physiol., Land. 209, 701-709. 20. Round J. M., Mathews Y. and Jones D. A. (1980) A quick simple and reliable histochemical method for ATPase in human muscle preparations. J. Histochem. 12, 701-710. 21. Sackman B. and Brenner H. R. (1978) Changes in synaptic channel gating during neuromuscular development. Nafure 276, 401-402. 22. Srihari T. and Vrbova G. (1978) The role of muscle activity in the differentiation of NMJs in slow and fast chick muscles. J. Neurocytol. 7, 5299540. 23. Thompson W. J. (1983) Lack of segmental selectivity in elimination of synapses from soleus muscle of new-born rats. J. Physiol. 335, 345-352. 24. Thompson W. J. and Jansen J. K. (1977) The extent of sprouting of remaining motor units in partially denervated immature and adult rat soleus muscle. Neuroscience 2, 523-535. (Accepted 4 September 1989)
Neura~e~~ceVol. 34, No. 2, pp. 525-532, 1990 Printedin Great Britain
THE EFFECT OF MUSCLE ACTIVITY ON MOTOR UNIT SIZE IN PARTIALLY DENERVATED RAT SOLEUS MUSCLES A. L.
CONNOLD
and G.
VRBOVA*
Department of Anatomy and Developmental Biology, Centre for Neuroscience, University College London, Gower Street, London WCIE 6BT, U.K. Ahstraet-4 has been shown previously that after section of L5 ventral ramus at 5 days the intact axons of L4 ventral ramus retain their large neonatal peripheral field in the rat soleus muscles. Soleus muscles of S-day-old rats were partially denervated by section of their major neural input, L5 ventral ramus, and in addition paralysed with a-bungarotoxin for 3-5 days. The motor unit size was examined 2 months later. The tension developed by individual motor units from muscles that were partially denervated and in addition temporally paralysed was much less than that after partial denervation Blone. This reduced tension output was not due to muscle atrophy but to a smaller number of muscle fibres supplied by individual axons. Thus, unlike after partial denervation only, motoneurons were unable to maintain their large neonatal territory when the muscle was temporarily paralysed and they were unable to reoccupy this territory after the muscles recovered from the paralysis. The possibility that arrested muscle maturation due to paralysis has a permanent effect on motor unit size is discussed.
The soleus muscle of the rat is innervated by axons from two lumbar spinal ventral roots, L4 and L5. In neonates, individual motor units are large and their axons contact up to five times as many muscle fibres as do the corresponding axons in adults3~“~” This large motor unit size is due to the fact that individual muscle fibres are contacted by several axons and motor unit territories overlap. A reduction of the number of synaptic contacts, on individual muscle fibres (elimination of polyneuronal innervation), occurs during the second and third week after birth in rats, and it is then that the adult size of the motor unit emerges.‘2*‘g~23 It is we11documented that the territory of a motor unit is not rigidly determined since, as well as having its size reduced during development,iJ an adult motoneuron has the ability to expand its peripheral field when a skeletal muscle becomes Moreover, the reduction partially denervated. 2~7~8~1’~‘2 in motor unit size during development is not a pre-programmed event since after partial denervation of the soleus in neonates it does not occur.8 It is known that neuromuscular activity during the period of rapid synapse elimination (postnatal days 9-12 in the rat) enhances the rate of elimination of polyneuronal innervationi while inactivity of the target at this time favours the persistence of neuromuscular contacts.s*22 Recent results show that while inactivity of the muscles during the time when synapse elimination is in progress (days g-16) does *To whom correspondence should be addressed. ACh, acetylcholine; AChE, acetylcholinesterase; a-BTX, a-bungarotoxin; EMG, electromyogram; SDH, succinate dehydrogenase; vr., ventral ramus.
Abbreuiurions:
preserve nerve-muscle contacts, it has the opposite effect during earlier stages of development (at birth), when paralysis of the muscle causes a dramatic decrease of neuromuscular contacts.” These results show that the target activity is im~rtant for the maintenance of immature synapses, and they suggest that the interaction between the nerve ending and the muscle is needed for further synapse development. In experiments where partial denervation of the soleus muscle is carried out by removing the L5 ventral ramus (vr.) the terminals of I.4 axons at the time of the operation are relatively ineficient but they develop with time and their quanta1 content becomes similar to that of terminals in a fully innervated soleus, even though their axons occupy a territory two to three times larger.12 It is likely that after partial denervation the overall activity of the muscle is reduced and this may allow the maintenance of neuromuscular contacts of L4 axons and hence the maintenance of a larger motor unit territory. Whether inactivity of the target and thus prevention of the interaction between nerve terminals and muscle fibres at a time when many of the L4 terminals are still weak will effect the territory of L4 axons after removal of L5 v.r. is studied in the present work. EXPERIMENTAL PROCEDURES Surgery
Section of the L5 v.r. was performed on one side in 5-day old Wistar Albino rats (University College London) under ether anaesthesia. Two days later a second operation was carried out on the same side, during which a silicone rubber strip containing a-bungarotoxin (20 ng a-BTX, Sigma or purified a-BTX, gift from Dr Dolly, and 0.3 mg NaCI/l mg of silicone) was implanted alongside the 525
526
A. L. CONNOLD and G. VRLIOVA
soleus muscle. The preparation of the strips is similar to that described previously.4 All the operations were carried out under sterile conditions In control experiments L5 v.r. section was followed by the implantation of a silicone rubber strip containing sodium chloride (0.3 mg NaCI/I mg of silicone). In some experiments the rats had only L5 v.r. section without any implant. The muscles were then tested 2 weeks or 2-10 months after the implantation of the strips. Tension recordings Two weeks or 2210 months after the operations, the rats were anaesthetized with chloralhydrate (4.5%, 1 ml/100 g body weight) and prepared for tension recording of their muscles. The soleus muscle was freed from the surrounding musculature and the Achilles tendon cut. The leg of the animal was rigidly fixed by pins. The tendon was then connected to a strain gauge via a silk thread. Soleus muscle contractions were elicited by stimulating its motor nerve with a pair of silver chloride electrodes using a pulse width of 0.0220.05 ms and the contractions were displayed on an oscilloscope screen (Tetronix R5113). The muscle length was adjusted to obtain maximum twitch tension at supramaximal stimulus intensity. The effect of the treatment was assessed by comparing the maximal tetanic tension developed by the operated muscle to that developed by its contralateral control and was expressed as a percentage in relation to the control, unoperated muscle. To test the possibility that denervated fibres were present at the time of the experiment, directly elicited tetanic contractions were also recorded. The two silver wire electrodes were placed either side of the soleus muscle and the muscle was stimulated by pulses of 2ms duration and supramaximal intensity. The maximum tetanic tension developed directly was compared to that obtained from
indirect stimulation, and expressed as a percentage of the directly elicited tension. Motor unit numbers and sizes To estimate the number of motor units in each muscle, the nerves to the operated and contralateral soleus muscles were stimulated by single pulses every 4 s. The stimulus strength was gradually increased to obtain stepwise increments of twitch tension, as individual motor axons become recruited. The number of stepwise increments was counted to give an estimate of the number of motor axons present in the nerve. Mean motor unit tension was obtained by dividing the maximum tetanic tension of each muscle by the number of motor units identified in it. The mean value obtained from an operated muscle was expressed as a percentage of the mean value obtained from its contralateral control muscle. After all tension measurements had been completed, the control and operated soleus muscles were dissected out from the rats, weighed, and frozen for subsequent histochemistry. Histochrmislry and morphomelry The dissected soleus muscles were stretched and secured along a pin at a length similar to that in the leg. The control and operated muscles were mounted side by side for each animal and frozen in isopentane cooled with liquid nitrogen. To determine the muscle fibre types present, fresh-frozen cross-sections (10 pm), cut from the middle of control and operated muscles, were collected on gelatinized (0.5%) glass slides and stained for succinate dehydrogenase” and alkalipreincubated myosin ATPasezO activities. The total number of muscle fibres and a random sample of their cross-sectional areas were also measured from these stained sections. The areas of 300&400 muscle fibres were obtained from a random sample using camera lucida drawings and a computer graphics programme. The sample was obtained by placing a grid of squares over each muscle section and including the muscle fibres in every sixth square.
Electromyogram recording To test whether the a-BTX applied in the manner described caused paralysis, electromyogram (EMG) recordings were carried out daily on treated and untreated soleus muscles as described by Navarette and Vrbov6.r4 In three previously unoperated rats that had electrodes implanted for EMG recording, a-BTX was applied to one soleus muscle on postnatal day 7.
RESULTS
The efect of temporary paralysis on partially denervated soleus muscles When partial denervation of rat soleus muscles is carried out by removing the L5 v.r. at 5 days the remaining L4 axons retain their relatively large territory so that 14 days later they still occupy 63% of the soleus muscle fibres. Without the removal of L5 v.r., L4 axons would have retracted and supplied only 20-30% of the muscles.‘2,24 To study the possibility that target activity is important for the maintenance of the enlarged territory of L4 axons after partial denervation, the muscles were paralysed by cr-BTX 2 days after section of the L5 v.r. and examined 2 weeks and 2210 months later. Muscle tensions and weights. Contractions of unoperated contralateral, and NaCI- and c(-BTXtreated partially denervated soleus muscles were elicited by stimulation of their motor nerves. The experiments were carried out either 16 days or 2 months after section of the L5 v.r. Figure 1 shows examples from two experiments examined 2 months after the operation. Traces from contralateral, untreated muscles are shown on the left, those from operated muscles on the right. The operated muscle at the top right was treated with NaCl, that at the bottom right with a-BTX. It is apparent that while both NaCl- and a-BTX-treated muscles are weaker than the unoperated contralateral, the reduction in force output is much greater after treatment with a-BTX. The results obtained in this manner are summarized in Table 1. In each animal the maximum tetanic tension of the NaCI- or a-BTX-treated muscle was expressed as a percentage of unoperated contralateral muscle. It is clear from Table 1 that the a-BTXtreated muscles developed a much smaller proportion of the tension of their unoperated contralateral muscle then either the untreated or the NaCl-treated muscles. This was so that no matter whether the muscles were tested 2 weeks or 2-10 months after a-BTX treatment. Thus the relative weakness of the treated muscles developed rapidly and was not temporary, but persisted. In some experiments, the ratio of directly to indirectly elicited maximum tetanic tension was also assessed in animals tested at 2-10 months. There was no difference between the tensions obtained by these two methods of stimulation, indicating that at 2-10 months after the rhizotomy there were no denervated fibres present in the operated muscles. These results
Muscle
activity
and motor
a
521
unit size
b
3OOm.s Fig. I. Records of single twitch and tetanic contractions (IO, 20, 40 and 80 Hz), from the soleus muscles of two rats elicited by stimulation of their motor nerves. Trace (b) is from an animal that had NaCl implanted in addition to L5 VI. section, while (a) represents the control unoperated muscle. Trace (d) is from an animal where a-BTX implant followed the L5 v.r. section, and (c) is a trace from the control soleus. could
not be due to the presence
strip
for,
as seen
in Table
of the silicone
1, if the
strip
rubber
contained
NaCl the tension produced by the partially denervated muscles was similar to that of rhizotomized muscles with no implants. Table I also shows that muscle weight of the a-BTX-treated muscles was reduced at 2-10 months, but not at 2 weeks. Motor unit number and sizes. The low tension output and weight of the paralysed partially denervated muscle is not due to a lower than usual number of motor axons in the L4 v.r. The number of motor units in the soleus muscles after LS v.r. section was assessed by counting the increments of twitch tension produced by stimulation of the soleus nerve by increasing current intensity. It varied between 5 and 13, with a mean of 8.7 f 1.4 S.E.M. (n = 6) in the partially denervated soleus muscles, and between 3 Table 1. Maximal tetanic tension and muscle weight of each operated muscle expressed as a percentage of its contralateral
unoperated
muscle
Treatment after L5 section
Tetanic tension (% op/con) 14 days 2 months
Muscle weight (% op/con) 14 days 2 months
None
69 + 1 (n = 10)
62+4 (n = 12)
92 + 4 (n = 12)
81 +3 (n = 8)
NaCl
65+ 11 (n = 3)
65 & 9 (n = 6)
88 + 5 (n = 3)
80 + 5 (n = 5)
a-BTX
31 f 6 (n = 11)
39 & 7 (n = 8)
13 +2
65 k 6 (n = 8)
(n = 6)
The operated muscles were from rats that had LS v.r. section at 5 days followed by either NaCl or a-BTX containing silicone implants. The muscles were examined 14 days or 2 months later. N = number of muscles in each group. Values are kS.E.M.
and 13, with a mean of 9.5 f 1.2 S.E.M. (n = 8), in the partially denervated a-BTX-treated muscles. These numbers of motor axons are not significantly different from each other. To determine changes of motor unit size in operated muscles, the mean motor unit tension was calculated for each muscle dividing the maximal tetanic tension by the number of motor units. The value obtained for the operated muscle was then expressed as a percentage of the mean motor unit tension of the contralateral control soleus. For normal muscles this value would be 100%. Table 2 shows
Table 2. Mean motor unit tension of operated muscle expressed as a percentage of the mean motor unit tension of the control unoperated muscle. Treatment after L5 v.r. section
Motor unit size (% op/con)
None NaCl a-BTX
225 & 13 (n = 8) 208 + 14 (n = 6) 121 + 23 (n = 8)
Mean motor unit tension calculated for each soleus muscle by dividing the maximal tetanic tension by the number of motor units present. The values obtained for operated muscle was then expressed as a percentage of the mean motor unit tension of the control unoperated soleus. First value represents results from animals with L5 rhizotomy at 5 days, values two and three represent L5 results where rhizotomy at 5 days was followed by a-BTX or NaCl treatment two days later. Experiments were carried out at 2 months. n-number of muscles in each group. Values are &S.E.M. con, control unoperated; op, operated.
528
A.L. CONNOLD and G. VRBOV/k Table 3. The numbers and areas of muscle fibres in control and operated muscles, from six rats aged 2 months, that had L5 ventral ramus section at 5 days followed by either NaCI or ~-BTX treatment 2 days later Treatment after L5 v.r. section
Muscle fibre number con op % op/con
Muscle fibre area ( ~ m 2) con op % op/con
NaC1
3118 2158 2679
2301 1699 2105
74 79 79
3228 4438 3839
3517 4708 4458
109 106 116
2246 2600 2484
1152 980 1367
51 38 55
4122 3590 3634
6339 3806 4912
154 106 135
-BTX
Values for individual muscles are given as well as calculated percentage of operated muscles in relation to their control. Con, control unoperated; op, operated.
that if L5 v.r. section is carried out at 5 days then 2 10 m o n t h s later this ratio is 225% + 13S.E.M., (n = 8). 8 W h e n L5 v.r. section was followed by NaCI t r e a t m e n t the value was 208% _+ 14 S.E.M. (n = 6), which is similar to that obtained from untreated rhizotomized muscles. If ~ - B T X is applied to the partially denervated muscles the m e a n m o t o r unit size, 2-10 m o n t h s later, is only 121% + 2 3 S.E.M. (n = 8), which is significantly less t h a n the m e a n m o t o r unit size of either partially denervated or N a C l - t r e a t e d partially denervated muscles. Muscle fibre development. The relatively smaller m o t o r unit sizes in the ~ - B T X - t r e a t e d muscles could be due to a p e r m a n e n t effect of the ~ - B T X t r e a t m e n t on the d e v e l o p m e n t and growth of muscle fibres. Muscle fibre n u m b e r s a n d areas were therefore measured in 2-10 m o n t h old rats. Table 3 summarizes the results obtained for all the muscles examined. Muscle fibre n u m b e r s for the operated muscles were expressed as a percentage of their contralateral controls. In three rats t h a t had L5 v.r. section a n d an N a C l implant, the values were 74, 79 a n d 79%. Thus some p e r m a n e n t loss of muscle fibres occurred due to the partial denervation. In three rats t h a t had L5 v.r. section followed by ~ - B T X implant, the n u m b e r of muscle fibres was 51, 38 and 55% only of the contralateral-control muscle, showing that by 2 m o n t h s the loss of muscle fibres is considerably greater in c¢-BTX-treated muscles t h a n in NaCItreated ones. The muscle fibre areas were also measured. Figure 2 c o m p a r e s the distribution of fibre areas in control soleus with those in (a) an N a C l - t r e a t e d and (b) an ~ - B T X - t r e a t e d partially denervated muscle. There is little difference between the distribution of muscle fibre area o f control and rhizotomized, NaC1treated soleus muscles (Fig. 2a), but after a - B T X t r e a t m e n t o f the partially denervated muscles the fibre area distribution was shifted to the right, due to the presence o f a greater n u m b e r of large fibres (Fig. 2b). This is consistent with results shown in Table 3, where the m e a n area o f N a C l - t r e a t e d muscles is only slightly increased above normal, whereas a more p r o n o u n c e d increase occurs in the ct-BTX-treated muscles.
These results, taken together, show that the difference in muscle tension between partially denervated muscles a n d partially denervated muscles t h a t had t e m p o r a r y paralysis is due to a p e r m a n e n t loss of muscle fibres and not to muscle fibre atrophy. Thus, a brief paralysis o f the developing muscle does not allow the m a i n t e n a n c e of the e x p a n d e d peripheral field of neonatal m o t o n e u r o n s that occurs in the absence of ct-BTX treatment.
Changes of contractile properties and histochemistry. W h e t h e r t r e a t m e n t with ~ - B T X affects the d e v e l o p m e n t of contractile characteristics was also
~ CONTROL ~2'~LS*NaCI
4(3
3(3
2(3
10
0
o~
b
D CONTROL ] LS+c~gTx
_~ 4c 3C 2C lC
5O00 Fibre area (j.im 2)
100O0
Fig. 2. Frequency distribution of muscle fibre areas. (a) Results from an animal that had L5 v.r. section at 5 days and NaC1 treatment 2 days later. (b) Results from an animal that had L5 v.r. section at 5 days and ~-BTX treatment 2 days later. In both (a) and (b) the empty areas represent the distribution of fibre areas from control muscles and the shaded areas represent those from operated muscles.
Muscle activity and motor unit size examined. Table 4 summarizes the results and shows that a-BTX has little effect on the time course of contraction. The effects of partial denervation and/or paralysis on muscle fibre type distribution were investigated by staining muscle sections for succinate dehydrogenase (SDH) and alkaline preincubated (fast) ATPase activities. In contralateral control soleus muscles stained for SDH, mainly large fibres interspersed with a few small, slightly darker stained fibres were seen (Fig. 3a). After LS v.r. section, these small dark fibres were conspicuously absent from the muscles. The absence of these small oxidative fibres was also seen in all the partially denervated muscles, no matter whether they had treatment with NaCl (Fig. 3b) or with a-BTX (Fig. 3d). The fast-myosin ATPase stain showed that the dark fast fibres present in the control soleus (Fig. 3e) were absent after L5 v.r. section and also after L5 v.r. section followed by NaCl or ol-BTX treatment (Fig. 3f and h). The eficts of treatment with u-bungarotoxin on the actiuity of soleus. To determine the degree and time course of the paralysis induced by the application of c(-BTX on the neonatal muscles, EMG recordings were carried out in freely moving animals with implanted electrodes. These records were taken from the control soleus and from soleus muscles that had had a-BTX applied to them. Three hours after the or-BTX implant was placed onto the muscle the control soleus showed considerable activity during locomotion (Fig. 4a) while almost no activity could be recorded from the operated muscle. Even 24 h later there was no EMG activity seen in the treated soleus (Fig. 4b) during spontaneous locomotion or when attempts were made to elicit activity by displacing the animal or moving its foot. These manipulations elicited vigorous activity in the control muscle, but none in the treated soleus. EMG activity was seen to gradually return, and 5 days after the operation the amounts of activity in the treated soleus were indistinguishable from those in the control muscle. Thus the paralysis recovered completely within 5 days. DISCUSSION
Following the removal of a major input to the rat soleus muscle the remaining motoneurons can successfully maintain a larger than normal peripheral field.*,‘* The present results show that this does not occur when the response of the target is temporarily prevented. Although the paralysis lasted for only 5 days the L4 axons were unable to maintain their enlarged neonatal territory. Neither were they able to reoccupy it later, when the muscle was no longer paralysed and the denervated fibres were still present. This is surprising, for these “vacated” muscle fibres were available for the recently disconnected terminals to establish contacts. The findings that (a) existing contacts are lost when the target is paralysed and (b) the denervated muscle fibres do not become
529
530
A. L. CONNOLV and G. VuaovA
Fig. 3. Examples of cross-sections of control and operated soleus muscles from 5-month-old rats. In ad the sections were stained for SDH activity and in e-h for alkaline preincubated ATPase activities. a, c, e and g were contralateral muscles to the soleus muscles that were treated in the following way: b and f, v.r. sectioned and NaCl implanted; and h, v.r. sectioned and a-BTX implanted.
Muscle activity and motor unit size Spontaneous activity
a Con
24h Fig. 4. EMG activity recorded from control and a-BTX treated soleus muscles in freely moving 7-day-old-rats. (a) The activity present in control and operated muscles 3 h after the application of GI-BTX. (b) The activity present in control and operated muscles 24 h after the application of a-BTX.
reinnervated, (see also Ref. 18), suggest that the reduction of the peripheral field after paralysis could be due to retarded development of those muscle fibres that at the time of paralysis are relatively immature, and lose their contact with the nerve terminal on
531
recovery from the paralysis. Development of muscle fibres is retarded as a result of the reduced input to the muscle, and additional paralysis. It could be that in muscle fibres where development is arrested the accumulation of a~tyIcholinesterase (AChE) does not take place. It has indeed been shown that paralysis of the embryonic muscies prevents the accumulation of AChE.9 In addition, the conversion of the slow neonatal acetylcholine (ACh) channels to the more mature form may also be halted,*’ for this transition too is activity-dependent,~ This could, during the period of recovery from the paralysis, lead to a situation seen in muscles treated with cr-BTX immediately after birth and examined 5 days later, where a very prolonged end-plate potential is seen.” In this case, too, immediately after recovery from the paralysis, a loss of many contacts occurs. A prolonged depola~zation causes disruption of neuromuscular connections, For exposure of soleus muscles to ACh or to inhibitors of AChE is known to bring about a reduciton of neuromuscular contacts.5.6.‘7 It is likely that in the present experiments in the paralysed muscles a number of muscle fibres have end-plates with low levels of AChE, and these endplates may then rapidly loose contact with their nerve terminals during the period of recovery from the paralysis. Moreover these fibres, for the same reasons, will then be unable to accept innervation during later stages of recovery. Such a mechanism could explain the ~rmanent nature of this short-term paralysis. Acknowledgements-The authors would like to thank the Vivian Smith Foundation, The National Fund for Research into Crippling Diseases and the Wellcome Trust for support. They are grateful to Dr Dolly for the gift of the purified a-bungarotoxin.
REFERENCES
1. Bagust J., Lewis D. M. and Westerman R. A. (1973) Polyneuronal innervation of kitten skeletal muscle. J. Physiol. 229, 241-255. 2. Brown M. C., Holland R. L. and Hopkins W. G. (1981) Restoration of focal multiple innervation in rat muscles by transmission block during a critical stage of development. J. Physiol., Land. 318, 355-364. 3. Brown M. C., Jansen K. K. S. and Van Essen D. (1976) Poiyneuronal innervation of skeletal muscle in new-born rats and its elimination during maturation. J. Physiof. 261, 387-422. 4. Connold A. L., Evers J. V. and VrbovB G. (1986) Effect of low calcium and protease inhibitor on synapse elimination during postnatal development in the rat soleus muscle. Devl Brain Res. 28, 99-107. 5. Duxson M. J. (1982) The effect of postsynaptic block on development of the neuromuscular junction in postnatal rats. J. Neurocytol. 11, 395-408. 6. Duxson M. J. and Vrbova G. (1984) Inhibition of acetylcholinesterase accelerates axon terminal withdrawal. 1. Neurocytol. 14, 337-363. 7. Edds M. V. and Small W. T. (1957) The behaviour of residual axons in partially denervated muscles of the monkey. J. exp. Med. 93, 207-215. 8. Fisher T. J., VrbovL G. and Widjetunge A. (1989) Partial denervation of the rat soleus muscle at two different development stages. Neuroscience 28, 755-763. 9. Gordon T., Tuffery A., Perry R. and Vrbovi G. (1974) Possible mechanisms determining synapse formation in developing skeletal muscles of the chick. Cell Tiss. Res. 155, 13-25. 10. Greensmith L., LeHouelleur J. and VrbovB G. (1988) The effect of early muscles paralysis on the distribution of inne~ation in the newborn rat. J. Physiol. 401, 52P. 11. Hoffman H. (1950) Local reinnervation of partially denervated muscle: a histophysiological study. Aust. J. exp. Biol. Med. Sci. 28, 383-397. 12. Lowrie M. B., O’Brien R. A. D. and Vrbovi G. (1985) the effect of altered peripheral field on motoneurone in developing rat soleus muscles. J. Physiol. 368, 513-524.
function
13. Nachlas M. M., Tson K. C., De Souza E., Chang C. H. and Se&man A. M. (1957) Cytochemical demonstration of succinate dehydrogenase by the use of a new p-nitrophenyl substituted ditetrazole. J. Hisfochem. Cytochem. S,420-436.
532
A. L. CONNOLD and G. VRBOV,~
14. Navarette R. and Vrbova G. (1983) Changes of activity patterns in slow and fast muscles during postnatal development. Devl Brain Res. 8, 11-19. 15. O’Brien R. A. D., &&erg A. J. C. and Vrbova G. (1978) Observations on the elimination of polyneuronal innervation in developing mammalian skeletal muscle. J. Physiol. 282, 571-582. 16. O’Brien R. A. D., &berg A. J. C. and Vrbova G. (1980) The effect of acetylcholine on the function and structure of the developing mammalian neuromuscular junction. Neuroscience 5, 1367-1379. 17. O’Brien R. A. D., &berg A. J. C. and Vrbova G. (1984) Protease inhibitors reduce the loss of nerve terminals induced by activity and calcium in developing rat soleus muscles in uitro. Neuroscience 12, 637-646. 18. Pestronk A. and Drachman D. B. (1985) Motor nerve terminal outgrowth and acetylcholine receptors: inhibition of terminal outgrowth by u-bungarotoxin and anti-acetylcholine receptor antibody. J. Neurosci. 5, 751-758. 19. Redfern P. A. (1970) Neuromuscular transmission in newborn rats. J. Physiol., Land. 209, 701-709. 20. Round J. M., Mathews Y. and Jones D. A. (1980) A quick simple and reliable histochemical method for ATPase in human muscle preparations. J. Histochem. 12, 701-710. 21. Sackman B. and Brenner H. R. (1978) Changes in synaptic channel gating during neuromuscular development. Nafure 276, 401-402. 22. Srihari T. and Vrbova G. (1978) The role of muscle activity in the differentiation of NMJs in slow and fast chick muscles. J. Neurocytol. 7, 5299540. 23. Thompson W. J. (1983) Lack of segmental selectivity in elimination of synapses from soleus muscle of new-born rats. J. Physiol. 335, 345-352. 24. Thompson W. J. and Jansen J. K. (1977) The extent of sprouting of remaining motor units in partially denervated immature and adult rat soleus muscle. Neuroscience 2, 523-535. (Accepted 4 September 1989)
