- Email: [email protected]
The Final Common Pathway in Postural Control—Developmental Perspective
Neuroscience and Biobehavioral Reviews, Vol. 22, No. 4, pp. 479–484, 1998 䉷 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain 0149-7634/98 $32.00 + .00
Pergamon
PII: S0149-7634(97)00033-X
The Final Common Pathway in Postural Control—Developmental Perspective D. KERNELL1 Department of Medical Physiology, University of Groningen, Bloemsingel 10, 9712 KZ Groningen, Netherlands
KERNELL, D. The final common pathway in postural control—developmental perspective. NEUROSCI BIOBEHAV REV 22(4)479– 484, 1998.—A brief review is given concerning postural specialisations among mammalian muscle fibres and motor units. Most skeletal muscles contain a mixture of fibres with different characteristics, and their slow-twitch (S) units are well-known to possess properties suitable for postural tasks: they are highly fatigue-resistant, well equipped for oxidative metabolism, and their slowness makes them energetically cheap in (semi-)isometric contractions. These features are adequately employed in motor behaviour owing to characteristics of the associated motoneurones. In adult mammals, the way in which a muscle is used can influence its proportion of S units. This adjustment occurs within a restricted ‘adaptive range’ which differs between muscles and animal species, presumably being preset at an early age. In the course of early foetal development, part of the slow vs. fast differentiation of muscle fibre properties can take place independently of innervation. Once innervation has taken place, however, motoneurones influence the differentiation in various ways. On the whole, a well coordinated timing seems to exist between the early differentiation of central motor mechanisms and of the peripheral machinery, largely causing the neuromuscular system to be/become ready for use when the brain needs it. 䉷 1998 Elsevier Science Ltd. All rights reserved. Skeletal muscle
Motoneurone
Motor unit
Posture
Speed
Plasticity
Development
Instead of having individual fibres combining all the necessary properties for these functional extremes, the contrasting muscular requirements in posture and movement are largely taken care of by muscle fibres with correspondingly contrasting properties. In fact, there is a wide and rather continuous range of variation in the contractile requirements, from near-isometric postural contractions lasting for hours (e.g. many antigravity muscles) to ballistic contractions providing maximal acceleration of limbs (e.g. in a jump). Correspondingly in individual muscles there is also a wide and rather continuous range of variation in many of the muscle fibre properties. Still, it is descriptively feasible and practical to subdivide this continuum into a number of major groups, such as the slow (S) fibres which are optimised for postural use and the fast (F) ones which are optimised for movements (comprising subvarieties with different degrees of fatigue sensitivity). In this brief review, neuromuscular function and differentiation will only be discussed using the simple S vs. F dichotomy. The following questions will be dealt with:
INTRODUCTION
IN GENERAL, the control of posture is necessarily strongly linked with that of movement: dynamic motor actions cannot be performed without (first) stabilising sections of the body that have to act as pivots for the moving parts. Furthermore, in behaving terrestrial animals there is also a permanent fight against the force of gravity, which is opposed by the continuous production of postural counter-force. Movements and posture do, however, put partly contrasting requirements on the final common pathway in motor control, the motoneurones and their muscle fibres. For the muscles such requirements include the following. 1. In posture: finely graded contractions of weak to moderate force, often of a (semi-)isometric nature. The contractile duration may be very long; hence, energy substrates should continuously be delivered from the blood and a good capacity for oxidative metabolism and an adequate capillary supply is essential. 2. In movements: rapidly regulated and brief (but often intermittently repeated) contractions, with possibilities available for the production of high force and power. Hence, energy will often have to be rapidly mobilised from internal stores (glycogen), sometimes under (semi-) anaerobic conditions due to the temporary occlusion of blood vessels. 1
Fatigue resistance
1. How S motoneurones and their muscle fibres (i.e. S muscle units) are optimised for postural functions. 2. How the percentage of S fibres and units in individual muscles may be made to vary in response to long-term changes of adult muscle use.
Tel.: þ 31-50-3632660; fax: þ 31-50-3632751; e-mail: [email protected].
479
480
KERNELL
3. How postural aspects of the neuromuscular system develop and differentiate in early life. NEUROMUSCULAR SPECIALISATIONS AND POSTURAL NEEDS
A crucial aspect of posture is that it often has to be continued for a long time, i.e. a high degree of endurance is absolutely essential. This property is optimised in the slow and oxidative S fibres to a remarkable degree compared with F fibres of the same muscle (for review and references see (6,20,25)). The slow contractile speed itself contributes to the high postural endurance because the maintenance of isometric contractile force is metabolically cheaper in slow than in fast fibres (e.g. (10)). S fibres (histochemically corresponding to ‘type I’ fibres) may be identified and counted/measured in histochemical preparations stained for myofibrillar ATPase. Muscles with marked postural functions have large numbers of S fibres. Muscles consisting only of S units are, however, rare exceptions (e.g. cat’s soleus muscle); practically all skeletal muscles are of a mixed composition. An appropriate employment of S units for posture, including the adequate use of their potentially high endurance, requires specialisations also in the neural component of the ‘final common path’—the motoneurones. In posture as well as in movement, the central nervous system controls force by (1) varying the number of active motoneurones (‘recruitment gradation’) and (2) varying the discharge rate of already recruited units (‘rate gradation’). The manner in which these gradation procedures are used together is important for optimising endurance in postural control (cf. (2)). In this context, the differentiation of membrane properties between S and F motoneurones is important. 1. The intrinsic membrane properties of S motoneurones give them a high electrical excitability (low current threshold for activation, see (25)) which makes them easier to activate synaptically than is the case for F motoneurones. Partly for this reason, the relatively small-axoned S motoneurones are usually preferentially recruited in weak contractions, including those used for postural control. This corresponds to recruitment according to Henneman’s ‘size principle’ (20); neuronal size itself does not, however, have a known causal role in this context. 2. Contractions can generally be better maintained at low than at high relative force levels of the individual muscle fibres. Therefore, it is important for endurance that steadily activated S motoneurones have properties that make them capable of discharging regularly at rates for which the twitches of their muscle fibres hardly fuse; such slow rates will occur when these motoneurones are just barely recruited. This ‘speed-match’ between minimum firing rate and muscle unit contractile speed reflects a similarity between the duration of a muscle unit twitch and the duration of the motoneuronal afterhyperpolarisation (AHP; (1,24)). Owing to the asynchronous activity of different motoneurones in the same pool, the sum of markedly unfused contractions in several recruited S units may result in a relatively steady muscle force. ADULT NEUROMUSCULAR PLASTICITY
It is well-known that the properties of skeletal muscles are modifiable by training (sport, rehabilitation, etc.).
Recruitment according to the ‘size principle’ automatically leads to different levels (and patterns) of activity for F vs. S muscle units. To what extent are the functional differences between these brands of units a long-term consequence of their different levels of ‘natural’ training? If an innervated fast-mixed muscle is subjected to extensive daily amounts (about 5% or more of total time) of electrical activation over several weeks, it becomes slower, more fatigue-resistant and, somewhat paradoxically, its muscle fibres also may become thinner and weaker (e.g. (4,12,13,26,28,37); for reviews and further references see (16,25)). The strengthening of normal muscles seems to require smaller amounts of daily training, including contractions of great force. The effects on the various contractile properties depend, in different ways, on the stimulation parameters. The total daily duration of activation is important for all effects. High pulse rates (i.e. strong evoked forces) have generally been found to be beneficial for the maintenance of muscle strength. Pulse rates have not, however, been found to be of critical importance for the effect of chronic stimulation on the speed of mixed-fast muscles (cf. (27,28)). Following cessation of chronic stimulation, muscle alterations revert back toward normal. This reversion happens more quickly for the slowed isometric speed and the decreased muscle force than for the increased fatigue resistance (4,27). It is clear that many of the striking consequences of chronic activation on muscle properties are expressed via effects on the machinery for gene transcription and RNAmediated protein synthesis. It is still unknown, however, how motoneuronal and contractile activity is linked to such effects and which intramuscular messenger molecules may be involved. Several of the effects can also be produced by direct electrical stimulation of denervated muscle fibres (e.g. (21,30,43)); thus, these long-term effects do, at least partly, come about as a direct consequence of the evoked muscle fibre activity. It is still unknown what the additional role might be of an activity-dependent synaptic release of particular molecules at the end-plate. There is much evidence indicating that, in the adult animal, muscle fibres are pre-specified with regard to the range within which their properties may be readily changed by alterations of use (‘adaptive range’). Thus, when comparing the long-term consequences of chronic stimulation between different experiments it has repeatedly been noted that the amount of effect can be very different for different species of animals (e.g. (38)) and for different muscles in the same animal (e.g. (43)). Conversely, a marked decrease in the amount of normal daily activity does not necessarily cause marked changes to take place in physiological and histochemical muscle fibre properties (e.g. (12,13,35)). Re-innervation experiments have shown that, although motoneurones may re-specify properties of many of the muscle fibres, biochemical characteristics of some fibres are markedly resistant to such neuronal influence ((40), cf. (22)). Besides long-term influences from motoneurones on muscle fibres (partly caused via motoneuronal effects on muscle fibre activity), there are also differentiating mechanisms working in the opposite direction. Thus, under certain conditions slow type-I fibres of adult animals even seem capable of changing the properties of re-innervating motoneurones from ‘fast’ to ‘slow’ with regard to their AHP and
NEUROMUSCULAR DIFFERENTIATION AND POSTURE
input resistance (for review see (31)). The molecular mechanisms involved in such actions are still unknown. Also in adult animals, motoneurone-independent mechanisms and agents may operate to change the distribution of muscle fibre properties within mixed muscles. A well-known example is thyroid hormone, which also has effects on denervated muscles (34), tending to downregulate the expression of type-I myosin and upregulate that of certain kinds of type-II myosin. In addition, muscle bulk may, of course, be markedly influenced by other hormones (e.g. growth hormone, anabolic steroids). NEUROMUSCULAR DEVELOPMENT
General course of events Mature muscle fibres arise from myotubes which come about from the fusion of mononuclear myoblasts. In normal development this process typically takes place in two main phases (for review and references see (23)): 1. first the primary myotubes arise, in the rat from at least about foetal day 12 (E12); 2. thereafter, secondary myotubes emerge using the primary generation of cells as a ‘scaffolding’. This process continues for varying times, in some species and muscles until well after birth. In man, much of the myotube and myofibre formation takes place in gestational weeks 7–19; already during this period different immunocytochemical fibre types can be distinguished. Different histochemical types start appearing around weeks 16–17 (quadriceps femoris, myosin ATPase; for further information and references, see ref. (7)). Motoneurones belong to the earliest nerve cells to appear within the spinal cord and their axons reach the developing muscle at the time of formation of the primary myotubes. About 50% of all the initial motoneurones ultimately die; in rat hindlimb pools, this process of cell death is completed prior to birth (23). Surviving neurones are thought to profit from beneficial retrograde effects from their target cells. The initial innervation of myotubes and muscle fibres is such that each muscle cell receives terminals from several motoneurones. In hindlimbs of rats, this situation persists until about 2–3 weeks postnatally. For instance, after about postnatal day 15 (P15) most soleus muscle fibres receive innervation from only one motoneurone, i.e. typically the motor units no longer overlap. One of the earliest physiological observations on muscular development concerned the early time course of changes of isometric speed. In rats as well as cats, the muscles that later develop into ‘slow’ (typically soleus) and those ultimately becoming predominantly ‘fast’ (the majority of mixed muscles) have a similarly slow isometric twitch at birth (e.g. (5,8)). Both sets of muscles rapidly become faster during the initial postnatal weeks. Subsequently, the two kinds of muscle deviate: soleus starts slowing down again while the ‘fast’ muscles remain fast. The changes of isometric twitch speed probably primarily reflect changes in the calcium kinetics associated with fibre activation and say little about alterations in the speed of shortening. When directly measured, the maximum speed of shortening has also been found to increase during the early postnatal period; in the rat this increase was, however,
481
not seen in slow but only in fast muscle (8). Myosin composition runs through a complex sequence of changes (e.g. (36)), including embryonic and neonatal myosin heavy-chain isoforms that are normally not found in adult muscle fibres. Role of motoneurones for the (early) differentiation of muscle fibre properties There is now much evidence that (part of) the differentiation of muscle fibres into different biochemical categories or ‘types’ can take place independently of the motoneurones. Also under aneural conditions different fibres will, for instance, express different final brands of myosin (rats: (9,22); chicken: (32)). Part of this process is related to the generations of initial myotubes, the first generation being preferentially slow and the second one initially fast (but adjustable in later life). Motoneuronal innervation is thought to be important for the appearance and ripening of the secondary fibres. It is largely unknown which other mechanisms may be responsible for the early differentiation sequence; thyroid hormone plays a role ((36)), at least for the speed at which changes occur ((11)). It is still unknown how the ‘adaptive ranges’ of the future adult muscle fibres are set. Ingrowing motoneurones apparently recognise different classes of emerging muscle fibres such that, ultimately, functionally homogeneous muscle units may arise because of an early matching based on cell recognition ((23)). Thus, in the early postnatal period of polyneuronal innervation, fibres within overlapping muscle units all tend to belong to the same ‘type’ ((39)). In the early postnatal period the regression of polyneuronal innervation is critically dependent on the presence of motoneuronal activity (for review see (23,33)). Activity also plays a complex role in the competition between different motoneurones for keeping a given muscle fibre. In the postnatal development of contractile muscle properties, the long-lasting tonic discharges of ‘slow’ motoneurones apparently exert a slowing effect on their muscle fibres. The secondary slowing of isometric contraction in the cat’s soleus muscle did not take place when the animals were made less tonically active by high-lumbar transection of the spinal cord ((5)). Similarly, in rats the percentage of slow postural fibres in the soleus gradually increases as the animal grows older and heavier ((29)). Thus, in these cases, postural needs apparently serve to adjust the properties of a markedly postural muscle already in early life. Within the limits somehow set by ‘adaptive ranges’, such adjustments remain possible throughout adult life (see above). Does the rate of early neuromuscular differentiation constitute a ‘bottleneck’ in the development of motor behaviour? In many species of animals, the neuromuscular properties have not yet reached their adult level of specialisation at birth. Most of the available knowledge concerns rats and cats who both are ‘altricial’ animals, being relatively helpless for some time after birth (as is also, obviously, the case for humans). The late development of their motor behaviour is probably, to a dominating extent, reflecting the rate of development of the central neuronal mechanisms
482
responsible for driving the spinal motoneurones. Still, this development seems to happen with a relatively parallel time course for the central mechanisms as for the neuromuscular system. Hence, it is of some interest to consider in which respects an immature neuromuscular system might be a limiting factor in the timing of appearance of adequate postural behaviour. Lifting the body off the ground during locomotion might be considered a postural ‘mile stone’. For rats, this maturation of locomotory posture takes place from about postnatal days P10–11 ((42); see also (15,17)); for cats, a secure weight-bearing posture is achieved at around P21 ((3)). Was this milestone reached at a given time because only then had the neuromuscular properties become ready for adequate postural use? Or were the required neuromuscular properties already there, waiting for the brain to put them to work? Below, such questions will be discussed in relation to a small selection of all the potential problems involved. ‘Tonic’ activity in motoneurones At around P10–P12 the motor activity of ‘postural’ muscles (e.g. soleus) in the rat starts to change from being dominated by brief bursts of ‘phasic’ activity to also showing long-lasting ‘tonic’ discharges, as would be needed for postural stabilisation ((33)). There is, however, no evidence that the generally phasic behaviour of neurones in younger animals depends on them being intrinsically incapable of delivering a long-lasting repetitive firing; such ‘tonic’ discharges have been evoked with injected currents in motoneurones of rats at day P8 or earlier ((14)). Speed-matching between motoneurone and muscle If the AHPs of neonatal motoneurones were as brief as those of adults, then they would not be well matched to the prolonged neonatal twitch durations; the minimum steady discharge rates of neonatal motoneurones would then be too high for the production of well maintained contractions (i.e. too much twitch fusion). This highly unmatched situation apparently does not appear because not only the twitches but also the AHPs are longer in neonatal than in adult rats. In the data of ref. (14) (calculated from their table 1), the average (⫾SD) AHP duration of unspecified neonatal lumbar motoneurones was 120 ⫾ 26 ms (n ¼ 33) and, during the period covered by these measurements (P3–P12), there was no significant correlation with age (r ¼ 0.03, n ¼ 33). In adult rats, the average AHP duration for gastrocnemius medialis and tibialis anterior motoneurones is 54 ⫾ 14 ms (n ¼ 125), which is significantly shorter than the neonatal value (P p 0.001; adult data from ref. (1)). Thus, there is apparently a general tendency also for a certain degree of AHP vs. twitch duration matching in neonatal animals. However, it is still unknown how well this match is set and maintained throughout the postnatal period for the slow postural units. In the cat, an evident co-variation between twitch speed and AHP duration is present from about 6 weeks of age, not just after birth ((19)). In predominantly fast rat muscles, twitch duration seems to speed up by a factor of about 4 from birth to adulthood (see illustrations of (8)); this change appears greater than the average shortening of AHP duration (the data above suggest an AHP shortening by about 120/54 ¼ 2.2). However, during the first couple of weeks, the relationship between twitch and AHP duration will quickly readjust
KERNELL
itself toward more adult-like conditions because in this initial period there is already a marked speeding-up of the twitch ((8)), without as yet any simultaneous shortening of the AHP ((14); cf. data from cat in (19)). Such a sequence of events (twitch shortening prior to AHP shortening) would be consistent with the presence of retrograde ‘speed-setting’ effects of muscle fibres onto their motoneurones (cf. (31)). Multiply innervated units Even though multiply innervated units may be functionally relatively homogeneous ((39)), the state of polyneuronal innervation would be highly undesirable in postural functions. In centrally evoked activity, endurance would be endangered by using single muscle fibres at too high levels of contractile force, such as those caused by the combined effects of slow but asynchronous discharges from several innervating motoneurones. Hence, it seems of direct relevance that, in the rat, the gradual disappearance of polyneuronal innervation of soleus (P9–P16; (23)) almost precisely overlaps with the period of gradual development of locomotory posture (P11–P15; (42)). Development of fatigue resistance The fatigue resistance of developing hindlimb muscles has not often been directly measured by methods such as those regularly employed in adult muscle, using electrical stimulation of the muscle nerve for standardised fatigue tests ((6)). Some such data are available for the cat, indicating that both future ‘slow’ and ‘fast’ muscles have a high fatigue resistance initially (measured from postnatal day P6–P9; (18)). Thus, ‘postural’ muscles apparently do not need (much) postnatal activity (‘training’) to acquire their normally high degree of fatigue resistance. In the first few postnatal weeks of kittens, endurance remains high for the ‘postural’ soleus muscle while the fast-mixed tibialis anterior becomes gradually more sensitive to fatigue (18). Force? One factor of crucial importance for the postural (antigravity) use of neonatal limb muscles concerns their capacity for generating sufficient force (torque) in long-lasting contractions, i.e. in contractions weak enough not to block circulation or otherwise compromise endurance. Whether an immaturity of force-production is a limiting factor of importance for the timing of the early appearance of motor behaviour is still unknown. Providing an answer to this question is not a simple task; extensive biomechanical measurements and calculations for all the relevant muscles, joints and moment-arms would be required. EPILOGUE
The data reviewed in Section 4 suggest that there is, on the whole, a well coordinated timing between the early differentiation of the central motor mechanisms and the peripheral machinery, largely causing the neuromuscular system to be/become ready for use when the brain needs it. This is further illustrated by comparisons between the time course of neuromuscular changes in altricial (e.g. cats, rats; helpless at birth) and precocial animals (e.g. sheep; can walk from birth). For instance, the early speeding-up of the isometric twitch of fast muscles, which occurs during the
NEUROMUSCULAR DIFFERENTIATION AND POSTURE
initial postnatal weeks in cats and rats, takes place well before birth in the precocial sheep (41). In these latter investigations, there was no evidence indicating that foetal movements were of importance for foetal muscle differentiation. Hence, this is yet another observation stressing the
483
largely ‘preprogrammed’ nature of early neuromuscular development. The role of motoneuronal activity and ‘training’ seems typically that of producing auxiliary (but potentially very important) adjustments rather than being a primary shaping force.
REFERENCES 1. Bakels, R. and Kernell, D., ‘Average’ but not ‘continuous’ speedmatch between motoneurons and muscle units of rat tibialis anterior. Journal of Neurophysiology, 1993, 70, 1300–1306. 2. Bakels, R. and Kernell, D., Threshold spacing in motoneurone pools of rat and cat: possible relevance for manner of force gradation. Experimental Brain Research, 1994, 102, 69–74. 3. Bradley, N. S. and Smith, J. L., Neuromuscular patterns of stereotypic hindlimb behaviors in the first two postnatal months. III. Scratching and the paw-shake response in kittens. Developmental Brain Research, 1988, 38, 69–82. 4. Brown, J. M. C., Henriksson, J. and Salmons, S., Restoration of fast muscle characteristics following cessation of chronic stimulation— physiological, histochemical and metabolic changes during slow-tofast transformation. Proceedings of the Royal Society of London, Series B, 1989, 235, 321–346. 5. Buller, A. J., Eccles, J. C. and Eccles, R. M., Differentiation of fast and slow muscles in the cat hindlimb. Journal of Physiology (London), 1960, 150, 399–416. 6. Burke, R. E., Motor units: anatomy, physiology and functional organization. In Handbook of Physiology—1, Vol. II, Part 1, ed. V. B. Brooks. American Physiological Society, Bethesda, MD, 1981, pp. 345–422. 7. Butler-Browne, G. S., Barbet, J. P. and Thornell, L.-E., Myosin heavy and light chain expression during human skeletal muscle development and precocious muscle maturation induced by thyroid hormone. Anatomy and Embryology, 1990, 181, 513–522. 8. Close, R., Dynamic properties of fast and slow skeletal muscles of mammals. In Exploratory Concepts in Muscular Dystrophy and Related Disorders, ed. A. T. Milhorat. Excerpta Medica, Amsterdam, 1967, pp. 142–149. 9. Condon, K., Silberstein, L., Blau, H. M. and Thompson, W. J., Differentiation of fiber types in aneural musculature of the prenatal rat hindlimb. Developmental Biology, 1990, 138, 275–295. 10. Crow, M. T. and Kushmerick, M. J., Chemical energetics of slow- and fast-twitch muscles of the mouse. Journal of General Physiology, 1982, 79, 147–166. 11. d’Albis, A., Chanoine, C., Janmot, C., Mira, J. C. and Couteaux, R., Muscle-specific response to thyroid hormone of myosin isoform transitions during rat postnatal development. European Journal of Biochemistry, 1990, 193, 155–161. 12. Donselaar, Y., Eerbeek, O., Kernell, D. and Verhey, B. A., Fibre sizes and histochemical staining characteristics in normal and chronically stimulated fast muscle of cat. Journal of Physiology (London), 1987, 382, 237–254. 13. Eerbeek, O., Kernell, D. and Verhey, B. A., Effects of fast and slow patterns of tonic long-term stimulation on contractile properties of fast muscle in the cat. Journal of Physiology (London), 1984, 352, 73–90. 14. Fulton, B. P. and Walton, K., Electrophysiological properties of neonatal rat motoneuones studied in vitro. Journal of Physiology (London), 1986, 370, 651–678. 15. Geisler, H. C. and Gramsbergen, A., Motor development after vestibular deprivation in rats. Neurosci. Biobehav. Rev., 1998, 22, 565–569. 16. Gordon, T., Fatigue in adapted systems. Overuse and underuse paradigms. In Fatigue. Neural and Muscular Mechanisms, ed. S. C. Gandevia, R. M. Enoka, A. J. McComas, D. G. Stuart and C. K. Thomas. Plenum, New York, 1995, pp. 429–456. 17. Gramsbergen, A., Posture and locomotion in the rat: inter- or independent development. Neurosci. Biobehav. Rev., 1998, 22, 547–553. 18. Hammarberg, C. and Kellerth, J.-O., The postnatal development of some twitch and fatigue properties of the ankle flexor and extensor muscles of the cat. Acta Physiologica Scandinavica, 1975, 95, 166–178. 19. Hammarberg, C. and Kellerth, J.-O., The postnatal development of some twitch and fatigue properties of single motor units in the ankle muscles of the kitten. Acta Physiologica Scandinavica, 1975, 95, 243– 257.
20. Henneman, E. and Mendell, L. M., Functional organization of motoneuron pool and its inputs. In Handbook of Physiology—1, Vol. II, Part 1, ed. V. B. Brooks. American Physiological Society, Bethesda, MD, 1981, pp. 423–507. 21. Henriksson, J., Galbo, H. and Blomstrand, E., Role of the motor nerve in activity-induced enzymatic adaptation in skeletal muscle. American Journal of Physiology, 1982, 242, C272–C277. 22. Hoh, J. F. Y., Myogenic regulation of mammalian skeletal muscle fibres. News in Physiological Sciences, 1991, 6, 1–6. 23. Jansen, J. K. S. and Fladby, T., The perinatal reorganization of the innervation of skeletal muscle in mammals. Progress in Neurobiology, 1990, 34, 39–90. 24. Kernell, D., The limits of firing frequency in cat lumbosacral motoneurones possessing different time course of afterhyperpolarization. Acta Physiologica Scandinavica, 1965, 65, 87–100. 25. Kernell, D., Organized variability in the neuromuscular system: A survey of task-related adaptations. Archives Italiennes de Biologie, 1992, 130, 19–66. 26. Kernell, D., Donselaar, Y. and Eerbeek, O., Effects of physiological amounts of high- and low-rate chronic stimulation on fast-twitch muscle of the cat hindlimb. II. Endurance-related properties. Journal of Neurophysiology, 1987, 58, 614–627. 27. Kernell, D. and Eerbeek, O., Recovery following intense chronic stimulation: A physiological study of cat’s fast muscle. Journal of Applied Physiology, 1991, 70, 1763–1769. 28. Kernell, D., Eerbeek, O., Verhey, B. A. and Donselaar, Y., Effects of physiological amounts of high- and low-rate chronic stimulation on fast-twitch muscle of the cat hindlimb. I. Speed- and force-related properties. Journal of Neurophysiology, 1987, 58, 598–613. 29. Kugelberg, E., Adaptive transformation of rat soleus motor units during growth. Journal of Neurological Sciences, 1976, 27, 269–289. 30. Lømo, T., Westgaard, R. H. and Dahl, H. A., Contractile properties of muscle: control by pattern of muscle activity in the rat. Proceedings of the Royal Society of London, Series B, 1974, 187, 99–103. 31. Mendell, L. M., Collins, W. F. III and Munson, J. B., Retrograde determination of motoneuron properties and their synaptic input. Journal of Neurobiology, 1994, 25, 707–721. 32. Miller, J. B. and Stockdale, F. E., What muscle cells know that nerves don’t tell them. Trends in Neuroscience, 1987, 10, 325–329. 33. Navarrete, R. and Vrbova´, G., Activity-dependent interactions between motoneurones and muscles: their role in the development of the motor unit. Progress in Neurobiology, 1993, 41, 93–124. 34. Nwoye, L., Mommaerts, W. F. H. M., Simpson, D. R., Seraydarian, K. and Marusich, M., Evidence for a direct action of thyroid hormone in specifying muscle properties. American Journal of Physiology, 1982, 242, R401–R408. 35. Pierotti, D. J., Roy, R. R., Bodine-Fowler, S. C., Hodgson, J. A. and Edgerton, V. R., Mechanical and morphological properties of chronically inactive cat tibialis anterior motor units. Journal of Physiology (London), 1991, 444, 175–192. 36. Rubinstein, N. A., Lyons, G. E. and Kelly, A. M., Hormonal control of myosin heavy chain genes during development of skeletal muscles. In Plasticity of the Neuromuscular System. Ciba Foundation Symposium 138. Wiley, Chichester, 1988, pp. 35–51. 37. Salmons, S. and Vrbova´, G., The influence of activity on some contractile characteristics of mammalian fast and slow muscles. Journal of Physiology (London), 1969, 201, 535–549. 38. Simoneau, J.-A. and Pette, D., Species-specific effects of chronic nerve stimulation upon tibialis anterior muscle in mouse, rat, guinea pig, and rabbit. Pflu¨gers Archiv, 1988, 412, 86–92. 39. Thompson, W. J., Soileau, L. C., Balice-Gordon, R. J. and Sutton, L. A., Selective innervation of types of fibres in developing rat muscle. Journal of Experimental Biology, 1987, 132, 249–263. 40. Unguez, G. A., Roy, R. R., Pierotti, D. J., Bodine-Fowler, S. and Edgerton, V. R., Further evidence of incomplete neural control of
484 muscle properties in cat tibialis anterior motor units. American Journal of Physiology, 1995, 268, C527–C534. 41. Walker, D. W. and Luff, A. R., Functional development of fetal limb muscles: a review of the roles of activity, nerves and hormones. Reproduction, Fertility and Development, 1995, 7, 391–398.
KERNELL 42. Westerga, J. and Gramsbergen, A., Development of locomotion in the rat. Developmental Brain Research, 1990, 57, 163–174. 43. Westgaard, R. H. and Lømo, T., Control of contractile properties within adaptive ranges by patterns of impulse activity in the rat. Journal of Neuroscience, 1988, 8, 4415–4426.
Pergamon
PII: S0149-7634(97)00033-X
The Final Common Pathway in Postural Control—Developmental Perspective D. KERNELL1 Department of Medical Physiology, University of Groningen, Bloemsingel 10, 9712 KZ Groningen, Netherlands
KERNELL, D. The final common pathway in postural control—developmental perspective. NEUROSCI BIOBEHAV REV 22(4)479– 484, 1998.—A brief review is given concerning postural specialisations among mammalian muscle fibres and motor units. Most skeletal muscles contain a mixture of fibres with different characteristics, and their slow-twitch (S) units are well-known to possess properties suitable for postural tasks: they are highly fatigue-resistant, well equipped for oxidative metabolism, and their slowness makes them energetically cheap in (semi-)isometric contractions. These features are adequately employed in motor behaviour owing to characteristics of the associated motoneurones. In adult mammals, the way in which a muscle is used can influence its proportion of S units. This adjustment occurs within a restricted ‘adaptive range’ which differs between muscles and animal species, presumably being preset at an early age. In the course of early foetal development, part of the slow vs. fast differentiation of muscle fibre properties can take place independently of innervation. Once innervation has taken place, however, motoneurones influence the differentiation in various ways. On the whole, a well coordinated timing seems to exist between the early differentiation of central motor mechanisms and of the peripheral machinery, largely causing the neuromuscular system to be/become ready for use when the brain needs it. 䉷 1998 Elsevier Science Ltd. All rights reserved. Skeletal muscle
Motoneurone
Motor unit
Posture
Speed
Plasticity
Development
Instead of having individual fibres combining all the necessary properties for these functional extremes, the contrasting muscular requirements in posture and movement are largely taken care of by muscle fibres with correspondingly contrasting properties. In fact, there is a wide and rather continuous range of variation in the contractile requirements, from near-isometric postural contractions lasting for hours (e.g. many antigravity muscles) to ballistic contractions providing maximal acceleration of limbs (e.g. in a jump). Correspondingly in individual muscles there is also a wide and rather continuous range of variation in many of the muscle fibre properties. Still, it is descriptively feasible and practical to subdivide this continuum into a number of major groups, such as the slow (S) fibres which are optimised for postural use and the fast (F) ones which are optimised for movements (comprising subvarieties with different degrees of fatigue sensitivity). In this brief review, neuromuscular function and differentiation will only be discussed using the simple S vs. F dichotomy. The following questions will be dealt with:
INTRODUCTION
IN GENERAL, the control of posture is necessarily strongly linked with that of movement: dynamic motor actions cannot be performed without (first) stabilising sections of the body that have to act as pivots for the moving parts. Furthermore, in behaving terrestrial animals there is also a permanent fight against the force of gravity, which is opposed by the continuous production of postural counter-force. Movements and posture do, however, put partly contrasting requirements on the final common pathway in motor control, the motoneurones and their muscle fibres. For the muscles such requirements include the following. 1. In posture: finely graded contractions of weak to moderate force, often of a (semi-)isometric nature. The contractile duration may be very long; hence, energy substrates should continuously be delivered from the blood and a good capacity for oxidative metabolism and an adequate capillary supply is essential. 2. In movements: rapidly regulated and brief (but often intermittently repeated) contractions, with possibilities available for the production of high force and power. Hence, energy will often have to be rapidly mobilised from internal stores (glycogen), sometimes under (semi-) anaerobic conditions due to the temporary occlusion of blood vessels. 1
Fatigue resistance
1. How S motoneurones and their muscle fibres (i.e. S muscle units) are optimised for postural functions. 2. How the percentage of S fibres and units in individual muscles may be made to vary in response to long-term changes of adult muscle use.
Tel.: þ 31-50-3632660; fax: þ 31-50-3632751; e-mail: [email protected].
479
480
KERNELL
3. How postural aspects of the neuromuscular system develop and differentiate in early life. NEUROMUSCULAR SPECIALISATIONS AND POSTURAL NEEDS
A crucial aspect of posture is that it often has to be continued for a long time, i.e. a high degree of endurance is absolutely essential. This property is optimised in the slow and oxidative S fibres to a remarkable degree compared with F fibres of the same muscle (for review and references see (6,20,25)). The slow contractile speed itself contributes to the high postural endurance because the maintenance of isometric contractile force is metabolically cheaper in slow than in fast fibres (e.g. (10)). S fibres (histochemically corresponding to ‘type I’ fibres) may be identified and counted/measured in histochemical preparations stained for myofibrillar ATPase. Muscles with marked postural functions have large numbers of S fibres. Muscles consisting only of S units are, however, rare exceptions (e.g. cat’s soleus muscle); practically all skeletal muscles are of a mixed composition. An appropriate employment of S units for posture, including the adequate use of their potentially high endurance, requires specialisations also in the neural component of the ‘final common path’—the motoneurones. In posture as well as in movement, the central nervous system controls force by (1) varying the number of active motoneurones (‘recruitment gradation’) and (2) varying the discharge rate of already recruited units (‘rate gradation’). The manner in which these gradation procedures are used together is important for optimising endurance in postural control (cf. (2)). In this context, the differentiation of membrane properties between S and F motoneurones is important. 1. The intrinsic membrane properties of S motoneurones give them a high electrical excitability (low current threshold for activation, see (25)) which makes them easier to activate synaptically than is the case for F motoneurones. Partly for this reason, the relatively small-axoned S motoneurones are usually preferentially recruited in weak contractions, including those used for postural control. This corresponds to recruitment according to Henneman’s ‘size principle’ (20); neuronal size itself does not, however, have a known causal role in this context. 2. Contractions can generally be better maintained at low than at high relative force levels of the individual muscle fibres. Therefore, it is important for endurance that steadily activated S motoneurones have properties that make them capable of discharging regularly at rates for which the twitches of their muscle fibres hardly fuse; such slow rates will occur when these motoneurones are just barely recruited. This ‘speed-match’ between minimum firing rate and muscle unit contractile speed reflects a similarity between the duration of a muscle unit twitch and the duration of the motoneuronal afterhyperpolarisation (AHP; (1,24)). Owing to the asynchronous activity of different motoneurones in the same pool, the sum of markedly unfused contractions in several recruited S units may result in a relatively steady muscle force. ADULT NEUROMUSCULAR PLASTICITY
It is well-known that the properties of skeletal muscles are modifiable by training (sport, rehabilitation, etc.).
Recruitment according to the ‘size principle’ automatically leads to different levels (and patterns) of activity for F vs. S muscle units. To what extent are the functional differences between these brands of units a long-term consequence of their different levels of ‘natural’ training? If an innervated fast-mixed muscle is subjected to extensive daily amounts (about 5% or more of total time) of electrical activation over several weeks, it becomes slower, more fatigue-resistant and, somewhat paradoxically, its muscle fibres also may become thinner and weaker (e.g. (4,12,13,26,28,37); for reviews and further references see (16,25)). The strengthening of normal muscles seems to require smaller amounts of daily training, including contractions of great force. The effects on the various contractile properties depend, in different ways, on the stimulation parameters. The total daily duration of activation is important for all effects. High pulse rates (i.e. strong evoked forces) have generally been found to be beneficial for the maintenance of muscle strength. Pulse rates have not, however, been found to be of critical importance for the effect of chronic stimulation on the speed of mixed-fast muscles (cf. (27,28)). Following cessation of chronic stimulation, muscle alterations revert back toward normal. This reversion happens more quickly for the slowed isometric speed and the decreased muscle force than for the increased fatigue resistance (4,27). It is clear that many of the striking consequences of chronic activation on muscle properties are expressed via effects on the machinery for gene transcription and RNAmediated protein synthesis. It is still unknown, however, how motoneuronal and contractile activity is linked to such effects and which intramuscular messenger molecules may be involved. Several of the effects can also be produced by direct electrical stimulation of denervated muscle fibres (e.g. (21,30,43)); thus, these long-term effects do, at least partly, come about as a direct consequence of the evoked muscle fibre activity. It is still unknown what the additional role might be of an activity-dependent synaptic release of particular molecules at the end-plate. There is much evidence indicating that, in the adult animal, muscle fibres are pre-specified with regard to the range within which their properties may be readily changed by alterations of use (‘adaptive range’). Thus, when comparing the long-term consequences of chronic stimulation between different experiments it has repeatedly been noted that the amount of effect can be very different for different species of animals (e.g. (38)) and for different muscles in the same animal (e.g. (43)). Conversely, a marked decrease in the amount of normal daily activity does not necessarily cause marked changes to take place in physiological and histochemical muscle fibre properties (e.g. (12,13,35)). Re-innervation experiments have shown that, although motoneurones may re-specify properties of many of the muscle fibres, biochemical characteristics of some fibres are markedly resistant to such neuronal influence ((40), cf. (22)). Besides long-term influences from motoneurones on muscle fibres (partly caused via motoneuronal effects on muscle fibre activity), there are also differentiating mechanisms working in the opposite direction. Thus, under certain conditions slow type-I fibres of adult animals even seem capable of changing the properties of re-innervating motoneurones from ‘fast’ to ‘slow’ with regard to their AHP and
NEUROMUSCULAR DIFFERENTIATION AND POSTURE
input resistance (for review see (31)). The molecular mechanisms involved in such actions are still unknown. Also in adult animals, motoneurone-independent mechanisms and agents may operate to change the distribution of muscle fibre properties within mixed muscles. A well-known example is thyroid hormone, which also has effects on denervated muscles (34), tending to downregulate the expression of type-I myosin and upregulate that of certain kinds of type-II myosin. In addition, muscle bulk may, of course, be markedly influenced by other hormones (e.g. growth hormone, anabolic steroids). NEUROMUSCULAR DEVELOPMENT
General course of events Mature muscle fibres arise from myotubes which come about from the fusion of mononuclear myoblasts. In normal development this process typically takes place in two main phases (for review and references see (23)): 1. first the primary myotubes arise, in the rat from at least about foetal day 12 (E12); 2. thereafter, secondary myotubes emerge using the primary generation of cells as a ‘scaffolding’. This process continues for varying times, in some species and muscles until well after birth. In man, much of the myotube and myofibre formation takes place in gestational weeks 7–19; already during this period different immunocytochemical fibre types can be distinguished. Different histochemical types start appearing around weeks 16–17 (quadriceps femoris, myosin ATPase; for further information and references, see ref. (7)). Motoneurones belong to the earliest nerve cells to appear within the spinal cord and their axons reach the developing muscle at the time of formation of the primary myotubes. About 50% of all the initial motoneurones ultimately die; in rat hindlimb pools, this process of cell death is completed prior to birth (23). Surviving neurones are thought to profit from beneficial retrograde effects from their target cells. The initial innervation of myotubes and muscle fibres is such that each muscle cell receives terminals from several motoneurones. In hindlimbs of rats, this situation persists until about 2–3 weeks postnatally. For instance, after about postnatal day 15 (P15) most soleus muscle fibres receive innervation from only one motoneurone, i.e. typically the motor units no longer overlap. One of the earliest physiological observations on muscular development concerned the early time course of changes of isometric speed. In rats as well as cats, the muscles that later develop into ‘slow’ (typically soleus) and those ultimately becoming predominantly ‘fast’ (the majority of mixed muscles) have a similarly slow isometric twitch at birth (e.g. (5,8)). Both sets of muscles rapidly become faster during the initial postnatal weeks. Subsequently, the two kinds of muscle deviate: soleus starts slowing down again while the ‘fast’ muscles remain fast. The changes of isometric twitch speed probably primarily reflect changes in the calcium kinetics associated with fibre activation and say little about alterations in the speed of shortening. When directly measured, the maximum speed of shortening has also been found to increase during the early postnatal period; in the rat this increase was, however,
481
not seen in slow but only in fast muscle (8). Myosin composition runs through a complex sequence of changes (e.g. (36)), including embryonic and neonatal myosin heavy-chain isoforms that are normally not found in adult muscle fibres. Role of motoneurones for the (early) differentiation of muscle fibre properties There is now much evidence that (part of) the differentiation of muscle fibres into different biochemical categories or ‘types’ can take place independently of the motoneurones. Also under aneural conditions different fibres will, for instance, express different final brands of myosin (rats: (9,22); chicken: (32)). Part of this process is related to the generations of initial myotubes, the first generation being preferentially slow and the second one initially fast (but adjustable in later life). Motoneuronal innervation is thought to be important for the appearance and ripening of the secondary fibres. It is largely unknown which other mechanisms may be responsible for the early differentiation sequence; thyroid hormone plays a role ((36)), at least for the speed at which changes occur ((11)). It is still unknown how the ‘adaptive ranges’ of the future adult muscle fibres are set. Ingrowing motoneurones apparently recognise different classes of emerging muscle fibres such that, ultimately, functionally homogeneous muscle units may arise because of an early matching based on cell recognition ((23)). Thus, in the early postnatal period of polyneuronal innervation, fibres within overlapping muscle units all tend to belong to the same ‘type’ ((39)). In the early postnatal period the regression of polyneuronal innervation is critically dependent on the presence of motoneuronal activity (for review see (23,33)). Activity also plays a complex role in the competition between different motoneurones for keeping a given muscle fibre. In the postnatal development of contractile muscle properties, the long-lasting tonic discharges of ‘slow’ motoneurones apparently exert a slowing effect on their muscle fibres. The secondary slowing of isometric contraction in the cat’s soleus muscle did not take place when the animals were made less tonically active by high-lumbar transection of the spinal cord ((5)). Similarly, in rats the percentage of slow postural fibres in the soleus gradually increases as the animal grows older and heavier ((29)). Thus, in these cases, postural needs apparently serve to adjust the properties of a markedly postural muscle already in early life. Within the limits somehow set by ‘adaptive ranges’, such adjustments remain possible throughout adult life (see above). Does the rate of early neuromuscular differentiation constitute a ‘bottleneck’ in the development of motor behaviour? In many species of animals, the neuromuscular properties have not yet reached their adult level of specialisation at birth. Most of the available knowledge concerns rats and cats who both are ‘altricial’ animals, being relatively helpless for some time after birth (as is also, obviously, the case for humans). The late development of their motor behaviour is probably, to a dominating extent, reflecting the rate of development of the central neuronal mechanisms
482
responsible for driving the spinal motoneurones. Still, this development seems to happen with a relatively parallel time course for the central mechanisms as for the neuromuscular system. Hence, it is of some interest to consider in which respects an immature neuromuscular system might be a limiting factor in the timing of appearance of adequate postural behaviour. Lifting the body off the ground during locomotion might be considered a postural ‘mile stone’. For rats, this maturation of locomotory posture takes place from about postnatal days P10–11 ((42); see also (15,17)); for cats, a secure weight-bearing posture is achieved at around P21 ((3)). Was this milestone reached at a given time because only then had the neuromuscular properties become ready for adequate postural use? Or were the required neuromuscular properties already there, waiting for the brain to put them to work? Below, such questions will be discussed in relation to a small selection of all the potential problems involved. ‘Tonic’ activity in motoneurones At around P10–P12 the motor activity of ‘postural’ muscles (e.g. soleus) in the rat starts to change from being dominated by brief bursts of ‘phasic’ activity to also showing long-lasting ‘tonic’ discharges, as would be needed for postural stabilisation ((33)). There is, however, no evidence that the generally phasic behaviour of neurones in younger animals depends on them being intrinsically incapable of delivering a long-lasting repetitive firing; such ‘tonic’ discharges have been evoked with injected currents in motoneurones of rats at day P8 or earlier ((14)). Speed-matching between motoneurone and muscle If the AHPs of neonatal motoneurones were as brief as those of adults, then they would not be well matched to the prolonged neonatal twitch durations; the minimum steady discharge rates of neonatal motoneurones would then be too high for the production of well maintained contractions (i.e. too much twitch fusion). This highly unmatched situation apparently does not appear because not only the twitches but also the AHPs are longer in neonatal than in adult rats. In the data of ref. (14) (calculated from their table 1), the average (⫾SD) AHP duration of unspecified neonatal lumbar motoneurones was 120 ⫾ 26 ms (n ¼ 33) and, during the period covered by these measurements (P3–P12), there was no significant correlation with age (r ¼ 0.03, n ¼ 33). In adult rats, the average AHP duration for gastrocnemius medialis and tibialis anterior motoneurones is 54 ⫾ 14 ms (n ¼ 125), which is significantly shorter than the neonatal value (P p 0.001; adult data from ref. (1)). Thus, there is apparently a general tendency also for a certain degree of AHP vs. twitch duration matching in neonatal animals. However, it is still unknown how well this match is set and maintained throughout the postnatal period for the slow postural units. In the cat, an evident co-variation between twitch speed and AHP duration is present from about 6 weeks of age, not just after birth ((19)). In predominantly fast rat muscles, twitch duration seems to speed up by a factor of about 4 from birth to adulthood (see illustrations of (8)); this change appears greater than the average shortening of AHP duration (the data above suggest an AHP shortening by about 120/54 ¼ 2.2). However, during the first couple of weeks, the relationship between twitch and AHP duration will quickly readjust
KERNELL
itself toward more adult-like conditions because in this initial period there is already a marked speeding-up of the twitch ((8)), without as yet any simultaneous shortening of the AHP ((14); cf. data from cat in (19)). Such a sequence of events (twitch shortening prior to AHP shortening) would be consistent with the presence of retrograde ‘speed-setting’ effects of muscle fibres onto their motoneurones (cf. (31)). Multiply innervated units Even though multiply innervated units may be functionally relatively homogeneous ((39)), the state of polyneuronal innervation would be highly undesirable in postural functions. In centrally evoked activity, endurance would be endangered by using single muscle fibres at too high levels of contractile force, such as those caused by the combined effects of slow but asynchronous discharges from several innervating motoneurones. Hence, it seems of direct relevance that, in the rat, the gradual disappearance of polyneuronal innervation of soleus (P9–P16; (23)) almost precisely overlaps with the period of gradual development of locomotory posture (P11–P15; (42)). Development of fatigue resistance The fatigue resistance of developing hindlimb muscles has not often been directly measured by methods such as those regularly employed in adult muscle, using electrical stimulation of the muscle nerve for standardised fatigue tests ((6)). Some such data are available for the cat, indicating that both future ‘slow’ and ‘fast’ muscles have a high fatigue resistance initially (measured from postnatal day P6–P9; (18)). Thus, ‘postural’ muscles apparently do not need (much) postnatal activity (‘training’) to acquire their normally high degree of fatigue resistance. In the first few postnatal weeks of kittens, endurance remains high for the ‘postural’ soleus muscle while the fast-mixed tibialis anterior becomes gradually more sensitive to fatigue (18). Force? One factor of crucial importance for the postural (antigravity) use of neonatal limb muscles concerns their capacity for generating sufficient force (torque) in long-lasting contractions, i.e. in contractions weak enough not to block circulation or otherwise compromise endurance. Whether an immaturity of force-production is a limiting factor of importance for the timing of the early appearance of motor behaviour is still unknown. Providing an answer to this question is not a simple task; extensive biomechanical measurements and calculations for all the relevant muscles, joints and moment-arms would be required. EPILOGUE
The data reviewed in Section 4 suggest that there is, on the whole, a well coordinated timing between the early differentiation of the central motor mechanisms and the peripheral machinery, largely causing the neuromuscular system to be/become ready for use when the brain needs it. This is further illustrated by comparisons between the time course of neuromuscular changes in altricial (e.g. cats, rats; helpless at birth) and precocial animals (e.g. sheep; can walk from birth). For instance, the early speeding-up of the isometric twitch of fast muscles, which occurs during the
NEUROMUSCULAR DIFFERENTIATION AND POSTURE
initial postnatal weeks in cats and rats, takes place well before birth in the precocial sheep (41). In these latter investigations, there was no evidence indicating that foetal movements were of importance for foetal muscle differentiation. Hence, this is yet another observation stressing the
483
largely ‘preprogrammed’ nature of early neuromuscular development. The role of motoneuronal activity and ‘training’ seems typically that of producing auxiliary (but potentially very important) adjustments rather than being a primary shaping force.
REFERENCES 1. Bakels, R. and Kernell, D., ‘Average’ but not ‘continuous’ speedmatch between motoneurons and muscle units of rat tibialis anterior. Journal of Neurophysiology, 1993, 70, 1300–1306. 2. Bakels, R. and Kernell, D., Threshold spacing in motoneurone pools of rat and cat: possible relevance for manner of force gradation. Experimental Brain Research, 1994, 102, 69–74. 3. Bradley, N. S. and Smith, J. L., Neuromuscular patterns of stereotypic hindlimb behaviors in the first two postnatal months. III. Scratching and the paw-shake response in kittens. Developmental Brain Research, 1988, 38, 69–82. 4. Brown, J. M. C., Henriksson, J. and Salmons, S., Restoration of fast muscle characteristics following cessation of chronic stimulation— physiological, histochemical and metabolic changes during slow-tofast transformation. Proceedings of the Royal Society of London, Series B, 1989, 235, 321–346. 5. Buller, A. J., Eccles, J. C. and Eccles, R. M., Differentiation of fast and slow muscles in the cat hindlimb. Journal of Physiology (London), 1960, 150, 399–416. 6. Burke, R. E., Motor units: anatomy, physiology and functional organization. In Handbook of Physiology—1, Vol. II, Part 1, ed. V. B. Brooks. American Physiological Society, Bethesda, MD, 1981, pp. 345–422. 7. Butler-Browne, G. S., Barbet, J. P. and Thornell, L.-E., Myosin heavy and light chain expression during human skeletal muscle development and precocious muscle maturation induced by thyroid hormone. Anatomy and Embryology, 1990, 181, 513–522. 8. Close, R., Dynamic properties of fast and slow skeletal muscles of mammals. In Exploratory Concepts in Muscular Dystrophy and Related Disorders, ed. A. T. Milhorat. Excerpta Medica, Amsterdam, 1967, pp. 142–149. 9. Condon, K., Silberstein, L., Blau, H. M. and Thompson, W. J., Differentiation of fiber types in aneural musculature of the prenatal rat hindlimb. Developmental Biology, 1990, 138, 275–295. 10. Crow, M. T. and Kushmerick, M. J., Chemical energetics of slow- and fast-twitch muscles of the mouse. Journal of General Physiology, 1982, 79, 147–166. 11. d’Albis, A., Chanoine, C., Janmot, C., Mira, J. C. and Couteaux, R., Muscle-specific response to thyroid hormone of myosin isoform transitions during rat postnatal development. European Journal of Biochemistry, 1990, 193, 155–161. 12. Donselaar, Y., Eerbeek, O., Kernell, D. and Verhey, B. A., Fibre sizes and histochemical staining characteristics in normal and chronically stimulated fast muscle of cat. Journal of Physiology (London), 1987, 382, 237–254. 13. Eerbeek, O., Kernell, D. and Verhey, B. A., Effects of fast and slow patterns of tonic long-term stimulation on contractile properties of fast muscle in the cat. Journal of Physiology (London), 1984, 352, 73–90. 14. Fulton, B. P. and Walton, K., Electrophysiological properties of neonatal rat motoneuones studied in vitro. Journal of Physiology (London), 1986, 370, 651–678. 15. Geisler, H. C. and Gramsbergen, A., Motor development after vestibular deprivation in rats. Neurosci. Biobehav. Rev., 1998, 22, 565–569. 16. Gordon, T., Fatigue in adapted systems. Overuse and underuse paradigms. In Fatigue. Neural and Muscular Mechanisms, ed. S. C. Gandevia, R. M. Enoka, A. J. McComas, D. G. Stuart and C. K. Thomas. Plenum, New York, 1995, pp. 429–456. 17. Gramsbergen, A., Posture and locomotion in the rat: inter- or independent development. Neurosci. Biobehav. Rev., 1998, 22, 547–553. 18. Hammarberg, C. and Kellerth, J.-O., The postnatal development of some twitch and fatigue properties of the ankle flexor and extensor muscles of the cat. Acta Physiologica Scandinavica, 1975, 95, 166–178. 19. Hammarberg, C. and Kellerth, J.-O., The postnatal development of some twitch and fatigue properties of single motor units in the ankle muscles of the kitten. Acta Physiologica Scandinavica, 1975, 95, 243– 257.
20. Henneman, E. and Mendell, L. M., Functional organization of motoneuron pool and its inputs. In Handbook of Physiology—1, Vol. II, Part 1, ed. V. B. Brooks. American Physiological Society, Bethesda, MD, 1981, pp. 423–507. 21. Henriksson, J., Galbo, H. and Blomstrand, E., Role of the motor nerve in activity-induced enzymatic adaptation in skeletal muscle. American Journal of Physiology, 1982, 242, C272–C277. 22. Hoh, J. F. Y., Myogenic regulation of mammalian skeletal muscle fibres. News in Physiological Sciences, 1991, 6, 1–6. 23. Jansen, J. K. S. and Fladby, T., The perinatal reorganization of the innervation of skeletal muscle in mammals. Progress in Neurobiology, 1990, 34, 39–90. 24. Kernell, D., The limits of firing frequency in cat lumbosacral motoneurones possessing different time course of afterhyperpolarization. Acta Physiologica Scandinavica, 1965, 65, 87–100. 25. Kernell, D., Organized variability in the neuromuscular system: A survey of task-related adaptations. Archives Italiennes de Biologie, 1992, 130, 19–66. 26. Kernell, D., Donselaar, Y. and Eerbeek, O., Effects of physiological amounts of high- and low-rate chronic stimulation on fast-twitch muscle of the cat hindlimb. II. Endurance-related properties. Journal of Neurophysiology, 1987, 58, 614–627. 27. Kernell, D. and Eerbeek, O., Recovery following intense chronic stimulation: A physiological study of cat’s fast muscle. Journal of Applied Physiology, 1991, 70, 1763–1769. 28. Kernell, D., Eerbeek, O., Verhey, B. A. and Donselaar, Y., Effects of physiological amounts of high- and low-rate chronic stimulation on fast-twitch muscle of the cat hindlimb. I. Speed- and force-related properties. Journal of Neurophysiology, 1987, 58, 598–613. 29. Kugelberg, E., Adaptive transformation of rat soleus motor units during growth. Journal of Neurological Sciences, 1976, 27, 269–289. 30. Lømo, T., Westgaard, R. H. and Dahl, H. A., Contractile properties of muscle: control by pattern of muscle activity in the rat. Proceedings of the Royal Society of London, Series B, 1974, 187, 99–103. 31. Mendell, L. M., Collins, W. F. III and Munson, J. B., Retrograde determination of motoneuron properties and their synaptic input. Journal of Neurobiology, 1994, 25, 707–721. 32. Miller, J. B. and Stockdale, F. E., What muscle cells know that nerves don’t tell them. Trends in Neuroscience, 1987, 10, 325–329. 33. Navarrete, R. and Vrbova´, G., Activity-dependent interactions between motoneurones and muscles: their role in the development of the motor unit. Progress in Neurobiology, 1993, 41, 93–124. 34. Nwoye, L., Mommaerts, W. F. H. M., Simpson, D. R., Seraydarian, K. and Marusich, M., Evidence for a direct action of thyroid hormone in specifying muscle properties. American Journal of Physiology, 1982, 242, R401–R408. 35. Pierotti, D. J., Roy, R. R., Bodine-Fowler, S. C., Hodgson, J. A. and Edgerton, V. R., Mechanical and morphological properties of chronically inactive cat tibialis anterior motor units. Journal of Physiology (London), 1991, 444, 175–192. 36. Rubinstein, N. A., Lyons, G. E. and Kelly, A. M., Hormonal control of myosin heavy chain genes during development of skeletal muscles. In Plasticity of the Neuromuscular System. Ciba Foundation Symposium 138. Wiley, Chichester, 1988, pp. 35–51. 37. Salmons, S. and Vrbova´, G., The influence of activity on some contractile characteristics of mammalian fast and slow muscles. Journal of Physiology (London), 1969, 201, 535–549. 38. Simoneau, J.-A. and Pette, D., Species-specific effects of chronic nerve stimulation upon tibialis anterior muscle in mouse, rat, guinea pig, and rabbit. Pflu¨gers Archiv, 1988, 412, 86–92. 39. Thompson, W. J., Soileau, L. C., Balice-Gordon, R. J. and Sutton, L. A., Selective innervation of types of fibres in developing rat muscle. Journal of Experimental Biology, 1987, 132, 249–263. 40. Unguez, G. A., Roy, R. R., Pierotti, D. J., Bodine-Fowler, S. and Edgerton, V. R., Further evidence of incomplete neural control of
484 muscle properties in cat tibialis anterior motor units. American Journal of Physiology, 1995, 268, C527–C534. 41. Walker, D. W. and Luff, A. R., Functional development of fetal limb muscles: a review of the roles of activity, nerves and hormones. Reproduction, Fertility and Development, 1995, 7, 391–398.
KERNELL 42. Westerga, J. and Gramsbergen, A., Development of locomotion in the rat. Developmental Brain Research, 1990, 57, 163–174. 43. Westgaard, R. H. and Lømo, T., Control of contractile properties within adaptive ranges by patterns of impulse activity in the rat. Journal of Neuroscience, 1988, 8, 4415–4426.