Book review |
THE CIRCUITRY OF THE HUMAN SPINAL CORD: ITS ROLE IN MOTOR CONTROL AND MOVEMENT DISORDERS
By Emmanuel Pierrot-Deseilligny and David Burke
2005
Cambridge: Cambridge University Press
Price: £110.00
ISBN: 978-0-521-82581-8 ‘One small step for man’
|
). To paraphrase Sherrington, we would now emphasize that reflex action and the control of movement in general are the results of coordinated activity in the brain's motor network. There are many different, highly interconnected parts of the motor network, including the spinal cord, basal ganglia, cerebellum and motor areas of the cerebral cortex. The network as a whole is responsible for all aspects of a voluntary movement, including its planning, preparation and execution; other important features such as motor learning (the acquisition of motor skill) and the monitoring of sensorimotor performance are also shared functions. It is difficult to assign to any one part of the network the complete responsibility for any single aspect of motor control.
A key part of the motor network is of course the neuronal machinery for movement that is located within the spinal cord. For many years the spinal cord was studied in the decerebrate preparation and, despite Sherrington's warnings, a tradition grew up whereby it was scientifically and even clinically acceptable to think of the spinal cord as having an existence quite separate from that of the supraspinal centres. Yet we know that the consequences of severing the cord from the supraspinal centres are far more devastating in humans than in animal models.
In 2006 the search is on for a new metaphor of how such a distributed motor system might work: we have had analogies drawn with electronic circuits, with digital computers and even the world-wide web. Western society has now opted to provide higher and higher monetary rewards for those that can entertain us with outstanding motor skills and so perhaps it is fitting to turn to football (soccer) to provide us with a simple analogy. Although each of the 11 members of the team network shows different types of activity, they all subserve a single objective, conveniently referred to as scoring a goal. The activity of the entire team is needed to produce this final outcome. Interestingly, in special circumstances, such as the sending off of a player for violent conduct, the team may have to reorganize rapidly its activity in order to continue to function. Each player must now adopt a different pattern of activity, compensating for the absent player. Observing the performance of the remaining team provides little direct insight into the functions carried out by the player who was dismissed. Thus, studying motor performance after one part of the motor network has been damaged by neurological disease, will not necessarily provide useful insights into the functions of the damaged structure. As von Monakow first pointed out, that performance is probably the result of compensatory plastic changes within the remaining members of the network.
Emmanuel Pierrot-Deseilligny and David Burke's important new book represents a review of many recent developments in the workings of the human spinal cord. The work reviewed is very much centred on the outstanding achievements that these two international leaders have contributed to the field, with Pierrot-Deseilligny based at the Salpêtrière in Paris and Burke in Sydney. There have also been frequent collaborative experimental studies and now they have joined forces to produce this fine book. What emerges from its pages are the major advances in our understanding of the spinal circuitry of the human spinal cord, examined, as the subtitle of the book indicates, in both health and disease. This book will become a unique resource, making available in one volume so many important published studies. The original papers are so careful and so detailed, that only when they are all brought together can one appreciate the comprehensiveness and coherence, understand the balance of the different findings or go straight to the result one needs. The critical insights into the clinical implications of each chapter and that devoted to pathophysiology are also unique.
Although Sherrington discovered the principle of reflex action and established most of the basic rules governing reflex activity of the spinal cord in the 1890s, it was not until 1910 that physiological evidence for a monosynaptic reflex in humans was first reported by Paul Hoffmann. Sherrington's myographic methods in animals concentrated on the biomechanical consequences of reflex action and it is remarkable that in his long treatise on the stretch reflex, published with Liddell in 1924, he does not mention Hoffmann's work. Stressing that he felt the tonic stretch reflex to be more important than the phasic knee jerk, he wrote: ‘Study of the stretch reflex in its greater completeness and entirety has been but recent. There can, however, be little doubt that the "knee jerk", a reaction long familiar to the physician, is a fractional manifestation of it’ (Liddell and Sherrington, 1924).
Paul Hoffmann visited Sherrington and was probably hurt by the lack of interest that he showed in Hoffmann's work. But although Hoffmann's H-reflex can actually only be observed in a limited number of muscles, his work laid the foundation for modern electrophysiological investigation and Pierrot-Deseilligny and Burke's book beautifully illustrates how far that investigative journey has progressed. This impressive collection of work demonstrates that, using a number of carefully designed and highly developed experimental techniques, it has been possible to tease apart several basic circuits in the human spinal cord.
The earlier chapters cover transmission in the major reflex circuits involving afferent inputs from proprioceptors in the upper and lower limbs that arise from group Ia, group II receptors in muscle spindles, Ib receptors in tendon organs and from cutaneous and flexor reflex afferents. There are also chapters dealing with important mechanisms that influence reflex transmission, including recurrent inhibition in motoneurons (the area in which Pierrot-Deseilligny first made his mark upon human neurophysiology) and presynaptic inhibition, a fundamental mechanism for controlling the barrage of information that confronts the CNS after every movement that we make.
The human spinal cord is a very complex piece of neuronal machinery and it is a major challenge to communicate the overriding principles of its functional organization, without getting too easily lost in the wealth of experimental data that are now available. The success of the authors is largely down to the organization of their book, which is superb. They took the brave decision to start with an account of methodology providing a summary of the main electrophysiological techniques used to study circuitry of the human spinal cord. Both authors have made major contributions to this methodology; this chapter provides an essential primer allowing the reader of subsequent chapters to appreciate the advantages and limitations of the non-invasive approaches they describe. These methods range from the long-established testing of H-reflexes, the recording of responses in single motor units and non-invasive stimulation of the motor cortex.
There is, as yet, no non-invasive method to assess the ongoing natural activity of the spinal cord itself: rather, this has to be deduced by probing it with a variety of inputs, generated either at the sensory periphery or by activation of descending pathways from supraspinal centres. Many of these ‘indirect but valid’ probes utilize unnatural electrical or magnetic stimuli, which generate ‘phasic artificial volleys’. The authors rightly caution that matters might be very different in the normal modus operandi, with tonic inputs occurring during ongoing natural movements, non-linear input–output relationships in the motoneuron pool and the tantalizing role of plateau potentials.
The more recent development of EEG and EMG coherence analysis in the frequency domain provides an approach that does not involve ‘external interference’. While the newer methods of functional MRI, PET and MEG have made enormous contributions to our knowledge of the supraspinal and especially cortical mechanisms, the small size of the cord, combined with other technical difficulties, has meant that these methods have so far added very little to the knowledge of natural activity in the spinal cord.
Our ability to interpret and understand the data obtained by indirect methods in healthy volunteers and patients depends upon animal studies which have used much more invasive methods. Each of the main chapters begins with a summary of the relevant principles and mechanisms that have been derived from animal experiments. The human experimental material is all clearly presented, with many of the original experimental data and circuit diagrams elegantly redrawn to give a great sense of uniformity to the work. Each chapter is clearly subdivided, with sections on critical evaluation of the special methodology needed to study a particular circuit (e.g. reciprocal Ia inhibition), the organization and pattern of connections of the particular circuit being addressed; next, the physiological significance of this circuit is assessed, together with its involvement in different motor tasks. A final section in each chapter deals with studies on patients and clinical implications and each chapter has a clear set of conclusions with an extensive résumé.
Throughout the book the authors are rigorous in their critical approach to poor experimental technique, erroneous conclusions and worn-out hypotheses. They provide useful information. Drawbacks and limitations of different methodologies are honestly and clearly explained. These include the use of antidromic volleys to assess recurrent inhibition or the use of electrical stimulation over tendons to excite group Ib fibres.
The authors highlight in several chapters the major importance of the inhibitory processes in spinal control: these include the group Ia reciprocal inhibition, the Ib and group II inhibitory pathways and their state-dependency, which allows rapid change in the gain and even the sign of the reflex effect during different phases of a motor task, such as locomotion.
The book concludes with two important chapters: the first is a necessarily speculative but nevertheless insightful synthesis of the role of the different spinal circuits in specific motor tasks. Here the authors bring together interaction of the circuits in control of movements at hinge (e.g. ankle, elbow) and ball (e.g. wrist, shoulder) joints. We can now see that the operational rules are highly specific and, clearly, concepts have moved on from the time when experimentalists attempted to apply the same set of reflex mechanisms to every muscle, joint and motor task. Finally, the authors turn to the pathophysiology of spinal circuits associated with spasticity and Parkinson's disease. Although the latter disorder has long been known to result from a neurodegenerative process in the basal ganglia, the importance of the motor network concept is again reinforced by the widespread changes seen in a number of spinal circuits. Further, we now know that deep brain stimulation, an effective symptomatic treatment for some Parkinson's Disease patients, can affect reflex circuits, such as that transmitting Ib autogenetic inhibition.
For me, the most controversial Chapter is that on the propriospinal relay for descending motor commands. This is the area in which Pierrot-Deseilligny has been most active in recent years. The pioneering work of the Lundberg school in Gothenburg showed that, in the cat, there is a system of propriospinal neurons, located in the upper cervical cord, which transmit commands, excitatory and inhibitory, to cervical motoneurons innervating forelimb muscles. A major input to this system comes from the corticospinal tract, although other descending pathways, such as fibres in the rubrospinal and reticulospinal tracts, converge upon the same C3-C4 neurons, which then project monosynaptically to motoneurons in the cervical enlargement. The same propriospinal neurons receive afferent inputs from the periphery. This type of control is referred to as ‘premotoneuronal’ because the integration of the command signals in these descending pathways occurs before the ‘final common path’—the alpha motoneuron. A signal advantage of such an organization is that it ensures that descending commands in different motor systems are not in conflict; it is proposed that such a system would also allow for rapid updating of commands before they reach the motoneurons.
Taking the cat propriospinal system for corticospinal transmission as his model, Pierrot-Deseilligny and his colleagues in Paris embarked on a long series of intricate experiments to demonstrate the existence of such a system in humans. First, he used newly developed experimental and analytical approaches to demonstrate that, in addition to the well-known monosynaptic inputs from group 1 afferents, it was possible to reveal non-monosynaptic responses in motor units in many upper limb muscles. The central delay involved was interpreted as being due to transmission through an oligosynaptic (probably disynaptic) pathway, involving afferent projections up to the third and fourth cervical segments (C3-C4) propriospinal neurons, which were excited and, in turn, projected their influence to the cervical motor nuclei, just as in cat. This model was supported by the finding that the central delay was greater for motor nuclei located more caudally (e.g. triceps brachii) than those more rostrally (e.g. biceps). The next step was to use TMS to demonstrate that these same oligosynaptic effects could be strongly modulated by the cortex and this in turn suggested that the pathway could, just as in the cat, transmit corticospinal commands to upper limb motoneurons.
There is evidence for increased excitation of this propriospinal system in hemiplegia, highlighting an important possible role in recovery after stroke. We also need to know more about this system after spinal injury, because it could provide a means by which even small amounts of regeneration in descending pathways could be relayed to spinal centres below the level of the lesion.
The experimental animal model upon which this research was constructed is the cat. Of course, the cat lacks the direct cortico-motoneuronal (CM) system, one of the distinguishing features of the human motor system that is shared with some, but not all, primates. This existence of the CM system continues to challenge our ideas about premotoneuronal control: clearly CM commands from cortex can influence motoneurons directly, free from presynaptic control. We also know that rapid updating of commands can occur at both the cortical and spinal level.
We now need to come to terms with the parallel existence of these direct and indirect systems (Lemon et al., 2004). The indirect systems are likely to be many in number and the degree of overemphasis given to the cervical propriospinal system is largely due to the strict adherence to the cat model, which as I shall discuss below, is probably unwise. In any event, the authors give a somewhat unbalanced view of the relative importance of the two systems and much future work will be needed to assess the relative contributions that monosynaptic and oligosynaptic transmission make. First, these contributions are likely to vary between muscles; for example, the cervical propriospinal system is relatively unimportant for control of intrinsic hand muscles, but these receive the largest CM input in primates. Secondly, we know that both systems possess strong state-dependency: while the CM system in the monkey has been repeatedly shown to be particularly active for fine digit movements, there is evidence that the propriospinal system in the cat is particularly important for reaching and less so for grasp.
One might be disappointed that the high level of rigour elsewhere in the book might be seen to have lapsed with uncritical acceptance of some recent studies claiming, on the basis of incomplete sections of the corticospinal tract in the monkey, that the propriospinal system is involved in precision grip. This conclusion is completely at odds with the finding of a reach-related propriospinal system in the cat and the absence of propriospinal effects on hand muscles in humans. The powerful feedforward inhibition of the macaque propriospinal system is such that it cannot be revealed without use of systemic strychnine, an extreme measure that still raises serious doubts as to its contribution under normal conditions.
The extrapolation from animal models to the human is now under intense scrutiny for a variety of ethical and scientific reasons. All the animal models in use exhibit highly successful and distinct patterns of motor behaviour; while some have argued that all the basic patterns are constrained across all vertebrae species, one can also see evidence for highly developed specializations that presumably have been favoured and selected by the evolutionary process. Thus in the chapter on monosynaptic Ia excitation, the authors note the importance of determining ‘how the different feedback systems have become adapted to the evolutionary demands for changed movement patterns’. They go on to summarize the rather striking and varied differences in the pattern of Ia excitation that is exerted on heteronymous muscles in cat, baboon and human. There are far more transjoint interactions in the upper limb of the human than in baboon or cat, possibly related to the highly developed functions of the human arm and hand such as throwing, reaching, grasping and manipulation.
Given these differences, one should surely not be surprised by significant changes across species for inputs derived from descending motor pathways. The authors highlight the increased level of feedforward and feedback inhibition in the propriospinal transmission system of the primate versus that of the cat. They suggest that this feature might be useful in ‘focusing’ corticospinal drive on a selective set of muscles during a motor act, although there is no evidence yet for such a mechanism. Actually, as the authors themselves state on page xvii, the evidence points to a much greater degree of specialization across species, with different patterns of organization making it difficult to extrapolate directly from circuits in animals to those in humans. For, just as the CM system is now known to be highly developed in humans and absent in all non-primates and even some primates, recent studies have shown that the C3-C4 propriospinal system, so well-developed in cat, appears to be absent in both the mouse and the rat, under the experimental conditions used. This same system is also very differently organized in the two different non-human primates that have been studied to date. Therefore, while the cat provided an excellent starting point for the imaginative and pioneering work on the human cervical interneuronal system, it seems likely that the detailed organization of this system will finally turn out to show significant differences between the cat and human, reflecting their very different, but successful patterns of motor activity.
We have barely started to understand the spinal circuits that subserve the most distinctive aspects of human motor behaviour—not least, our amazing ability to learn new motor behaviours and use a bewildering variety of tools to bring about major changes to our earthly environment and even beyond. Surely the sight of two astronauts, far out in space, tiling and grouting the exterior of their space shuttle represents one of the peaks of human motor development. This book elegantly illuminates the fundamental circuits that provide the physiological underpinnings of that achievement.
Sobell Department of Motor Neuroscience and Movement Disorders Institute of Neurology, University College London London, UK
References
Lemon RN, Kirkwood PA, Maier MA, Nakajima K, Nathan P. Direct and indirect pathways for corticospinal control of upper limb motoneurones in the primate. Prog Brain Res 2004; 143: 263–79.[Web of Science][Medline]
Liddell EGT, Sherrington CS. Cited in: Denny–Brown D, editor. Selected writings of Sir Charles Sherrington. London: Hamish Hamilton Medical Books; 1924. p. 394.
Sherrington CS. The integrative action of the nervous system. New Haven: Yale University Press; 1906. p. 7–8.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||