This is a review of our investigations into the neuronal mechanisms of functional recovery after spinal cord injury (SCI) in a non-human primate model. In primates, the lateral corticospinal tract (l-CST) makes monosynaptic connections with spinal motoneurons. The existence of direct cortico-motoneuronal (CM) connections has been thought to be the basis of dexterous digit movements, such as precision gripping. However, recent studies have shown that after lesion of the direct CM connections, by a l-CST lesion at the C4/C5 level, precision gripping is initially impaired, but shows remarkable recovery with training within several weeks. Plastic changes of the neural circuits underlying the recovery occur at various levels of the central nervous system. In the subcortical networks, intracellular recordings from the motoneurons in anesthetized animals demonstrated that transmission through the disynaptic pathways from the CST was enhanced, presumably mediated by the propriospinal neurons in the mid-cervical segments. The γ-band musculo-muscular coherence (MMC), with a peak frequency around 30 Hz, appeared over a wide range of forelimb muscles and was strengthened in parallel to the recovery of the precision grip. Appearance of the γ-band MMC also paralleled the change in the activation pattern of forelimb muscles; muscles which were antagonists before the lesion showed co-activation after recovery. Such γ-band MMC is thought to originate in the subcortical network, presumably in the brainstem or spinal cord. In the cortical networks, a combination of positron emission tomography and reversible inactivation techniques has shown that the bilateral primary motor cortex (M1) and ventral premotor cortex (PMv) have different contributions to functional recovery depending on the recovery stage; the bilateral M1 plays a major role in early stage recovery (<1 month), whereas the contralateral M1 and bilateral PMv are the prominent contributors to the later stages (3-4 months). Such changes in cortical activity in M1 and PMv have been shown to accompany changes in the expressions of plasticity-related genes, such as GAP-43. Changes in the dynamic properties of neural circuits, both at the cortical and subcortical levels, are time-dependent. Multidisciplinary studies to clarify how the changes in the dynamic properties of individual components of the large-scaled networks are coordinated during recovery will help to develop effective therapeutic strategies to recovery from SCI.
Copyright © 2011 Elsevier Inc. All rights reserved.