Restoring Voluntary Control of Locomotion after Paralyzing Spinal Cord Injury

Half of human spinal cord injuries lead to chronic paralysis. Here, we introduce an electrochemical neuroprosthesis and a robotic postural interface designed to encourage supraspinally-mediated movements in rats with paralyzing lesions. Despite the interruption of direct supraspinal pathways, the cortex regained the capacity to transform contextual information into task-specific commands to execute refined locomotion. This recovery relied on the extensive remodeling of cortical projections, including the formation of brainstem and intraspinal relays that restored qualitative control over electrochemically enabled lumbosacral circuitries. Automated treadmill-restricted training, which did not engage cortical neurons, failed to promote translesional plasticity and recovery. By encouraging active participation under functional states, our training paradigm triggered a cortex-dependent recovery that may improve function after similar injuries in humans.

Corticolumbar dynamics during recovery after spinal cord injury

Selin Anil

In humans, a severe spinal cord contusion interrupts the vast majority of supraspinal projections to the spinal cord below the lesion. Permanent paralysis results from the chronic failure of these spared projections to engage lumbar circuits producing leg movement. We modeled these injuries in rodents, and showed the immediate access of cortical circuits onto the electrochemically-enabled lumbar spinal cord. Neuroprosthetic rehabilitation leveraged spared motor circuits elements, ultimately leading to the establishment of a cortico-reticulo-spinal relay that mediated the recovery of voluntary control over the legs. Cortical fibers underwent a drastic anatomical remodeling, contacting glutamatergic reticulospinal neurons in the brainstem which, despite lesion variability, consistently retained synaptic connections onto lumbar circuits. This circuit-level reorganization mediated a cortex-dependent recovery of natural locomotor behaviors. In the aim of refining our understanding of the cortical contribution to the restoration of leg motor control with neuroprosthetic rehabilitation, we established a methodology to translate to rats a technique of endoscopic calcium imaging in freely-moving animals. We increased the stability of the implant, reaching unprecedented chronic degrees of longevity. Using this newly implemented technology, we tracked the activity of single corticolumbar cells during natural walking in intact rats, showing a strong representation of motor behavioral features in the population. We additionally highlighted the stability of locomotion-related classifications of individual neurons, suggesting a segregation of movement-related representations. The contusion was accompanied by important anatomical changes of corticolumbar cell bodies, preventing the longitudinal identification of single neurons across injury. However, population-level analyses highlighted a maintained strong representation of locomotion at late stages after injury and training. We did not find any dynamic change in corticolumbar population activity correlating with the recovery of voluntary leg motor control, which reflected in a stability of the fractions of movement-related neurons. These early results start shedding light on possible mechanisms of functional plasticity accompanying recovery after spinal cord injury, as a peculiar form of motor learning. Concurrently to the functional reorganization of corticolumbar neurons, we investigated the anatomical plasticity of this specific population with contusion and training. Neuroprosthetic rehabilitation led to a strengthening of collateral connections onto the thoracic spinal segments above the lesion. However, we found no effect of training on the anatomical remodeling throughout the brain and brainstem. Notably, the absence of sprouting of corticolumbar collaterals in the brainstem reticular formation leads us to question the identity of the cortical population involved in the formation of the cortico-reticulo-spinal relay.

EPFL2020

Axon regeneration and circuit reorganization after complete spinal cord injury

Sabry Leonardo Barlatey

First human trials involving neuroprosthetic rehabilitation demonstrated recently that significant functional benefits can be achieved with lumbosacral neuromodulation and reorganized spared projections. However, complete spinal cord injuries (SCI) wholly isolate infralesional circuits from any supraspinal control, leading to irreversible motor deficits that neuroprosthetics still fails to address. As axons fail to regrow across the lesion site, neuroregenerative research after SCI has been dogmatically focusing attention on minimizing the environmental inhibitions to axon regeneration. In the present work, we demonstrate that spontaneous axon regeneration failure can be reversed, and that propriospinal axons are able to grow across complete non-neural lesion cores when the required facilitators are provided. We identified three mechanisms needed for propriospinal regrowth : i) enhancing neuronal intrinsic growth capacities using viral technologies, ii) remodeling the lesion core with growth factors, in order to densify permissive substrates, and iii) guiding axons chemically across the SCI site. Propriospinal neurons - that coordinate different spinal segments in healthy conditions - are known to support recovery of locomotion in incomplete models of spinal cord injury. However, restoring a robust descending bridge of propriospinal axons does not by itself promote any functional benefit. We hypothesize that a lack of somatosensory feedback to the motor cortex (M1), together with insufficient descending motor control, may prevent coherent exchange of information between lumbosacral and cortical motor centers. Therefore, we explored the neuroregenerative potential of reticulospinal axons - which are projected from the brainstem to the spinal cord - as a second relay for the descending motor cortical command. Together with axon guidance and lesion remodeling, our viral manipulation of intrinsic growth programs induced limited yet significant reticulospinal regeneration into the lesion core. In order to enhance this reticulospinal regrowth, we explored activity-dependent mechanisms of neurite growth control. As we observed that activity does not recruit growth programs after SCI, we revealed a possible divergence between the mechanisms that underly reticulospinal axon regrowth and neurite sprouting. In the mean time, we are developing a somatosensory interface that aims at restoring sensory feedback to the motor cortex after complete SCI. Based on hindlimb electromyographic activity, we linked locomotion to optogenetic neuromodulation of the motor thalamus, and were able to evoke responses in M1 by selectively activating thalamocortical projections, both before and after complete spinal lesion. Such technology enables a direct recruitment of thalamic nuclei that gate motor learning and motor planning at a cortical level. Coming experiments will test wether a strategy combining propriospinal regeneration with locomotion-dependent thalamocortical modulation is sufficient to restore voluntary stepping after complete SCI.

EPFL2019

Combining Reflexes and External Sensory Information in a Neuromusculoskeletal Model to Control a Quadruped Robot

Auke Ijspeert, Azhar Aulia Saputra

This article examines the importance of integrating locomotion and cognitive information for achieving dynamic locomotion from a viewpoint combining biology and ecological psychology. We present a mammalian neuromusculoskeletal model from external sensory information processing to muscle activation, which includes: 1) a visual-attention control mechanism for controlling attention to external inputs; 2) object recognition representing the primary motor cortex; 3) a motor control model that determines motor commands traveling down the corticospinal and reticulospinal tracts; 4) a central pattern generation model representing pattern generation in the spinal cord; and 5) a muscle reflex model representing the muscle model and its reflex mechanism. The proposed model is able to generate the locomotion of a quadruped robot in flat and natural terrain. The experiment also shows the importance of a postural reflex mechanism when experiencing a sudden obstacle. We show the reflex mechanism when a sudden obstacle is separately detected from both external (retina) and internal (touching afferent) sensory information. We present the biological rationale for supporting the proposed model. Finally, we discuss future contributions, trends, and the importance of the proposed research.