HYPOTHESIS AND THEORY published: 12 April 2021 doi: 10.3389/fnins.2021.643463 Engaging Spinal Networks to Mitigate Supraspinal Dysfunction After CP V. Reggie Edgerton 1,2,3,4 , Susan Hastings 5 and Parag N. Gad 1,6,7* 1 Department of Neurobiology, University of California, Los Angeles, Los Angeles, CA, United States, 2 Department of Neurosurgery, University of California, Los Angeles, Los Angeles, CA, United States, 3 Brain Research Institute, University of California, Los Angeles, Los Angeles, CA, United States, 4 Institut Guttmann, Hospital de Neurorehabilitació, Institut Universitari Adscrit a la Universitat Autònoma de Barcelona, Barcelona, Spain, 5 SH Pediatric Physical Therapy, San Jose, CA, United States, 6 Rancho Research Institute, Downey, CA, United States, 7 SpineX, Inc., Los Angeles, CA, United States Although children with cerebral palsy seem to have the neural networks necessary to generate most movements, they are markedly dysfunctional, largely attributable to abnormal patterns of muscle activation, often characterized as spasticity, largely reflecting a functionally abnormal spinal-supraspinal connectivity. While it is generally assumed that the etiologies of the disruptive functions associated with cerebral palsy can be attributed primarily to supraspinal networks, we propose that the more normal connectivity that persists between peripheral proprioception-cutaneous input to the spinal networks can be used to guide the reorganization of a more normal Edited by: spinal-supraspinal connectivity. The level of plasticity necessary to achieve the required Enrico Rejc, reorganization within and among different neural networks can be achieved with a University of Louisville, United States combination of spinal neuromodulation and specific activity-dependent mechanisms. Reviewed by: By engaging these two concepts, we hypothesize that bidirectional reorganization of Lynn Bar-On, Amsterdam University Medical proprioception-spinal cord-brain connectivity to higher levels of functionality can be Center, Netherlands achieved without invasive surgery. Andrew Gordon, Columbia University, United States Keywords: cerebral palsy, spinal cord, brain, muscle, EMG *Correspondence: Parag N. Gad paraggad@ucla.edu THE PROBLEM Specialty section: It is generally assumed that the primary pathology of the nervous system that leads to This article was submitted to cerebral palsy (CP) is located within and among different combinations of supraspinal networks Autonomic Neuroscience, and these pathologies can be due to multiple etiologies. In most cases, however, it appears a section of the journal that these supraspinally occurring pathologies also will be necessarily manifested as spinally Frontiers in Neuroscience mediated dysfunctions, affecting multiple peripheral sensory-motor systems involving equilibrium Received: 18 December 2020 posture, locomotion, and trunk and head control (Smith and Gorassini, 2018). Although a Accepted: 22 March 2021 high level of functionally immature connections are normally formed early in development, Published: 12 April 2021 the pruning of neurons and synaptic connections occur subsequently using multiple, largely Citation: unknown, guidance mechanisms that result in the more effective connections between spinal Edgerton VR, Hastings S and networks and descending axons and proprioceptive afferents. But, if functionally immature or Gad PN (2021) Engaging Spinal Networks to Mitigate Supraspinal abnormal connections persist at the end of the early developmental phase, the supraspinal Dysfunction After CP. and propriospinal connectivity will result in abnormal sensory-motor responses. Persistence Front. Neurosci. 15:643463. of these functionally abnormal synaptic connections are reinforced postnatally throughout the doi: 10.3389/fnins.2021.643463 critical period of development and into adulthood, resulting in the commonly recognized Frontiers in Neuroscience | www.frontiersin.org 1 April 2021 | Volume 15 | Article 643463 Edgerton et al. Hypothesis: Spinal Networks Mitigate CP Dysfunction neuromuscular disorders associated with CP, with the aberrant brain-spinal connections to a greater prominence of most common symptoms being collectively diagnosed as functionally normal connections. We reasoned that we could do neuromuscular spasticity and stiffness of joints. More specifically, this by maximizing the dominance of proprioception and the however, there is a persistent pervasiveness of poor coordination spinal networks that translate this sensory input and minimizing of the motor pools. Given the concept of coordination is the pathology of the brain, in controlling posture and so central to our basic hypothesis, it is important that the locomotion. Cappellini et al. (2016, 2020) noted how much the meaning of this concept be clearly defined. Often, a coordinated dysfunction of gait in children with CP can be related to spinal movement is one in which there is reciprocity of the dynamics neuronal networks vs. supraspinal dysfunction. A more thorough of the temporal patterns of activation and deactivation of knowledge about pattern generation circuitries in infancy may flexor and extensor motor pools that generate the movement. improve our understanding of developmental motor disorders, “Normal” movements, however, are generated with a continuum highlighting the necessity for regulating the functional properties of differing degrees of overlap and changes in levels of activation of abnormally developed neuronal locomotor networks as a target and deactivation. It is useful to be aware that the motor task for early sensorimotor rehabilitation. Similarly a very tight link that is being generated is defined largely by the temporal was described between the activity patterns of populations of patterns of activation of motor pools. For example, many of the pyramidal tract neurons and the biomechanics of unconstrained same muscles are activated when stepping forward, backward locomotion in cats (Prilutsky et al., 2005). Versteeg et al. and sideways, but the patterns of muscles activated differ (2021) reported that cuneate nucleus neurons have muscle- substantially (Shah et al., 2012). The degree of reciprocity vs. like properties that have a greater sensitivity to active than co-activation varies considerably among the uninjured general passive movements of the upper limb and that their receptive population. In individuals with symptoms of spasticity, there fields resemble single muscles. These observations suggest that is an unusually greater coincidence of co-contractions of flexor muscle specific signals proprioceptive input could have an and extensor motor pools movements that would be considered activity-dependent impact on supraspinal networks that could poorly coordinated. Theoretically, the greater the number of transform dysfunctional neural networks to a more functional connections that develop between the brain and spinal cord state. There is extensive evidence that the neural networks that are functionally aberrant, the fewer normal targets that of an individual with CP can learn more effective movement remain accessible (Bennett et al., 1983; Callaway et al., 1989; skills as it does after spinal injury (Dewar et al., 2015; Morgan Alexeeva et al., 1997; Maegele et al., 2002). A fundamental and et al., 2015; Reid et al., 2015). Thus, we hypothesize that spinal essential driver in the biological design of our nervous systems, neuromodulation in concert with proprioceptive-driven activity- phylogenetically, ontogenetically and epigenetically in reaching dependent mechanisms of spinal networks can transform the the normal targets undoubtedly has been earth’s gravitational supraspinal-spinal dysfunctional connectivity of CP into highly vectors (Edgerton et al., 2000). We reason that this fundamental functional connections, improving the functionality of networks. feature in the strategy of our sensory-motor design highlighted The result should be the recovery of motor tasks that are more here is that the difficulty in accommodating gravity is consistently forceful, powerful, efficient and display finer control as needed in and pervasively revealed in individuals with CP, as children and a 1G environment. This hypothesis is counter to the predominant in adults. So many of their sensory-motor challenges are linked thinking that all of the “motor” functions, noted above are to maintaining equilibrium while moving effectively in a 1G controlled largely by the brain, rather than the spinal cord. environment (Recktenwald et al., 1999). It is clearly evident Ironically, proprioception is considered to be a major contributor in the motor behaviors of individuals with CP that the neural to the spasticity in CP, and, therefore, is a primary target to reduce connectivity did not develop appropriately for this uniformly the tonic stiffness by performing selective dorsal root rhizotomy present fundamental gravitational challenge. For the reasons (Mortenson et al., 2021). However, selective rhizotomy minimizes noted above our interventional strategy as presented has been the sensor input, which normally plays a prominent role in focused functionally on the necessity of realigning the sensory- controlling movements and can disrupt autonomic functions. motor connectivity to accommodate to normal gravitational Multiple clues led us to the logic applying a specialized vectors. Theoretically, to regain a normal supraspinal-spinal neuromodulatory technique in combination with subject specific connectivity to earth’s gravitational forces, the earlier maladaptive activity-dependent rehabilitation. First, our studies, and that of state that was learned postnatally to sustain equilibrium while in many other labs for decades clearly tell us that proprioception an abnormal state, must be re-transformed to achieve a normal plays a more important role than occasionally correcting translation of sensory input in a 1G environment (Smith and mistakes, as in tripping, when there is not enough time to Gorassini, 2018; Cappellini et al., 2020). adjust the planned movements (Gerasimenko et al., 2017). Proprioception plays a central role in the details of controlling posture, locomotion, and fine motor tasks, even when there SOLUTION is no connectivity between the spinal networks and the brain (Grillner and Zangger, 1975; de Leon et al., 1999). Some specific Given our experience with spinal cord injury (Gerasimenko Y. observations suggesting a possible crucial role for spinal networks et al., 2015; Reggie et al., 2018), which has similar, and in in individuals with CP to facilitate functional recovery are: (1) some ways more severe functional aberrations than in CP, After a complete, mid-thoracic spinal transection a mammal can we developed interventions designed to transform functionally step forward, backward or sideways and at speeds appropriate Frontiers in Neuroscience | www.frontiersin.org 2 April 2021 | Volume 15 | Article 643463 Edgerton et al. Hypothesis: Spinal Networks Mitigate CP Dysfunction to the speed of the treadmill belt. (2) Humans that have lost Figure 1 illustrates the experimental outcomes that we proprioception as an adult are essentially, functionally paralyzed, hypothesize can be obtained with a patient with CP. We at least for months, even though all descending motor pathways hypothesize that applying a combination of non-invasive spinal remain intact. (3) Given that spinal networks can readily learn neuromodulation and specific activity-dependent mechanisms to to perform a new motor task without help from the brain, we an individual with CP can trigger a bidirectional reorganization reasoned that the potential level of plasticity among spinal and of proprioception-spinal cord-brain connectivity to a functional supraspinal networks in the presence of spinal neuromodulation state that can generate near normal postural and locomotor combined with skilled therapy can be robust enough to supplant functions. An individual free of abnormalities are continuously the original supraspinal disruptive connectivity and reinforce signaling sensory-motor information bi-directionally between more normal functional connections with a greater presence of the brain and spinal cord and the kinetic and kinematic sensors normal sensory input associated with routine motor tasks via (proprioception) in muscle, tendons, and skin (Figure 1A). CP activity-dependent mechanisms. (4) Another clue, suggests that is generally considered to be a cerebral pathology (Figure 1B) much of the aberrant connectivity is inextricably linked to skills which will send abnormal motor signals due to the spinal associated with equilibrium, i.e., with gravity being of a constant networks which in turn will project abnormal signals to presence in the evolution of all life on Earth (Wallard et al., 2014, predominately flexor or extensor motor pools and muscles. As 2018). It seems obvious that the network connectivity design of opposed to the spinal networks generating a highly coordinated sensory motor functions that has evolved phylogenetically, has agonist-antagonist pattern (Figure 1A), the disruptive, aberrant, specifically accommodated to earth’s gravity, and that there are descending commands from the brain generates abnormal numerous motor dysfunctions that reflect that this feature was levels of co-contraction signals among the spinal interneuronal not carried out normally, either ontogenetically or epigenetically networks that induces high levels of co-contractions of flexors in many CP cases. And finally, and (5) could the fact that and extensor muscles. The first key point here is that the CP being a developmental phenomenon be an advantage in disruptive signals from the brain, then spinal, then muscular, using interventional strategies to take advantage of higher levels results in abnormal proprioceptive signal to the spinal networks of neuroplasticity during the critical period of development of which we now know sends muscle specific signals to multiple motor primitives as suggested by Bizzi and colleagues (Overduin supraspinal nuclei. Thus, basically a continuous loop of abnormal et al., 2015) and hypothesized by Edelman (1993) as Neural signals imposes persistent abnormal sensory-motor signals that Darwinism and Neuronal Group Selection, presumably of the results in a reorganization of networks that reflect the initial brain, and Edgerton and colleagues as a similar phenomenon in supraspinal pathology. Fortunately, there is reason to predict the spinal cord (Edgerton et al., 2001). that there are several key points that can be used to interrupt the abnormal sensory-motor loop to a much greater level of normality. These points are: (1) The disruptive spinal networks TESTING THE HYPOTHESIS can be transformed to a more normal physiological state with newly developed electrical neuromodulation combined with While the biological bases of the underlying concepts may activity-dependent modulation via a near normal pattern of seem sound, there has not, however, been a single systems- proprioceptive ensembles project to the spinal networks. (2) level concept relevant to the etiology of CP as we have Proprioceptive input has the potential to overcome the disruptive proposed. Technically it is now feasible to induce systems descending input and reorganize the connectivity into a state level changes in specific functional connectivity of networks that generates coordinated movements. And, (3) There is strong patterns of sensory input that translates to a predictable evidence that a similar normalizing process can occur in motor output by repetitively activating specific proprioception the more normal ascending input to supraspinal centers can patterns as occurs during stand training that enables standing then form more normal networks in the descending motor ability but not stepping ability (de Leon et al., 1999), facilitates signals. Experiments in which CP patients attempt to perform changes at the level of synapses and neural networks and motor task in the presence and absence of neuromodulation changes the expression of genes that could mediate the while recording supraspinal and spinally evoked potentials synaptic adaptations to locomotor training after a spinal and recording EMG activities of different combinations of hemisection. More comprehensive, quantitative assays under muscles and kinematics and kinetics to characterize the motor clinically controlled conditions on a limited population of task will provide the data that can determine whether the subjects could be performed with present-day technologies. proposed hypothesis should be accepted, at least from a Clear evidence of the validity of the proposed mechanisms at functional, systems level. a systems level of understanding seems to be not only feasible with proper resources, but in addition a case can be made for the urgency of such an effort based on it’s potential efficacy DISCLOSURE and safety as demonstrated from more than a decade of studies of human subjects with a similar non-invasive interventional VE holds shareholder interest in NeuroRecovery Technologies strategy in recovering functions after SCI (Gerasimenko Y. P. and hold certain inventorship rights on intellectual property et al., 2015; Rath et al., 2018; Sayenko et al., 2018; Gad et al., 2020; licensed by the Regents of the University of California to Kreydin et al., 2020). NeuroRecovery Technologies and its subsidiaries. VE and PG Frontiers in Neuroscience | www.frontiersin.org 3 April 2021 | Volume 15 | Article 643463 Edgerton et al. Hypothesis: Spinal Networks Mitigate CP Dysfunction FIGURE 1 | Schematic representation of supraspinal-spinal-muscle connectivity in normal cerebral development, abnormal cerebral development and in abnormal cerebral development after spinal neuromodulation and activity-based therapy (A) normal brain-spinal networks and muscles with sensors, bidirectionally communicating input and output signals forming a complete loop, including reciprocal EMG of agonist and antagonist muscles, see two channels at the bottom. (B) A region of supraspinal pathology (triangle) resulting in aberrant descending signals causing disruptive degrees of co-contractions of flexor and extensor motor pools and muscles. (C) A remodeling process can of supraspinal and spinal networks can begin with a combination of a non-invasive electrical neuromodulation technique that empowers the spinal networks to begin to assume a dominating control of normalizing the coordination of flexor and extensor motor pools. With repetitive practice in the presence of neuromodulation, we propose that there will be significant reorganization toward a gradually occurring normalization of supraspinal and spinal networks. holds shareholder interest in SpineX, Inc. and hold certain AUTHOR CONTRIBUTIONS inventorship rights on intellectual property licensed by the Regents of the University of California to SpineX, Inc. All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication. DATA AVAILABILITY STATEMENT FUNDING The original contributions presented in the study are included in the article/supplementary material, further inquiries can be The authors declare that this study received funding from directed to the corresponding author. by Walkabout Foundation, Dana & Albert R. Broccoli Frontiers in Neuroscience | www.frontiersin.org 4 April 2021 | Volume 15 | Article 643463 Edgerton et al. Hypothesis: Spinal Networks Mitigate CP Dysfunction Charitable Foundation and Nanette, Burt Forester, including Pediatrics (CTIP). The funders were not involved in the matching by PwC LLP, Roberta Wilson, BEL13VE in Miracles study design, collection, analysis, interpretation of data, Jack Jablonski Foundation, Brain Recovery Project, and The the writing of this article, or the decision to submit it West Coast Consortium for Technology & Innovation in for publication. REFERENCES Morgan, C., Novak, I., Dale, R. C., and Badawi, N. (2015). Optimising motor learning in infants at high risk of cerebral palsy: a pilot study. BMC Pediatr. Alexeeva, N., Broton, J. G., Suys, S., and Calancie, B. (1997). Central cord syndrome 15:30. doi: 10.1186/s12887-015-0347-2 of cervical spinal cord injury: widespread changes in muscle recruitment Mortenson, P., Sadashiva, N., Tamber, M. S., and Steinbok, P. (2021). Long-term studied by voluntary contractions and transcranial magnetic stimulation. Exp. upper extremity performance in children with cerebral palsy following selective Neurol. 148, 399–406. doi: 10.1006/exnr.1997.6689 dorsal rhizotomy. Childs Nerv. Syst. doi: 10.1007/s00381-020-05018-2 [Epub Bennett, M. R., McGrath, P. A., Davey, D. F., and Hutchinson, I. (1983). Death ahead of print]. of motorneurons during the postnatal loss of polyneuronal innervation of rat Overduin, S. A., d’Avella, A., Roh, J., Carmena, J. M., and Bizzi, E. (2015). muscles. J. Comp. Neurol. 218, 351–363. doi: 10.1002/cne.902180311 Representation of Muscle Synergies in the Primate Brain. J. Neurosci. 35, Callaway, E. M., Soha, J. M., and Van Essen, D. C. (1989). Differential loss of 12615–12624. doi: 10.1523/jneurosci.4302-14.2015 neuromuscular connections according to activity level and spinal position of Prilutsky, B. I., Sirota, M. G., Gregor, R. J., and Beloozerova, I. N. (2005). neonatal rabbit soleus motor neurons. J. Neurosci. 9, 1806–1824. doi: 10.1523/ Quantification of motor cortex activity and full-body biomechanics during jneurosci.09-05-01806.1989 unconstrained locomotion. J. Neurophysiol. 94, 2959–2969. doi: 10.1152/jn. Cappellini, G., Ivanenko, Y. P., Martino, G., MacLellan, M. J., Sacco, A., Morelli, 00704.2004 D., et al. (2016). Immature Spinal Locomotor Output in Children with Cerebral Rath, M., Vette, A. H., Ramasubramaniam, S., Li, K., Burdick, J., Edgerton, Palsy. Front. Physiol. 7:478. doi: 10.3389/fphys.2016.00478 V. R., et al. (2018). Trunk Stability Enabled by Noninvasive Spinal Electrical Cappellini, G., Sylos-Labini, F., Dewolf, A. H., Solopova, I. A., Morelli, D., Stimulation after Spinal Cord Injury. J. Nneurotrauma 35, 2540–2553. doi: Lacquaniti, F., et al. (2020). Maturation of the Locomotor Circuitry in Children 10.1089/neu.2017.5584 With Cerebral Palsy. Front. Bioeng. Biotechnol. 8:998. doi: 10.3389/fbioe.2020. Recktenwald, M. R., Hodgson, J. A., Roy, R. R., Riazanski, S., McCall, G. E., 00998 Kozlovskaya, I., et al. (1999). Effects of spaceflight on rhesus quadrupedal de Leon, R. D., Tamaki, H., Hodgson, J. A., Roy, R. R., and Edgerton, V. R. (1999). locomotion after return to 1G. J. Neurophysiol. 81, 2451–2463. doi: 10.1152/ Hindlimb locomotor and postural training modulates glycinergic inhibition jn.1999.81.5.2451 in the spinal cord of the adult spinal cat. J. Neurophysiol. 82, 359–369. doi: Reggie, E. V., Yury, G., Parag, G., and Dimitry, S. (2018). Basic concepts underlying 10.1152/jn.1999.82.1.359 activity-dependent mechanisms in the rehabilitation of sensory-motor function Dewar, R., Love, S., and Johnston, L. M. (2015). Exercise interventions improve after spinal cord injury. Spinal Cord Med. 54, 897–911. postural control in children with cerebral palsy: a systematic review. Dev. Med. Reid, L. B., Rose, S. E., and Boyd, R. N. (2015). Rehabilitation and neuroplasticity Child Neurol. 57, 504–520. doi: 10.1111/dmcn.12660 in children with unilateral cerebral palsy. Nat. Rev. Neurol. 11, 390–400. doi: Edelman, G. M. (1993). Neural Darwinism: selection and reentrant signaling in 10.1038/nrneurol.2015.97 higher brain function. Neuron 10, 115–125. doi: 10.1016/0896-6273(93)90 Sayenko, D. G., Rath, M., Ferguson, A. R., Burdick, J., Havton, L. A., Edgerton, V. R. 304-a et al. (2018). Self-assisted standing enabled by non-invasive spinal stimulation Edgerton, V. R., Roy, R. R., and De Leon, R. D (2001). “Neural Darwinism in after spinal cord injury. J. Neurotrauma. 36, 1435–1450. doi: 10.1089/neu.2018. the mammalian spinal cord,” in Spinal Cord Plasticity: Alterations in Reflex 5956 Function, eds M. M. Patterson and J. W. Graw (Boston: Kluwer Academic), Shah, P. K., Gerasimenko, Y., Shyu, A., Lavrov, I., Zhong, H., Roy, R. R., et al. 185–206. doi: 10.1007/978-1-4615-1437-4_8 (2012). Variability in step training enhances locomotor recovery after a spinal Edgerton, V. R., Roy, R. R., Hodgson, J. A., Day, M. K., Weiss, J., Harkema, S. J., cord injury. Eur. J. Neurosci. 36, 2054–2062. doi: 10.1111/j.1460-9568.2012. et al. (2000). How the science and engineering of spaceflight contribute to 08106.x understanding the plasticity of spinal cord injury. Acta Astronaut. 47, 51–62. Smith, A. T., and Gorassini, M. A. (2018). Hyperexcitability of brain stem pathways doi: 10.1016/s0094-5765(00)00009-6 in cerebral palsy. J. Neurophysiol. 120, 1428–1437. doi: 10.1152/jn.00185.2018 Gad, P., Kreydin, E., Zhong, H., and Edgerton, V. R. (2020). Enabling Versteeg, C., Chowdhury, R. H., and Miller, L. E. (2021). Cuneate nucleus: The respiratory control after severe chronic tetraplegia: an exploratory case study. somatosensory gateway to the brain. Curr. Opin. Physiol. 20, 206–215. doi: J. Neurophysiol. 124, 774–780. doi: 10.1152/jn.00320.2020 10.1016/j.cophys.2021.02.004 Gerasimenko, Y., Gorodnichev, R., Moshonkina, T., Sayenko, D., Gad, P., Wallard, L., Dietrich, G., Kerlirzin, Y., and Bredin, J. (2014). Balance control in gait and Reggie Edgerton, V. (2015). Transcutaneous electrical spinal-cord children with cerebral palsy. Gait Posture 40, 43–47. doi: 10.1016/j.gaitpost. stimulation in humans. Ann. Phys. Rehabil. Med. 58, 225–231. doi: 2014.02.009 10.1016/j.rehab.2015.05.003 Wallard, L., Dietrich, G., Kerlirzin, Y., and Bredin, J. (2018). Effect of robotic- Gerasimenko, Y. P., Lu, D. C., Modaber, M., Zdunowski, S., Gad, P., Sayenko, assisted gait rehabilitation on dynamic equilibrium control in the gait of D. G., et al. (2015). Noninvasive Reactivation of Motor Descending Control children with cerebral palsy. Gait Posture 60, 55–60. doi: 10.1016/j.gaitpost. after Paralysis. J. Neurotrauma 32, 1968–1980. doi: 10.1089/neu.2015.4008 2017.11.007 Gerasimenko, Y., Sayenko, D., Gad, P., Liu, C. T., Tillakaratne, N. J. K., Roy, R. R., et al. (2017). Feed-Forwardness of Spinal Networks in Posture Conflict of Interest: PG was employed by SpineX. SH was employed by Susan and Locomotion. Neuroscientist 23, 441–453. doi: 10.1177/107385841668 Hastings Pediatric Clinic. 3681 Grillner, S., and Zangger, P. (1975). How detailed is the central pattern generation The remaining author declares that the research was conducted in the absence of for locomotion? Brain Res. 88, 367–371. doi: 10.1016/0006-8993(75)90401-1 any commercial or financial relationships that could be construed as a potential Kreydin, E., Zhong, H., Latack, K., Ye, S., Edgerton, V. R., and Gad, P. (2020). conflict of interest. Transcutaneous Electrical Spinal Cord Neuromodulator (TESCoN) Improves Symptoms of Overactive Bladder. Front. Syst. Neurosci. 14:1. doi: 10.3389/fnsys. Copyright © 2021 Edgerton, Hastings and Gad. This is an open-access article 2020.00001 distributed under the terms of the Creative Commons Attribution License (CC BY). Maegele, M., Muller, S., Wernig, A., Edgerton, V. R., and Harkema, S. J. The use, distribution or reproduction in other forums is permitted, provided the (2002). Recruitment of spinal motor pools during voluntary movements versus original author(s) and the copyright owner(s) are credited and that the original stepping after human spinal cord injury. J. Neurotrauma 19, 1217–1229. doi: publication in this journal is cited, in accordance with accepted academic practice. No 10.1089/08977150260338010 use, distribution or reproduction is permitted which does not comply with these terms. Frontiers in Neuroscience | www.frontiersin.org 5 April 2021 | Volume 15 | Article 643463