Neural Adaptation with Locomotor Training in the Functionally Isolated Human Spinal Cord

S.J. Harkema^1,2, J. Beres-Jones¹, C.K. Ferreira^2
1Department of Neurology - Assistant Professor, ²Brain Research Institute, University of California, Los Angeles, CA 90095

 

Table of Contents ::Home ::Introduction ::Methods ::Results ::Conclusions ::References ::Acknowledgements

::Introduction The theoretical framework for the spinal interneuronal reorganization presented here is drawn from decades of studies of spinal interneuronal systems and observations from from the functionally isolated human spinal cord after spinal cord injury (1-16). Clonus, one manifestation of spasticity, produces involuntary neuromuscular responses that can interfere with the ability to walk after spinal cord injury (SCI). We propose that the central oscillators responsible for clonus interact with the interneuronal circuits that generate standing and stepping. After human SCI, reconfiguration of these interneuronal circuits may occur not only from loss of supraspinal input but also from chronic unloading of the legs. This interneuronal reorganization then routinely results in clonic efferent output, rather than more functional tonic efferent output. Thus, re-introducing loading of the legs should modulate clonus after severe SCI, possibly even reducing the amount of clonus. Standing and stepping are two functional tasks that provide load not experienced during manually induced stretch. These motor tasks provide different repetitive afferent information to the spinal cord; for example, bilateral leg loading during standing versus alternate leg loading with stepping. The specific afferent input associated with loading of these different motor tasks should differentially modulate the oscillatory clonic efferent output.   Figure 1. EMG activity (microvolts, ?V) from the right soleus (RSOL) of a SCI subject during unsustained (left) and sustained (right) clonus. The black line represents muscle length and “up” is muscle stretch. Arrows identify the refractory period where no activity occurs, although the stretch is maintained. Clonus results in dispersed EMG activity of approximately 58 ms in duration followed by a refractory period of approximately 135 ms (1-3). If stretch is maintained, then clonic bursts will reoccur at a frequency of 5-8 Hz. Studies have detailed the role of the Ia afferents and gamma system contribution to clonus (8). Further studies outlined the importance of a spinal oscillator that generates rhythmicity once activated by peripheral events (2,3). We suggest that sustained ankle clonus requires both continuous Ia afferent input and an elevated central state of excitability which maintains the activity of a central oscillator that provides oscillatory excitatory and inhibitory drive to the motor nuclei Figure 2. EMG activity (microvolts, ?V) and muscle length (black line, “up” represents muscle lengthening or stretch) with sequential theoretical model of interneuronal organization after severe human spinal cord injury (Dimitrijevic et. al. 1967-1980).   ::Methods Thirty individuals with a severe SCI were studied during manually induced stretch of the plantarflexors, stepping and standing using body weight support on a treadmill with manual assistance. EMG activity (microvolts ?V) was obtained from the soleus (SOL), medial gastrocnemius (MG), tibialis anterior (TA), medial hamstrings (MH) and vastus lateralis (VL). We calculated the co-activity (the relative time that two muscles are active and inactive simultaneously) between agonist and antagonist muscle pairs (Johnson 2001). In brief, we calculate quantities to describe the co-activity between two muscles that result in values between 0.0 and 1.0. That is, muscle pairs that are exactly 180o out of phase with each other will result in a value of 0.0, and completely synchronized activity in a value of 1.0. Generally, we consider a value <0.5 as co-activation of two muscles and a value >0.5 as alternation between the muscles. This method also calculates clonicity (relative amount of clonus to total EMG activity). Clonicity values fall within a range of 0.0 and 1.0 with a higher value representing more clonic EMG activity.   ::Results Figure 3. EMG activity (microvolts, uV) during sustained clonus from the left and right soleus (L SOL, R SOL) from a SCI subject during sustained stretch of the left plantarflexors. Mean co-activation value is 0.33 +/- 0.02. Figure 4. (Left) EMG activity (microvolts, ?V) during sustained clonus from the left and right soleus (LSOL, RSOL), tibialis anterior (LTA, RTA) and from the left SOL, TA and vastus lateralis (LVL) from a SCI subject during sustained stretch of the left plantarflexors. Mean coactivity values (left graph).LSOL/LTA 0.76 +/- 0.02; RSOL/RTA 0.60 +/-0.03; RSOL/LSOL 0.23 +/-0.01; LTA/RTA 0.35 +/- 0.02; (right graph) LSOL/LTA 0.79 +/- 0.04; LSOL/LVL 0.67 +/- 0.05;LTA/LVL 0.63 +/- 0.04 Figure 5. EMG activity (microvolts; uV) from the vastus lateralis (VL), soleus (SOL), and medial gastrocnemius (MG) from the left (L) and right (R) limbs of a spinal cord injured subject during bilateral stepping (left) and standing (right) using body weight support and manual assistance. The black horizontal bar signifies the footswitch (FS) data, which indicates foot-ground contact. Clonic EMG is co-activated among ipsilateral and contralateral extensors. Mean co-activity values for stepping in red and standing in blue are: LSOL/LMG: 0.78 +/- .03, 0.89 +/- .02; LSOL/LVL: 0.68 +/- .04, 0.63 +/- .02; L MG/LVL: 0.62 +/- .03, 0.67 +/- .02; RSOL/RMG: 0.69 +/- .03, 0.88 +/- .01; RSOL/RVL: 0.78 +/- .03, 0.60 +/- .02; RMG/RVL: 0.78 +/- .03, 0.64 +/- .02; LSOL/RSOL: 0.34 +/- .08, 0.72 +/- .02; LMG/R MG: 0.37 +/- .07, 0.65 +/- .01; LVL/RVL: 0.58 +/- .1, 0.73 +/- .02;   ::Conclusion Interaction of peripheral stimuli and central oscillators generate clonus. Clonus can be used as an experimental tool to understand neural plasticity after neurologic injury. Propriospinal circuits can be activated and result in the spreading of clonic activity across several segments (L2-S1) without changing the mechanics of the system. Sustained ankle clonus may require both continuous Ia afferent input and an elevated CSE, which maintains the activity of a central oscillator that provides oscillatory excitatory and inhibitory drive to the motor nuclei Afferent input related to loading and rate of stepping modulates the features and can reduce clonic EMG in individuals with severe SCI. The timing of the clonic EMG activity is influenced by the phase of the step cycle and the associated afferent input (e.g. load cutaneous and stretch). The presentation of long-term task-specific repetitive afferent information can decrease clonogenic efferent output. Functional reorganization of spinal neural networks occur with task specific training.   ::References 1. Beres-Jones JA, Johnson TD, Harkema SJ. Clonus after human spinal cord injury cannot be attributed solely to recurrent muscle-tendon stretch. Exp Brain Res 149(2): 222-236, 2003. 2. Dimitrijevic MR, Nathan PW. Studies of spasticity in man: analysis of reflex activity evoked by noxious cutaneous stimulation. Brain XCI: 350-368, 1967. 3. Dimitrijevic MR, Nathan PW, Sherwood AM. Clonus: the role of central mechanisms. Journal of Neurology 43: 321-332, 1980. 4. Duysens J, van de Crommert H. Neural control of locomotion; The central pattern generator from cats to humans. Gait and Posture 7: 131-141, 1998. 5. Gossard J-P, Hultborn H: in Wernig A (ed): Plasticity of Motoneural Connections. Elsevier Science Publishers BV, 1991, pp 385-404. 6. Grillner S, Wallén Peter. Central pattern generators for locomotion, with special reference to vertebrates. Ann Rev Neurosci 8: 233-261, 1985. 7. Grillner S, Zangger P. On the central generation of locomotion in the low spinal cat. Experimental Brain Research 34: 241-261, 1979. 8. Hagbarth KE, Wallen G, Leofstedt L. Muscle spindle activity in man during voluntary fast alternating movements. Journal of Neurology, Neurosurgery and Psychiatry 38(7): 625-635, 1975. 9. Hultborn H. State-dependent modulation of sensory feedback. Journal of Physiology 533(Pt 1): 5-13, 2001. 10. Jankowska E. Interneuronal relay in spinal pathways from proprioceptors. Prog Neurobiol 38(4): 335-378, 1992. 11. Jankowska E. Spinal interneuronal systems: identification, multifunctional character & reconfigurations in mammals. J Physiol 533(Pt 1): 31-40, 2001. 12. Johnson TD, Elashoff RM, Harkema SJ. A bayesian change-point analysis of electromyographic data: detecting muscle activation patterns and associated application. Biostatistics 4(1): 143-164, 2002. 13. Lundberg A. Convergence of excitatory and inhibitory action on interneurones in the spinal cord. UCLA Forum Med Sci 11: 231-265, 1969. 14. McCrea DA: Spinal circuitry of sensorimotor control of locomotion, Journal of Physiology 533(1): 41-50, 2001. 15. Pearson KG. Neural Adaptation in the generation of rhythmic behavior. Annu.Rev.Physiol. 62: 723-753. 2000. 16. Pearson KG, Rossignol S. Fictive motor patterns in chronic spinal cats. Journal of Neurophysiology: 1874-1887, 1991.   ::Acknowledgements Top Row: Amy Budovitch, Janell Beres-Jones, Christie Ferreira, Michelle Adams Middle Row: Bruce Chorney, Patrick Hu, Sonia Newman, Jacquelyn Leong, Luis Alvarado Bottom Row: Jennifer Cobanov, Edward Lan, Jeanine Yip, Mary Tran We would like to thank the research volunteers for their dedication and valuable contributions to these studies. This work was supported by NIH grants NS36854, NS16333, and RR00865; The Christopher Reeve Foundation; and the Roman Reed Spinal Cord Injury Research Fund of California.