Sensory-based rehabilitation facilitates functional improvements after spinal cord injury (SCI) in animals and humans. Clinical programs for human patients utilize combinatorial Forced Exercise (FE) and Non-Forced Exercise (NFE) training programs to facilitate functional improvements (think treadmill vs. walking on a track). For FE, it is believed that incoming environmental cues work locally on the spinal level to mitigate functional deficits. While NFE, which enables for trial and error, introduces motor variability into the neural system. Variability helps fine tune descending brain commands to improve motor performance. However, key questions remain: How do different training paradigms change spinal cord circuits and contribute to motor recovery? Answers will help to fine tune these sensory-rehabilitation strategies to optimize functional recovery. The hypothesis of this proposal is that FE and NFE initiate distinct neural rewiring strategies and characterizable differences in functional recovery following SCI. I will use mouse genetic strategies to illuminate anatomical changes in rewiring. Specifically, I will examine the changes in neural circuitry of a premotor network, Deep Dorsal horn spinal cord ParvalBumin+ neurons (dPVs). dPVs receive convergent touch, proprioceptive, and supraspinal information, and are differentially engaged during task-dependent motor behaviors. Changes in inputs onto or outputs from dPVs therefore reflect the contributions of paradoxical training (FE/NFE) in mediating recovery. Specific Aim 1 will test the hypothesis that FE will evoke elevated sensory neural rewiring, while NFE will result in increased descending brain inputs. I will couple genetic approaches with quantitative synaptic analysis to anatomically map corticospinal and touch/proprioceptive inputs onto dPVs. Aim 1’s training potential lies in learning mouse genetics, high resolution quantitative synaptic analysis, and to further hone my biostatistics training. Specific Aim 2 will test the hypothesis that NFE facilitates smooth naturalistic motor movements and behavioral state maps more closely related to the preinjury condition than FE. I will couple muscle activity recordings with highly sensitive computer vision/machine learning to investigate the influence of training (1) granularly on muscle responses (EMGs) and joint activity and (2) holistically on naturalistic behavior. I will also characterize dPVs outputs onto motor neurons using genetic approaches and label affected motor neurons innervating leg muscles with implanted electrodes. Aim 2’s training potential is rooted in cutting-edge computational techniques, muscle recordings, and data interpretation. The collective results will provide an understanding of the (1) supraspinal and sensory integration and evoked motor responses involved in NFE and (2) functional benefits of NFE in complex motor behaviors. My research will break down current barriers of translational preclinical research to help instruct clinical rehabilitation. The high training potential for these Aims has been carefully designed to fill my gap-based knowledge. The impact of this fellowship will foster my successful, impactful, and enduring independent research career in the field of SCI.
