Chemogenetic excitation and play promote functional recovery in juvenile spinal cord injury

Functional recovery after spinal cord injury depends on whether reorganizing descending circuits form adaptive, task-specific relays rather than maladaptive ones. The corticospinal tract undergoes competitive, activity-dependent refinement during a defined juvenile window, so the timing and polarity of cortical activity may dictate whether injured circuits rewire usefully. Whether biasing cortical activity during this window can steer reorganization toward recovery, and whether naturalistic rehabilitation supplies the activity such steering requires, remain unknown. In juvenile rats we delivered excitatory or inhibitory chemogenetic receptors, or a control vector, to the forelimb motor cortex at postnatal day 5, induced a right cervical hemisection at postnatal day 14, and gave oral clozapine-N-oxide during nightly rough-and-tumble play from postnatal day 20 to 50. Play, the rehabilitation common to all groups, did not differ between them, isolating cortical-activity polarity as the manipulated variable. Outcomes spanned forelimb dexterity, reaching kinematics, cortical output by intracortical microstimulation, descending architecture by intersectional viral tracing, and frequency-dependent depression of the Hoffmann reflex. In untreated injured animals, cortical output re-lateralized maladaptively toward the intact forelimb, driving the ipsilateral (intact) limb in 11 of 15 evoked responses and 3 of 5 animals. Excitation preserved selective drive to the impaired forelimb (38 of 39 responses; concordant-site proportion Kruskal-Wallis H = 9.08, exact P = 0.0031), and pure ipsilateral re-lateralization was absent from all nine treated animals (3 of 5 control versus 0 of 9 treated; Fisher's exact P = 0.027). Across behavioural, kinematic, and anatomical measures a consistent excitatory > control > inhibitory gradient emerged. Excitation enhanced wrist elevation and forward extension during reaching, restored frequency-dependent depression of the Hoffmann reflex, and selectively recruited three parallel lesion-bypassing relays: interhemispheric cortico-cortical-spinal, cortico-reticulo-spinal, and cortico-propriospinal (ipsilesional cortical relay neurons F(2,14) = 56.96, P < 0.0001). Inhibition yielded the weakest behavioural recovery and least relay recruitment, and did not restore contralesional cortical drive. Cortical-activity polarity during a juvenile window thus acts as a developmental switch governing whether descending reorganization is adaptive or maladaptive. Guided excitation, paired with behaviourally meaningful activity, reshapes diffuse sprouting into structured, lesion-bypassing relays and corrects the spontaneous re-lateralization that otherwise misdirects recovery. These findings provide a translational rationale for activity-paired neuromodulation in paediatric spinal cord injury.