GDA transplantation suppresses atrophy of red nucleus neurons and promotes robust behavioral recovery. (a) Injured left-side red nuclei in control rats contained an average of 52% of the neurons counted in uninjured right-side red nuclei at 5 weeks after injury. The numbers of neurons in the injured left-side red nuclei of GDA-transplanted animals, however, was 81% of total neuron numbers in uninjured right-side nuclei (*p < 0.01). (b) Grid-walk analysis of locomotor recovery. Graph showing the average number of mistakes per experimental group at different time points after injury for GDA-transplanted rats versus the control-lesion and sham-operated groups. GDA-transplanted animals (green) performed significantly better than lesioned controls at all post-injury time points (p < 0.05). (c) Transplanted GRPs do not promote locomotor recovery. Graph showing the average number of grid-walk mistakes per experimental group from 1 day before injury (baseline pre-lesion) to 2 weeks after injury for a separate series of matched RST-lesioned rats that received either GRP or GDA transplants versus lesion-only control rats. Note the complete failure of locomotor recovery in GRP-transplanted animals compared with lesion-only controls at all time points and confirmation of significant locomotor recovery in response to GDA transplantation (p < 0.05). cs, cyclosporine.