A graded contusion spinal cord injury (SCI) was created in the adult rat spinal cord using the Infinite Horizons (IH) impactor to study the correlation between injury severity and anatomical, behavioral, and electrophysiological outcomes. Adult Fisher rats were equally divided into five groups and received contusion injuries at the ninth thoracic level (T9) with 100, 125, 150, 175, or 200 kdyn impact forces, respectively. Transcranial magnetic motor-evoked potentials (tcMMEPs) and BBB open-field locomotor analyses were performed weekly for 4 weeks postinjury. Our results demonstrated that hindlimb locomotor function decreased in accordance with an increase in injury severity. The locomotor deficits were proportional to the amount of damage to the ventral and lateral white matter (WM). Locomotor function was strongly correlated to the amount of spared WM, which contains the reticulospinal and propriospinal tracts. Normal tcMMEP latencies were recorded in control, all of 100-kdyn-injured and half of 125-kdyn-injured animals. Delayed latency responses were recorded in some of 125-kdyn-injured and all of 150-kdyn-injured animals. No tcMMEP responses were recorded in 175- and 200-kdyn-injured animals. Comparison of tcMMEP responses with areas of WM loss or demyelination identified the medial ventrolateral funiculus (VLF) as the location of the tcMMEP pathway. Immunohistochemical and electromicroscopic (EM) analyses showed the presence of demyelinated axons in WM tracts surrounding the lesion cavities at 28 days postinjury. These data support the notion that widespread WM damage in the ventral and lateral funiculi may be a major cause for locomotor deficits and lack of tcMMEP responses after SCI.

Most in vivo spinal cord injury (SCI) experimental models use rodents. Due to the anatomical and functional differences between rodents and humans, reliable large animal models, such as non-human primates, of SCI are critically needed to facilitate translation of laboratory discoveries to clinical applications. Here we report the establishment of a controlled spinal contusion model that produces severity-dependent functional and histological deficits in non-human primates. Six adult male rhesus macaque monkeys underwent mild to moderate contusive SCI using 1.0 and 1.5mm tissue displacement injuries at T9 or sham laminectomy (n=2/group). Multiple assessments including motor-evoked potential (MEP), somatosensory-evoked potential (SSEP), MR imaging, and monkey hindlimb score (MHS) were performed. Monkeys were sacrificed at 6 months post-injury, and the lesion area was examined for cavitation, axons, myelin, and astrocytic responses. The MHS demonstrated that both the 1.0 and 1.5mm displacement injuries created discriminative neurological deficits which were severity-dependent. The MEP response rate was depressed after a 1.0mm injury and was abolished after a 1.5mm injury. The SSEP response rate was slightly decreased following both the 1.0 and 1.5mm SCI. MRI imaging demonstrated an increase in T2 signal at the lesion site at 3 and 6months, and diffusion tensor imaging (DTI) tractography showed interrupted fiber tracts at the lesion site at 4h and at 6 months post-SCI. Histologically, severity-dependent spinal cord atrophy, axonal degeneration, and myelin loss were found after both injury severities. Notably, strong astrocytic gliosis was not observed at the lesion penumbra in the monkey. In summary, we describe the development of a clinically-relevant contusive SCI model that produces severity-dependent anatomical and functional deficits in non-human primates. Such a model may advance the translation of novel SCI repair strategies to the clinic.