A Cervical Spinal Cord Hemi-Contusion Injury Model Based on Displacement Control in Non-Human Primates (Macaca fascicularis)

Abstract

Non-human primate (NHP) spinal cord injury (SCI) models can be informative in the evaluation of treatments that show promise in rodent models prior to translation to humans. In the present study, we aimed to establish a cervical spinal hemi-contusion model with controlled displacement and evaluate the abnormalities in behavior, electrophysiology, histology, and magnetic resonance imaging. Twelve adult NHPs were divided into an SCI group (n = 8, 24 and 48 weeks) and a control group (n = 4). An impactor (Φ = 4 mm) was driven to compress the left C5 cord at 800 mm/sec. The contusion displacement and peak force was 4.08 ± 0.17 mm and 19.8 ± 4.6 N. The behavioral assessment showed a consistent dysfunction below the wrist and spontaneous recovery of limb function after injury. Lesion length and lesion area at the epicenter based on T2 hyperintensity were 5.68 ± 0.47 mm and 5.99 ± 0.24 mm2 at 24 weeks post-injury (wpi), and 5.29 ± 0.17 mm and 5.95 ± 0.24 mm2 at 48 wpi. The spared spinal cord area immuno-positive for glial fibrillary acidic protein was significantly reduced, while the staining intensity increased at 24 wpi and 48 wpi, compared with the sham group. Ipsilateral somatosensory and motor evoked potentials were dynamic, increasing in latency and decreasing in amplitude compared with pre-operative values or the contralateral values, and correlated to varying degrees with behavioral outcomes. A shift in size–frequency distribution of sensory neurons of the dorsal root ganglia (DRG) was consistent with a loss of large-diameter cells. The present study demonstrated that the NHP SCI model resulted in consistent unilateral limb dysfunction and potential plasticity in the face of loss of spinal cord and DRG tissue.