Abstract
Because the presence of pain impedes motor recovery in individuals with spinal cord injury (SCI), it is necessary to understand their supraspinal substrates in translational animal models. Using functional magnetic resonance imaging (fMRI) in a rat model of hemicontusion cervical SCI, supraspinal changes were mapped and correlated with sensorimotor behavioral outcomes. Female adult rats underwent sham or SCI using a 2.5 mm impactor and 150 kdyn force. SCI permanently impaired motor activity in only the ipsilesional forelimb along with thermal hyperalgesia at 5 and 6 weeks. Spinal MRI at 8 weeks after SCI showed ipsilateral T1 and T2 lesions with no discernable lesions across shams. fMRI mapping during electrical forepaw stimulation indicated SCI-induced sensorimotor reorganization with an expansion of the contralesional forelimb representation. Resting state fMRI-based functional connectivity density (FCD), a marker of regional neuronal hubs, increased or decreased across brain regions involved in nociception. FCD increases after SCI were in the primary and secondary somatosensory cortices (S1 and S2), anterior cingulate cortex (ACC), insula, and the pre-frontal cortex (PFC), and decreases were across the hippocampus, thalamus, hypothalamus, and amygdala in SCI. Resting state functional connectivity (RSFC) assessments from the FCD altered regions of interest indicated cortico-cortical RSFC increases and cortico-insular, cortico-thalamic, and cortico-hypothalamic RSFC decreases after SCI. Hippocampus, amygdala, and thalamus showed decreased RSFC with most cortical regions and between themselves except the hippocampus-amygdala network, which showed increased RSFC after SCI. Whereas select nociceptive region's intrinsic activity associated strongly with evoked pain behaviors after SCI (e.g., PFC, ACC, hippocampus, thalamus, hypothalamus, M1, and S1(BF)) other nociceptive regions had weaker associations (e.g., amygdala, insula, auditory cortex, S1(FL), S1(HL), S2, and M2), but differed significantly in their intrinsic activities between sham and SCI. The weaker associated nociceptive regions may possibly encode both the evoked and affective components of pain.