Effects of focal injury on locomotor network activity below lesioned segments: a study of the rat spinal cord in vitro model

We have recently shown that a transient, combined application of the glutamate agonist kainate with a solution containing toxic metabolites (reactive radicals in hypoxic/aglycaemic conditions) induces a large injury to the rat spinal cord in vitro with irreversible loss of locomotor network activity. Clinical spinal lesions are often focally restricted to few spinal segments with later extension of the functional and structural damage to spinal areas initially spared by the primary insult. The present study intended to investigate the functional changes developing below and above a focal lesion by following up the activity of locomotor networks and correlating it with histological changes for up to 24 h later. For this purpose, we applied to an area of the spinal cord comprised between T11 and L1 a solution containing kainate and toxic metabolites for 1 h, washed it out and kept the preparation in normal saline for 24 h at which time it was fixed and sectioned for histochemical analysis. Initial experiments indicated that sham preparations not exposed to the toxic insult retained their electrophysiological and histological characteristics for up to 24 h in vitro. The toxic solution evoked a large depolarization of spinal networks and fully abolished reflex and locomotor activity of the segments within the barriers: such segments did not show any functional recovery with large loss of neurons and glia. These observations suggested that the lesion protocol was adequate to produce an effective focal injury. Below the lesion site, polysynaptic reflexes were transiently depressed (in conjunction with network depolarization), yet fully recovered after washout. Despite the large focal lesion, it was possible to elicit polysynaptic responses from ventral roots by applying stimuli that activated pathways across the lesioned area. This result indicated that the limited number of surviving axons cursing through injured segments could still functionally connect segments above and below the lesion with appropriate electrical stimuli. Fictive locomotion evoked by NMDA and serotonin applied to segments below the lesion was transiently abolished on the same day of the insult and recovered the day after with typical characteristics of motor pool alternation and periodicity. Conversely, trains of electrical stimuli applied to dorsal roots below the lesion could induce cumulative depolarization of segmental networks without triggering fictive locomotion. Notwithstanding the large number of surviving cells outside the lesioned area, microglia activation was detected with engulfing of spinal neurons, indicating the onset of a delayed, potentially damaging process of spinal networks. The present data provide a simple model to monitor the early evolution of focal injury with particular emphasis on locomotor networks and their cellular constituents.