Search

Clinical spinal cord injury (SCI) is accompanied by comorbid peripheral injury in 47% of patients. Human and animal modeling data have shown that painful peripheral injuries undermine long-term recovery of locomotion through unknown mechanisms. Peripheral nociceptive stimuli induce maladaptive synaptic plasticity in dorsal horn sensory systems through AMPA receptor (AMPAR) phosphorylation and trafficking to synapses. Here we test whether ventral horn motor neurons in rats demonstrate similar experience-dependent maladaptive plasticity below a complete SCI in vivo. Quantitative biochemistry demonstrated that intermittent nociceptive stimulation (INS) rapidly and selectively increases AMPAR subunit GluA1 serine 831 phosphorylation and localization to synapses in the injured spinal cord, while reducing synaptic GluA2. These changes predict motor dysfunction in the absence of cell death signaling, suggesting an opportunity for therapeutic reversal. Automated confocal time-course analysis of lumbar ventral horn motor neurons confirmed a time-dependent increase in synaptic GluA1 with concurrent decrease in synaptic GluA2. Optical fractionation of neuronal plasma membranes revealed GluA2 removal from extrasynaptic sites on motor neurons early after INS followed by removal from synapses 2 h later. As GluA2-lacking AMPARs are canonical calcium-permeable AMPARs (CP-AMPARs), their stimulus- and time-dependent insertion provides a therapeutic target for limiting calcium-dependent dynamic maladaptive plasticity after SCI. Confirming this, a selective CP-AMPAR antagonist protected against INS-induced maladaptive spinal plasticity, restoring adaptive motor responses on a sensorimotor spinal training task. These findings highlight the critical involvement of AMPARs in experience-dependent spinal cord plasticity after injury and provide a pharmacologically targetable synaptic mechanism by which early postinjury experience shapes motor plasticity.

Most experimental studies of spinal cord injury (SCI) in rats damage the thoracic cord, with the consequent functional loss being due to interruption of long tracts connecting the caudal spinal cord to the rostral nervous system. Less work has been done evaluating injury to the cervical cord, even though it is the most common level of human SCI. In addition to the long tracts, the cervical spinal cord contains the sensory and motor neurons responsible for upper extremity function. The purpose of this study was to further develop a rat model of cervical spinal cord contusion injury using a modified NYU/MASCIS weight drop device. Mild (6.25 mm) and moderate (12.5 mm) C5 unilateral injuries were produced. Behavioral recovery was examined using a grooming test, a paw preference test, a walkway test (The Catwalk), and a horizontal ladder test. Histological outcome measures included sparing at the lesion epicenter, sparing throughout the extent of the lesion, quantification of myelin loss rostral and caudal to the lesion, and motor neuron counts. Compared to controls, animals receiving SCI exhibited injury severity-specific deficits in forelimb, locomotor, and hindlimb function persisting for 6-weeks post-SCI. Histological analysis revealed ipsilateral containment of the injury, and differentiation between groups on all measures except motor neuron counts. This model has many advantages: (1) minimal animal care requirements post-SCI, (2) within subject controls, (3) functional loss involves primarily the ipsilateral forelimb, and (4) it is a behavioral and histological model for both gray and white matter damage caused by contusive SCI.

The present study analyzed the anatomical plasticity of serotonergic immunoreactive projections to external anal sphincter (EAS) motoneurons, and the behavioral plasticity of EAS reflexes, penile erection, and locomotion in rats with spinal contusion injury (SCI) or complete spinal cord transection (TX). Electromyographic activity of the EAS, penile erection latency, and BBB locomotor score exhibited parallel recovery over the 6-week recovery period after contusion SCI. This pattern of recovery was not observed in TX animals. While locomotor scores demonstrated a small increase after TX, erectile and anorectal function remained at abnormal levels established immediately after injury. Serotonergic immunofluorescent (5-HT-IF) staining at the lesion site identified a small number of fibers spared after SCI that may provide a substrate for functional recovery. Pixel density measurements of 5-HT-IF in the vicinity of retrogradely labeled EAS and unlabeled pudendal motoneurons necessary for penile erection provide indirect evidence of serotonergic sprouting that parallels the observed functional recovery in animals with SCI. No 5-HT-IF was detected caudal to the injury site in TX animals. These studies indicate: (1) lumbosacral eliminative and reproductive reflexes provide a valid means of studying the mechanisms of post-SCI plasticity; (2) the similar recovery curves suggest similar return of descending control, perhaps through sprouting of descending serotonergic fibers; (3) the observed deficits after TX likely represent the permanent removal of descending inhibition and reflect reorganization of segmental circuitry.

Previous physiological and behavioral studies have shown that the nucleus raphe obscurus (nRO) modulates pelvic floor reflex function (Yamanouchi and Kakeyama [1992] Physiol. Behav. 51:575-579; Beattie et al. [1996] Soc. Neurosci. Abstr. 22:722.4; Holmes et al. [1997] Brain Res. 759:197-204). In the present study, small injections of fluorescent tracers were used to investigate direct descending projections from the rostral and caudal portions of the brainstem nRO to retrogradely labeled pudendal motoneurons (MN) in the male rat. The caudal nRO projects into the ventral and lateral funiculi of the spinal cord, with arborizations in the thoracic intermediolateral cell column; in laminae VII, IX, and X of the lumbosacral cord; and in the sacral parasympathetic nucleus (SPN). Many identified external anal sphincter and ischiocavernosus MNs appeared to be in direct apposition with fibers originating from the caudal nRO; and more than half of the bulbospongiosus MNs that were identified appeared to receive such descending input. In addition to the nRO spinal autonomic and pudendal motoneuronal targets, projections were observed to regions of the intermediate gray that contain interneurons organizing the pelvic floor reflexes and to MN pools that are involved in functionally related somatic activities. Finally, several neurons in the lumbar enlargement were labeled retrogradely with FluoroRuby after injections into the nRO and the immediately adjacent reticular formation. Thus, the nRO may be in a position to modulate the coordinated actions of autonomic preganglionic and functionally related skeletal MN activity involved in sexual and eliminative reflex functions.

Previous research has demonstrated that anorectal contractions in the rat are modulated by activation of spinal autonomic circuits. In the present study, anterograde tracing of descending pathways originating from the caudal nucleus raphe obscurus (nRO) revealed that this nucleus projects to cells within the intermediolateral (IML) cell column of the thoracic cord and the sacral parasympathetic nucleus (SPN). These anatomical studies suggested that the nRO may influence the regulation of spinal reflexes of the pelvic floor. In a second set of experiments, acute rat preparations were used to investigate changes in anorectal motility during electrical stimulation of the nRO. Anorectal contractions were measured by a fluid-filled manometer. Electrical stimulation of the nRO significantly reduced spontaneous anorectal activity when compared to baseline contractions recorded for 1 min prior to stimulation. Stimulation sites outside the nRO did not affect anorectal contractions when compared to either (a) the 1-min pre-stimulation baseline for that site or (b) the 1-min stimulation period for sites within the nRO. Stimulation of caudal portions of the nRO were more likely than the rostral nRO to reduce anorectal contractions. Given that the SPN contains preganglionic neurons which may be involved in control of anorectal contractions (mediated via the pelvic nerve), the studies presented here suggest a functional role for nRO regulation of preganglionic motoneurons innervating the distal gut of the rat.

The locations and dendritic arbors of sacral motoneurons (MNs) innervating the striated external anal and urethral sphincters (EAS and EUS, respectively) of the cat were investigated using muscle injections of cholera toxin-conjugated HRP (CT-HRP). Serial reconstructions showed that all cells labelled after EAS injections were located in the dorsomedial (DM) subdivision of Onuf's nucleus, whereas all cells labelled from the EUS were located in the ventrolateral (VL) subdivision. The dendritic arbors of EAS and EUS MNs were very different, suggesting differences in afferent control. In addition, prominent bundles of dendrites extended preferentially into the regions occupied by functionally appropriate preganglionic neurons (PGNs) in the sacral parasympathetic nucleus (SPN) which innervate the colon and bladder.

The relationship of lumbar dorsal root afferents to lateral motor column motoneurons was studied using anterograde injury filling of dorsal roots and retrograde injury filling of ventral roots with horseradish peroxidase. At the light microscopic level, horseradish peroxidase labelled dorsal root axons were observed to separate into a medial division of large diameter axons which enter the dorsal funiculus and a lateral division of small diameter axons which form a compact bundle in the dorsolateral funiculus which may be homologous to the mammalian tract of Lissauer. Within the spinal gray, primary afferents terminate in two distinct regions. The more ventral of these terminal fields, which receives collaterals of primary afferent axons in the dorsal funiculus, overlaps the dendritic arborizations of the lateral motor column motoneurons. Some axons leave the ventral terminal field to enter the dorsal lateral motor column. Here they terminate on the primary dendrites and somata of lateral motor column motoneurons. At the electron microscopic level, labelled primary afferent terminals were seen to synapse upon lateral motor column motoneuron dendrites as well as upon the somata of dorsally positioned lateral motor column motoneurons. These terminals contain small spherical vesicles and occasional dense-cored vesicles. The synaptic specializations are characterized by a small amount of postsynaptic material. The lateral motor column may be divided into dorsal and ventral portions on the basis of the primary afferent distribution and this is in accord with functional, physiological and developmental data.

Dorsal and ventral roots of Rana catesbeiana tadpole lumbar spinal cord were labeled with horseradish peroxidase during development of the hindlimb. Labeled motoneurons in the process of migrating were found in the region of the developing terminal arbors of dorsal root axons. Electron microscopy showed that some of these dorsal root terminals contacted the migrating motoneurons.

The effects of testosterone administration on penile reflexes, and on the motoneurons of the spinal nucleus of the bulbocavernosus which innervate perineal muscles involved in these reflexes, were investigated in castrated male rats. Penile reflexes were restored following 48 h of testosterone administration initiated 6 weeks after castration. The amount of synaptic input to the identified motoneurons was increased following short term testosterone treatment, compared to that seen in animals receiving no testosterone, albeit to a lesser extent than that seen in animals receiving long term testosterone treatment. This increase in synaptic inputs in the short term testosterone group occurred despite the lack of an increase in somatic area. Thus, plasticity of the synaptic input to these neurons, as well as recovery of penile reflexes, occurred as a result of alterations in the hormonal state of the animal, and such changes occurred relatively rapidly.