This thesis examined two aspects of information processing by the feline spinal cord during centrally-evoked motor activity: 1) the modification of transmission from different sensory afferents and 2) the neuronal elements of reflex pathways from group II muscle afferents during fictive motor behaviours (i.e motoneuron activity under neuromuscular blockade). Fictive locomotion was evoked by electrical stimulation in the midbrain and fictive scratch was triggered by stimulation of the skin covering the ears following curare application to cervical dorsal roots in decerebrate in vivo feline preparations. Both monosynaptic and longer latency components of muscle and cutaneous afferent-evoked field potentials were reduced in amplitude during fictive locomotion and scratch, but field potentials evoked by muscle group II afferents were suppressed more than those evoked by cutaneous and group I muscle afferents recorded at the same spinal locations. The novel finding, that field potentials evoked at the same spinal locations by muscle and cutaneous afferents are suprressed differently, suggests that there is a preferential and non-uniform control of transmission from muscle and cutaneous fibres during motor activity. Extracellular recordings from neurons within the lumbar spinal segments showed that suppression of group II afferent input during fictive motor activity results in a powerful reduction of the activation of neurons with input from muscle group II afferents in 93% of the examined neurons after short trains of stimuli were delivered to peripheral nerves. However, more neurons remained recruitable by group II intensity stimulation if train duration was sufficiently long with only 33% showing a reduction in sensory-evoked firing. The majority of the neurons that remained responsive to muscle group II afferent input during fictive locomotion had axonal projections to supralumbar, or supraspinal areas and showed spontaneous, often rhythmic, firing activity. Overall, the studies presented in this thesis provide insights into the mechanisms by which the mammalian spinal cord processes sensory information and on how sensory input is able to control motor activity in spite of suppressive control provided by the nervous system. / October 2006

Spinal cord injuries can cause the catastrophic loss of motor and sensory function. The neurological deficits that result are the consequence of not only the primary injury to the spinal cord, but also a complex milieu of secondary pathological processes that are now beginning to be understood. The major mechanisms that underlie this secondary pathology include vascular disruption, ischemia, oxidative stress, excitotoxicity, and inflammation. In light of this, the fact that this secondary pathology occurs after the initial impact makes it potentially amenable to therapeutic intervention. Pharmacotherapies may attenuate some of these processes and minimize secondary damage. Some of the promising treatments that are emerging for acute spinal cord injury are drugs that are already used by physicians for the treatment of unrelated diseases. These drugs, which have already been established to be safe for humans, offer the unique advantage over other novel therapeutic interventions that have yet to be tested in humans. This would save a tremendous amount of time and money needed for human safety studies, if considered as a treatment for spinal cord injury. Examples of such drugs include minocycline (an antibiotic), erythropoietin (a recombinant hormone used to treat anemia), and statins (a popular class of blood cholesterol reducers), all of which have demonstrated the ability to attenuate the various pathophysiological processes initiated after trauma to the central nervous system. In a series of studies, erythropoietin, darbepoetin, atorvastatin, simvastatin, and minocycline were all evaluated for their ability to improve neurologic recovery in a clinically relevant model of spinal cord injury. My experiments revealed that erythropoietin, darbepoetin, atorvastatin and minocycline did not significantly improve neurological recovery. These negative results were in stark contrast to the positive findings which had been published in the literature suggesting that differences in experimental models and methodology influence the neuroprotective efficacy of these drugs. Simvastatin, on the other hand, demonstrated significant improvements in locomotor and histological outcomes. Although this is indeed exciting, the results were modest at best. My results highlight the need for further preclinical work on the above treatments to refine and optimize them prior to proposing them for human testing.

Previous research in our laboratory has shown that the spinal cord is capable of a simple form of instrumental learning. Spinally transected rats that receive controllable shock to an extended hindlimb exhibit a progressive increase in flexion duration that reduces net shock exposure. Subjects that receive uncontrollable shock, on the other hand, do not exhibit an increase in flexion duration, and are unable to produce this instrumental response even when they are later tested with controllable shock. This behavioral deficit can also be elicited by intermittent shock to the tail, and as little as 6 minutes of this shock is sufficient to produce a deficit that can last up to 48 hours as shown by Crown, Ferguson, Joynes, and Grau in 2002. Instrumental training has been shown to provide a number of beneficial effects. The instrumental training regimen produces a lasting effect that enables learning when subjects are later tested with a more difficult response criterion. Similarly, instrumental training can provide protection against the deleterious effects of uncontrollable shock as shown by Crown and Grau in 2001. The present study aims to determine the role of brain-derived neurotrophin factor (BDNF) in the beneficial effects of instrumental training. Experiments 1 and 2 examined the role of BDNF in the facilitory effect of instrumental training. Through the inhibition of endogenous BDNF, Experiment 1 showed that BDNF is necessary for the facilitation effect. Experiment 2 demonstrated that exogenous BDNF can produce the facilitation effect in dose-dependent fashion. Experiment 3 showed that the inhibition of BDNF attenuates the protective effect of instrumental training. Likewise, Experiment 4 showed that exogenous BDNF can substitute for instrumental training, and produce this protective effect. Experiment 5 showed that exogenous BDNF can block the development of the deficit when given immediately after uncontrollable shock. Experiment 6 showed that exogenous BDNF can block the expression of the deficit. Taken together, these experiments outline a major role for BDNF in mediating the beneficial effects of instrumental learning in the rat spinal cord.

A substantial body of work exists to suggest that brain and spinal mechanisms react differently to nociceptive information. The current experiments were design to identify parallels and differences in the way the spinal cord processes nociceptive information, as compared to intact animals. In addition, pharmacological manipulations were employed to identify the opioid receptors activated by continuous shock, and to decipher at what synaptic level (e.g. pre or post synaptically) intermittent shock affects the release of endogenous opioids. A common dependent variable was used in all experiments to assess changes in nociceptive reactivity, the tail-flick test. The results revealed that intermittent and continuous stimulation have an opponent relationship on nociceptive processing in the isolated spinal cord. Continuous stimulation (3, 25-s continuous 1.5 mA tail-shocks) induced an antinociceptive response that was attenuated by prior exposure to brief (80 ms) intermittent shock (Experiment 1). When intermittent shock was given after continuous shock, intermittent shock failed to attenuate continuous shock-induced antinociception (Experiment 2). The impact of intermittent shock on continuous-shock induced antinociception decayed after 24 hours (Experiment 3). Intermittent and continuous shock enhanced the antinociceptive consequences of a moderate dose of systemic morphine (5 mg/kg) (Experiment 4). Continuous shock-induced antinociception was attenuated by equal molar concentrations of CTOP (µ opioid antagonist) and Nor-BNI (κ opioid antagonist), but not naltrindole (δ opioid antagonist) (Experiment 5). Intermittent shock failed to attenuate the antinociception induced by DAMGO (µ opioid agonist) or Dynorphin A (κ opioid agonist).

Research has shown that exposure to just 6 minutes of uncontrollable shock 24 hours following contusion injury impairs locomotor recovery and leads to greater tissue loss at the injury epicenter. Uncontrollable shock is known to elevate corticosterone levels in intact rats and corticosterone exacerbates cell death in the hippocampus following injury, suggesting the effects may be related to a stress-induced release of corticosterone. Uncontrollable shock also affects other indices of stress including, spleen weight and norepinephrine, and has been shown to elevate pro-inflammatory cytokines. The present experiments were designed to assess whether uncontrollable shock has similar effects after contusion injury. Experiment 1 examined whether injury itself produced a stress response. Subjects received anesthesia alone, a laminectomy, or a contusion injury. Twenty-four hours later, they were restrained for 6 minutes and blood was collected from the leg. They were sacrificed 24 hours later and spleens were weighed, and plasma corticosterone and norepinephrine were assessed using ELISAs. IL-1! and IL-6 levels at the injury site were also measured using an ELISA. Contusion injury had no impact on any of the biological outcomes. For Experiment 2, subjects received 6 minutes of uncontrollable tailshock or an equivalent amount of restraint. Subjects were sacrificed 6, 24, 72, or 168 hours later. Uncontrollable shock caused a decrease in spleen weight and increased plasma corticosterone within 24 hours. Increases in IL-1! and IL-6 were also seen. Morphine was used in Experiment 3 to block the “psychological” component of uncontrollable shock. Subjects received morphine (20 mg/kg; i.p.) or saline 30 minutes prior to uncontrollable shock and were sacrificed 24 hours later. Morphine did not prevent the consequences of uncontrollable shock and, in some cases, potentiated its effects. The effect of controllability was examined in Experiment 4. After receiving a contusion injury, subjects received either controllable (master) or uncontrollable (yoked) legshock over the course of 2 days. A third group served as unshocked controls. Master subjects did not differ from yoked subjects on any of the biological outcomes measured. Unshocked subjects, however, exhibited an increase in corticosterone, IL-6, and blood monocytes.

Spinal cord neurons can support a simple form of instrumental learning that can be used to assess behavioral potential (plasticity) within this system. In this paradigm, subjects completely transected at the second thoracic vertebra learn to minimize shock exposure by maintaining a hindlimb in a flexed position. Preexposure to uncontrollable shock (shock independent of leg position) disrupts this learning. Activation of opioid receptors seems to contribute to the expression of the behavioral deficit observed after uncontrollable shock. Intrathecal application of naltrexone, a nonselective opioid receptor antagonist, blocked the expression, but not the induction, of the deficit. Treatment with nor-BNI, a kappa receptor antagonist, prior to testing had a similar effect, whereas mu (CTOP) and delta (naltrindole) receptor antagonists did not block the deficit. These findings suggest that prior exposure to uncontrollable shock induces a kappa opioid mediated event that inhibits learning. The current study examined the role of the kappa receptor in the behavioral deficit. Only GR89696, a selective kappa-2 receptor agonist, inhibited learning. This impairment was dose-dependent and, at the highest dose (30 nmol), inhibited learning for 96 hours. However, GR89696 and uncontrollable shock did not interact in an additive fashion. Instead, an intermediate dose attenuated the induction of the deficit. These findings suggest that activation of kappa receptors, specifically the kappa-2 subtype, inhibit instrumental learning and block the induction of the learning deficit. Both effects may be linked to the inhibition of NMDA-mediated plasticity.

Spinal cord injuries can cause the catastrophic loss of motor and sensory function. The neurological deficits that result are the consequence of not only the primary injury to the spinal cord, but also a complex milieu of secondary pathological processes that are now beginning to be understood. The major mechanisms that underlie this secondary pathology include vascular disruption, ischemia, oxidative stress, excitotoxicity, and inflammation. In light of this, the fact that this secondary pathology occurs after the initial impact makes it potentially amenable to therapeutic intervention. Pharmacotherapies may attenuate some of these processes and minimize secondary damage. Some of the promising treatments that are emerging for acute spinal cord injury are drugs that are already used by physicians for the treatment of unrelated diseases. These drugs, which have already been established to be safe for humans, offer the unique advantage over other novel therapeutic interventions that have yet to be tested in humans. This would save a tremendous amount of time and money needed for human safety studies, if considered as a treatment for spinal cord injury. Examples of such drugs include minocycline (an antibiotic), erythropoietin (a recombinant hormone used to treat anemia), and statins (a popular class of blood cholesterol reducers), all of which have demonstrated the ability to attenuate the various pathophysiological processes initiated after trauma to the central nervous system. In a series of studies, erythropoietin, darbepoetin, atorvastatin, simvastatin, and minocycline were all evaluated for their ability to improve neurologic recovery in a clinically relevant model of spinal cord injury. My experiments revealed that erythropoietin, darbepoetin, atorvastatin and minocycline did not significantly improve neurological recovery. These negative results were in stark contrast to the positive findings which had been published in the literature suggesting that differences in experimental models and methodology influence the neuroprotective efficacy of these drugs. Simvastatin, on the other hand, demonstrated significant improvements in locomotor and histological outcomes. Although this is indeed exciting, the results were modest at best. My results highlight the need for further preclinical work on the above treatments to refine and optimize them prior to proposing them for human testing.

No description available.

Prior work has shown that spinal neurons are capable of discriminating between temporally regular and temporally irregular stimulation. These effects have been observed using an in vivo assay of spinal plasticity based on an instrumental learning task, in which response-contingent leg shock produces an increase in flexion duration. Exposure to temporally regular stimulation (fixed spaced stimulation; FT) promotes learning, and temporally irregular stimulation produces a learning deficit. The experiments in this dissertation were designed to test other properties of fixed spaced shock that promote spinal plasticity and the structure responsible for the FT effect. Experiment 1 focused on the minimum number of stimulations necessary to re-establish the capacity to learn (a component of the “FT effect”), finding that180-360 shocks produced a learning deficit and that additional training (540-900 shocks) allowed learning. Experiment 2 found that shock number, not duration of exposure determined whether the FT effect emerged. Experiment 3 investigated if the FT effect emerges after shock was presented in two sessions separated by 24 hrs, and showed that two bouts of 360 shocks yielded the FT effect. Further, the initial bout of fixed spaced shock had a long-term benefit (Experiment 4). The results of Experiment 5 suggested that omitting shocks from a train of FT stimulation has little effect on the benefit of fixed spaced shock treatment. Experiment 6 replicated this observation, showing that randomly deleting half of the shocks (from a 720 FT shock series) had no effect on learning. Further, this schedule also induces a lasting protective effect, blocking the learning deficit produced by variable spaced shock (Experiment 7). To explore whether a central system or a peripheral filter mediates the FT effect, Experiment 8 challenged spinal neurons by phase shifting the relation between fixed spaced stimulation applied to two dermatomes. The FT effect only emerged when stimuli occurred in an alternating pattern across dermatomes, implying regularity is abstracted by a central system. Experiment 9 surgically isolated central pattern generator (L1-L2) from the portion of the spinal cord that mediates instrumental learning (L4-S2), finding that disrupting the connections between these two regions eliminated the FT effect.

No description available.