Secondary hypoxemia exacerbates the reduction of visual discrimination accuracy and neuronal cell density in the dorsal lateral geniculate nucleus resulting from fluid percussion injury.

The purpose of this study was to determine the impact of secondary hypoxemia on visual discrimination accuracy after parasagittal fluid percussion injury (FPI). Rats lived singly in test cages, where they were trained to repeatedly execute a flicker-frequency visual discrimination for food. After learning was complete, all rats were surgically prepared and then retested over the following 4-5 days to ensure recovery to presurgery levels of performance. Rats were then assigned to one of three groups [FPI + Hypoxia (IH), FPI + Normoxia (IN), or Sham Injury + Hypoxia (SH)] and were anesthetized with halothane delivered by compressed air. Immediately after injury or sham injury, rats in groups IH and SH were switched to a 13% O2 source to continue halothane anesthesia for 30 min before being returned to their test cages. Anesthesia for rats in group IN was maintained using compressed air for 30 min after injury. FPI significantly reduced visual discrimination accuracy and food intake, and increased incorrect choices. Thirty minutes of immediate posttraumatic hypoxemia significantly (1) exacerbated the FPI-induced reductions of visual discrimination accuracy and food intake, (2) further increased numbers of incorrect choices, and (3) delayed the progressive recovery of visual discrimination accuracy. Thionine stains of midbrain coronal sections revealed that, in addition to the loss of neurons seen in several thalamic nuclei following FPI, cell loss in the ipsilateral dorsal lateral geniculate nucleus (dLG) was significantly greater after FPI and hypoxemia than after FPI alone. In contrast, neuropathological changes were not evident following hypoxemia alone. These results show that, although hypoxemia alone was without effect, posttraumatic hypoxemia exacerbates FPI-induced reductions in visual discrimination accuracy and secondary hypoxemia interferes with control of the rat's choices by flicker frequency, perhaps in part as a result of neuronal loss and fiber degeneration in the dLG. These results additionally confirm the utility of this visual discrimination procedure as a sensitive, noninvasive means of assessing behavioral function after experimental traumatic brain injury.

Arteether-induced brain injury in Macaca mulatta. I. The precerebellar nuclei: the lateral reticular nuclei, paramedian reticular nuclei, and perihypoglossal nuclei.

Malaria poses a threat across several continents: Eurasia (Asia and parts of Eastern Europe), Africa, Central and South America. Bradley (1991) estimates human exposure at 2,073,000,000 with infection rates at 270,000,000, illnesses at 110,000,000, and deaths at 1,000,000. Significant mortality rates are attributed to infection by the parasite Plasmodium falciparum, with an estimated 90% among African children. A worldwide effort is ongoing to chemically and pharmacologically characterize a class of artemisinin compounds that might be promising antimalarial drugs. The U.S. Army is studying the efficacy and toxicity of several artemisinin semi-synthetic compounds: arteether, artemether, artelinic acid, and artesunate. The World Health Organization and the U.S. Army selected arteether for drug development and possible use in the emergency therapy of acute, severe malaria. Male Rhesus monkeys (Macaca mulatta) were administered different daily doses of arteether, or the vehicle alone (sesame oil), for a period of either 14 days, or 7 days. Neuropathological lesions were found in 14-day arteether treated monkeys in the precerebellar nuclei of the medulla oblongata, namely: (1) the lateral reticular nuclei (subnuclei magnocellularis, parvicellularis, and subtrigeminalis), (2) the paramedian reticular nuclei (subnuclei accessorius, dorsalis, and ventralis), and the perihypoglossal nuclei (n. intercalatus of Staderini, n. of Roller, n. prepositus hypoglossi). The data demonstrate that the simina meduallry precerebellar nuclei have a high degree of vulnerability when arteether is given for 14 days at dose levels between 8mg/kg per day and 24 mg/kg per day. The neurological consequences of this treatment regimen could profoundly impair posture, gait, and autonomic regulation, while eye movement disorders might also be anticipated.

Behavioral and neural toxicity of arteether in rats.

Repeated administration of the artemisinin antimalarial compound, 3-arteether (AE) (25 mg/kg, i.m.) was evaluated in rats using a two-choice, discrete trial, auditory discrimination task and subsequent neurohistology. Rats were trained to choose one of two response levers following presentation of white noise or a tone + white noise. Increasing and decreasing the intensity of the tone increased and decreased discriminability, respectively, and differential reinforcement density produced systematic changes in response bias. AE (n = 5) or vehicle (n = 5) was injected daily (9-12 days). Initial injections of AE did not affect behavioral performance. Continuing daily injections produced significant decreases in choice accuracy and significant increases in choice reaction time. When overt signs of severe toxicity were observed, rats were sacrificed and significant neural pathology was observed in the nucleus trapezoideus of AE-treated rats. In a subsequent experiment, AE was injected for 3 (n = 5), 5 (n = 5), or 7 (n = 5), consecutive days and performance was examined for an additional 7 days. Behavioral disruption was only observed in rats receiving AE for 7 days and the greatest degree of disruption occurred after AE injections were completed. Histopathological examination showed significant neural pathology in the nuclei trapezoideus, superior olive, and ruber of rats receiving 7- and 5-day AE regimens, and in the nucleus trapezoideus of rats receiving the 3-day regimen. Thus, behavioral disruption reflected, but did not predict, neuropathology. These results confirm and extend earlier results demonstrating neurotoxicity of AE in rats. Further, these results demonstrate that the auditory discrimination task provides an objective behavioral measure of AE neurotoxicity, and thus, can serve as a valuable tool for the safety development of AE and other artemisinin antimalarial compounds.

Visual system degeneration induced by blast overpressure.

The effect of blast overpressure on visual system pathology was studied in 14 male Sprague-Dawley rats weighing 360-432 g. Blast overpressure was simulated using a compressed-air driven shock tube, with the aim of studying a range of overpressures causing sublethal injury. Neither control (unexposed) rats nor rats exposed to 83 kiloPascals (kPa) overpressure showed evidence of visual system pathology. Neurological injury to brain visual pathways was observed in male rats surviving blast overpressure exposures of 104-110 kPa and 129-173 kPa. Optic nerve fiber degeneration was ipsilateral to the blast pressure wave. The optic chiasm contained small numbers of degenerated fibers. Optic tract fiber degeneration was present bilaterally, but was predominantly ipsilateral. Optic tract fiber degeneration was followed to nuclear groups at the level of the midbrain, midbrain-diencephalic junction, and the thalamus where degenerated fibers arborized among the neurons of: (i) the superior colliculus, (ii) pretectal region, and (iii) the lateral geniculate body. The superior colliculus contained fiber degeneration localized principally to two superficial layers (i) the stratum opticum (layer III) and (ii) stratum cinereum (layer II). The pretectal area contained degenerated fibers which were widespread in (i) the nucleus of the optic tract, (ii) olivary pretectal nucleus, (iii) anterior pretectal nucleus, and (iv) the posterior pretectal nucleus. Degenerated fibers in the lateral geniculate body were not universally distributed. They appeared to arborize among neurons of the dorsal and ventral nuclei: the ventral lateral geniculate nucleus (parvocellular and magnocellular parts); and the dorsal lateral geniculate nucleus. The axonopathy observed in the central visual pathways and nuclei of the rat brain are consistent with the presence of blast overpressure induced injury to the retina. The orbital cavities of the human skull contain frontally-directed eyeballs for binocular vision. Humans looking directly into an oncoming blast wave place both eyes at risk. With bilateral visual system injury, neurological deficits may include loss or impairments of ocular movements, and of the pupillary and accommodation reflexes, retinal hemorrhages, scotomas, and general blindness. These findings suggest that the retina should be investigated for the presence of traumatic or ischemic cellular injury, hemorrhages, scotomas, and retinal detachment.

Exposure to sublethal blast overpressure reduces the food intake and exercise performance of rats.

Exposure to blast overpressure can typically inflict generalized damage on major organ systems, especially gas-containing organs such as the lungs and the gastrointestinal tract. The purpose of the present study was to use rat's food intake and exercise wheel running as behavioral correlates of the perhaps more subtle damage to these organ systems induced by sublethal blast overpressure. Toward this end, all rats were exposed to a 12-h light/dark cycle and food was available only in the dark period. Prior to exposure, rats in the (E)xercise group were required to execute five rotations of an activity wheel for a food pellet; wheel turns that occurred at times other than when a rat was feeding were recorded separately and labeled exercise running. In the (S)edentary and (A)nesthesia groups, wheel running was not possible and rats were required to execute five leverpresses for a single pellet. A compressed air-driven shock tube was used to expose rats to a supra-atmospheric wave of air pressure. The tube was separated into two sections by a polyester membrane, the thickness of which determined peak and duration of overpressure. All rats were anesthetized with 50 mg/kg of phenobarbital. After reaching a deep plane of anesthesia, they were individually tied in a stockinet across one end of the shock tube. In preliminary tests, the membrane thickness was 1000 (A)ngstroms and rats in Group L(ethality) were exposed to a 129 kPa (peak amplitude) wave of overpressure. Three of six rats survived exposure to this peak pressure; pathology was evident in the lungs and gastrointestinal tract of all non-survivors. Rats in Groups E and S were tested with a 500 A membrane, which resulted in an 83 kPa peak amplitude. All rats survived exposure to this lower peak pressure. On the day of exposure to blast, the relative reduction of intake during the first 3 h of the dark period was significantly greater for Group E than for Groups S and A; the intake of Groups E and S remained reduced for four additional recovery days. Bodyweight was not significantly affected. Exercise wheel running also was reduced significantly on the day of exposure and during subsequent recovery days. These preliminary findings suggest that exposure to sublethal blast overpressure can reduce food consumption and exercise performance, perhaps as a consequence of damage to the gastrointestinal tract and lungs.

Arteether: risks of two-week administration in Macaca mulatta.

Male rhesus monkeys (Macaca mulatta) were administered daily doses of the antimalarial drug arteether. The 14-day treated group received either 24 mg/kg/day, 16 mg/kg/day, or 8 mg/kg/day. The seven-day treatment group received either 24 mg/kg/day or 8 mg/kg/day. All control cases in each group received the sesame oil vehicle alone. Neurologic signs were absent for animals in the seven and 14-day treatment groups except for one monkey which showed diffuse piloerection on day 14, and another monkey receiving 24 mg/kg/day for seven days showed mild lethargy after the fourth day. Mild, sporadic anorexia was noted in all animals by day 14, and a single animal showed diffuse piloerection on day 14. Surgical anesthesia preceded killing by exsanguination and was accompanied by perfusion fixation of the central nervous system. Brain sections were cut and then stained for study by light microscopy. Evidence of neuronal pathology, both descriptive and numerical, was collected. The neuroanatomic and neuropathologic findings demonstrated that arteether produced extensive brainstem injury when administered for 14 days. The magnitude of brainstem neurotoxicity was dose-dependent, where injury was greatest at the 24 mg/kg/day dose level, less at the 16 mg/kg/day dose level, and least at the 8 mg/kg/day dose level. Arteether induced multiple systems injury to brainstem nuclei of 1) the reticular formation (cranial and caudal pontine nuclei, and medullary gigantocellular and paragigantocellular nuclei); 2) the vestibular system (medial, descending, superior, and lateral nuclei); and 3) the auditory system (superior olivary nuclear complex and trapezoid nuclear complex). The vestibular nuclei and the reticular formation were most severely injured, with the auditory system affected less. The cranial nerve nuclei (somatic and splanchnic) appeared to escape damage, with the exception of the abducens nerve nucleus. The same brainstem nuclear groups of seven-day treated monkeys appeared normal. The statistical data are concordant with the descriptive data in demonstrating neurotoxic effects. In summary, no neurologic deficits were detected in any of the vehicle control monkeys (14-day and seven-day cases). Monkeys in the 14-day treatment group were free of clinical neurologic signs throughout the first week. At day 14, fine horizontal nystagmus was seen in one monkey, and another monkey exhibited diffuse piloerection. Monkeys in the seven-day treatment group were free of clinical neurologic signs except for one case. This monkey was treated with 24/mg/kg/day of arteether and exhibited lethargy after the fourth day. These indications of dysfunction arose too late to be practical indicators of neurotoxicity.

Phospholipase A2-induced neurotoxicity in vitro and in vivo in rats.

The present study evaluated the neurotoxic potential of phospholipase A2 (PLA2) in in vitro (primary neuronal cultures) and in vivo (EEG and behavior) rat models of CNS excitability. In vitro, PLA2 (0.0038-5.8 nM) or melittin (a potent activator of endogenous PLA2; 100-5000 nM), were highly neurotoxic, causing approximately 500 units/ml LDH release. The neurotoxic EC50s for PLA2 and melittin were 1.8 (1.4-2.3) and 848 (501-1280) nM, respectively. Neurotoxic concentrations of PLA2 stimulated neuronal release of [3H]AA. Preliminary in vitro experiments evaluating changes in neuronal calcium flux indicated that PLA2 caused transient, and melittin sustained, increases in [Ca2+]i. In vivo, PLA2 (0.5-5 micrograms i.c.v.) or melittin (2.5-20 micrograms i.c.v.) produced nonconvulsive EEG seizures, which generalized to status epilepticus. While the onset of seizure development was markedly delayed for PLA2 (1.5-4.5 h), the seizure inducing effects of melittin were evident within 3.5 +/- 0.2 min and more severe. Both PLA2 and melittin were lethal, exhibiting LD50s of 0.62 micrograms and 8.4 micrograms, respectively. Pretreatment with (+)-MK801 (5 micrograms, i.c.v.) significantly attenuated melittin, but not PLA2, in vivo neurotoxicity. PLA2 induced neuropathology in surviving rats revealed extensive cortical and subcortical injury to forebrain neurons and fibre pathways. Collectively, these results demonstrate the potent neurotoxic potential of PLA2, the delayed clinical nature of its in vivo neurotoxicity and the applicability of these model systems to future studies on mechanisms of PLA2 neurotoxicity and the development of potential PLA2 antagonists.

Fatal neurotoxicity of arteether and artemether.

Artemisinin (qinghaosu) and several derivatives have been developed and are in use as antimalarial drugs but scant information is available regarding animal or human toxicity. Following a eight-day, multiple-dose, pharmacokinetic study of arteether (AE) (10 mg/kg/day [n = 6] and 20 mg/kg/day [n = 6]) in dogs, all high-dose animals displayed a progressive syndrome of clinical neurologic defects with progressive cardiorespiratory collapse and death in five of six animals. Neurologic findings included gait disturbances, loss of spinal and pain response reflexes, and prominent loss of brain stem and eye reflexes. Animals had prolongation of QT interval corrected for rate (QTc) on electrocardiograms (ECGs) with bizarre ST-T segment changes. Prominent neuropathic lesions were noted to be primarily limited to the pons and medulla. Similar lesions with dose-related severity were noted in eight other dogs studied in a second study with intramuscular (IM) administration of AE in sesame oil during a 28-day, dose-ranging study using 5, 10, 15, and 20 mg/kg/day. Injury, graded by a pathologist blinded to the dose group, showed a dose-related, region-specific injury in all animals that was most pronounced in the pons. Further studies in Sprague-Dawley rats using IM administration of AE and artemether (AM) at a dose of 12.5-50 mg/kg/day for 28 days confirmed the onset of a clinical neurologic syndrome with dose-related changes in body weight, activity, and seizure-like activity, stereotypic movement disorders, and ECG changes.(ABSTRACT TRUNCATED AT 250 WORDS)

Neurotoxicity in animals due to arteether and artemether.

Several artemisinin (qinghaosu) derivatives have been developed and are in use as antimalarial drugs but scant animal or human toxicity data are available. We noted a progressive syndrome of clinical neurological defects with cardio-respiratory collapse and death in 5/6 dogs dosed daily for 8 d with intramuscular arteether (AE) at 20 mg/kg/d in a pharmacokinetic study. Neurological findings included gait disturbances, loss of spinal reflexes, pain response reflexes and prominent loss of brain-stem and eye reflexes. Electrocardiography showed prolongation of the QT interval corrected for rate (QTc). Prominent neuropathic lesions were sharply limited to the pons and medulla. Neurological injury, graded by a pathologist 'blinded' to dose group, showed a dose-related region-specific injury which was most pronounced in the pons and medulla in all animals. Rats treated with AE and artemether (AM) at 12.5 to 50 mg/kg/d for 28 d confirmed clinical neurological abnormalities with high doses (> 25 mg/kg/d) after 6-14 d. Neuropathological examination of rat brain sections at 5 levels from the rostral cerebrum to the caudal medulla showed a dose-related pattern of injury characterized by hyalinized neuron cell bodies and loss of Nissl substance; changes congruent with those noted in dogs. No significant difference was noted in the extent, type, or distribution of lesions in the brains of rats treated with equivalent doses of AE or AM.(ABSTRACT TRUNCATED AT 250 WORDS)

Neurology and neuropathology of Soman-induced brain injury: an overview.

Battlefield use of nerve agents poses serious medical threats to combat troops and to civilians in the immediate or adjacent environment. The experiments reported herein were carried out in the 1980s to help to define both the neurological and neuropathological consequences of exposure to the organophosphate nerve agent Soman. These data contributed to the scientific foundation for a program of drug development to find agents that would prevent or reduce the risk of injury to the central nervous system and specifically pointed to the importance of including an anticonvulsant in the treatment of agent exposure. Since these experiments were conducted, research efforts have continued to improve pretreatment and treatment, such as the inclusion of the anticonvulsant diazepam in the medical treatment of exposed personnel.

Evaluation of transient forebrain ischemia induced by four vessel occlusion using schedule-controlled behavior.

Transient (5-min) global forebrain ischemia, induced by four- vessel occlusion, was assessed using a multiple fixed-ratio, fixed-interval schedule of food presentation in five rats. Under control conditions, the schedule produced distinctive response rates in each schedule component. Initially, ischemia disrupted responding under both schedule components, and to approximately the same degree. In general, total session responses returned to pre-occlusion levels during the course of 45 post-occlusion days, however, response rates under the fixed-interval component showed slightly less recovery than those under the fixed-ratio component. Histological assessment revealed considerable variability in hippocampal damage between rats. Severe damage in the CA1, CA2, and CA3 formations was observed in a single rat, and that rat also showed the greatest degree of response disruption. These results suggest that schedule-controlled responding may be a valuable method for assessing the effects of ischemic injury, and thus, putative neuroprotective compounds, on complex behavior.

Sensory innervation of the canine esophagus, stomach, and duodenum.

The sensory innervation of the postpharyngeal foregut was investigated by injecting the enzyme horseradish peroxidase (HRP) into the walls of the esophagus, stomach, or duodenum. The transported HRP was identified histochemically, labeled neurons in the spinal and vagal ganglia were counted, and the results were plotted using an SAS statistical program. The spinal sensory fields of each viscus were defined using three determinations: craniocaudal extent, principal innervation field, and peak innervation field. The data revealed that innervation fields are craniocaudally extensive, the sensory field of each viscus overlaps significantly with its neighbor, yet each viscus can be characterized by a field of peak innervation density. Craniocaudal innervation of the esophagus spans as many as 22-23 paired spinal ganglia (C1-L2). There are two peak innervation fields for the cervical (C2-C6 and T2-T4) and for the thoracic (T2-T4 and T8-T12) sectors of the esophagus. The sensory innervation of the stomach extends craniocaudally over as many as 25 paired spinal ganglia (C2-L5). The peak innervation field of the stomach spans a large area comprising the cranial, middle, and the immediately adjoining caudal thoracic ganglia (T2-T10). The duodenum is innervated craniocaudally by as many as 15 paired thoracolumbar ganglia (T2-L3). Peak innervation originates in the middle and caudal thoracic ganglia and cranial lumbar (T6-L1) ganglia. There is a recognizable viscerotopic organization in the sensory innervation of the postpharyngeal foregut; successively more caudal sectors of this region of the alimentary canal are supplied with sensory fibers from successively more caudal spinal dorsal root ganglia. Vagal afferent innervation of the esophagus, stomach, and duodenum is bilateral and originates predominantly, but not exclusively, from vast numbers of neurons in the nodose (distal) ganglia. The esophagus is innervated bilaterally and more abundantly by jugular (proximal) ganglia neurons than is either the stomach or duodenum. The physiological significance of the findings are discussed in relation to the phenomena of visceral pain and referred pain.

Neurological dysfunction after intrathecal injection of dynorphin A (1-13) in the rat. II. Nonopioid mechanisms mediate loss of motor, sensory and autonomic function.

The kappa opioid agonist dynorphin A (Dyn A) (1-13) produced dose-related neurological deficits after subarachnoid injection in the lumbar spinal cords of rats. Whereas the neurological dysfunctions produced by low doses of Dyn A (1-13) were transient, higher doses caused persistent deficits, characterized by motor and nociceptive impairment in hindlimbs and tail, hindlimb edema, priapism, bladder infarction and atony and urinary incontinence. These deficits appeared to result from nonopioid actions of Dyn A (1-13), as they were: 1) not blocked by the opioid antagonists naloxone or WIN 44,441-3; 2) shared by Dyn A (3-13), which lacks opioid activity; and 3) not produced or altered by the selective kappa opioid agonist U 50,488. Coinjection of a combination of peptidase inhibitors, shown previously to enhance the actions of Dyn A fragments in vitro, significantly increased the paralytic actions of Dyn A (1-13). The peptidase inhibitors did not by themselves cause neurological dysfunction, and they did not alter the paralytic potency of the peptidase-resistant delta opioid antagonist ICI 174864. These findings indicate that Dyn A effects were: 1) limited appreciably by its rapid enzymatic degradation after injection and 2) most likely the result of actions of the intact peptide rather than proteolytic products generated after injection. Neuroanatomical evaluations revealed extensive neuronal and axonal injury in the lumbosacral spinal cords of rats injected with 25 nmol of Dyn A (1-13). Collectively, these results indicate that Dyn A (1-13) acts through nonopioid mechanisms to cause the injury and death of neurons involved in diverse spinal cord functions.

Neurologic deficits and neuronal injury in rats resulting from nonopioid actions of the delta opioid receptor antagonist ICI 174864.

The delta opioid receptor antagonist ICI 174864 produces postural abnormalities and barrel rolling after i.c.v. injection and hindlimb and tail flaccidity after spinal subarachnoid injection in rats. These effects appear to result from nonopioid characteristics of ICI 174864 because they are neither shared nor blocked by other opioid antagonists (naloxone, ICI 154129 and WIN 44,441-3) and are produced by two compounds (ICI 174644 and ICI 178173) that are structurally related to ICI 174864 but lack its delta antagonist properties. Barrel rolling and hindlimb paralysis are also produced by dynorphin A-related peptides; however, rats failed to demonstrate tolerance or cross-tolerance to the hindlimb paralytic actions of ICI 174864 or dynorphin A (1-13) after 7 days of continuous spinal intrathecal infusion of either of these compounds. Whereas hindlimb responses to low doses of ICI 174864 (1.6-6.2 nmol intrathecally) were usually transient, higher doses (6.2-25 nmol intrathecally) produced persistent hindlimb motor dysfunction, altered nociception, priapism, hindlimb edema, bladder infarction and atony and urinary incontinence. Neuronal and axonal changes in the lumbosacral spinal cords of rats with persistent and transient neurologic deficits provided direct evidence of the neuropathologic actions of ICI 174864 (3.1 and 6.2 nmol) and ICI 174644 (25 nmol). These results indicate that 1) use of ICI 174864 as a selective delta opioid receptor antagonist is potentially compromised by its nonopioid neuropathologic actions and 2) ICI 174864 and dynorphin A-related peptides are unique among opioid agonists and antagonists in sharing barrel rolling and hindlimb paralytic effects. A similar mechanism of action may underlie the shared nonopioid actions of these peptides.

Endogenous opioids in spinal cord injury: a critical evaluation.

Based upon evidence that opioid antagonists improve neurological outcome following either traumatic or ischemic spinal cord injury, endogenous opioids have been implicated in the pathophysiology of these disorders. Naloxone improved both spinal cord perfusion and neurological function following traumatic spinal cord injury in cats, and was subsequently observed to improve neurological outcome following ischemic spinal cord injury in rabbits. Using several opioid antagonists with varied selectivities for different types of opioid receptors, it was suggested that kappa opioid receptors are involved in both these models of spinal cord injury. In addition, spinal cord trauma in rats is associated with increased concentrations of the endogenous kappa agonist dynorphin A, and increased kappa opioid receptor binding capacity localized to the injury site. Furthermore, dynorphin A induces hindlimb and tail flaccidity following intrathecal injection in rats. Thus, the pathophysiological effects of endogenous opioids in spinal cord injury have been proposed to involve dynorphin A interactions with kappa opioid receptors. However, disparities between the actions of intrathecally injected dynorphin A in rats and the presumed actions of endogenous dynorphin A in cat and rabbit spinal cord injury have been revealed in recent experiments. Paralysis resulting from intrathecal dynorphin A is not altered by opioid receptor antagonists or TRH, produced by non-opioid dynorphin A fragments but not by other selective kappa opioid agonists, and associated with non-opioid mediated reductions in spinal cord blood flow. Furthermore, despite reports of endogenous opioid changes following rat spinal cord trauma, in contrast to cats and rabbits, naloxone failed to improve neurological outcome following traumatic rat spinal cord injury. Thus, the specific endogenous opioids and opioid receptor types involved in spinal cord injury remain to be resolved, and do not appear to be universal among different models of spinal cord injury in different species. Additionally, dynorphin A may participate in spinal cord injury mechanisms in the rat through non-opioid actions.