Abstract The first part of this research was focused on the relationship between injury severity and cell death mechanisms after spinal cord ischemia-reperfusion injury. The major blood supply to the thoracolumbar spinal cord comes from the segmental arteries originating from the thoracoabdominal aorta. Paraplegia cause by spinal cord ischemia is a devastating complication of thoracoabdominal aortic surgery. Previous studies indicated that ischemia-reperfusion injury of the central nervous system causes two distinct types of cell death, necrosis and apoptosis. It was also implicated that the intensity of injury can somehow affect the cell death mechanisms. In the first series of our experiments, by occluding the descending thoracic aorta with or without simultaneously inducing hypovolemic hypotension in rats, we established a model of experimental spinal cord ischemia-reperfusion (SCIR) in which the injury severity can be controlled. Recordings of carotid blood pressure (CBP) and spinal cord blood flow (SCBF) showed that aortic occlusion induced dramatic CBP elevation but SCBF drop in both the normotensive (NT) and hypotensive (HT) groups. However, the HT group demonstrated significantly lower SCBF during aortic occlusion, and much slower elevation of SCBF after reperfusion, and extremely poor neurological performance. Spinal cord lesions were characterized by infarction associated with extensive necrotic cell death, but little apoptosis and caspase-3 activity. In contrast, in the NT group, SCIR resulted in minor tissue destruction associated with persistently abundant apoptosis, augmented caspase-3 activity, and favorable functional outcome. The relative sparing of motoneurons in the ventral horns from apoptosis might have accounted for the minor functional impairment in the NT group. The severity of ischemia-reperfusion (I/R) injury was found to have substantial impact on the histopathological changes and cell death mechanisms, which correlated with neurological performance. These findings implicate that injury severity and duration after injury are two critical factors to be considered in therapeutic intervention. Based on the knowledge that bPrevious studies have implicated both excitotoxicity and apoptosis are involved in the pathogenesis of SCIR injury, we proposedtested the possibility that the N-methyl-D-aspartate (NMDA) receptor antagonist (dizocilpine maleate: (MK801) and the protein synthesis inhibitor (cycloheximide) would produce a synergic effect in the treatment of SCIR injury. In the second series of experiments, I/R iSpinal cord ischemia-reperfusion injury was induced by a thoracic aortic occlusion and blood volume reduction, followed by reperfusion and volume restoration. ischemia-reperfusion Rats were treated with vehicle, MK801, cycloheximide, or combination of MK801 and cycloheximide in combination. The MK801 and combined therapy group got a better recovery of hHind limb motor function recovery was better in the MK801 and combined-therapy groups than in the control and cycloheximide groups. On the 7th day after ischemia-reperfusion injury, all three treated groups showed significantly higher neuronal survival rates (NSR) than that of the control group. Among the three treated groups, the combined-treatment group showed the highest NSR. In addition, the Ttherapeutic effect of the combined-treatment group (27.4% increase of NSR) iwas better than the anticipated by the addition of MK801 and cycloheximide based on NSR data group. The number of apoptotic cells of was significantly reduced in the cycloheximide group and the combined-treatment group, as compared to that of the control group. It was unchanged in the MK-801 group. These results suggest that combined treatments directed at blocking both NMDA receptor-mediated excitotoxic necrosis and caspase-mediated apoptosis might have synergic therapeutic potential in reducing SCIR injury. Mitogen-activated protein kinases (MAPKs) including c-Jun N-terminal kinases (JNK), p38, and extracellular signal-regulated kinases (ERK), play important roles in the transduction of stressful signals and the integration of cellular responses. Although it has been generally held that the JNK and p38 pathways are related to cell death and degeneration, while the ERK pathway, cell proliferation and survival, controversy still exists. The roles of the ERK pathway in I/R injury of the CNS, in particular, remain to be clarified, because contradictory data have been reported by different investigators. Given this controversy, in the third series of experiments, we examined in injured spinal cords the temporal and spatial profiles of ERK1/2 activation following SCIR, and the effects of inhibiting the kinase that phosphorylates ERK1/2, MEK. The results showed that I/R injury induced an immediate phosphorylation of ERK1/2 in the spinal cord, which was alleviated by a MEK inhibitor, U0126. The control group was characterized by poorer neurological outcome, more severe tissue destruction, pronounced apoptosis, and lower neuronal survival. In contrast, the U0126-treated group demonstrated more apparent improvement of hind limb motor function, less tissue destruction, lack of apoptosis, and higher neuronal survival. In addition, administration of U0126 also significantly increased the activation of nuclear factor-£eB (NF-£eB) and the expression of cellular inhibitor of apoptosis protein 2 (c-IAP2). These findings implicate that the mechanisms underlying the neuroprotection afforded by ERK1/2 inhibition may be through the NF-£eB-c-IAP2 axis. The activation of the MEK-ERK signaling pathway appeared to be harmful in SCIR injury. Strategies aimed at blocking this pathway may bear potential therapeutic benefits in the treatment of SCIR injury. The second part of the research was focused on the pathophysiology of surgery-associated paraspinal muscle injury and measures to protect surgically violated paraspinal muscles. The wide dissection and forceful retraction of paraspinal muscles which are often required for posterior spinal sugery may severely jeopardize the muscles structurally and functionally. Immediate posteoperative pathological changes in the surgically violated paraspinal muscles may cause severe pain and a delay of patient ambulation. Long-term sequelae of surgical injury of paraspinal muscles include chronic back pain and impaired back muscle strength. Ironically, being a common complication of posterior spinal surgery, paraspinal muscle injury is so often neglected. Limited previous data indicate that the underlying pathophysiology of muscle damage involve both mechanical and ischemic mechanisms. We hypothesized that surgical dissection and retraction may produce oxidative stress within the paraspinal muscles. Meanwhile, we also proposed that the oxidative stress may trigger certain protective mechanisms within the insulted muscles. The first part of our study was a human study conducted to assess the significance of oxidative stress, and the relationship between it and the stress response mediated by heat shock protein 70 (HSP70) induction within paraspinal muscles under intraoperative retraction. A group of patients with lumbar spondylolisthesis treated with posterolateral lumbar spinal fusion, pedicle fixation and laminectomy were enrolled. Multifidus muscle specimens were harvested intraoperatively before, at designated time points during, and after surgical retraction. Muscle samples were analyzed for HSP70 and malondialdehyde (MDA) levels. Both HSP70 expression and MDA production within multifidus muscle cells were increased significantly by retraction. HSP70 expression then dropped after a peak at 1.5 hr of retraction, whereas MDA levels remained elevated even after release of retractors for reperfusion of the muscles. Histopathological and immunohistochemical evidence indicated that the decline of HSP70 synthesis within muscle cells after prolonged retraction was the result of severe muscle damage. These results highlighted the noxious impact of intraoperative retraction on human paraspinal muscles, and the significance of oxidative stress at the cellular and molecular levels. It is also implicated that intraoperative maneuvers aimed at reducing the oxidative stress within the paraspinal muscles may help attenuating surgery-associated paraspinal muscle damage. Given the findings of the first part of our study, and the knowledge that inflammation is a major postoperative pathological finding in surgically injured paraspinal muscles, we proceeded to examine the roles of two important inflammatory mediators, cyclooxygenase (COX)-2 and nuclear factor (NF)-£eB, in the pathogenesis of retraction-associated paraspinal muscle injury. A rat model of paraspinal muscle dissection and retraction that mimicks the conditions in human posterior spinal surgery was established. In the control group, paraspinal muscles were dissected from the spine through a dorsal incision, and then laterally retracted. Paraspinal muscle specimens were harvested before, and at designated time points during and after persistent retraction. The time course of NF-£eB activation as well as the expression of COX-2 were examined. Severity of inflammation was evaluated based on histopathology and myeloperoxidase (MPO) activity. NF-£eB activation was inhibited by the administration of pyrrolidine dithiolcarbamate (PDTC) in the PDTC-treated group. In the control group, retraction induced an early increase of NF-£eB/DNA binding activity in paraspinal muscle cells, which persited throughout the whole course of retraction. COX-2 expression was not detectable until 1 day after surgery, and reached a peak at 3 days. The time course of COX-2 expression correlated with that of inflammatory pathology and MPO activity. Extensive muscle fiber loss and collagen fiber replacement were observed at 7 days after surgery. Pretreatment with PDTC inhibited intraoperative NF-£eB activation and greatly downregulated postoperative COX-2 expression and inflammation in the muscles. Fibrosis following inflammation was also significantly abolished by PDTC administration. These findings indicate that NF-£eB-regulated COX-2 expression and inflammation play an important role in the pathogenesis of surgery-associated paraspinal muscle injury. Therapeutic strategies involving NF-£eB inhibition may be applicable to the prevention of such injury.