The Dual Role Of Tumor Necrosis Factor-alpha In The Pathophysiology Of Spinal Cord Injury

PartⅠEstablishment of spinal cord injury model in rats and evaluation of limb locomotor function and histopathology of ratsafter spinal cord injuryObjective:To establish the model of spinal cord injury in rats, and to evaluate the limb locomotor function and histopathology in this model.Methods:1. Adult, male, TNF-αtg rats and wt littermates (7-8 weeks old and weight 300±20 g) were anesthetized by intraperitoneal injection of ketamine (90 mg/kg) / xylazine (10 mg/kg). All rats underwent a complete single level laminectomy at the 10th thoracic vertebra. Establishing animal model of SCI with a pneumatic impactor device: a computer-controlled impact by a rod was administered at 150 kilodyns onto the T10 segment with the dura intact.2. Rats were tested for motor function on days 1, 3, 7, 14, 21, 28, 35, 42, 49 and 56 post-SCI. Recovery from motor disturbance was graded using the 21-point murine Basso, Beattie, and Bresnahan (BBB) hind limb locomotor rating scale.3. At 6, 24 and 72 hours post-SCI, TNF-αtg and wt rats (n=5 per time point, respectively) were anesthetized with sodium pentobarbital (100 mg/kg) by I.P. injection and sacrificed. Each 1-cm spinal cord encompassing the injured epicenter was cut sagittally in consecutive 20μm sections on the freezing microtome. Tissue sections were selected from the epicenter of the impacted spinal cord, and the terminal deoxynucleotidyl transferase-mediated dUTP-digoxygenin nick-end labeling (TUNEL) technique was used to detect apoptotic cells in the spinal cord tissue.4. On days 7, 14 and 56 after SCI, TNF-αtg and wt rats (n=10 per time point, respectively) were anesthetized and sacrificed. The preparation and selection of tissue cryosections were same as described above. The tissue sections were stained with luxol fast blue and luxol fast red. Each section was photographed by using Nikon microscope, and the area of tissue loss was measured by using NIH image J 1.37v analysis software.Results:1. In the acute phase post-SCI (≤day 3), TNF-αtg rats showed worse hindlimb motor function than wt rats (p < 0.01). However, on day 7 post-SCI, TNF-αtg rats had better recovery of motor function compared with wt rats (p < 0.01). TNF-αtg rats continued to recover from dysfunction with time and, by 56 days post-injury, some of them reached perfect scores (21) on the BBB rating scale. The other TNF-αtg rats also had significant improvement in their motor function. In contrast, although some improvement was observed in wt rats by 56 days post-injury, there were still obvious deficits in their hindlimb motor function (p < 0.01).2. At 6 h post-SCI, apoptotic cells began to appear in the border of the lesion in TNF-αtg and wt rats, and apoptosis reached the maximal level at 24 h post-injury. At 72 h post-SCI, apoptosis in the spinal cord tissue of both groups began to decrease. Moreover, during this process, the level of apoptosis was always higher in TNF-αtg rats than in wt rats (p < 0.01).3. In the chronic phase post-SCI (≥day 7), the areas of tissue loss in TNF-αtg rats became smaller with time (p < 0.01), while wt rats had no significant change in the areas of tissue loss. In addition, the areas of tissue loss in wt rats were always higher than in TNF-αtg rats at each time point (on day 7, p < 0.05; on day 14 and 56, p < 0.01).PartⅡThe expression of TNF-αmRNA and protein in the injuredspinal cord after spinal cord injuryObjective:To detect the expression of TNF-αmRNA and protein in the injured spinal cord after spinal cord injury. Methods: 1. Establishing the model of SCI in rats (same as described above).2. At 1, 4, 8, 24, 72 hours, and on days 7 and 14 after SCI, TNF-αtg and wt rats (n=5 per time point, respectively) were anesthetized and sacrificed. A 1-cm segment of spinal cord encompassing the lesion epicenter was dissected out from each rat. The expression of TNF-αmRNA in the injured spinal cord tissue was detected by real-time polymerase chain reaction (rtPCR).3. At 1, 4, 8, 24, 72 hours, and on days 7 and 14 after SCI, TNF-αtg and wt rats (n=5 per time point, respectively) were anesthetized and sacrificed. A 1-cm segment of spinal cord encompassing the lesion epicenter was dissected out from each rat and separately placed in 500ul tissue protein extraction reagent to homogenize. The tissue homogenate was centrifuged at 12000 g for 20 min at 4℃, and the concentration of TNF-αin the supernatant was determined using an enzyme linked immunosorbent assay (ELISA) kit for rat TNF-α.Results:1. The expression of TNF-αmRNA increased rapidly in the injured spinal cord of TNF-αtg and wt rats after SCI, and reached the maximal level at 1 h post-injury, and then gradually decreased over time. At 72 h post-injury, the TNF-αmRNA in the injured spinal cord of both groups declined to their levels before injury. Moreover, the level of TNF-αmRNA in the injured spinal cord was much higher in TNF-αtg rats than in wt rats at each time point (p < 0.01).2. The concentration of TNF-αrapidly increased in the injured spinal cord of TNF-αtg and wt rats after SCI, reaching the peak level at 1 h post-injury, and then gradually decreased with time. By 7 days post-injury, the level of TNF-αin the injured spinal cord of both groups returned to their levels before injury (TNF-αtg rats maintained lower and baseline level of TNF-α, and TNF-αin wt rats was undetected when measured by ELISA kit). Furthermore, in this process, the concentration of TNF-αin the injured spinal cord was always higher in TNF-αtg rats than in wt rats (p PartⅢThe changes of neurocytes and the expression of BDNF in theinjured spinal cord after spinal cord injuryObjective:To detect the changes of neurocytes and the expression of BDNF in the injured spinal cord after spinal cord injury.Methods:1. Establishing the model of SCI in rats (same as described above).2. On days 7, 14 and 56 after SCI, TNF-αtg and wt rats (n=5 per time point, respectively) were anesthetized and sacrificed. The preparation and selection of tissue cytosections was same as mentioned above. Immunohstochemistry for neurons, astrocytes and microglia was performed on spinal cord tissue sections with mouse anti-NeuN, mouse anti-GFAP and rabbit anti-Iba1 primary antibodies, respectively.3. On days 1, 7, 14 and 56 after SCI, TNF-αtg and wt rats (n=5 per time point, respectively) were anesthetized and sacrificed. The preparation of spinal cord samples and the extraction of total protein from spinal cord tissue were same as assay of TNF-αprotein in the injured spinal cord. The concentration of BDNF in the supernatant was detected with BDNF E_max^? ImmunoAssay System.Results:1. After SCI, the number of neurons in the border of the lesion in SD rats significantly decreased with time. Compared with SD rats, although the number of neurons in TNF-αtg rats also decreased, the level of decrease was lower, and TNF-αtg rats still remained more neurons in the border of the lesion than SD rats in the chronic phase post-SCI (on day 56, p < 0.01).2. The expression of Iba1 and GFAP were significantly increased in the injured spinal cord of TNF-αtg and wt rats after SCI. Ibal and GFAP reached the peak level on days 7 and 14, respectively, and then gradually decreased with time. On day 56, the expression of Ibal and GFAP in the injured spinal cord of wt rats nearly returned to the levels before injury, while TNF-αtg rats still maintained higher levels of Ibal and GFAP in the injured spinal cord than their uninjured controls. Furthermore, in the chronic phase post-SCI, the levels of Iba1 and GFAP in the injured spinal cord were always higher in TNF-αtg rats than in wt rats (p < 0.05).3. On day 7 post-SCI, the concentration of BDNF in the injured spinal cord significantly increased in both TNF-αtg and SD rats (p < 0.01), reached their maximal levels on day 14 after SCI (p < 0.01), and then declined with time. On day 56 post-SCI, the concentration of BDNF in SD rats returned to their level before injury, while TNF-αtg rats still maintained a high level. Moreover, the level of BDNF in the injured spinal cord was always higher in TNF-αtg rats than in SD rats at each point post-SCI (p < 0.05).PartⅣThe expression of TNFR1 and TNFR2 in the injured spinalcord after spinal cord injuryObjective:To detect the expression of TNFR1 and TNFR2 in the injured spinal cord after spinal cord injury.Methods:1. Establishing the model of SCI in rats (same as described above).2. At 6 hour, and on days 7, 14 and 56 after SCI, TNF-αtg and SD rats (n=5 per time point, respectively) were anesthetized and sacrificed. The preparation and selection of tissue cytosections were same as in situ apoptosis assay. Immunohistochemistry for TNFR1 and TNFR2 was performed on spinal cord tissue sections with rabbit anti-TNFR1 and goat anti-TNFR2 primary antibodies, respectively.Results:1. The protein expression and immunoreactivity of TNFR1 and TNFR2 significantly increased in the injured spinal cord tissue in both SD and TNF-αrats post-SCI, and reached their peak levels on day 7 after SCI. Then the protein expression and immunoreactivity of TNFR1 decreased in the injured spinal cord in both groups with time, and hardly returned to their levels before injury on day 56 post-SCI. Although the protein expression and immunoreactivity of TNFR2 also decreased in the injured spinal cord in both groups, their levels of decrease were lower than TNFR1, so TNFR2 still retained high expression and immunoreactivity in the injured spinal cord in both groups on day 56, and there were no differences compared with the levels on day 7 post-SCI (p > 0.05).Conclusions1. SCI can induce the expression of TNF-αmRNA and protein in the injured spinal cord.2. In the acute phase of SCI, TNF-αcan induce apoptosis in the injured spinal cord; In the chronic phase of SCI, TNF-αcan activate and recruit astrocytes and microglia to mediate neuroprotective effects.3. SCI can promote the expression of TNFR1 and TNFR2 in the injured spinal cord, and there are significant differences in the expression between TNFR1 and TNFR2 in the chronic phase of SCI.4. TNF-αplays a dual role in the pathophysiology of SCI, i.e. it shows neurodestructive effects in the acute phase (≤day 3), and exhibits neuroprotective effects in the chronic phase (≥day 7); the mechanism may relate to the expression of TNFR1 and TNFR2, as well as the concentration of TNF-αin the injured spinal cord.