Dual neuronal response to tumor necrosis factor-alpha following spinal cord injury

BACKGROUND： Numerous studies have shown that tumor necrosis factor α （TNF-α） is closely correlated with spinal cord injury （SCI）, but the mechanisms of TNF-α and therapeutic treatments for SCI are still poorly understood. OBJECTIVE： To determine the role of TNF-α in the pathogenesis of SCI. DESIGN, TIME AND SETTING： An in vivo experiment based on genetically engineered animals was performed at the Medical University of South Carolina, Charleston, South Carolina, USA, between June 2007 and October 2008. MATERIALS： TNF-α transgenic rats （Xenogen Biosciences in Cranbury, New Jersey, USA） were utilized in this study. METHODS： TNF-α transgenic （tg） and wild-type （WT） rats underwent a complete single-level laminectomy at the 10^th thoracic vertebra （T10）. MAIN OUTCOME MEASURES： Motor function of rat hindlimb was assessed using the Basso, Beattie, and Bresnahan hindlimb locomotor rating scale. Histological evaluation of spinal cord tissue loss was conducted. Immunohistochemistry for astrocytes, microglia/macrophages, and TNF receptors （TNFRs） was performed on spinal cord tissue sections. TNF-α mRNA expression was detected by real-time polymerase chain reaction. The concentrations of nerve growth factor （NGF） and brain-derived neurotrophic factor （BDNF） in the supernatant were determined using an enzyme-linked immunosorbent assay kit for rat NGF or BDNF, respectively. The rats were injected subcutaneously with etanercept to verify that TNF-α was the direct effect of the modulation of behavioral and neurodegenerative outcomes in the TNF-α tg rats. RESULTS： TNF-α tg rats showed higher expression of TNF-α mRNA in the spinal cord prior to SCI. TNF-α tg rats showed worse motor deficits than WT rats in the acute period （〈 3 days） after SCI （P 〈 0.01）, while in the chronic period, TNF-α tg rats exhibited persistent elevated baseline levels of TNF-α mRNA and improved recovery in motor function and tissue healing compared to WT rats （P 〈 0.01 ）. Following SCI, the number of microglia/macrophages in TNF-α tg rat was always greater than in WT rat （P 〈 0.01）. There were no significant differences in NGF and BDNF levels in the supernatant of spinal cord homogenates. TNFR1 expression was significantly greater in the TNF-α tg rats compared to the WT rats （P 〈 0.01）. However, TNFR2 expression did not reveal a significant increase in the TNF-α tg rats compared to the WT rats. Finally, treatment with etanercept reduced injury acutely, but exacerbated the injury chronically. CONCLUSION： Overexpression of TNF-α is deleterious in the acute phase, but beneficial in the chronic phase in the response to SCI. The role of TNF-α post-injury may depend on TNF-α expression in the spinal cord and its differential binding to TNFRI. Our observations may have clinical relevance that antagonists or inhibitors of TNF-α could be administered within the early time window post-injury, and appropriate amounts of TNF-α could be administered during the chronic stage, in order to improve the final neurological recovery in patients with SCI.