Abstract: BACKGROUND: Treadmill training is one of the effective ways to promote the recovery of motor function after spinal cord injury. Treadmill training can promote neurogenesis, but the effect of different intensities of treadmill training on the activation of endogenous stem cells is still unclear.
OBJECTIVE: To analyze the activation effect of different intensities of treadmill training on endogenous neural stem cells in the spinal cord of mice after spinal cord injury.
METHODS: Fifty female C57BL/6J mice were divided into control group, spinal cord injury group, low-, moderate-, and high-intensity exercise groups with 10 mice in each group by random number table method. T10 segment spinal cord injury model was constructed by the clamp method in spinal cord injury group, low-, moderate-, and high-intensity exercise groups. On day 7 after spinal cord injury, mice in the low-, moderate-, and high-intensity exercise groups were respectively trained on the treadmill with corresponding intensity, 3 times /d, 10 min/ times, 6 times a week for 28 consecutive days. At 3, 7, 14, 21, and 28 days after treadmill training, the hind limb motor function was evaluated by BMS score. At 28 days after treadmill training, the spinal cord tissue of the injured area was obtained, and the expression of epidermal growth factor receptor, glial fibrillary acidic protein, and 5-Ethynyl-2'-deoxyuridine (EdU), a proliferative marker, was detected. Hematoxylin-eosin staining was used to observe the morphology of spinal cord. 
RESULTS AND CONCLUSION: (1) The BMS score of mice in the spinal cord injury group was lower than that in the control group (P < 0.05). With the extension of treadmill training time, the BMS scores of mice with spinal cord injury gradually increased, and the BMS scores of mice in moderate-intensity exercise group on days 14 and 21 after treadmill training were higher than those in spinal cord injury group and low- and high-intensity exercise groups (P < 0.05). The BMS score of mice in moderate- and high-intensity exercise group was higher than that in spinal cord injury group and low-intensity exercise group at 28 days after treadmill training (P < 0.05). (2) Compared with the control group, the proportion of epidermal growth factor receptor and EdU positive cells was increased in spinal cord injury group (P < 0.05). Compared with spinal cord injury group, the proportion of epidermal growth factor receptor and EdU positive cells was increased in low-, moderate-, and high-intensity exercise groups (P < 0.05), and the highest was found in moderate-intensity exercise group. Compared with control group, the proportion of glial fibrillary acidic protein positive cells was increased in spinal cord injury group (P < 0.05). Compared with spinal cord injury group, the proportion of glial fibrillary acidic protein positive cells was lower in low-, moderate-, and high-intensity exercise groups (P < 0.05), and the moderate-intensity exercise group was the lowest. (3) Hematoxylin-eosin staining showed that a large cavity was formed in the injured area of mice with spinal cord injury, and the cavity in the injured area of mice with spinal cord injury decreased after different intensities of treadmill training, and the decrease was most obvious in the moderate-intensity exercise group. (4) These results indicate that low-, moderate-, and high-intensity treadmill training can promote the recovery of motor function of mice with spinal cord injury by activating endogenous neural stem cells, and the effect of moderate-intensity exercise training is the most obvious.

Key words: 脊髓损伤, 跑台训练, 脊髓神经干细胞, 脊髓内源性神经干细胞, 运动训练