PLGA复合有序多孔胶原蛋白支架制备人工神经修复大鼠坐骨神经缺损

Fabricating a biomimetic artificial nerve with polylactic acid glycolic acid copolymer composite ordered multi tunnel collagen scaffold to repair the sciatic nerve defects in rats

目的利用聚乳酸-羟基乙酸共聚物（PLGA）复合有序多孔胶原蛋白支架制备人工神经并探讨其修复大鼠坐骨神经缺损的效果。方法通过真空冷冻干燥及定向拉制方法制备有序多孔胶原蛋白支架用以模拟神经束膜，使用PLGA制备导管用以模拟神经外膜，然后在体视显微镜下将有序多孔胶原蛋白支架装入PLGA导管（5：1）制备人工神经。将制备好的64只坐骨神经缺损模型大鼠按随机数字表法分为4组：人工神经组、单纯PLGA组、同种异体神经组、对照组，每组16只，其中人工神经组、单纯PLGA组、同种异体神经组分别用制备的人工神经、PLGA空导管及同种异体坐骨神经桥接缺损神经，对照组则不做移植处理。术后11周时对各组大鼠进行行为学、神经电生理检测及荧光金逆行束路示踪。术后12周时采用苏木素染色及四甲基罗丹明-α-银环蛇毒素（TMR-α-BTX）染色观察靶区肌组织（腓肠肌），采用NF200免疫组化染色观察再生轴突及采用Nissl荧光染色观察再生脊髓前角运动神经元。结果术后11周时同种异体神经组、人工神经组、单纯PLGA组和对照组大鼠运动功能（前后脚间距、脚掌与前进方向的夹角）恢复情况依次变差，各组间差异均有统计学意义（P〈0．05）；神经电生理检测显示，无论是潜伏期还是复合动作电位波幅，均以同种异体神经组的恢复效果最好，人工神经组次之，而单纯PLGA组的潜伏期最长、复合动作电位波幅最小，各组间差异均有统计学意SO（P〈0．05）。术后12周时各组大鼠间肌肉湿重比及肌纤维横截面积比、运动终板面积比较差异均有统计学意义伴0．05），其中人工神经组大鼠的腓肠肌萎缩程度较单纯PLGA组有显著改善，但尚未达到同种异体神经组的水平；NF200免疫组化染色显示．人工神经组移植区可见大量的NF200阳性轴突，但数量略少于同种异体神经组，而单纯PLGA组中NF200阳性轴突数量最少，且主要分布于导管近侧端；单纯PLGA组仅有（19．33±6．73）％的再生脊髓前角运动神经元被荧光金标记，而人工神经组、同种异体神经组的荧光金阳性率分别达到（42．67±7．45）％、（50．13±4．33）％，各组间差异均有统计学意义（P〈0．05）。结论PLGA复合有序多孔胶原蛋白支架构建的人工神经能够有效地修复大鼠坐骨神经缺损。但其效果仍与同种异体神经移植有一定差距。

Objective To investigate the potential of polylactic acid glycolic acid copolymer （PLGA） composite ordered multi tunnel collagen scaffold in fabricating a biomimetic artificial nerve graft to repair the sciatic nerve defects in rats. Methods The ordered multi tunnel collagen scaffold was prepared by vacuum freeze-drying and directional drawing method to simulate the epineurium; the outer conduit was prepared by PLGA to simulate the epineurium; and then, the ordered multi tunnel collagen scaffolds were loaded in the PLGA conduit （5：1） under a stereomicroscope to develop a novel biomimetic artificial nerve. Sixty-four rats were randomly divided into four groups： artificial nerve group, PLGA group, peripheral nerve group, and non-graft control group （n=16）; rats in the artificial nerve group, PLGA group, and peripheral nerve group were repaired with artificial nerve graft, hollow PLGA conduit and allogeneic sciatic nerve to bridge the sciatic nerve defect, while the sciatic nerve with the gap in rats of the control group was without any grafting. After 11 weeks of operation, the hind limbs of rats in each group were detected by behavioral test, electrophysiological examination and Fluoro-Gold retrograde tracing method. The changes of muscle tissues （gastrocnemius） were observed by hematoxylin staining and TMR-α-BTX staining, and the regenerated axons were observed by immunohistochemical staining with NF200 and the regenerative spinal anterior horn motor neurons were observed by Nissl fluorescence staining 12 weeks after operation. Results After 11 weeks of operation, the recoveries of the motor functions （the distance between the injured hindlimb and forelimb, the rotation angle of the injured foot） in the peripheral nerve group, artificial nerve group, PLGA group and control group were significantly deteriorated in turn, and the differences were statistically significant （P〈0.05）. Electrophysiological examination showed that the recovery effect of peripheral nerve group was the best in both latency and amplitude of the compound muscle action potential, followed by artificial nerve group. The latency of PLGA group was the longest and the amplitude of compound action potential was the smallest; significant differences were noted between each two groups （P〈0.05）. At 12 weeks after operation, the wet weight ratio of muscle fibers, area of muscle fibers and neuromuscular junction area were significantly different between each two groups （P〈0.05）; the degree of gastrocnemius atrophy in the artificial nerve group was significantly improved than that in the PLGA group, but not yet reached the level of peripheral nerve group. NF200 immunohistochemical staining showed that a large number of NF200-positive axons were seen in the grafts of the artificial nerve group, but the number was slightly smaller than that of the peripheral nerve group; the number of regenerated axons in the PLGA group was the smaller and mainly distributed near the proximal side. In the PLGA group, only （19.33 ±6.73）% regenerated spinal anterior horn motor neurons were labeled with Fluoro-Gold, while the positive rates of Fluoro-Gold in the artificial nerve group and peripheral nerve group were （42.67±7.45）% and （50.13±4.33）%; the differences between each two groups were statistically significant （P〈0.05）. Conclusion The biomimetic artificial nerve made of PLGA conduit and ordered multi tunnel collagen scaffold can efficiently reconstruct the defected peripheral nerve with guiding axonal regeneration and promoting functional restoration in rats; however, its effect is poor than peripheral nerve grafting.