【Introduction】
Chronic nerve injuries show poor prognosis to nerve reconstruction. The assumption is that motor end plates in the target muscle degenerate irreversibly, and does not respond to incoming nerve signals. Little is known about the morphological changes and the functional response of the neuromuscular junction (NMJ) in time-variable nerve injuries.

【Materials and Methods】
Sciatic nerve axotomy were first performed in C57BL/6 mice. In group 1, the nerve is repaired primarily, whereas in group 5 the nerve is not repaired. In group 2, the nerve is reconstructed immediately with a 1cm autograft. In group 3, the nerve is reconstructed with the autograft at two weeks after the axotomy; in group 4 the nerve is reconstructed four weeks after the axotomy. NMJ was analyzed by immunohistochemical staining of the target muscles. Functional reinnervation was determined by 1) treadmill running exercise and 2) compound muscle action potential (CMAP) of the target muscle.

【Results】
Treadmill exercise showed substantial difference between group 4 to group 1 (p = 0.054) and to group 2 (p = 0.005) at eight weeks. CMAP showed only significant difference between group 4 (2.64 mV) and group 1 (5.44 mV) at eight weeks. Immunohistochemical staining of the NMJ at eight weeks showed greater than 95% of motor end plate innervation for groups 1 to 4, but group 4 had notably higher % of partially innervated NMJ. Motor end plate fragmentation was different between the groups but not significant. % of nonmyelinating SCs with morphological response to injury at the NMJ was significantly lower in groups 4 and 5 at four weeks; by eight weeks, groups 1 to 4 showed no obvious differences (Figure 1).

【Conclusion】
The mice models accurately represented the clinical scenarios of nerve grafting in acute and chronic nerve injuries. The main morphological response in correlation to the functional outcomes was displayed by the SCs and the innervating axons at the NMJ. This highlights the importance of glial cells in the NMJ as a marker for functional reinnervation.