The peripheral nerve traction injury is one of the most common iatrogenic injuries in clinic (Schwartzman and Grothusen, 2008). The traction-related injury of the peripheral nerve, including brachial plexus, recurrent laryngeal nerve, and sciatic nerve, results in neuropathic pain, dysfunction of sensory, and motor systems (Liu H. et al., 2020). The tractive damage on the nerve takes place mainly at the outer epineurium and perineurium, whereas the structure inside the endoneurium remains relatively intact (Chiang et al., 2008). These morphologic findings suggest that the nerve palsies from traction injuries are temporary, and the electromyography (EMG) would gradually gain partial recovery (Brauckhoff et al., 2018). However, if the traction is prolonged or repeated, the resulted nerve root avulsion or widespread longitudinal damage would restrict the EMG recovery due to the limited possibilities for nerve reconstructions (Malessy et al., 2004). The current standard of nerve repair for severe traction injury is autografting (Zheng et al., 2018). This procedure involves the harvest of a section of donor nerve, and the transplantation into the defect (Ray and Mackinnon, 2010). However, the autografting will cause an additional surgery to excise the peripheral nerve, which increases the chance of surgical complications. Additionally, the donor nerve may be insufficient, and the autografts are also restricted to the diameter mismatch between the donor and acceptor nerves (Cheah et al., 2017). Hence, there is a strong demand to find a new way for nerve repair after severe traction injury.

The nerve guidance conduits (NGCs) have shown great promise in the regeneration and functional recovery of peripheral nerves (Zhang et al., 2020a). However, NGCs are rarely used for the nerve traction injury repair, and the underlying impact needs to be elucidated. The morphologic characteristics of nerve traction injury are different from the transection or thermal injury, which are described as neural fiber nonlinearity, myelin degeneration, and internal structural collapse (Gluck et al., 2018). Although at a later stage the nerves experience partial structural and myelin regeneration, the continuity, linearity, and density of myelin cannot fully be recovered (Lopez-Silva et al., 2021). To create an appropriate microenvironment for nerve regeneration, a graphene oxide/poly(D,L-lactide-co-caprolactone) (PLCL) nerve conduit has been developed (Zhang et al., 2020b). Besides the surface chemical property, the surface morphology of NGCs is sculptured by longitudinally oriented structures, which has been found helpful in the orientation and motility of nervous system cells (Zhang et al., 2018a). Moreover, neural recovery can be also improved by using bioactive molecules such as cell growth factors and peptides (Xia et al., 2017; Sarker et al., 2018). For instance, KHIFSDDSSE (KHI) peptides, derived from neural cell adhesion molecules (NCAM), are immobilized onto a substrate, showing efficiency in selective guidance of migration of Schwann cells (SCs) (Ren et al., 2015).

In this study, a porous PLCL film with longitudinally oriented structure and KHI peptides/nerve growth factor (NGF) loading is designed to explore its applicability for the nerve repair after traction injury. The PLCL film mixed with NaCl particles is micropatterned by using a polydimethylsiloxane (PDMS) stamp under heating. After removal of the NaCl particles via water rinsing, the obtained porous micropatterned PLCL film is aminolyzed and grafted with the KHI peptides via glutaraldehyde coupling. The NGF is grafted/loaded on the film by reaction with the unreacted aldehyde groups and physical adsorption. The SCs are cultured on the modified PLCL film to investigate their adhesion, ratio of length/width (L/W), and elongation behavior (Figure 1). Moreover, the NGCs manufactured from the modified PLCL film are used in the sciatic nerve repair in vivo, whose overall performance is evaluated by EMG analysis, morphological observation, and neuromuscular junction (NMJ) during the process of reinnervation after traction injury.

Complete recovery of neural functions after severe traction injury is a great challenge in clinic (Mahan et al., 2019). The treatment by using artificially engineered NGCs can provide an appropriate peripheral nerve growth environment via the improvement of physicochemical properties, surface longitudinally oriented microstructures, and bioactive molecule modification (Carvalho et al., 2019; Vijayavenkataraman, 2020). It was found that the NGCs having aligned topology and bioactive molecules on their inner surface can effectively control cell behaviors and nerve regeneration (Oh et al., 2018; Shah et al., 2019; Samadian et al., 2020). However, the effects of combination in the repair of nerve traction injury are still unknown. The aim of this work is to explore the beneficial combinations of KHI peptide modification, NGF loading, and surface linear micropatterns in the reinnervation after severe traction.

The biodegradation of PLCL is very slow and would not take place apparently within 21 days (Zhang et al., 2015; Zhang et al., 2020b). The macroprous structures on the surface of PLCL films were obtained after leaching of the NaCl porogen particles (Supplementary Figure S1). The porous structures with a suitable pore size were beneficial for cell loading and adhesion, which is consistent with previous report (Sun et al., 2021). The 3D and porous structures offer a larger surface area, enabling the efficient grafting/loading of KHI peptides and NGF. Besides, the porous structure allows better supply of nutrients and oxygen via the pores on the NGCs, which are extremely important for cell proliferation and nerve regeneration inside the NGCs. Nonetheless, the overall linear (grooves and ridges) pattern structure was maintained although the patterns on the porous regions were distorted or absent.

The groove and ridge morphologic surface belongs to one of the longitudinally oriented microstructures, which can affect the nerve cellular behaviors and guide the directional migration of SCs (Wang et al., 2015). It has been found that the surface anisotropic stripes of grooves and ridges can bring meaningful improvement to nerve repair by adapting the morphology, orientation, and motility of nervous system cells (Tonazzini et al., 2015). Besides, Zhang et al. (2018b) found that SCs cultured on 20/40 μm polyester migrate along the stripes, with a higher ratio of L/W compared to those on the flat film. Furthermore, the migration rate of SCs was also increased significantly when being cultured on the 3/3-μm parallel stripe PLCL, which may promote the M2 polarization of macrophage (Zhang et al., 2020a). In this study, the microstructures of parallel grooves and ridges were created on the surface of the PLCL film, which could similarly guide the direction of SCs and enhance the ability of migration even though the macroprous structure was further implemented. Moreover, these microstructures were also sculptured on the surface of NGCs and implemented in the recovery of severe nerve traction injury in vivo. The results reveal that the microstructures of parallel grooves and ridges on the NGCs were helpful in alleviating the nonlinear structures at the injured point after severe traction injury.

The KHI peptide (sequence Lys-His-Ile-Phe-Ser-Asp-Asp-Ser-Ser-Glu-Lys-Pra) is derived from an NCAM, which is a transmembrane protein specifically expressed on neural cells, including neuron, SCs, and other neuroglial cells (Kam et al., 2002; Turner et al., 2019). In our previous study, we developed a complementary density gradient of KHI peptides and demonstrated that the SCs spread better and directionally migrate to the direction of a higher concentration of KHI peptides (Ren et al., 2015). After KHI peptides are immobilized on substrates, the activated NCAM sites can trigger the formation of focal adhesion, improving SC migration (Maness and Schachner, 2007; Lehembre et al., 2008). Furthermore, it was also reported that the KHI peptides can promote the adhesion of neuroglial cells but not affect the adhesion of fibroblasts (Kam et al., 2002). These results prevent the consequence of aberrant reinnervation, which leads to synkinesis rather than physiologic motion (Crumley, 2000). In the present study, the KHI peptides were modified on the surface of a PLCL film, showing a beneficial effect in promoting SC adhesion in vitro. Moreover, the NGCs modified with KHI peptides were also firstly implemented to repair peripheral nerve traction injury in vivo, showing much effective in elevating the level of recovery and accelerating the process of nerve reinnervation.

The NGF is a well-known neurotrophic factor and widely used in the clinical treatment for peripheral nerve regeneration (Liu M.-Y. et al., 2020). However, a platform for controlled delivery is required because of its short half-life in vivo and its potential to impede axonal regeneration when used in supraphysiological doses (Lackington et al., 2019). Hence, we introduced a PLCL porous film loaded with NGF to investigate its effect on the promotion of nerve regeneration. It has been found that KHI-peptide modification did not influence on NGF release. Besides, the porous property of the films would prolong the release time of NGF till reaching 1 week. As shown in Supplementary Figure S2, the loaded NGF could be released in a sustained manner, and the cumulative dose of NGF reached up to 35 ng. The loading of NGF was found beneficial in nerve regeneration on the PLCL film having KHI peptides and micropatterns. As shown in Figure 4, the proliferation and elongation of SCs were increased significantly in the MPLCL-peptide-NGF group compared with those in the PLCL group. Furthermore, the MPLCL-peptide-NGF conduits were also found more beneficial in nerve regeneration after severe traction injury compared with the PLCL group in vivo. The results reveal that NGF loading is helpful to nerve reinnervation after severe traction injury, and the molecular mechanism would be investigated in our following work.

The macroprous structures, groove and ridge morphologic surface, KHI peptide, and NGF modifications were found helpful in nerve regeneration after severe traction injury in vivo. As shown in Figure 5, the animal model of peripheral nerve severe traction injury was validated. The injured peripheral nerves were treated by the NGCs of PLCL, MPLCL, MPLCL-peptides, and MPLCL-peptide-NGF respectively for 4 weeks. It was found that the neural EMG functions were recovered best in the group of MPLCL-peptide-NGF, including the amplitude increase and NCV acceleration (Figure 6). The gait analysis has been performed widely for evaluating the recovery of motor function as a result of functional recovery after peripheral nerve injury in the rat (Varejao et al., 2001). However, it had been reported that paw contraction and autotomy confounded the measurement of toe spread, a factor that is given particular weighting in the formula for sciatic function index determination (Ao et al., 2011). The conduction velocity is an objective and reliable index for evaluation of the conduction of action potential in peripheral nerves (Jiao et al., 2009). Therefore, we chose neural monitoring and nerve conductive velocity as the indexes to evaluate nerve functional recovery. It can be also found that the fiber spacing was reduced, and the ratio of fiber fragment was significantly decreased in the MPLCL-peptide group and MPLCL-peptide-NGF group (Figure 7). Besides, the atrophic manifestation of gastrocnemius muscle was significantly decreased in the MPLCL-peptide and MPLCL-peptide-NGF groups, showing the weight of gastrocnemius muscle gaining and the diameter of muscle bundle increasing (Figure 8). As a result, the combination of KHI peptide modification, NGF loading, and parallel linear micropattern surface could promote the reinnervation of the sciatic nerve after severe traction injury in vivo.

The PLCL films/NGCs were successfully modified with KHI peptides, NGF loading, and parallelly linear micropatterns. The KHI peptide modification and NGF loading were beneficial in SC proliferation, migration, and adhesion. The groove and ridge morphologic surface on the NGCs were helpful in alleviating the nerve nonlinear structures at the injured point after severe traction injury. The KHI peptide modification, NGF loading, and parallel linear micropattern surface could promote the adhesion and elongation of SCs in vitro and the reinnervation of the sciatic nerve after severe traction injury in vivo. Nonetheless, the underlying molecular mechanisms need to be elucidated further in the future.