Peripheral nerve injury continues to be a major challenge in reconstructive neurosurgery. Owing to huge clinical demand, peripheral nerve regeneration, particularly larger gap injuries, has become a prime focus of basic and clinical research. Accelerating axonal regeneration to promote reinnervation and improve functional recovery after PNI is a clinical necessity and an experimental challenge (Goel et al., 2009; Cotter et al., 2010; Tang et al., 2013).

For surgical repair of PNI with substantial neural defects, the current gold standard is to bridge the defect with an autologous nerve graft that is obtained from another part of the body. In terms of inevitable drawbacks associated with autologous nerve graft, the development of artificial substitutes for autologous nerve grafts is an urgent need in the field of regenerative medicine. NTE has been shown to satisfy this need. It is typically comprised of a neural scaffold with incorporated biochemical cues. The scaffold is prepared with a variety of synthetic or natural biomaterials through well-defined fabrication techniques. Among a large body of scaffold biomaterials, Poly(lactic-co-glycolic acid) (PLGA) (Bini et al., 2004), comprised of Polylactide (PLA) and Polyglycolide (PGA), shows excellent neural affinity and biocompatibility with cells. PLGA not only can be used to control the release of inosine, NGF and insulin-like growth factor (IGF), but also combine to seed cells such as Schwann cells, bone marrow stromal cells (BMSCs) or neural stem cells (NSCs), or can be processed into a graft for guiding peripheral nerve regeneration.

Seed cells, as an important component of NTE, provide an optional strategy for NTE with an enhanced ability to repair PNI. Attractively, EPI-NCSCs, combined with the advantage of adult stem cell and embryonic stem cell, represent a unusual type of multipotent adult stem cell and is a suitable candidate for cell-based therapy (Hu et al., 2006; Sieber-Blum et al., 2006). Accumulating evidences (Amoh et al., 2005) suggest that EPI-NCSCs can be induced to differentiate into Schwann cells and might secrete growth factors to modulate the behavior of Schwann cells. Our previous reports have showed that EPI-NCSCs could reduce inflammation (Li et al., 2017) and promote the segmental recovery of PNI (Zhang et al., 2014). Moreover, OECs are specialized glial cells somewhat similar to Schwann cells and astrocytes, which secrete NGF and BDNF, suggesting that NTFs produced by OECs might enhance the survival of damaged axons (Marshall et al., 2006). Indeed, abundant studies have reported that OECs are able to promote axonal regeneration and remyelination after SCI (Gomes et al., 2018; Gómez et al., 2018; Wright et al., 2018) and PNI (Radtke et al., 2009; Guérout et al., 2011). Importantly, the synergic effects of EPI-NCSCs and OECs improved locomotor function of contused spinal cord of rats and enhanced the expression of NTFs (Zhang et al., 2015). Based on these, the combination of EPI-NCSCs and OECs may synergistically repair PNI. Therefore, we will probe the mechanism that co-transplantation of EPI-NCSCs and OECs have reparative effects on PNI.

In the present study, we engineered nerve grafts by incorporating EPI-NCSCs and OECs as seed cells into PLGA to bridge a 10 mm long sciatic nerve defect in rats, and investigated therapeutic effects of cells co-transplantation by using IHC, electrophysiological recording in vivo and neural function test.

The poor outcome after PNI has provoked researchers to improve methodologies for peripheral nerve regeneration (Xu et al., 2012; Hu et al., 2013). In this study, we used a rat sciatic nerve defect model to demonstrate that EPI-NCSC and OEC co-transplantation may ameliorate PNI. We further employed histology (TEM, IHC), electrophysiology in vivo, behavioral tests and retrograde tracing methods to identify the facilitation of peripheral nerve regeneration and nerve function recovery after co-transplantation.

Peripheral nerve system has a great regeneration potential, particularly when there is an appropriate microenvironment, such nerve conduits applied to a nerve defect that provide a guide and a biological environment for nerve regeneration. During the past few years, studies have been focused on various conduit materials, particularly biodegradable polymers such as poly(glycolic acid) (PGA) (Waitayawinyu et al., 2007), poly(L-lactic acid) (PLLA) (Hood et al., 2009), polycaprolactone (PCL) (Mligiliche et al., 2003) and Poly(lactic-co-glycolic acid) (PLGA) (Bini et al., 2004). These studies have indicated that the conduit itself does not have a pronounced effect on nerve repair. Thus, approaches to nerve repair are now focused on the optimization of the combination of nerve conduits such as NTFs (Tang et al., 2013), ECM (Li et al., 2010, 2017), and seed cells (Gu et al., 2014). We have reported that PLGA is strong (Figure 2C, 3A), and has neural affinity and biocompatibility with cells (Zhang et al., 2014), which are ideal graft properties for nerve regeneration (Li et al., 2010; Zhang et al., 2014). Interestingly, ECM provides well microenvironment, which regulate cell adhesion, spreading, and proliferation, and promote cell survival and myelination (Chernousov et al., 2008). Therefore, in this study transplanted cells combined with ECM may facilitate transplanted cell adhesion, spreading, proliferation, and cell survival. These are in line with the results in our study (Figure 3G,H).

Cell transplantation provides great potential for enhancement of nerve regeneration in view of NTFs, anti-inflammatory effect, as well as axon regeneration. EPI-NCSCs is a preferred candidate for cell transplantation, due to multipotent potential, autologous tissue without immune issue, and easy obtainment. What’s more, it combines the virtue of adult stem cells and embryonic stem cells. EPI-NCSCs, originated from the embryonic neural crest, can be easily obtained from the bulge of hair follicles without harmlessness for the donor (Figure 1A–D), which expand quickly in vitro. Recent studies have demonstrated the utilization of stem cells for peripheral nerve regeneration (Amoh et al., 2005). OECs secrete many NTFs including BDNF and NGF, which is often used to repair PNI (Guérout et al., 2011; Zhu et al., 2014; Ruiz-Mendoza et al., 2016; Wright et al., 2018). Based on the above, we co-implanted EPI-NCSCs and OECs into nerve conduits that were applied to a rat sciatic nerve defect model, demonstrating that EPI-NCSC and OEC co-transplantation increased the number of myelin sheathes (Figure 6D,I,K) and facilitated functional nerve recovery (Figure 5G,H).

The recovery of locomotor function is assessed by the SFI (de Medinaceli et al., 1982). In the present study, the SFI of the EPI-NCSC+OEC group was higher than that the DMEM/F₁₂ group 8 weeks after transplantation (Figure 5G). This is consistent with report that transplanted OECs in a transected sciatic nerve model improved the motor function 3 weeks after surgery (Radtke et al., 2009). In the sensory function test, withdrawal time from hot water was employed to assess the recovery of sensory function (Röyttä et al., 1999). A previous study demonstrated that transplanted EPI-NCSCs promoted sensory recovery in response to warm water withdrawal (Li et al., 2017); we showed a similar but modest effect, whereas we observed a quicker withdrawal response in the EPI-NCSC+OEC group when compared with the other groups, except that the withdrawal response never came to reaching the control group (Figure 5H). This indicated that EPI-NCSC+OEC co-transplantation promoted the partial recovery of motor and sensory function. We further assessed the function of the regenerating sciatic nerves by electrophysiology in vivo. The results indicated that NCV, cMAPs and latency in the EPI-NCSC+OEC group were improved compared with the individual cell group (Figure 4). Previous studies have reported that transplanted neural stem cells improve NCV in a rat sciatic nerve transection model (Xu et al., 2012), that EPI-NCSCs promote the recovery of sciatic nerve cMAPs (Li et al., 2017), and that transplanted OECs improved sciatic nerve NCV and cMAPs in a rat sciatic nerve defect model (Li et al., 2010). We further performed gross observations and histological analyses by HE and TEM, which showed that the number, and thickness of myelin sheathes in the EPI-NCSC+OEC group were greater than in the individual cell group (Figure 6), similar to a previous study showing that transplanted OECs promoted the recovery of myelin sheathes in a rat sciatic nerve defect model (Franklin et al., 1996; Li et al., 2010). However, it reported that OECs was capable of remyelinating demyelinated CNS axons following transplantation into rat spinal cord injuries (Barnett et al., 2000; Kato et al., 2000) and promoted axon sprouting in the lesioned spinal cord (Richter et al., 2005), which indicated that OECs might play a crucial role in the regeneration of myelin sheathes and axon. To assess the accuracy of axon regeneration by EPI-NCSC+OEC co-transplantation, we used DiI retrograde tracing methods. The results indicated that the number of DiI-labeled motoneurons in SCL_4-6 in the EPI-NCSC+OEC group was greater than in the DMEM/F₁₂ group (Figure 5). Further, EPI-NCSC+OEC co-transplantation reduced motoneuron apoptosis in SCL_4-6 8 weeks after implantation (Figure 7), similar to a previous study using horseradish peroxidase retrograde tracing that demonstrated greater neuron survival following NCSCs transplantation than in the control group 52 weeks after transplantation (Lin et al., 2009). Strikingly, triple fluorescent retrograde tracing determined that OECs promoted the recovery of facial motor nerves by stimulating axonal sprouting (Guntinas-Lichius et al., 2001). Additionally, the resuming ratio of gastrocnemius in the EPI-NCSC+OEC group was higher than the EPI-NCSC, OEC, and DMEM/F₁₂ groups (Figure 7L) which further indicated that EPI-NCSC+OEC co-transplantation promoted neurotrophic support to the recovery of gastrocnemius. Overall, our results indicate that EPI-NCSC+OEC co-transplantation may promote axonal regeneration and the recovery of nerve function.

However, how can functional nerve and axon regeneration be improved? Research to date indicates the importance of NTFs, particularly BDNF and NGF, during recovery after PNI (Sendtner et al., 1992; Zhang et al., 2000; Shakhbazau et al., 2012a; Tang et al., 2013). A growing amount of evidence indicates that, in addition to providing structural support for growing axons by the expression of NTFs (Stoll and Müller, 1999), OECs release many NTFs, including NGF, BDNF and GDNF (Bunge et al., 1989; Brown et al., 1991). NGF is known to guide axons (Yu et al., 2010), promote axonal sprouting (Tuszynski et al., 1996), and cell migration (Cao et al., 2007; Figure 3G, 8N), stimulate myelination (Chan et al., 2004), improve the regeneration of sensory neurons and reduce denervated muscle atrophy (Crowley et al., 1994; Shakhbazau et al., 2012a; Figure 7L, 8N), and eventually to improve functional recovery after injury (Röyttä et al., 1999; Kemp et al., 2011). In our study, the expression of NGF in the EPI-NCSC+OEC group was higher than in the other group (Figure 8G–L,N), indicating that NGF might participate in nerve repair. Up-regulation of NGF might also play a vital role in nerve repair by increasing the number of myelin sheathes, which was greater in the EPI-NCSC+OEC group than in the DMEM/F₁₂, EPI-NCSC or OEC groups (Figure 6F). Besides, BDNF is important for stimulating axonal elongation (Acheson et al., 1991; Zhang et al., 2000; English et al., 2005; Wilhelm et al., 2012) and for survival of motoneurons (Sendtner et al., 1992; Yan et al., 1992; Koliatsos et al., 1993). We also found that the expression of BDNF in the EPI-NCSC+OEC group was greater than in the other group (Figure 8A–F,M). Likewise, it reported that OEC-M treatment after contusive SCI increased BDNF levels and then improved the function recovery and promoted the axonal regeneration (Pastrana et al., 2007; Gu et al., 2017). Herein, high expression level of BDNF and NGF in the EPI-NCSC+OEC group might improve nerve function and promote the axonal regeneration. Therefore, in this study we speculated that the mechanism that EPI-NCSC+OEC co-transplantation repaired PNI might be up-regulation of BDNF and NGF, but BDNF and NGF might derive from: (1) OEC might secrete, which have been reported to express BDNF and NGF and facilitate axonal regeneration after SCI (Lipson et al., 2003); (2) Schwann cells might secrete BDNF and NGF after PNI in vivo (Shakhbazau et al., 2012b); (3) EPI-NCSC might secrete little BDNF and NGF (Sieber-Blum et al., 2006), but the combination of EPI-NCSC and OEC might heighten the expression level of BDNF and NGF, which have been reported that co-culture of Schwann cells and adult stem cells led to synergistic neurotrophic effects (BDNG and NGF) in PNI (Dai et al., 2013) and synergic effects of EPI-NCSCs and OECs increased the expression of BDNF and GDNF in SCI (Zhang et al., 2015). One maybe that the combination of cells promote the secretion of BDNF and NGF from Schwann cells of donor; another maybe that EPI-NCSC prompts the expression of BDNF and NGF from OECs. However, specific origin of BDNF and NGF remain to further investigate.

Overall, our results unveiled a beneficial effect of co-transplantation of EPI-NCSCs and OECs after PNI, whereas many issues remained to be determined such as the differentiation of stem cells in vivo. EPI-NCSC and OEC co-transplantation was a promising and easily transformable approach that could lead to significant amelioration of patients suffering from PNI.

Our findings indicated that EPI-NCSC and OEC co-transplantation promoted sciatic nerve regeneration and improves nerve function. Moreover, the mechanism of PNI improved by EPI-NCSC and OEC co-transplantation was likely to up-regulate the expression of BDNF and NGF. The application of EPI-NCSC and OEC co-transplantation in clinical trials might improve clinical outcomes and provided a new methods for PNI.

The raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any qualified researcher.

All experimental procedures with animals were approved by the local institution review board and were carried out according to the guidelines of the Third Military Medical University (Chongqing, China) for the care and use of laboratory animals.

LZ performed the experiment, conceived the study design, analyzed the data, and drafted the manuscript. BLi and BLiu participated in conception and design of the experiments. ZD modified the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.