Combining chitin biological conduits with small autogenous nerves and platelet‐rich plasma for the repair of sciatic nerve defects in rats

Abstract Aims Peripheral nerve defects are often difficult to recover from, and there is no optimal repair method. Therefore, it is important to explore new methods of repairing peripheral nerve defects. This study explored the efficacy of nerve grafts constructed from chitin biological conduits combined with small autogenous nerves (SANs) and platelet‐rich plasma (PRP) for repairing 10‐mm sciatic nerve defects in rats. Methods To prepare 10‐mm sciatic nerve defects, SANs were first harvested and PRP was extracted. The nerve grafts consisted of chitin biological conduits combined with SAN and PRP, and were used to repair rat sciatic nerve defects. These examinations, including measurements of axon growth efficiency, a gait analysis, electrophysiological tests, counts of regenerated myelinated fibers and observations of their morphology, histological evaluation of the gastrocnemius muscle, retrograde tracing with Fluor‐Gold (FG), and motor endplates (MEPs) distribution analysis, were conducted to evaluate the repair status. Results Two weeks after nerve transplantation, the rate and number of regenerated axons in the PRP‐SAN group improved compared with those in the PRP, SAN, and Hollow groups. The PRP‐SAN group exhibited better recovery in terms of the sciatic functional index value, composite action potential intensity, myelinated nerve fiber density, myelin sheath thickness, and gastrectomy tissue at 12 weeks after transplantation, compared with the PRP and SAN groups. The results of FG retrograde tracing and MEPs analyses showed that numbers of FG‐positive sensory neurons and motor neurons as well as MEPs distribution density were higher in the PRP‐SAN group than in the PRP or SAN group. Conclusions Nerve grafts comprising chitin biological conduits combined with SANs and PRP significantly improved the repair of 10‐mm sciatic nerve defects in rats and may have therapeutic potential for repairing peripheral nerve defects in future applications.


| INTRODUC TI ON
Peripheral nerve damage often leads to pain and severe disability. In the United States, approximately 50,000 people undergo surgery to repair peripheral nerve damage each year. The causes of injury include traffic accidents, fractures, explosive injuries, and iatrogenic injuries. Only a small proportion of these patients achieve full functional recovery after treatment. [1][2][3][4][5][6] When the peripheral nerve defect is larger than 3 cm, the current gold standard clinical treatment is autogenous nerve transplantation. The main problems with autogenous nerve transplantation are that it requires the use of two surgical sites and the donor area is left denervated. Many new therapeutic strategies to improve nerve repair are being developed in basic, preclinical, and clinical trials.
Tissue-engineered nerve grafts exhibit the most potential as a replacement for autogenous nerve grafts and are widely used to repair peripheral nerve defects. [7][8][9][10] The core factors for constructing tissue-engineered nerve grafts include biological scaffolds, seed cells, and various growth factors. 11 Most tissue-engineered nerve grafts are constructed from nerve conduits. Nerve conduits grafted between nerve defects provide a suitable microenvironment for regenerating axons, thus improving the efficiency of nerve regeneration and restoring impaired motor and sensory functions. Chitin biological absorbable conduits exhibit good biological properties in the peripheral nerve defect repair model. 12,13 A single proximal axon produces multiple lateral buds during nerve regeneration, a phenomenon known as the "multiple amplification" effect, which leads to a maximum power rate of approximately 3.3 for nerve regeneration. 14 More importantly, small autogenous nerves (SANs) secrete neurotrophic factors and promote Schwann cell (SC) proliferation during Wallerian degeneration. 15,16 SANs with a complete structure provide a hierarchically aligned structure and an optimal bridge for the rapid growth of axons. Therefore, tissue-engineered nerve grafts can be constructed using SANs. In addition, the application of SANs overcomes the problem of insufficient sources for autologous nerve transplantation by increasing the supply of autogenous nerves and reduces the level of denervation at the donor nerve site. [17][18][19] Neurotrophic factors are important components of tissueengineered nerve grafts. Soluble neurotrophic factors can be directly integrated into nerve conduits. These nutritional factors include nerve growth factor (NGF), insulin-like growth factor (IGF), fibroblast growth factor, brain-derived neurotrophic factor (BDNF), and glial cell line-derived neurotrophic factor (GDNF). 20 -22 Platelet-rich plasma (PRP) contains a variety of neurotrophic factors and cytokines that participate in the regulation of early inflammation, angiogenesis, fibrogenesis, and macrophage polarization during repair. Platelet-derived growth factors, such as transforming growth factor-microRNA, IGF-1, platelet-derived growth factor-AB, vascular endothelial growth factor, GDNF, NGF, and platelet-derived body signal molecules such as microRNA, promote the distal growth of axons and accelerate the migration and proliferation of SCs. 23,24 All of these processes are key to nerve functional recovery. Moreover, activated PRP is a gel and can be used as a framework for nerve regeneration. Thus, PRP provides the appropriate biomimetic microenvironment for nerve regeneration. 25,26 Because of the potential roles PRP and SANs play in promoting nerve regeneration, nerve grafts were constructed from chitin biological conduits combined with SANs and PRP to repair 10-mm sciatic nerve defects in rats in this study. The effectiveness of nerve injury repair was systematically evaluated using animal experimental methods.

| Preparationofchitosanneuralconduits
Each chitin biological conduit was prepared according to a previously described protocol (National Invention patent no. ZL01136314.2: Partial deacetylated chitin biological conduit guiding and promoting effective nerve regeneration and its manufacturing method). 12,18

| PreparationandinjectionofautogenousPRP
After 1 week of acclimation, the rats were anesthetized with 2.5% isoflurane gas, and 1 ml of whole blood was collected from the posterior orbital venous plexus of each rat and placed in a collection vessel containing 3.8% (w/v) sodium citrate. The whole blood was centrifuged at 400 × g and 22℃ for 10 min and divided into three layers thereafter. The bottom layer comprised red blood cells, the upper layer comprised acellular plasma, and the middle layer comprised platelets and was known as the white membrane. The platelet-containing plasma was transferred to a sterile centrifuge tube without anticoagulant and centrifuged at 800× g for 10 min to obtain the platelet concentrate. The upper two-thirds of the plasma were removed, leaving the lower one-third that comprised approximately 50 µl of PRP, which was resuspended by gently shaking in a centrifuge tube ( Figure 1A). The red blood cells, white blood cells, and platelets in the whole blood and PRP were counted using an automatic counter. The PRP was activated with 10% calcium chloride (Sigma-Aldrich, St. Louis, MO, USA) and a bovine thrombin (Sigma-Aldrich) mixture to obtain the PRP gel.   Table 1.

| Histochemical staining of regenerated nerve fibers
Three rats in each group were randomly euthanized with excess sodium pentobarbital solution at 3 weeks after nerve transplantation, and the nerve grafts were extracted. After fixation, the nerve grafts were cut into 12-µm longitudinal sections using a frozen-section machine. Six slides from each group underwent immunofluorescence staining (NF200/S100/DAPI) and hematoxylin and eosin (HE) staining. A vertical scanner (Zeiss, Jena, Germany) was used to observe the regeneration of the nerve fibers and inflammation at 200× magnification. The sections were stained with NF200 (Sigma-Aldrich) to reveal the regenerated axons and with S100 (Sigma-Aldrich) to show SCs, following a previously described method. Briefly, the specimens were incubated with NF200 and S100 primary antibodies overnight at 4℃ and washed three times with PBS for 5 min. The mixed secondary antibodies Alexa488 and Alexa594 (Abcam, Cambridge, UK) were added, and the specimens were incubated at room temperature for 1 h and then dyed with DAPI for 5 min.

| Gait analysis
The CatWalk XT 10.6 gait analysis system (Noldus, Wageningen, the Netherlands) was used to evaluate the recovery of muscle motor function at 2, 4, 6, 8, 10, and 12 weeks after nerve transplantation.
The animals in each group were placed in a closed passage consisting of a glass platform and black plastic walls. The sciatic functional index (SFI) was calculated using the following formula: where ETS is the experimental distance from the first to the fifth toe, NTS is the normal first-to-fifth-toe distance, EPL is the experimental distance from the heel to the top of the third toe, NPL is the normal heelto-third-toe distance, EITS is the experimental distance from the second to the fourth toe, and NITS is the normal second-to-fourth-toe distance.

| Nerve electrophysiological detection
The rats in each group were tested with a neurophysiological instrument (Keypoint, Nørresundby, Denmark) under general anesthesia at 12 weeks after nerve transplantation to determine the recovery of sciatic nerve conduction at the affected side. The gastrocnemius and sciatic nerve at the affected side were carefully exposed during the operation. Two stimulating electrodes were placed at the proximal and distal ends of the nerve graft, and a recording electrode was inserted into the abdomen of the gastrocnemius. Compound muscle action potentials (CMAPs) when the nerve was electrically stimulated at 3.0 mA and 1 Hz were recorded. The delay and peak amplitude of the CMAPs were compared among the different groups.

| Morphological evaluation of regenerated nerve fibers
The experimental animals were sacrificed by intraperitoneal injection of an excess of 3% pentobarbital sodium solution at 12 weeks after nerve transplantation. The 5-mm middle segment of the regenerated nerve and the portion located 2 mm away from the distal end of the nerve graft were obtained. The middle segment of the nerve graft was crosscut for immunofluorescence staining (NF200/S100/DAPI), using the procedure described in section 2.3.1. The distal 2 mm of the regenerated nerve was fixed in 2.5% (v/v) glutaraldehyde for 6 hours after the adherent tissue was trimmed and then fixed with 1% osmium tetroxide. The specimens were cut into 1-µm-thick semithin cross-sectional slices and 70-nm-thick ultrathin cross-sectional slices with a Leica EM UC7 ultramicrotome (Wetzlar, Germany). The specimens were stained with 1% (w/v) toluidine blue/1% (w/v) borax solution for 5 min, and then with lead citrate and uranyl acetate. The images were collected using a BX51 microscope with a DP71 camera (Olympus, Tokyo, Japan). Five sections were randomly selected in each group, and five fields were randomly selected from each section to count the mean density of myelinated nerve fibers by IPP 6.0 observed at 400× magnification. The ultrathin sections were observed using transmission electron microscopy (Philips, Best, The Netherlands). Five sections were randomly selected in each group, and five fields were randomly selected from each section to measure the mean diameter of myelinated nerve fibers and the mean thickness of the myelin sheath by IPP 6.0 at 5000× magnification.

| Gastrocnemius recovery
The gastrocnemius of the affected and healthy sides was removed and weighed at 12 weeks after nerve transplantation. The affected gastrocnemius muscles from each group were immersed in 4% paraformaldehyde, embedded in paraffin, and cut into 7-µm slices. All

| Dataanalyses
Blinding was used for all image analysis and behavior assessments.

| PRPcharacteristics
Rat whole-blood and PRP cell counts are summarized in

| Chemicalstainingofregeneratednervetissue
Representative results of immunofluorescence (NF200/S100/ DAPI) and HE staining of nerve grafts for each group at 3 weeks after nerve transplantation are shown in Figures 2 and 3, respectively. The immunofluorescence staining showed that green

| Electrophysiologytest
The rats in each group were subjected to electrophysiological tests at 12 weeks after nerve transplantation to assess the recovery of nerve conduction, as shown in Figure 5. The CAMP peak amplitude is related to the number of innervated muscle fibers, whereas CAMP delay time is related to the myelin sheath thickness of regenerated nerves. Figure 5A shows the representative CMAP waveforms for each group. Although the CMAP peak amplitude in the PRP-SAN group was significantly lower than that in the Autograft group (p < .01), it was higher than those in the PRP, SAN, and Hollow groups (p < .01) ( Figure 5C). In addition, the CMAP delay time in the PRP-SAN group was less than that in the Autograft group (p < .05), but significantly greater than those in the PRP, SAN, and Hollow groups (p < .01) ( Figure 5B).

| Degreeofaxonregenerationandmyelination
Immunofluorescence staining of the middle cross-sections of the rat grafts in each group was performed at 12 weeks after nerve transplantation, and the results are shown in Figure 6. Green fluorescence represents regenerating axons, red fluorescence represents SCs, and blue fluorescence represents nuclei. Our observations revealed that more regenerated axons were found in the Autograft and PRP-SAN groups than in the PRP, SAN, and Hollow groups.
Results of toluidine-blue staining and electron microscopic examination of transverse sections of the rat grafts are shown in Figure 7. Figure 7A shows representative images of toluidine-blue staining for each group and the number of myelinated nerve fibers in a unit field of vision. The results revealed that the number of regenerated myelinated nerve fibers in the PRP-SAN group was not significantly different from that in the PRP group, but higher than those in the SAN and Hollow groups (p < .05), and significantly lower than that in the Autograft group (p < .01) ( Figure 7D). Figure 7B shows representative transmission electron microscopy images of ultrathin sections of the distal grafts for each group. The myelin sheath thickness of the regenerated nerve fibers in the PRP-SAN group was significantly higher than those in the PRP, SAN, and Hollow groups (p < .01) but significantly lower than that in the Autograft group (p < .01) ( Figure 7E). In addition, the perimeter-based g-ratio value was calculated to further evaluate the degree of myelination of the regenerated nerve fibers. The g-ratio of the regenerated nerve fibers was significantly lower in the PRP-SAN group than in the PRP Green fluorescence (NF200) shows axons, red fluorescence (S100) shows Schwann cells (SCs), and blue fluorescence (DAPI) shows the nucleus. Scale bar =100 μm and SAN groups but higher than in the Autograft group. No difference was observed between the PRP and SAN groups ( Figure 7F). Figure 7C shows the local enlargement of the myelin sheath of representative regenerated nerve fibers in each group.

| Gastrocnemiusmusclerecovery
The gastrocnemius muscles of two rats in each group were removed and weighed at 12 weeks after nerve transplantation, and Masson's trichrome staining was performed on the affected muscles. The results are shown in Figure 8. Representative images of the affected and healthy gastrocnemius muscles in each group are shown in Figure 8A. The wet weight ratio of affected-to-healthy gastrocnemius muscle in the PRP-SAN group was significantly lower than that in the Autograft group (p < .01), but higher than those in the PRP, SAN, and Hollow groups (p < .05). No difference was observed between the PRP and SAN groups ( Figure 8C). In addition, Figure 8B shows that in the Autograft group (p < .01) but significantly higher than those in the other three groups (p < .01) ( Figure 8D).

| FGretrogradetracing
The numbers of DRG sensory neurons and anterior spinal motor neurons growing through the nerve defects in each group were evaluated at 12 weeks after nerve transplantation using the FG retrograde tracing method to analyze the effects of nerve grafts on nerve regeneration in each group. Figure 9A shows representative FGlabeled sensory neurons for each group. The number of FG-labeled sensory neurons in the PRP-SAN group was higher than those in the PRP, SAN, and Hollow groups (p < .05), but significantly lower than that in the Autograft group (p < .01) ( Figure 9D). Figure 9B shows representative FG-labeled motor neurons for each group. The number of FG-labeled motor neurons in the PRP-SAN group was significantly higher than those in the PRP, SAN, and Hollow groups (p < .01) but significantly lower than that in the Autograft group (p < .01) ( Figure 9E). No difference was observed between the PRP and SAN groups. Figure 9C shows a locally enlarged view of representative motor neurons at the anterior horn of the spinal cord for each group.

| MEPsevaluation
The number of MEPs labeled with BTX 647 reagent at 12 weeks after the operation was assessed to evaluate the number of relationships established between motor nerve fibers and muscle fibers. Figure 10A shows representative transparent images of the flexor digitorum longus for each group. Figure 10B shows the three-dimensional MEPs model of the flexor digitorum longus for each group, indicating the three-dimensional distribution of MEPs in the muscles. Figure 10C shows are considered most conducive for promoting tissue healing and repair. 33 In this study, the platelet enrichment level of the PRP was 4.89, which is within the optimal range.
In addition, SANs are readily available during clinical surgery without causing additional sensory or motor impairments due to their extraction. SANs secrete neurotrophic factors and promote SC proliferation during Wallerian degeneration. Moreover, small nerves with a complete structure afford a hierarchically aligned structure, which provides the best bridge for the rapid growth of axons.
The nerve grafts constructed in this experiment were used to evaluate the promotion of nerve repair using a 10-mm sciatic nerve defect rat model. The results of the histochemical evaluation at 2 weeks after nerve transplantation showed that the number of regenerated axons in the PRP-SAN and PRP groups was greater than that in the Hollow group, while regenerated axons in the PRP-SAN and SAN groups grew faster than those in the Hollow group. The Axons can transport material both anteriorly and in a retrograde manner. The axon transport function is lost once it is damaged. If the continuity of a damaged axon can be restored, the axon transport function can also be restored. The axon transport function is an important indicator reflecting the structure, metabolism, and functional integrity of a nerve, and the restoration of axon transport function after nerve injury repair is powerful evidence for nerve regeneration. [34][35][36] In this study, the distal part of the transplanted nerve was infiltrated with FG, which can be transported to the body of the neuron by way of retrograde axoplasmic transport. This allowed us to assess the number of axons recovered from axoplasmic transport in the transplanted nerve segment. The results showed that nerve axon recovery in the PRP-SAN, PRP, and SAN groups was better than that in the Hollow group, and nerve axon recovery in the PRP-SAN group was better than that in the PRP and SAN groups, which was consistent with expectations.
In addition, BTX 647 reagent was used in an experiment to mark MEPs, which indicate the locations where peripheral nerves exchange information with skeletal muscle. Restoration of the MEPs structure is crucial for the recovery of skeletal muscle function. 37,38 The distribution of the MEPs layer of skeletal muscle reflects the repair outcome of peripheral nerves. According to the results, the MEPs distribution in the PRP-SAN, PRP, and SAN groups was denser than that in the Hollow group.
These results indicate that although chitin biological conduits combined with PRP alone or SANs alone can promote peripheral nerve regeneration, the combination of PRP and SANs had a greater facilitating effect than either one alone on peripheral nerve regeneration. At the same time, the SANs and PRP used in this experiment were obtained from the body in a quick, safe, and convenient manner. Therefore, the protocol described here can be a new method for peripheral nerve defect repair in the future.

| CON CLUS IONS
In summary, nerve grafts comprising chitin biological conduits combined with SANs and PRP significantly promoted the repair of 10mm sciatic nerve defects in rats. PRP and SANs in the graft were completely self-derived and obtained quickly and safely. These results will contribute to the development of an effective strategy for repairing peripheral nerve defects.

ACK N OWLED G M ENTS
We thank Richard Turner for editing the English version of a draft of this manuscript. The English in this document has been checked by at least two professional editors, both native speakers of English. For a certificate, please see: http://www.textc heck.com/certi ficat e/0AXSBi.

D I SCLOS U R E
The authors declare no conflict of interest.

AUTH O RCO NTR I B UTI O N S
Chang-Feng Lu contributed to investigation, methodology, data curation, software, validation, and writing-original draft. Bo Wang contributed to project administration, resources, and formal analysis. Pei-Xun Zhang contributed to supervision and resources. Shuai Han contributed to software and resources. Wei Pi contributed to methodology and validation. Yu-Hui Kou contributed to supervision, writing-review and editing, and funding acquisition. Bao-Guo Jiang contributed to supervision and funding acquisition.

DATAAVA I L A B I L I T YS TAT E M E N T
The data are available from the corresponding author upon reasonable request.