Development of a novel method for decellularizing a nerve graft using a hypertonic sodium chloride solution

Introduction

A peripheral nerve injured by trauma or tumor resection needs to be reconstructed to bridge the gap between the two sections. Autologous nerve transplantation is often used in clinical practice for this purpose; however, it necessitates sacrifice of a donor site, and the harvesting of a nerve with the appropriate caliber can be difficult (1). Although artificial conduits including silicon tubes (2), collagen tubes (3), and polyglycolic acid (PGA) (4-5) have been used as substitute nerves, they lack three-dimensional (3D) conformations, which restricts their ability to bridge large gaps. The basal lamina is required to provide a scaffold for the infiltration of Schwann cells and to promote axonal regeneration in the right direction. Schwann cells express nerve growth factors (NGF) and stimulate axon extension (6). Thus, the induction of Schwann cells is crucial for extending the axons of the artificial nerves (6-7-8-9). Acellular nerve grafts are potentially useful because they are equipped with scaffolding (10-11-12-13). Current decellularizing processes involve freeze-thaw methods (10-11) and detergent processing (12-13-14); however, there are limitations to these approaches. For example, a single cycle freeze processing might not adequately remove cells, leading to a rejection response induced by the membranes of remnant cells (15). Moreover, although detergent processing can efficiently remove cells, it can also destroy the extracellular matrix (ECM) and fine three-dimensional structure, thus destroying the scaffold required for Schwann cell infiltration and efficient axon extension (16). Detergent processing also involves many cytotoxic processes (16-17).

In an attempt to improve current nerve reconstruction techniques, we harvested the sciatic nerve of one rat, created acellular nerve grafts by treating the tissue with hypertonic sodium chloride solution, and replanted the graft in another rat. We then evaluated the success of the acellular nerve graft by hematoxylin-eosin (H-E) staining, immunostaining, anterograde nerve tracing, and electron microscopy images. As controls, we assessed nerve grafts that were freeze-thawed, while nerves that had undergone detergent processing were also evaluated by H-E staining, immunostaining, anterograde nerve tracing, and electron microscopy images.

Materials and Methods

Results

Evaluation of acellular nerve

H-E staining revealed no nucleoli in the nerve grafts processed with our method, and the ECM was preserved (Figs. 1a and b). Immunohistochemical analysis revealed that laminin, an architectural component of the neural basal laminae, maintained its spatial structure (Figs. 1e and f). Additionally, evaluation with the electron microscope revealed that most cells were removed but some myelin remained (Figs. 2a and b).

Fig. 1 -

Hematoxylin-eosin (H-E) staining and anti-laminin antibody labeling showing nerves before and after decellularization, a) intact nerve, b) acellular nerve treated by our method, c) freeze-thaw processed nerve, d) detergent processed nerve, e) laminin-labeled intact nerve, f) laminin-labeled acellular nerve treated by our method, g) laminin-labeled freeze-thaw processed nerve, h) laminin-labeled detergent-processed nerve, scale bar: a-d: 100 μm; e-h: 20 μm H-E-stained acellular nerves treated by our method and detergent-processed nerves contain no cells, and the extracellular matrices are preserved. In contrast, the freeze-thaw-processed nerve contains numerous cells. Immunostaining (anti-laminin antibody) image: the neural basal laminae are preserved in the nerves treated by our method and the freeze-thaw-processed nerves. The structure of the neural basal lamina is destroyed in detergent-treated nerves.

Fig. 2 -

Electron microscopy images before and after nerve decellularization. Schwann cells and myelin sheaths (arrow heads) are almost absent from the acellular nerve treated by our method, and the neural basal lamina (arrows) is preserved. Many cells (Schwann cells and myelin sheaths) remain in the freeze-thaw-processed nerve. Cells are absent, but the neural basal lamina (arrows) is destroyed in the detergent-processed nerve. (Scanning electron microscopy showing a) intact nerve, b) acellular nerve treated by our method, c) freeze-thaw-processed nerve, d) detergent-processed nerve; arrow heads indicate Schwann cells and myelin sheaths arrows indicate neural basal laminae).

Immunostaining of freeze-thaw-processed nerves confirmed preserved basal laminae (Fig. 1g); however, H-E staining (Fig. 1c) and electron microscopy (Fig. 2c) revealed that many cells remained. In contrast, H-E (Fig. 1d) and electron microscopy (Fig. 2d) of nerves treated with detergent confirmed successful decellularization, but immunostaining demonstrated that the basal laminae were broken down (Fig. 1h).

Histological images two months after transplantation

H-E staining showed that grafts treated with our method contained many cells (Fig. 3a). Extended neurofilaments were stained with the anti-neurofilament antibody throughout the graft, and S-100-positive Schwann cells were observed along and surrounding the neurofilaments (Fig. 3b).

Fig. 3 -

Two months after the transplantation: (a, c, e) H-E-stained images and (b, d, f) immunostained images. Although no inflammatory cells are apparent, many cells infiltrated throughout the graft, but the number of visible cells is small in the freeze-thaw-processed nerve. Scale bar: 100 μm. (b, d, f) Neurofilaments run throughout the graft (green: anti-neurofilament antibody) and Schwann cells exist along neurofilaments (red: anti-S-100 antibody). Some Schwann cells surround the neurofilaments. Blue shows DAPI (4’,6-diaminodino-2-phenylindole) staining.

H-E-stained images of freeze-thaw and detergent-processed grafts also exhibited many cells (Figs. 3c and e), while neurofilaments and Schwann cells were observed with fluorescent immunostains (Figs. 3d and f).

Regeneration of neurofilaments observed by anterograde nerve tracing

Following tracing, neurofilaments were fluorescent red at the injection site and extended in the distal direction. The tracer method allowed for visual observation of extending neurofilaments. In rats that received grafts prepared with our method, several neurofilaments ran through the graft from the proximal side of the native nerve (Fig. 4a) and also overrode the suture from the graft side to the peripheral side (Fig. 4b).

Fig. 4 -

Anterograde nerve tracing showing neurofilaments regeneration. Neurofilaments run throughout the graft treated by our method (a), and some filaments are seen across the neural suture (b). Neurofilaments are also observed in the freeze-thaw and detergent processed grafts. (Neurofilaments: arrow heads, scale bar; 200 µm).

Neurofilaments were also observed at the distal ends of freeze-thaw and detergent-processed grafts (Figs. 4c and d), but they did not cross the suture.

Electron microscopy images eight months after transplantation

At eight months after nerve grafts were transplanted using our technique, the graft had not been absorbed by living tissue. However, we observed Schwann cells and myelinated axons infiltrating the graft. Schwann cells induced the formation of neurofilaments and myelin sheaths. Additionally, the neurofilaments and Schwann cells were surrounded by endoneurium, and bundles of neurofilaments were surrounded by perineurium (Figs. 5a and b), resulting in the generation of a structure that was almost the same as that of the intact nerve.

Fig. 5 -

TEM images 8 months after transplantation. Schwann cells infiltrate the graft and induce the formation of neurofilaments and myelin sheaths. (SC: Schwann cells, M: myelin sheath, P: perineurium, E: epineurium).

Discussion

Peripheral nerve reconstruction in the clinical setting is necessary to repair damage due to trauma or surgical procedures. Reconstruction often requires a nerve autograft that can bridge the nerve gap and does not elicit a rejection response because it is autogenic. Nonetheless, sacrifice of the donor site is inevitable, and it is difficult to harvest a proper caliber nerve to match the recipient site (1, 17).

Numerous synthesized nerve grafts for restoring peripheral nerves are already used in clinical settings (18-19-20-21), including collagen tubes (3), polyglycolic acid (PGA) (4-5), and silicon tubes (2); however, these synthesized conduits do not hold biogenic structures. Synthesized nerve grafts can mimic biogenic nerves but cannot mimic microstructures in acellular nerves. The lack of biogenic structures might be one reason why synthesized nerves cannot bridge long gaps (7-8, 22). We hypothesized that the ideal artificial nerve graft would be equipped with scaffolding to induce Schwann cells to infiltrate the graft and promote nerve extension.

Some acellular tissues already used in the clinic, such as skin (23-24), nerve (20, 25), and cartilage (26), preserve the ECM, which is important for promoting cell binding and effective tissue regeneration. Nonetheless, tissues subjected to weak decellularization can cause rejection (12-13, 17).

Although the freeze-thaw (10-11) and detergent (12) acellular methods are the currently prevalent approaches for producing nerve grafts, the former method does not provide a satisfactory outcome in that cells remain in the graft. When graft transplants are performed with residual cells, they can induce adverse reactions. Therefore, the two methods are often combined (12-13-14). However, detergents are cytotoxic (18), may chemically damage the ECM (16, 27), and can destroy the scaffolding to host infiltrating cells.

An acellular nerve requires a structure that cells can infiltrate to synthesize the many axon growth factors needed for successful nerve repair (28-29). Schwann cells play the most important role in producing (9), NGF receptors (6, 9), and growth-associated protein (GAP)-43, which is a growth cone involved in axon growth (30). Macrophages derived from the blood also support these processes (31-32). Schwann cells infiltrate the acellular nerve and dedifferentiate; some become fixed, while others proliferate and promote axonal extension. Therefore, the ideal acellular nerve should be equipped with a scaffold that Schwann cells can permeate easily.

The principle behind our method is that cells subjected to intracellular dehydration are removed from the ECM by the hypertonic sodium chloride solution. Compared with other methods designed to preserve the ECM, this method is less invasive and does not require materials that are toxic to biogenic tissues.

After transplantation, immunostaining and electron microscopy revealed that Schwann cells in acellular nerve grafts generated using our method were almost the same as those in the intact peripheral nerve. Immunostaining revealed neurofilaments running through the graft and some Schwann cells surrounding the neurofilaments. Immunohistochemical examination of detergent and freeze-thaw-processed grafts demonstrated that neurofilaments ran through these grafts, and Schwann cells surrounded the neurofilaments. However, the three-dimensional conformation of freeze-thaw-processed grafts was rough, and H-E staining showed that the number of cells in the graft was low compared with the detergent-processed grafts two months after transplantation (the numbers of axons were not counted).

The anterograde tracing method used in this experiment has not been used for examining nerve regeneration in artificial nerve scaffolds. To date, nerve regeneration has been observed with the use of TEM or toluidine blue staining or indirectly with cross-sectional imaging. The nerve tracer used here specifically labels axons and aids in the direct observation of axons with fluorescent microscopy (33). This method is simple and proficient for demonstrating continuous nerve extension. Together with TEM imaging, anterograde tracers can confirm that myelinated axons are passing through grafts.

While artificial nerves can bridge only a few centimeters (2-3-4-5, 7-8, 22, 34), acellular nerves can bridge over 4.5 cm (35) because the ECM induces the fixation of Schwann cells in the graft. In the clinical setting, however, acellular nerves run the risk of rejection and contamination by cytotoxic substances that may damage cells. The decellularizing methods (freeze-thaw and detergent) have failed to achieve high decellularizing efficiency and low cytotoxicity (12-13, 17). We did not research the remaining antigenicity of the grafts produced with the different decellularizing method used in this study. Therefore, while it is possible that our method did not achieve complete decellularization, the grafts did not exhibit antigenicity at allotransplantation. Additionally, the nerve graft generated using our method maintained scaffolding and was not invaded by inflammatory cells. The results suggest that our nerve graft can serve as a functional graft and that the level of decellularization achieved by our method does not induce a rejection response or assimilation. It remains unknown how many remaining cells are needed to elicit a rejection response; this problem may be solved by counting the regenerated axons in the graft and examining the relationships between remaining cells and the regenerated axons. Lastly, this experimental system has only been tested in rats, and immunoreactions vary among species. In order to apply our novel graft to regenerative medicine in humans, heterologous transplantation will need to be considered. If proven safe for humans, our simple and harmless decellularizing method will be useful in clinical settings.