Introduction
Traumatic injuries to the nervous system affect millions of people worldwide every year. Severe injuries that cause substantial nerve defects may lead to life-long disability in patients, reduced quality of life, and heavy economic and social burdens [
1−
3]. Despite the intrinsic regenerative capacity of the adult mammalian peripheral nervous system (PNS), spontaneous nerve regeneration in PNS has been consistently linked to poor functional outcome if no therapeutic interference is adopted [
4]. Unlike PNS neurons, central nervous system (CNS) neurons cannot regenerate on their own after nerve injury because of the presence of inhibitors within myelin and the formation of a glial scar [
5,
6]. Consequently, clinical management of CNS injury (e.g., spinal cord injury) is still less promising, whereas clinical repair of peripheral nerve injuries has achieved considerable progress through concerted efforts for several hundred years [
7], particularly based on advancements in tissue engineering and regenerative medicine in recent decades.
A new field of neural tissue engineering has actively emerged, rapidly grown, and attracted the attention of researchers and clinicians. The development of tissue engineered nerve grafts allows the replacement of traditional neurorrhaphy by nerve grafting strategy to treat nerve injuries (mainly peripheral nerve injury). Several experimental and preclinical studies have reported the feasibility of using tissue engineered nerve grafts as an alternative to autologous nerve grafts to bridge peripheral nerve defects. Many excellent review articles that outline the abovementioned research activities have been published [
8–
14]. This review provides a brief summary of recent progress in the field of neural tissue engineering, especially in combination with the research experience of the author’s group. Some potential directions for new research are also suggested.
Innovative theory of tissue engineered nerves
Nerve regeneration and functional reconstruction is an extremely complex process that involves multifactorial mechanisms that need to be understood at the molecular, cellular, and tissue levels. Based on the accumulated knowledge on nerve regeneration, which includes long-term research work of the authors, an innovative theory on “biodegradable tissue engineered nerves” has been gradually established. A tissue engineered nerve is grafted into the injured nerve to bridge the nerve defect. In the present paper, the authors claim that an ideal nerve graft should have the following properties: (1) biodegradability at a controllable rate in the body; (2) biocompatibility with nerve tissues and cells; (3) reduced immunogenicity; (4) facilitates angiogenesis and metabolic exchange; (5) leads to less scar formation; (6) high availability of biomaterials for construction of the graft. The abovementioned basic requirements are generally accepted as guidelines for the development of tissue engineered nerve grafts. The essentials of our innovative theory about tissue engineered nerves include the following: (1) an ideal nerve graft with the abovementioned characteristics is prepared and used for bridging a nerve defect; (2) the nerve graft provides a favorable microenvironment, allows lesioned neurons in the proximal stump to grow neurites within the nerve graft, and guides the oriented regrowth of axons to reach target organs; and (3) along with gradual degradation and absorption of the nerve graft, the nerve defect is occupied by the regenerated nerves to achieve tissue regeneration and functional reconstruction (Fig. 1A−1D). The abovementioned theory has been described in the new textbook entitled “
Essential Biomaterials Science,” which is published by the University of Cambridge [
15].
Like other tissue engineered constructs, tissue engineered nerve grafts are typically composed of a neural scaffold (or template) added with molecular and/or cellular cues. In recent decades, great endeavors have been largely devoted to biomaterial selection, scaffold configuration, and scaffold fabrication for optimizing the generation of neural scaffolds. A variety of synthetic and natural biomaterials have been extensively used [
16], and even new inorganic-based materials with unique surface properties have started to attract the attention of researchers besides traditional biomaterials [
17−
23]. The scaffold configuration has been designed to consist of a nerve guidance conduit and a well-defined luminal structure; for example, physical fillers or multiple intraluminal channels in the scaffold lumen have been used [
24−
29]. Notably, for scaffold fabrication, nanotechnology has currently improved the performance of neural scaffolds by offering a nanoscale topography and mimicking the architecture of natural extracellular matrix (ECM) [
30,
31].
Moreover, as discussed in the following section, biological cues are often added to a neural scaffold during the construction of a typical nerve graft. After implantation, a neural scaffold guides axonal outgrowth and enhances Schwann cell migration, thereby allowing repair of peripheral nerve injury.
Translation of tissue engineered nerve grafts into clinic
As early as a decade ago, chitosan, a biodegradable natural polysaccharide, was observed for the first time to be quite suitable for neural tissue engineering applications because of its excellent biocompatibility with neural cells [
32]. The
in vivo biodegradation product of chitosan, chitooligosaccharide, was also shown to support neural cell adhesion and encourage neuronal differentiation and neurite outgrowth via upregulation of neurofilaments and N-cadherin [
33]. To decipher the mechanisms by which chitooligosaccharides promote nerve regeneration, we proposed that chitooligosaccharide stimulated the proliferation of Schwann cells via the miRNA-27a/FOXO1 axis [
34] (Fig. 1E).
Based on the abovementioned findings, chitosan-based nerve scaffolds were engineered using an injection molding method. The scaffolds were composed of a chitosan nerve guidance conduit filled with polyglycolic acid (PGA)/polylactic-co-glycolic acid (PLGA) filaments. The resulting nerve scaffolds were not only used to bridge a large peripheral nerve defect in large animal models, namely, the dog and the monkey (support cells are also used for monkeys) in experimental studies [
35,
36] (Fig. 2), but were also translated into the clinic and shown to return satisfactory outcomes [
37,
38] (Fig. 3). In addition, chitosan-based neural scaffolds have also been used to treat long-term delayed dog sciatic nerve defects [
39], and combined use of the resulting scaffolds with a growth factor (neurotrophin-3) was shown to enhance the repair of completely transected spinal cords in rats [
40].
The chitosan-based nerve scaffolds have been patented as inventions in China, and our cooperation with industrial enterprises has been helpful for the application of the nerve graft product in clinics. With approval from the Chinese State Food and Drug Administration (CFDA), the nerve graft products are currently being used in clinical trials, and a prospective randomized controlled multicenter study has been launched and is now underway in four Chinese public hospitals [
41]. A news article in the journal
Science reported this pioneer work and indicated that our research team was among the first in the world to develop new nerve grafts using chitosan and the first to use such grafts in the clinic setting [
42].
In addition, several absorbable nerve conduits and wraps, including NeuroGen
®, NeuroFlex
TM, NeuroMax
TM, NeuroWrap
TM, and NeuroMend
TM, have been approved for clinical repair of peripheral and cranial nerves by the US Food and Drug Administration (FDA) or Conformit Europe (CE) [
43,
44]. The abovementioned commercially available products are mainly composed of collagen or synthetic polymers.
Some new designs of tissue engineered nerve grafts are being applied in the repair of spinal cord injury. For example, a nerve graft comprising a linear-ordered complex of collagen scaffold and brain-derived neurotrophic factor was prepared and was shown to significantly promote functional recovery after completely transected spinal cord injury in canine. The successful animal experiment holds great promise in clinical applications for spinal cord injury paralysis or other movement disorders caused by neurological diseases [
45,
46].
Improved fabrication technologies for tissue engineered nerves
Traditional technology for processing polymers into porous scaffolds often depends on the use of common crosslinking agents, such as glutaraldehyde, which may impair the biocompatibility of biomaterials due to cytotoxicity and may affect the chemical structure of natural polymers. In our previous work, however, chitosan-based neural scaffolds were prepared through patented technology, thereby eliminating the need for crosslinking agent, ensuring non-toxicity and safety of nerve grafts in the body, and preserving the inherent properties of chitosan [
47]. A unique technology that controls the degradation rate of chitosan in the body by changing the degree of acetylation of chitosan has been invented. The technology helps solve the problem of matching the degradation rate of nerve grafts with the rate of nerve regeneration [
48]. A natural and low-toxicity crosslinking agent, genipin, is used to immobilize nerve growth factor (NGF), a neurotrophic factor, onto chitosan-based neural scaffolds to generate a novel nerve graft, which is beneficial for peripheral nerve repair [
49]. Moreover, joint use of multiple new technologies allows the generation of chitosan-based scaffolds with porous structure and good tensile strength. These scaffolds offer excellent biological properties to facilitate vessel growth and material exchange, while longitudinally aligned fibrous fillers in the lumen of chitosan-based scaffolds effectively guide the migration of Schwann cells and oriented axonal growth [
36]. Our group invent biomimetic silk fibroin-based never grafts, which have proven to exhibit even better regenerative performances than chitosan-based nerve grafts in some aspects [
50,
51].
In vivo interactions between neural scaffolds and host cells/tissue are bidirectional. The scaffolds could elicit cell and tissue responses, whereas host cells/tissue might change the local environment provided by the scaffold. Based on the abovementioned information, nanotechnological techniques, which are updated forms of traditional techniques, are being introduced to the fabrication of neural scaffolds because of nanoscale topography, which closely resembles the architecture of natural ECM and induces cell contact guidance in nerve regeneration [
31]. To date, nanostructures are largely incorporated into the common polymer-based neural scaffolds to improve bulk and surface properties. In many laboratories, diverse manufacturing methods, such as electrospinning [
52], self-assembly [
53], and surface micro/nanopatterning [
54], have been used to engineer novel neural scaffolds, and these efforts have achieved promising results for nerve repair.
Formation of regenerative microenvironment
In recent years, the establishment of a favorable regenerative microenvironment has been increasingly recognized as an essential element in neural tissue engineering. For a typical tissue engineered nerve graft, the neural scaffold is generally assumed to protect and guide axonal regrowth, while the added molecular and/or cellular cues are responsible for the formation of a regenerative microenvironment. Interaction of the scaffold with endogenous (from the injured site) or exogenous (from the added growth factors and/or support cells) biological cues also influences the regenerative microenvironment. In an old-fashioned perspective, a tissue engineering scaffold is merely an inert structure that is temporarily used to assist in the construction of inanimate objects. From an innovative perspective, however, the scaffold has to become an active vehicle for biological cues to participate in the regulation of mechanical and molecular signals [
55−
57]. Accordingly, the interaction between a neural scaffold and biological cues is a complex bidirectional and multidimensional process; the scaffold stimulates relevant molecular and cellular responses, while neural and non-neural cells affect the biocompatibility of the scaffold, with both contributing to the formation of a favorable regenerative microenvironment that should be close to the native microenvironment within PNS and CNS.
Support cells that are added to a tissue engineered nerve graft include Schwann cells, mononuclear cells, sheath cells, embryonic stem cells, neural stem cells, marrow mesenchymal stem cells (MSCs), skin-derived precursor cells, adipose-derived stem cells, olfactory ensheathing cells, and induced pluripotent stem cells (iPS). The abovementioned cells have supportive roles in nerve regeneration through (1) cell replacement in a direct or indirect (through differentiation into neural cells) manner, (2) release of growth factors, and (3) production of ECM.
The molecular cues within a tissue engineered nerve graft are furnished by added growth factors, which include NGF, brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3), glial cell line-derived neurotrophic factor (GDNT), ciliary neurotrophic factor (CNTF), and fibroblast growth factors (FGFs). Despite many studies that have investigated the use of growth factors for neural tissue engineering, the implications of growth factors in nerve regeneration, including the underlying mechanisms, are still quite unclear and even somewhat controversial. Moreover, some traditional Chinese medicine ingredients, such as
Achyranthes bidentata polypeptides, have been used by our group as an additive to nerve grafts and have shown the capability for nerve repair to a certain degree [
58−
62]. Now, research attention regarding the use of growth factors for nerve regeneration is turned to the development of sustained delivery strategies, in which nanoscale, gene-based, and cell-based delivery systems seem to merit more concern [
63].
After all, biological cues encapsulated in a tissue engineered nerve graft enable the introduction of molecular and/or cellular therapies into neural tissue engineering. To date, however, cellular and molecular therapies directed at peripheral nerve repair have not yet reached beyond the laboratory stage, and their applications in the clinical setting have been beset with numerous challenges, such as the type and quantity of cells or factors, their delivery, cell viability or factor activity, cell phenotypic stability, timing of treatment, regulatory issues, and high costs [
64,
65]. Therefore, new approaches must be developed as alternatives to a simple and direct addition of support cells or growth factors to a nerve graft. Mimicking of the native ECM in nervous system within a nerve graft is a promising choice among the alternatives. The ECM is composed of diverse biological macromolecules, such as proteins, glycoproteins, and glycosaminoglycans, and contains sufficient biological cues to regulate cell phenotype and function in biological processes, including nerve regeneration [
66]. In addition, the biomaterial-based neural scaffold usually interacts with biological cues through the ECM.
In fact, ECM-modified nerve grafts have been previously engineered using acellular allogeneic and xenogeneic nerve (or non-nerve) tissue and are often called acellular nerve grafts, in which the ECM is derived from nerve or non-nerve tissue. To overcome drawbacks linked to the tissue-derived ECM, such as tissue scarcity, pathogen transfer, insufficient mechanical properties, and uncontrollable degradation kinetics [
67], the development of cell-derived ECM to replace tissue-derived ECM has been initiated. For example, a chitosan/silk fibroin-based, Schwann cell-derived ECM modified nerve grafts have been used to bridge rat sciatic nerve defects [68]. A series of morphological and functional examinations show that the Schwann cell-derived ECM modified nerve grafts are close to the nerve tissue-derived ECM modified nerve grafts in terms of performance. Meanwhile, safety evaluation procedures, such as blood routine examination, biochemical examination, thrombosis assay, immunological and tumor marker detection, and the histopathological examination of multiple organs, suggest that the cell-derived ECM modified nerve grafts eliminate the possibility of pathogen transfer and exhibit tunable physiochemical properties as a result of suitable processing and reconstruction (Fig. 4).
Concluding remarks and perspectives
In the past decades, tissue engineered nerve grafts have been successfully generated and gradually perfected. Neural tissue engineering, however, is a still growing field that remains to be applied to the clinical setting. Many attempts are being made to push neural tissue engineering field to higher levels, and some suggestions for research directions are proposed below.
The development of newer and better biomaterials has always been the research focus of tissue engineering, and neural tissue engineering is not an exception. In more recent years, various new biomaterials have been used in engineering nerve grafts to treat spinal cord injury and peripheral nerve injury. For example, BDNF is bound to a newly prepared biomaterial, linear-ordered collagen scaffold (LOCS) fibers, by tagging a collagen binding domain (CBD). CBD is a peptide TKKTLRT deduced from collagenase. The resulting complex (LOCS+ CBD-BDNF) exhibits striking therapeutic effects on completely transected spinal cord in rats and dogs [
45,
69]. Consequently, the search for more suitable biomaterials will be consistently performed, and the
in vivo biocompatibility of biomaterials will be seriously examined to increase the bioactivity of neural scaffolds to allow positive interactions with biological cues.
To identify novel molecular targets for the treatment of CNS and PNS injuries, the use of microRNAs in nerve regeneration has begun to attract increasing interest. MicroRNAs (miRNAs, miRs) are non-coding, single-stranded, 21 bp to 23 bp-long RNAs that function in posttranscriptional regulation of gene expression. The importance of miRNAs for neural development and degeneration has been delineated [
70,
71], and the involvement of miRNAs in nerve injury and regeneration is being actively investigated [
72]. Our group has noted that many miRNAs, including miR-21, miR-221/222, miR-182, miR-9, and let-7, are differentially expressed in injured nerve tissue after peripheral nerve injury. The abovementioned miRNAs also have critical functions in neuronal survival, neurite outgrowth, and phenotype modulation of Schwann cells by targeting (negative regulation) respective target mRNAs [
73−
80]. To apply miRNAs in peripheral nerve regeneration, a silicone conduit added with miR-9 agomir (steroid-conjugated mimic) or let-7d inhibitor was implanted in rats to bridge the sciatic nerve defect. The
in vivo results show that overexpression of miR-9 [
79] or reduced expression of let-7d [
78] reduced or enhanced Schwann cell migration and axon outgrowth within a regenerative microenvironment, respectively, thereby suggesting the potential of miRNA-based strategies for clinical management of nerve injury (Fig. 5).
The safety and effectiveness of tissue engineered nerve grafts have to be evaluated before their translation into the clinic setting. Given the constantly updated knowledge on biomaterials and formulation of increasingly rigorous safety standards for clinically usable tissue engineered products, safety evaluation for biomaterials may be based on a new paradigm related to molecular expression modulation. Evaluation of effectiveness may be conducted at the gene, cellular, and tissue levels and may be based on the outlook of molecular expression and phenotype modulation.
However, the search and development of support cells and growth factors for neural tissue engineering applications may still be worth studying further. Since various cellular and molecular cues evoke distinct effects on the performance of nerve grafts, elucidating the orchestration of different cue-induced effects via comparative studies is very important. Meanwhile, new approaches to address the obstacles associated with translation of molecular and cellular therapy in a clinical setting should be explored [
81−
84].
Given the abovementioned considerations, perhaps a new perspective is required to apply biological cues in neural tissue engineering. The material, structure, and architecture of nerve grafts must be designed to closely resemble or mimic the native ECM. Studies must focus on the interaction of the nerve grafts with host cells/tissue is upon entering the body. In particular, the implanted nerve grafts may affect the immune system and hematopoietic system in the body and can interfere with processes, such as apoptosis and angiogenesis, during nerve regeneration and functional restoration. ECM is secreted from the cells and tissue to outside of the cell and is subsequently distributed on the cell surface or on the space between cells, thereby leading to tissue-specificity. The ECM consists of complex networks that support the organization and structure of cells and tissue and affect survival, proliferation, adhesion, differentiation, and other cell functions. As a consequence, ECM-modified nerve grafts hold great promise as alternatives to molecular and cellular therapy in the management of nerve injury. Future research must be directed to the development of ECM-modified nerve grafts.
Overall, the ultimate goal of neural tissue engineering is to facilitate the reconstruction of normal nerve tissue with the same structure and functions as those in a healthy individual after traumatic injuries to the nervous system. Tissue engineering is a vigorous field that is full of chances and challenges.
Higher Education Press and Springer-Verlag Berlin Heidelberg