Modern Trends for Peripheral Nerve Repair and Regeneration: Beyond the Hollow Nerve Guidance Conduit

Peripheral nerve repair and regeneration remains among the greatest challenges in tissue engineering and regenerative medicine. Even though peripheral nerve injuries (PNIs) are capable of some degree of regeneration, frail recovery is seen even when the best microsurgical technique is applied. PNIs are known to be very incapacitating for the patient, due to the deprivation of motor and sensory abilities. Since there is no optimal solution for tackling this problem up to this day, the evolution in the field is constant, with innovative designs of advanced nerve guidance conduits (NGCs) being reported every day. As a basic concept, a NGC should act as a physical barrier from the external environment, concomitantly acting as physical guidance for the regenerative axons across the gap lesion. NGCs should also be able to retain the naturally released nerve growth factors secreted by the damaged nerve stumps, as well as reducing the invasion of scar tissue-forming fibroblasts to the injury site. Based on the neurobiological knowledge related to the events that succeed after a nerve injury, neuronal subsistence is subjected to the existence of an ideal environment of growth factors, hormones, cytokines, and extracellular matrix (ECM) factors. Therefore, it is known that multifunctional NGCs fabricated through combinatorial approaches are needed to improve the functional and clinical outcomes after PNIs. The present work overviews the current reports dealing with the several features that can be used to improve peripheral nerve regeneration (PNR), ranging from the simple use of hollow NGCs to tissue engineered intraluminal fillers, or to even more advanced strategies, comprising the molecular and gene therapies as well as cell-based therapies.

Tissue engineering (TE) is a vital instrument in the field of regenerative medicine (RM), which has been the subject of dynamic scientific research in the last decades (Langer and Vacanti, 1993 ). The two terms have been referred to as TERM altogether (Furth et al., 2007 ). TERM tactics exist in a variety of human tissues and organs. The pivotal point is that these strategies bring new therapeutic possibilities, not only to general population, but more specifically to young patients and professional athletes and sportsman, allowing their reintegration and rebuilding of biological functions. Overall, TERM strategies assist in the re-establishment, support, regeneration, or replacement of injured tissues and organs (Furth et al., 2007 ). Traumatic offense, oncological resection, congenital malformations, or progressively degenerative diseases result on tissues and organs that need replacement. TERM strategies propose to use the ideologies of several fields, such as cell relocation, material science, nanotechnology and bioengineering to fabricate biological replacements of the damaged organs, eliminating the need and the wait of a transplant. Since its start, TE and now TERM, have been relying on three pillars: (i) scaffolds, (ii) cells, and (iii) growth factors (GF) (Langer and Vacanti, 1993 ). The basic strategies used in TERM approaches, applied to PNR, can be seen in Figure 3 . Figure 3 The classic TE model, where a triad of components interact with each other: scaffolds, cells and biological molecules. Overall, it includes the combination of living cells isolated from the patient donor tissue (nerve) and expanded in culture, with a natural, synthetic, or bioartificial matrix or scaffold. Together with the addition of biological stimuli such as growth factors, a 3D living construct that is structurally, mechanically, and functionally equal to a nerve tissue. The engineered construct can be implanted in the patient in order to restore the damaged tissue. To achieve the ambitioned regeneration, several strategies can be used. One of them is the use of acellular matrices, composed only of smart biomaterials, which can react to the fluctuations in the environment when one of its property changes by the exterior conditions. They can either be prepared by manufacturing artificial scaffolds or by preparing acellular materials after decellularization of tissues (Moore et al., 2011 ). Cellular components can also be added and advanced medicinal therapeutic products (ATMPS) are obtained (Hu et al., 2016 ). For instance, Hu et al. ( 2016 ) used a 3D-printing technology to manufacture a bio-conduit. It consisted of a cryo-polymerized gelatin gel to which was further added adipose-derived stem cells (ADSCs). In a different approach, Dai et al. co-cultured Schwann cells and dental pulp stem cells on poly(d,l-lactide) (PLA) NGCs and assessed their effectiveness for restoring a 15 mm long critical gap defect in the rat sciatic nerve. In both strategies, acellular or cellular materials, the biomaterial transformed in a scaffold must provide mechanical sustenance and proper features that contribute to tissue regeneration and formation, as the seeded cellular components cells proliferate and acquire the right morphology (Atala, 2004 ; Furth et al., 2007 ). Furthermore, nerves are exposed to several types of stress placed upon them, such as tensile, shear and compressive forces, which arise from postures or movements. Therefore, the scaffolds used in attempt to regenerate nerves must withstand such forces (Topp and Boyd, 2006 ). Materials used in this scope must have mechanical resistance to hold the regenerating nerve, however matching the mechanical and physical properties of the native nerve. Allied to that, properties such as tensile strength, suturability, and appropriate swelling and degradation must meet the required necessities (Nectow et al., 2012 ). Another TERM strategy focuses on the use of cells alone, without the use of biomaterials, in which cells organize by self-organization and self-assembly (Shimomura et al., 2018 ). Altogether, regardless of the strategy followed, TERM aims at the creation of regenerated tissue for the rebuilding of the damaged parts in the body, resulting in substantial therapeutic benefits for patients for whom there are not currently any clinically effective therapies. An immense body of research has been published regarding PNI and PNR (Mobini et al., 2017 ; Tajdaran et al., 2019 ; Tomita et al., 2019 ). However, regrettably, TERM strategies are not yet delivering significant progress in terms of clinical outcomes and commercialization, as there has not been a substantial translation from the bench to the clinics. This applies specifically to the case of PNR. Despite intensive research and plentiful approaches that have been published, none have achieved the desired results. The First Concept: Regeneration Within a Hollow Conduit The application of hollow conduits for nerve repair was

Topographical cues are influential controllers of the amount and orientation of neurite elongation (Thomson et al., 2017 ). One way to achieve an anisotropic 3D matrix is with controlled freeze-dried technique. More specifically, freeze-casting is a manufacturing method that allows the production of porous matrices containing a controlled and highly hierarchical architecture, which can be done with a variety of polymers (Scotti and Dunand, 2018 ). Since this is a rather well-known and simple approach, it has been widely applied to PNR. Singh et al. ( 2019 ) developed an antioxidant polyurethane NGC filled with aligned chitosan-gelatin cryogel filler. Only tested in vitro , neonatal DRGs and Schwann cells were seeded on the aligned scaffolds, which resulted in earlier migration and alignment to form “Bands of Bungner”-like structures. Following a stage-wise strategy, Huang L. et al. ( 2018 ) produced a directionally freezing oriented collagen-chitosan filler in a porous electrospun PCL conduit. The NGC was optimized to have blend of collagen/chitosan (1:1) as filler and a wall thickness of 400 μm. Such features allowed the NGC to shield growing axons from compression forces while, at the same time adding enough space for regenerating nerves. Furthermore, the anisotropic inner structure allowed Schwann cells and axons from DRGs to extend and migrate parallelly, in a significantly higher rate as compared to a isotropic substrate. Manoukian et al. ( 2019 ) took it a step further, and included interconnected longitudinally-aligned pores in a biodegradable chitosan NGC reinforced with drug-loaded halloysite nanotubes. The chitosan conduit allowed the sustained delivery of 4-Aminopyridine, a potassium-channel blocker, that has the capacity to modulate prolonged nerve action potentials and strongly promoted neurotransmitters release. The aligned and interconnected pores allowed for migration of Schwann cells, both longitudinally and horizontally. Furthermore, 4-Aminopyridine delivery resulted in substantial, dose-dependent upregulation of pivotal growth factors, namely nerve growth factor (NGF), myelin protein zero, and brain derived neurotrophic factor (BDNF). Also, it promoted nerve impulse conduction, being an attractive strategy for nerve repair and regeneration.

Unidirectionally aligned micro- or nano-filaments inside NGCs support axonal growth cones in identifying and physiologically answering to the surrounding environment and stimulating them to follow the path to the distal stump. Furthermore, such structures can allow increasing the available surface area to volume ratio, thus potentially enhancing cellular adhesion and proliferation. In one of the first studies of this kind, Matsumoto et al. ( 2000 ) added laminin coated fibers as NGC luminal filler, to bridge a 80 mm gap in a canine peroneal nerve model, however with poor outcomes. Quigley et al. ( 2013 ) developed an advanced and multi-modal conduit where aligned PLGA fibers are present in the lumen of a knitted PLA sheath coated with electrospun PLA nanofibers. To provide further support, the PLGA fibers are standing on an alginate hydrogel impregnated with several NTFs. The aligned PLGA fibers were remarkably helpful in guiding the growing axons. The fibers formulations was precisely chosen to encourage either axonal outgrowth or Schwann cell growth (75:25 for axons; 85:15 for Schwann cells). Furthermore, axonal outgrowths were found inside and around the NGC and most of the regenerated fibers were positively myelinated. Such type of strategy has been explored in numerous studies, with variable outcomes (Yoshii and Oka, 2001 ; Yoshii et al., 2003 ; Kim et al., 2008 ). What they have in common is that the addition of such filaments clearly extends the regeneration limits when compared to hollow NGCs. Later, studies were focused on the more suitable diameter of fibers as well as the density of fibers inside the NGC, in order to achieve the best possible outcomes (Jiang et al., 2014 ; Xie et al., 2014 ; Jia et al., 2019 ). Regarding the first parameter, the use of fibers in the “nanometer” scale proved to provide the best outcomes, elevating electrospinning as the technique of choice (Jiang et al., 2014 ; Jia et al., 2019 ). Regarding the density of the fibers, lower densities increase the ability to bridge a critical nerve gap (Ngo et al., 2003 ). Similar to what was found in the case of hydrogels, higher densities and occlusion of the lumen tend to lead to inhibition of regenerating nerves.

In terms of micro- or nano-patterned surfaces for PNR, there are two main approaches that can be used: (i) micro- or nano-patterned two dimensional surfaces that can be rolled up to form a conduit, and (ii) patterned micro or nanomaterials to be used as a filler material in NGCs. A vast selection of studies has used microgrooves to explore the most suitable sizes to encourage axon elongation, having in consideration the cell diameter (Huang C. et al., 2015 ; Davis et al., 2018 ; Li G. et al., 2018 ). Investigational data advocates that thin microgrooves (5–10 μm) confines the growth of cells and axon elongation in comparison to broader grooves (20–60 μm). On the other hand, the narrower microgrooves were found to improve axonal alignment, diminish the number of axon branches per cell, and decrease incorrect distal re-connection. Inclusively, longitudinal nanogrooves (200 nm) sustained functional recovery of sciatic nerves in rats (Huang C. et al., 2015 ) and enhanced the growth cone attachment and proliferation (Park et al., 2013 ). Mobasseri et al. ( 2015 ) modified the surface topography of PCL/PLA blending films to improve cellular guiding. The results obtained with the sloped walls grooved conduit with 70 μm wall thickness was similar autologous nerve graft, in a gap model of 10 mm in the rat sciatic nerve. Huang Y.-A. et al. ( 2018 ) reported that they were able to exactly place the neuronal cell body and control the direction of axonal growth by merging surface topography and a cell placement device technologies, creating the prospect of engineering complex tissues. Furthermore, they were able to create an on-site axotomy, studying how the topography can influence the initial regeneration of injured axons. Promising results obtained using anisotropic topographical cues as luminal fillers can be seen in Figure 6 . However, the incorporation of luminal fillers, regardless of their kind, is not always successful, as a recent study by Saltzman et al. demonstrated (Saltzman et al., 2018 ). This study aimed to compare the performance of a PGA conduit containing collagen fibers within the conduit, a hollow collagen conduit, and a nerve autograft. The results demonstrated that the nerve repair using the autologous nerve graft resulted in greater motor nerve salvage, followed by the hollow collagen conduit and only then the conduit containing luminal filler. Therefore, a wise selection of the luminal filler must be made since each strategy has its pros and cons. Figure 6 Anisotropic guiding cues have been successfully produced as NGCs luminal fillers. (AI) Transverse and longitudinal Micro-CT sections of the hollow conduit; (AII) Transverse and longitudinal Micro-CT sections of the oriented chitosan-gelatin cryogel as luminal filler; (BI) DRG explants seeded on the longitudinal sections of the directionally orientated collagen-chitosan filler, where neurites align in the matrix; (BII) 3D reconstruction of axonal regeneration and Schwann cell migration on the orientated collagen-chitosan filler; (CI) Schematic drawing of the peripheral nerve structure; (CII) SEM micrograph of the produced silk fibroin NGC fabricated incorporating microchannels, which looks like the depicted schematic; Scale bar 200 μm; (DI) DRG explants seeded in 0.25% volume of the anisogel, presenting isotropic structure, in which neurites do not orient; (DII) DRGs explants seeded in 1% anisogel in which neurites decide to orient; (E) Representative images of DRG explants neurites cultured on random patterns achieved with nanoimprinting lithography with metallic stampers made of three different spacings: (EI) on a flat surface; (EII) On a Blu-Ray disc spacing; (EIII) On a digital video disc spacing; and (EIV) On a compact disc spacing. Scale bar: 200 μm. Figures have been reprinted and adapted from: (A) (Singh et al., 2019 ), (B) (Huang L. et al., 2018 ), (C) (Dinis et al., 2015 ), (D) (Rose et al., 2017 ), and (E) (Huang L. et al., 2018 ). “The authors of this review, have, however, a criticism on how micro- and nano- scaled features are presented in the literature. Although measurements are identified in micrometers (μm), they are referred to as nanopatterning or nanogrooves. There is very little literature where the nanometer-sized features are explicit in nanometers (nm). Therefore, future works should consider to be more precise in terms of scaling nomenclature.”

Apart from the use of suitable NGCs and fillers, the conception of a more biologically appealing milieu is of high importance in PNR. Nerve GFs are molecules that are naturally released in the processes of injury and result in enhanced nerve regeneration. Therefore, it is important to mimic their release, which is vital for nerve growth, differentiation and expansion (Tajdaran et al., 2019 ). However, the artificial administration of GFs as a therapy is problematic duty to accomplish because of their high biological activity, which obliges to administer extremely small doses. Pleiotropic effects and short biological half-life are also other common constraints (Pfister et al., 2007 ). The fiasco of GF delivery may be credited to unsuitable release kinetics, as several delivery systems unveil an elevated initial burst release (Madduri et al., 2010b ). In an attempt to progress in terms of releasing profile, delivery systems that allow adjusting the release kinetics are being investigated (Pfister et al., 2007 ). The use of biodegradable biomaterials is advantageous in this specific case, since they can act as vehicles for GF delivery, allowing to manipulate specific biomaterials parameters to attain the desired rate of sustained release. Physical crosslinking (Madduri et al., 2010a ), chemical immobilization (Aebischer et al., 1989 ), polymer coating (Madduri et al., 2010a ), and nanoparticles (Giannaccini et al., 2017 ) are some of the strategies that are being used ( Figure 7 ). Figure 7 Different strategies for incorporation and delivery of GFs from NGCs. One of the simplest approaches is based on simply blending the NTFs on the polymer, with or without further crosslinking of the polymer. Microspheres containing NTFs can also be blended in the polymer. At the surface, NTFs can be found just after an adsorption process or conjugated with other molecules for stronger entrapment or covalent links. When considering the delivery of NTFs from the lumen, several approaches can be followed, such as using engineered cells, nanofibers or hydrogels capable of loading and releasing the NTFs. The various GFs have diverse actions in the nerve, as they can enhance functional regeneration, support axonal elongation, and Schwann cell migration, by means of acting as neuroprotective components through receptor-mediated activation of specific pathways. GFs regularly used to aid in PNR can be seen in Table 1 . NTFs normally used to improve nerve regeneration primarily belong to three separate groups (Deister and Schmidt, 2006 ): the neurotrophins, the glial-cell-line-derived neurotrophic factor family ligands, and the neuropoietic cytokines. The first group, neurotrophins, include NGF, BDNF, neurotrophin-3 (NT3) and neurotrophin-4 (NT4) (Peleshok and Saragovi, 2006 ). The second group includes the very glial cell derived neurotrophic factor (GDNF) and the ciliary neurotrophic factor (CNTF) (Rickert et al., 2014 ). Although they belong to distinct families, all above-mentioned GFs have affinity to different Receptor Tyrosine Kinase (RTKs) on cells, which are trans-membrane proteins that will activate a cascade of events after substrate binding, activating determined cellular responses (Boyd and Gordon, 2003 ). Furthermore, RTKs are known to be retrograde transported from the extremities to cell body, where they will once again activate transcription and translation of proteins, initiating signaling pathways to boost neuronal outgrowth (Hausott and Klimaschewski, 2016 ). The GFs of the neurotrophin family are structurally and functionally related peptides with active functions in both CNS and PNS, in particular, they mediate the effective survival and differentiation of several neuronal related cell populations. Table 1 The use of GFs and their effect on CNS and PNS (Terenghi, 1999 ). Neurotrophic factor Neural response Receptor and action NGF Sensory neuron survival Sensory neuron outgrowth Spinal cord regeneration TrkA/p75, receptors expressed in sympathetic/peripheral sensory neurons (Schwann cells upregulate NGF and p75 in response to PNS injury); Involved in survival signaling and neurite outgrowth; GDNF Motor neuron survival Motor neuron outgrowth Sensory neuron survival GFRa/Ret, receptors expressed in sensory/motor neurons, GDNF primarily produced by Schwann cells in development and plays an important role in sensory regeneration; BDNF Motor neuron survival Motor neuron outgrowth Sensory neuron outgrowth TrkB, BDNF mRNA upregulated in distal nerve stump after sciatic nerve transection; positive modulation of peripheral nerve myelination; CNTF Motor neuron survival Motor neuron outgrowth Spinal cord regeneration CNTFR, present in peripheral nerves and myelinating Schwann cells; promotes survival of motor neurons; NT-3 Motor neuron survival Motor neuron outgrowth Sensory neuron outgrowth Spinal cord regeneration TrkC, NT-3 mRNA downregulated in distal nerve stump after sciatic nerve transection; negative modulation of pe

NT-3 has overlapping neurotrophic activity with NGF, however studies suggest it has an broader specificity as compared to NGF and BDNF (Maisonpierre et al., 1990 ). Studies show the reinnervation of motor target tissues is also related to the presence of NT-3 and its physiological effects (Sterne et al., 1997 ). Few reports have been published regarding this growth factor. One of the most recent focuses on the use of NT-3 to overcome Charcot–Marie–Tooth neuropathies, which are a heterogeneous group of peripheral nerve disorders (Sahenk et al., 2014 ). In these disorders, Schwann cells are affected. However, continued treatment with NT-3 is not feasible due to its short half-life and nonexistence in the market. In a way to overcome this problem, Sahenk et al. ( 2014 ) hypothesized that the delivery of NT-3 via gene therapy with an adeno-associated virus. With such strategy and therapy, quantifiable NT-3 amounts were found in blood, in quantities enough to significantly promote nerve regeneration, which was verified by histopathology, and electrophysiology (Sahenk et al., 2014 ). Related to the levels of NT-3 after injury, it has been reported in literature that it's levels are keept unchanged after nerve transection (Omura et al., 2005 ). Interestingly, it has been reported that neurotrophin-4 (NT-4), a prominent indicator for neuron survival, was first found to be up-regulated and then down-regulated in the injured sciatic nerve stumps. Therefore, up-regulation of NT-4 after nerve crush injury might contribute to nerve regeneration (Zhang et al., 2019 ).

An advanced and upgraded strategy can be considered using NGCs and GFs, which is the use of a GF gradient along with a NGC device. GFs gradient at the surface of the conduit aims at promoting contact guidance along the structure, maintaining the controlled release properties (Tang et al., 2013 ). This technique is based on the fact that the growth cone at the tip of the axon has path-finding ability, aiming at crossing the gap in a precise direction, correctly reaching the distal target. At the same time which is increasing the speed of such phenomena, what is called growth cone chemotaxis (Mai et al., 2009 ). Axons extend in search of their appropriate targets, often over long distances with the assistance of growth cones detecting and following molecular gradients (Mortimer et al., 2008 ). Many studies have been focusing on this strategy of immobilized concentration gradients, either on films or NGCs, with different approaches (Moore et al., 2006 ; Yu et al., 2010 ; Lin et al., 2011 ; Tang et al., 2013 ). More recently, Uz et al. ( 2017 ) developed a PLLA porous film, that besides having longitudinal surface micropatterns, also presents a gradient of NGF. As a result, the existent surface gradient of NGF contributed to an early fast release from the surface film and enabled oriented neurite outgrowth of PC12 along with the longitudinal micropatterns. With this double strategy, the authors could control the concurrent precise release of neurotrophic factors as well as the directional neurite outgrowth in PC12 cells. In another study (Sun et al., 2019 ), Sun et al. introduced exogenous cells secreting GFs capable of spatial distribution along the conduit. Instead of using a normal cellular density and distribution along the conduit, the authors used encapsulated bone marrow MSCs (BMSCs) capable of producing high levels of BDNF. In fact, comparing with standard cell lumen injection, the conduits encapsulated with stem cells presented dissimilar cell attachment and distribution after 6 weeks, in vivo . Such a construct indorsed Schwann cell relocation from the center to the distal end. In another interesting and combinatorial approach, Chang et al. ( 2017 ) developed a natural biodegradable multi-channeled scaffold composed by oriented electrospun nanofibers containing a neurotrophic gradient. For the gradient, the authors followed two strategies: the NGF was simply blended with gelatin and BDNF was encapsulated in nanoparticles further embedded in gelatin hydrogel. The gelatin scaffold was divided into five regions from A1–A5 (low concentration to high concentration, from proximal to distal site). Their strategy promoted intense nerve regeneration in critical sciatic nerve defect in a rabbit model. Our group has also been developing a GF gradient strategy, by fabricating silk fibroin NGCs containing GF gradients in its walls, as can be seen in Figure 8 . However, unlike Tang et al. ( 2013 ), who simply used soaking and coating techniques in the silk fibroin conduits, our group's strategy is based on gradually entrapping the GFs in the conduit's wall. This is achieved by enzymatically crosslinking the polymer, mediated by a horseradish peroxidase/hydrogen peroxide reaction. Figure 8 Proof of concept regarding the fabrication of a silk fibroin NGCs incorporating a gradient of GFs. (A) Schematic of a NGC incorporating two different concentrations of GFs in the wall of the conduit. In principle, the gradient of GFs increases from proximal to the distal, therefore attracting the growing axons to reach their distal target. (B) Stereomicrograph of a silk fibroin NGC presenting a gradient along the walls. The orange color represents the chosen Concentration 1, followed by the white color, representing Concentration 2. As it can be assessed, there is no separation in the conduit between the different concentrations, as the conduit is totally uniform. Scale bar: 1,000 μm.

In a work developed by Ko et al. ( 2018 ), hepatocyte growth factor (HGF) has been recently found to be up-regulated, resulting in a higher expression of the referred HGF after a PNI, both at the injury and distal sites. The authors tested this specific growth factor, with known angiogenic activity and anti-inflammatory activity (Nakamura and Mizuno, 2010 ) in a model of PNI in mice. After the injury, not only HGF was highly expressed, but its receptor, c-met, was also found to be upregulated only in Schwann cells. In addition, exogenous administration of HGF at the injury site led to an increase of the myelin thickness and axon diameter in injured nerves. To further prove this fact, when mice were treated with a c-met inhibitor, the opposite happened, as myelin thickness and axon regrowth were diminished, indicating that the positive effect of HGF was hindered. In line with this findings, Boldyreva et al. ( 2018 ) also studied the effect of HGF in PNR. Gene therapy with HGF-bearing plasmid (pC4W-hHGF) led to the repair of nerve morphometry and functional recovery comparable to the autograft (positive control). Moreover, in HGF-treated mice, histological evaluation showed a three-fold intensification in axon quantification in the distal nerve end, when compared to control, indicating great potential in keeping a healthy distal target. Besides confirming the potential of the application of HGF in cases of PNI, gene-therapy itself proved to be effective and advantageous. In addition to confirmed beneficial effects in PNR, literature has revealed that this growth factor applies beneficial effects in motor, sensory, and parasympathetic neurons. Furthermore, the beneficial effects of this growth factor can be considered mitogenic, morphogenic, angiogenic, antiapoptotic, antifibrotic, and anti-inflammatory, which can act upon numerous tissues (Imamura and Matsumoto, 2017 ).

The importance of cells as crucial actors in the nerve regenerative process is evident. Consequently, cell-based therapy became a significant and evidenced intervention which improves the functional clinical outcome after nerve injury (Yousefi et al., 2019 ). Many approaches can be followed, from primary Schwann cells, neural stem cells, embryonic stem cells, bone marrow stromal cells (BMSCs) and TE cellular components. Some of the most recent and relevant published reports, where cell-therapies are an important tool for PNR, regarding the use of several cell types can be seen in Table 2 . Table 2 Cell-based therapies for PNR. Cell type Defect size, location, animal model Outcomes References Schwann cells 15 mm, sciatic nerve, rat Schwann cells overexpressing FGF-2 in a chitosan conduit supported the early regenerative process; Meyer et al., 2016a 5 mm, bilateral cavernous nerves, rat Simple Schwann cells or GDNF-transduced Schwann cells grafts led to 75 % and 94 % success rate, respectively, compared to the 25 % of autografts; May et al., 2016 In vitro , micro-patterned surface fabricated by laser ablation with NGF When co-culturing with Schwann cells, NSCs differentiated into neuronal cells with robust expression of βIII tubulin and microtubule-associated protein-2; Yeh et al., 2017 5 mm, laryngeal nerve, rat Laminin-chitosan-PLGA NGC combined with Schwann and NSC promoted significantly higher nerve regeneration when compared to acellular grafts; Li Y. et al., 2018 In vitro co-culture of Schwann cells and DRGs Schwann cells in co-culture with DRGs promoted longer neurite extension and formation of myelin around DRG neurites; Wu et al., 2018 Bone Marrow stem cells (BMSCs) 20 mm autograft, sciatic nerve, rat BMSCs can differentiate into Schwann cell-like phenotype and myelinate axons, also expressing neuronal markers such as GFAP and S100; Keilhoff and Fansa, 2011 10 mm, sciatic nerve, rat Tropomyosin receptor kinase A overexpression enhanced the efficacy of BMSCs on PNR and improved functional recover; Zheng et al., 2017 Contusion injury of the spinal cord, rat Intravenous delivered BMSCs exosomes tend to migrate into the injury site, where they exert their beneficial effects; Lankford et al., 2018 Undifferentiated adipose derived stem cells (ADSCs) 10 mm, sciatic nerve, rat Number and diameter of the myelinated fibers were significantly higher in the case of silicone NGC loaded with ADSCs; Santiago et al., 2009 6 mm, sciatic nerve, rat Decreased muscular atrophy and enhanced PNR when PCL conduits were loaded with ADSCs; Mohammadi et al., 2013b Blunted injury, sciatic nerve, mouse Transplanted ADSCs did not differentiate into Schwann cells but promoted PNR, since they encouraged axon regeneration, formation of myelin and restoration of denervated muscle atrophy; Sowa et al., 2016 15 mm, sciatic nerve, rat ADSCS injected directly in the muscles connected to the damaged nerve were found to have increased presence of IL−10 and Ki67, which helped in delaying the onset of muscular atrophy; Schilling et al., 2019 Differentiated adipose derived stem cells (ADSCs) 10 mm, sciatic nerve, rat Schwann cell-like differentiated ADSCs were found to express neurotrophic factors, namely NGF, BDNF, glial-GDNF, and NT4. The same study also reported an increase of anti-apoptotic m-RNA of Bcl-2 as well as a decrease of pro-apoptotic m-RNA Bax and caspase-3, which lead to a neuroprotective state; Reid et al., 2011 Human umbilical-cord stem cells (HUCMSCs) 10 mm, sciatic nerve, rat HUCMSCs increased the expression of neurotrophic and angiogenic factors, which led to a more favorable environment for nerve regeneration; Shalaby et al., 2017 10 mm, sciatic nerve, rat Wharton jelly-derived stem cells, in addition to an injection of dexamethasone resulted in advanced regeneration compared to the autograft; Moattari et al., 2018 Olfactory ensheathing cells (OECs) 8 mm, sciatic nerve, rat PLLA NGC seeded with OEC encouraged nerve regeneration similarly to the autograft group; Kabiri et al., 2015 In vitro , to test how OECs promote neurite outgrowth of cortical neurons in an inhibitory scar-like culture model It was found that OECs enhanced neurite elongation through direct contact and alignment of neuronal and OEC processes in scar-like cultures; Khankan et al., 2015 5 mm, facial nerve, rat OECs transplanted within the NGC improved regeneration of transected facial nerve, with large numbers of myelinated nerve fibers, crude fibers, larger myelin thickness and volume in the transplanted graft; Gu et al., 2019 Neural stem cells (NSCs) Intra-orbital crush, optic nerve, mouse Intravitreally grafted NSCs differentiated into astrocytes that survived in the host eyes, stably expressed CNTF and significantly attenuated the loss of the axotomized retinal ganglion. The CNTF-secreting NSCs also induced long-distance regrowth of the lesioned retinal ganglion axons; Flachsbarth et al., 2014 3 mm, sciatic nerve, mouse The addition of IL12p80 togethe

ADSCs isolated from adipose tissue are one of the most used stem cells. Many in vitro and in vivo studies have confirmed the ability of ADSCs to promote nerve regeneration, in both undifferentiated (Zhang and Rosen, 2018 ) and differentiated (Ching et al., 2018 ) conditions. Concerning undifferentiated ADSCs, the main concern is related to the possible differentiation that cells may undergo when implanted, originating non-desirable phenotypes, such as adipocytes. In fact, nerve regeneration may be delayed due to fat obstructing NGCs (Papalia et al., 2013 ). In respect to Schwann cell-like differentiated ADSCs, they were found to express a variety of intrinsic neurotrophic factors, namely NGF, BDNF, GDNF, and NT4 (Reid et al., 2011 ).

Still not widely recognized by the scientific community, Schwann cells and olfactory ensheathing cells (OECs) are two types of glial cells with crucial performance in reestablishing function and sensation after nerve injury within the PNS. OECs have similar capacities to Schwann cells, as they can reduce scar formation and promote regeneration (Yao et al., 2018 ). As a component of both PNS and CNS, reports on the use of OEC have mostly been focused on the CNS and insufficient assays have addressed their worth for PNI. One of the mechanisms by which OECs assist in PNR, is that they produce neurotrophins, such as NGF, BDNF, NT-3, NT-4/5, BDNF, and GDNF, leading to neuronal outgrowth and axonal regeneration (Woodhall et al., 2001 ). However, unlike Schwann cells, OECs are not capable of producing cytokines that attract macrophages (Su et al., 2009 ).

The dermis in the skin comprises neural crest related precursor cells, namely the skin-derived precursor cells (SKPCs). They can be differentiated into different cell types belonging to the neural crest, such as neurons and Schwann cells, to find their application in the PNS (McKenzie et al., 2006 ). For such, SKPCs are cultured in medium enriched with neural crest cues such as neuregulins, after which SKPCs morphology switches to Schwann cells-like. These cells are capable of inducing myelin proteins transcriptions when in contact with neurites (McKenzie et al., 2006 ).

Neuroplasticity plays a chief role in PNI patients and how their recovery therapy is carried. Briefly, the term neuroplasticity refers to the ability of the neuronal system to change (Navarro et al., 2007 ). It is now recognized that the brain is not a hard-wired unchanged circuit, but an adaptive system that may experience injury-induced changes. Research in this field has already proved and continues to deepen the available information regarding the interplay between the CNS and PNS throughout an entire lifespan (Taylor et al., 2009 ). Animal experiments have recognized that the so called neuroplasticity in the cortex begins instantly after peripheral nerve injury (Wall et al., 1986 ). More recently, resourcing to advanced imaging techniques, many reports established that a PNIs are linked to quantifiable structural changes in the human brains (Seil, 2014 ; Makin and Bensmaia, 2017 ; Nordmark et al., 2018 ). For instance, the persistent and chronic pain often present in PNI patients' leads to physical changes in the brain, such as insula thickness (Goswami et al., 2016 ). In another example, functional magnetic resonance image (fMRI) has shown that patients who had neuromuscular interventions control the reconstructed muscle with different parts of the brain when compared to the healthy muscle (Chen et al., 2003 ). After peripheral nerve transection and surgical repair, it has also been noticed that both gray and white matter suffer from physical and organizational irregularities in numerous cortical areas as well as thinner cortex, revealing the functional plasticity (Taylor et al., 2009 ). The literature concentrates mainly on measurable parameters and the progress of modern non-invasive neuroimaging techniques to scrutinize brain plasticity in humans. Among the first category, nerve conductions studies, quantitative sensory testing and neuronal responsiveness to stimulus are evaluated (Oberman and Pascual-Leone, 2013 ). Related to neuroimaging, fMRI is capable of measuring white/gray matter volume, density and integrity, as well as cortical thickness (Zatorre et al., 2012 ). Until today, in the clinical setting, the standard diagnosis as well as treatment of PNI patients' relies solely on the fusion of simple clinical examination as well as neuro-electrophysiology measurements (Tagliafico et al., 2010 ). Consequently, the analysis accuracy is frequently decreased since the two approaches mentioned before do not provide adequate information for the needed surgical repair or treatment. Several imaging techniques exist, such as diffusion tensor imaging (DTI), ultrasound and positron emission tomography (PET) (Rangavajla et al., 2014 ). However, assessments by fMRI usually achieve the highest resolutions and are capable of delivering detailed information and necessary quantifications (Rangavajla et al., 2014 ). This technique can, complimentary to the classical evaluation methods that give little information, provide satisfactory spatial and temporal resolution for evaluation of degeneration or regeneration after injury for a correct diagnosis or treatment. Not only physical changes can be detected, but physiological and emotional traits can also be recognized, such as depression or the phenomena of catastrophizing pain (Fernández-Muñoz et al., 2016 ). Maladaptive neuroplasticity also occurs very often. Chronic neuropathic pain is the result of pain-related changes as the answer to the injury (Jensen and Finnerup, 2014 ). Although not fully understood, it is suspected that neuroplasticity arises from deficient and incomplete nerve regeneration, to which the brain must physically adapt. Cortical changes following PNIs include topographic restructuring of the somatosensory and motor cortices, which will be long lasting. Overall, the importance of this specific field relies on the fact that considering peripheral and central changes that directly affect the patient recovery may lead to the advance of new healing and therapeutic approaches, including tailoring of timing and rehabilitation policies (Oberman and Pascual-Leone, 2013 ). A scheme representing neuroplasticity can be seen in Figure 10 . Figure 10 Scheme representing the neuroplasticity that occurs throughout the CNS after PNIs. (A) Healthy peripheral nerve being subjected to an injury. Immediately after the trauma, the functionality of the nerve is affected, and the correct neurotransmission is interrupted. Neuroplasticity that occurs in the CNS following PNI is thought to occur through several mechanisms, with two of the most prominent theories being: (B) Unmasking of existing synaptic connections. In this process, there is the unmasking of neural paths which were not normally used for a specific purpose, and new neural paths are activated when the normally used system fails; and (C) Sprouting of new nerve terminals, where there is collateral sprouting from intact healthy cellular components to a denervated region, in an attempt to reestablish the n

PNIs are permanently accompanied by high costs, translated to economical loss in the health care systems. In addition, there is the morbidity and the significant decrease in patient's quality of life. Therefore, improvements in this field are imperative. There is an increasing agreement that further progress in the field of PNR is urgent, and that the success in the area will no longer be dependent on the new and sophisticated microsurgical tools and techniques or simple hollow NGCs. Instead, due to tremendous quality research recently published, the scientific community is realizing that the best way to move forward is with a multi-combinatorial approach, where different disciplines and specialists may reach a new level of innovation. As such, any approach to peripheral nerve regeneration must integrate a comprehensive and multifaceted strategy, mimicking the natural process of nerve injury and regeneration. In vivo , a cascade of events takes place where cellular components, growth factors, topographical and biological cues act together and naturally induce a certain degree of regeneration. Therefore, developed NGCs, their inherent components and consequent activated mechanisms must enable and simplify the host regenerative process and stimulate regeneration, not only at the injury location, but also at both proximal and distal areas of the injury. Only a multi-factorial device will be capable of overcoming inherent obstacles to regeneration, such as denervation atrophy of muscle targets, lack of vascularization and excessive inflammation. This report focused on reviewing the diverse approaches that can be used in order to go beyond an empty tubular NGCs, similar to those that have been classically used in the clinics. In one hand, it has been proven that to be successful in treating critical gaps (lengthier than 10 mm, in rat), luminal fillers must be added, to encourage the growth cone in its path-finding quest, to reach the distal target. These luminal fillers comprising 3D structures and fabricated in a variety of ways are absolutely needed in the case of more severe injuries. In this scope, the clinical translation of NeuraGen 3D Nerve Guide Matrix® represented a great step toward the change that needed to be seen in this context. Furthermore, the creation of a favorable milieu in the injury site, created with the exogenous delivery of different biological molecules such as neurotrophic factors, other GFs or the presence of pro-regenerative miRNAs, is fundamental for nerve regeneration. In this area, a wide range of factors and delivery strategies can be engineered, maintaining the principle of a timely fashion release. Finally, tissue engineers have a panoply of cell-based therapies which can be adopted to amend clinical outcomes. Because of stem cells' potential, they have become a source of cells which act as an alternative to Schwann cells in PNR. Either differentiated or undifferentiated, the mechanisms by which they are valuable in the case of PNI were also explored. All of the above-mentioned features can and should be used to tackle the present hurdles of nerve regeneration. Such tools allow the medical community to perform a more precise work both at the diagnostic as well as in the treatment stages. Additionally, only by surpassing these obstacles and enhancing the velocity and quality of regeneration will allow a proper rehabilitation, where a motor and sensor recovery helps patients regain their quality of life. Overall, it is crucial to understand that a full functional recovery of PNIs is vital for the patient, to improve its mental health, daily performance, morale, general mobility, agility and capability. All the aforementioned technologies, some of which can already be found in the stage of clinical trials, have the power to go beyond the empty NGCs in terms of positive outcomes. Therefore, they should be adopted and adapted to have a future in the clinical setting, so that patient could benefit from them.