Nerve regeneration in the peripheral and central nervous systems

The critical role of Schwann cells in supporting nerve regeneration in the peripheral nervous system (PNS) and the contrasting inability of oligodendrocytes to do so in the central nervous system (CNS) has concerned neuroscientists since the seminal work of Ramon y Cajal (Cajal, 1928). Cajal's power of observation and capacity to capture the essence of, and to interpret his light microscopic slides of normal and injured nerves in the PNS and CNS, established these now well accepted concepts. Oligodendrocytes in the CNS share the capacity of Schwann cells to myelinate axons although their pattern of myelination is quite different; each oligodendrocyte protrudes cytoplasmic processes that enwrap and form myelin sheaths around several of the axons within the CNS whilst, each Schwann cell wraps around a single peripheral axon with the cell body being included in the wrap. In the rat, for example, an individual oligodendrocyte in the optic nerve forms as many as 20 myelin internodes (Ransom et al. 1991). The axons that are disconnected from the cell bodies or soma of the injured neurons in both nervous systems undergo Wallerian degeneration: the axons degenerate and the myelin fragments (Stoll et al. 2002; Vargas & Barres, 2007). The denervated Schwann cells proliferate and extend long cytoplasmic processes. The cells and their processes cross surgical sites of reunion of proximal and distal nerve stumps and, they line the endoneurial tubes as the Bands of Büngner to support and guide growing axons to denervated targets (Büngner, 1891; Cajal, 1928; Fu & Gordon, 1997). However, there is a limited time window in which the Schwann cells express growth-associated genes and, in turn, support axon regeneration (Fu & Gordon, 1995; Hoke et al. 2006; Gordon et al. 2011). The Schwann cell numbers decline and the cells undergo atrophy, progressively failing to support nerve regeneration (Fu & Gordon, 1995; Vuorinen et al. 1995a; Gordon et al. 2011). In the CNS, on the other hand, a programmed cell death is initiated when oligodendrocytes lose axonal contact, many of the cells entering into an 'atrophy-like resting state' (Vargas & Barres, 2007). Wallerian degeneration is not robust in the CNS as it is in the PNS: that the poor clearance of the myelin debris is likely to be an important contributing factor to failure of CNS nerves to regenerate after injury is illustrated by the remaining debris 3 years after degeneration of human CNS axons (Buss et al. 2004). The oligodendrocytes fail to support nerve growth in the CNS; the retraction bulbs of the axons proximal to CNS lesions (Erturk et al. 2007) and the inhibitory molecules of the myelin, including myelin-associated glycoprotein and NOG0 expressed on the glial cells, preclude regeneration, in contrast to the Schwann cells' support of nerve regeneration in the PNS (Fenrich & Gordon, 2004). This review introduces the topics that were covered by each of four contributors, Jessen, Bunge, Karimi-Abdolrezaee, and Chan, to the symposium entitled 'Axon regeneration and remyelination in the peripheral and central nervous systems' at the Physiology Society meeting in Cardiff in July 2015. Their topics are reviewed in four accompanying articles in this issue of the Journal of Physiology (Alizah & Karimi-Abdolrezaee, 2012; Bunge, 2016; Chan et al., 2016; Jessen & Mirsky, 2016). The developmental stages of the myelination of peripheral nerves, a very active subject of research over many years by Jessen and Mirsky in London, UK, provide the scaffold on which the investigators are currently studying the extent to which the developmental maturation of myelinating and non-myelinating (Remak) Schwann cells is reversed after PNS nerve injury. The switch of the adult myelinating Schwann cell to a non-myelinating growth supportive phenotype involves activation of thousands of genes, many of which are said to be 'repair-related' (Arthur-Farraj et al. 2012). Whilst this conversion was previously regarded as a process of dedifferentiation of the mature Schwann cell phenotype to a less mature phenotype (Jessen & Mirsky, 2005), the authors now provide evidence for the activation of a more distinct repair programme in the denervated Schwann cells that differs fundamentally from the development of the cells (Jessen & Mirsky, 2016). The denervated cells, but not immature cells, organize myelin clearance indirectly by macrophage recruitment and directly by autophagy of myelin (Gomez-Sanchez et al. 2015). The repair programme includes the upregulation of genes for transcription factors such as the helix–loop–helix factor, Olig 1, and sonic hedgehog (Shh), which are either not expressed or expressed at very low levels in immature Schwann cells (Arthur-Farraj et al. 2012). The expression of the transcription factor c-jun appears to be a key regulator of the distinct regeneration programme, c-jun directly or indirectly controlling the expression of ∼180 genes in the denervated Schwann cells and down-regulating myelin-associated genes (Jessen & Mirsky, 2008; Arthur-Farraj et al. 2012). The selective removal of c-jun from Schwann cells in transgenic mice detrimentally reduces survival of both motor and sensory neurons even though the Schwann cells developed and functioned normally in intact nerves (Fontana et al. 2012). After nerve injury in the adult, the denervated Schwann cells, which otherwise would upregulate c-jun, fail to express growth-associated proteins such as neurotrophic factors and cell surface proteins, fail to remove myelin from denervated nerve, do not align normally as the bands of Büngner, and in turn, fail to support axon regeneration. K. R. Jessen and R. Mirsky are currently investigating whether a progressive decline in c-Jun expression in chronically denervated Schwann cells (unpublished data) contributes to the previously demonstrated concurrent decline in the capacity of these cells to support axon regeneration (Fu & Gordon, 1995). The early in vitro work on Schwann cells by Jessen and Mirsky paralleled the in vitro work carried out by Mary and Richard Bunge in St Louis, MO, USA, each isolating the cells, the former concentrating on the development of Schwann cells, and the latter on adult Schwann cells. Over a period spanning from 1975 during their research in St Louis, Mary and Richard Bunge purified and expanded the numbers of Schwann cells in vitro, culminating in their initial joint effort at the Miami Project to Cure Paralysis, to bring the transplantation of purified human Schwann cells in rats to human transplantation into the CNS of spinally injured individuals (Bunge & Wood, 2012; Bunge, 2016). This objective is currently being brought to fruition with the first stage recognition of safety by the US Food and Drug Administration for the implantation of autologous human Schwann cells into injured spinal cord of patients (Bunge, 2016). This success built on the steady progress of developing and using an in vivo rat model of complete spinal cord transection to progressively optimize the conditions for implanting purified rat and human Schwann cells into the transection site. Inclusion of neurotrophic factors such brain derived neurotrophic factor and neurotrophin-3 (Xu et al. 1995; Blits et al. 2003), rolipram to enhance the growth capacity of the injured CNS neurons (Pearse et al. 2004; Fouad et al. 2005), and chondroitinase ABC to break down the inhibitory extracellular matrix molecules contributing to the glial scarring after spinal cord injury (Flora et al. 2013), contributed to the regeneration of CNS axons across the Schwann cell bridge and the moderate recovery of locomotion even after complete spinal cord transection. The idea of using these agents originated in the work of Filbin and colleagues who correlated the high levels of cAMP (Cai et al. 2001; Hannila & Filbin, 2008) with the capacity of immature CNS neurons to regenerate on growth inhibitory substrates and the finding that chondroitinase ABC was effective in promoting nerve regeneration in the PNS (Cai et al. 2001; Udina et al. 2010; Sabatier et al. 2012). The success of the implanted cocktail of cells and agents in the rat transected spinal cord was exemplified by the increased survival, regeneration and myelination of CNS axons as well as functional movements of the paralysed hindlimbs. Hence, whilst the seminal study of Richardson and colleagues demonstrated that damaged CNS neurons regenerated their axons within the growth permissive environment of the denervated Schwann cells of an inserted peripheral nerve (Richardson et al. 1980), the systematic Bunge approach of inserting Schwann cells into the CNS environment to promote CNS nerve regeneration, holds promise. Even in the PNS, Schwann cells must be within a nerve graft to support nerve regeneration and/or acellular conduits may be populated with migrating cells primarily from denervated nerve stumps and less so from the proximal nerve stump, in order to support regeneration of axons through the conduit (Pfister et al. 2011). The Bunge combination of rolipram with chondroitinase ABC to promote CNS nerve regeneration was also shown to reverse the inhibitory effects of the upregulated chondroitin sulfate proteoglycans (CSPGs) in injured rat spinal cord, their presence as the scar being the hallmark of CNS injuries (Karimi-Abdolrezaee & Billakanti, 2012). Karimi-Abdolrezaee and her colleagues later demonstrated that the CSPGs are responsible, at least in part, for the death of many oligodendrocytes after spinal cord injury (Alizah & Karimi-Abdolrezaee, 2016). In these experiments either the protein tyrosine phosphatase RPTP-σ receptors for the CSPGs on neural precursor cells were knocked out or both the RPTP-σ and leukocyte common antigen-related phosphatase receptor (LAR) for CSPGs were knocked down with interference RNA (Dyck et al. 2015). These experimental manipulations significantly increased the survival of the oligodendrocytes, providing evidence for the involvement of the CSPGs in oligodendrocyte demise. Importantly, the axon loss, probably occurring secondary to demyelination, was also alleviated by the experimental manipulations to reduce CSPG inhibitory effects. Karimi-Abdolrezaee and colleagues also considered the problems of the relatively low expression of growth factors that include neuregulin-1 in denervated oligodendrocytes and the inflammatory response of the injured CNS. They provided evidence for the potential for neuregulin-1 to counteract the limited differentiation of oligodendrocytes from endogenous precursor cells, namely that the depressed neuregulin-1 expression in the injured spinal cord correlates with the limited differentiation of the oligodendrocytes from oligodendrocyte and neural precursor cells, the increased astrocyte differentiation, and the formation of scar (Gauthier et al. 2013). Karimi-Abdolrezaee also pointed out that the pro-inflammatory microglia/macrophages within the injured spinal cord counteract survival and myelination of and by oligodendrocytes whilst the anti-inflammatory cytokines such as IL-10 released by the microglia/macrophages may oppose the inhibitory effects of CSPGs on axonal regeneration and remyelination (Pluchino & Martino, 2005; Butovsky et al. 2006). The high levels of cAMP that are effective in promoting nerve regeneration in the CNS (Pearse et al. 2004) are also important in the efficacy of brief low frequency electrical stimulation in promoting nerve regeneration and accelerating target reinnervation after PNS nerve injuries in animal models as well as in human patients (Gordon et al. 2009, 2010). Our animal studies demonstrated that a 1 h period of low frequency (20 Hz) continuous electrical stimulation accelerates the outgrowth of regenerating motor and sensory axons across a surgical gap between the proximal and distal nerve stumps to, in turn, culminate in accelerated target reinnervation (Al-Majed et al. 2000; Brushart et al. 2005; Eberhardt et al. 2006; Geremia et al. 2007). In the same manner as Bunge brought her rat data on Schwann cell support of CNS regeneration to clinical application, Chan and colleagues have brought the electrical stimulation paradigm to clinical application. The same brief (1 h) electrical stimulation was applied to the human median nerve above the site of a carpal tunnel release surgery in patients who suffered ∼50% loss of innervation of the hand muscles of the median eminence. This proof of principle study demonstrated that all the injured motor nerves reinnervated the muscles within 6–8 months of the brief electrical stimulation in contrast to there being an insignificant rise in numbers when no stimulation was applied (Gordon et al. 2010). Chan and colleagues have continued these studies to demonstrate functional outcomes after electrical stimulation following the surgical repair of the transected digital nerve (Wong et al. 2015) and of increased grasp after electrical stimulation of the ulnar nerve at the elbow immediately following surgical release (Chan, unpublished findings). The brief electrical stimulation is effective in accelerating target reinnervation under conditions where surgical repair is delayed in the case of the human nerves (Gordon et al. 2010) and in rat nerves (Elzinga et al. 2015). That the electrical stimulation was effective despite the regression of growth potential of chronically injured neurons and chronically denervated Schwann cells (Gordon et al. 2011) is very promising. In summary, a diverse group of neuroscientists have undertaken extensive research that has uncovered key molecular mechanisms that underlie the failure of nerve regeneration in the CNS as well as in the PNS under conditions of chronic injuries. The problem of axon guidance remains with random regeneration of PNS axons into denervated endoneurial tubes (Brushart & Musalem, 1980). Nonetheless, in the light of the evidence of preferential motor and sensory nerve reinnervation of motor and sensory nerve pathways, especially after brief electrical stimulation (Brushart, 1993; Burshart et al. 2005) and the key differences in the growth factor expression of Schwann cells within motor and sensory nerves that are now being recognised (Hoke et al. 2006), even the problems of misdirection of regenerating axons may progressively dissipate. Hence, overall, the research work to date is providing important directives to overcome the problems of poor regenerative capacities of both CNS and PNS neurons that are currently being translated into clinical reality. None declared.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.