Schwann cells and peripheral nerve injuries

Axonal repair in the central nervous system (CNS) is extremely limited after injury. In contrast, the PNS exhibits a high regenerative capacity. This ability is to a large extent due to the remarkable plasticity of SCs. During development, myelinating SCs form a one-to-one relationship with large caliber axons and wrap them in a myelin sheath, while non-myelinating SCs, also called Remak SCs, surround multiple small caliber axons without producing myelin. Upon axon injury, myelinating and non-myelinating SCs undergo extensive reprogramming that promotes and guides axonal repair. SCs lose contact with and demyelinate the distal stump axon and convert into a repair phenotype. This phenotypic transformation involves the downregulation of several pro-myelinating genes. Repair SCs are characterized by a specific profile which enables the regeneration process. SC reprogramming involves the upregulation of several genes and the activation of multiple transcriptional mechanisms [1,2,3]. Among the main players, c-Jun, mitogen-activated protein kinase (MAPK) pathways, Sonic Hedgehog (Shh) and chromatin modifications control and regulate the repair program. In the injured site, repair SCs participate in the disintegration and removal of damaged axons during the process of Wallerian degeneration and assist myelin debris clearance to create a regrowth favorable environment. Myelin debris clearance is achieved by the digestion of both intrinsic and extrinsic myelin fragments by means of myelinophagy and phagocytosis and the recruitment and activation of invading macrophages [4, 5]. Afterwards, SCs secrete trophic factors to support survival of damaged neurons and promote axon regrowth [1, 6]. Repair and Remak SCs also extend long parallel processes and align in tracts called bands of Büngner to guide the regrowing axon back to innervate its former target [3, 7]. Finally, SCs proliferate, upregulate pro-myelinating genes, re-differentiate into myelinating SCs and remyelinate the regenerated axon. An overview of the repair program is illustrated in Fig. 1. This repair machinery requires a dynamic and orchestrated regulation of SC plasticity and reprogramming following axon injury. Although the underlying molecular mechanisms remain still partially understood, several research groups have significantly contributed to our current understanding of how the key steps involved in the PNS repair program are controlled, which we will discuss in the next chapters of this review.

Fig. 1
figure 1

Repair program in the PNS. Illustration of the main steps of the repair program orchestrated by SCs after peripheral nerve injury. Each step shows a schematic representation of a single neuron (blue) interacting with SCs (light yellow) within an adult peripheral nerve undergoing a traumatic lesion. In step 4, macrophages (green cells) help SCs to clear axon and myelin debris

Full size image

Nerve injury and methods in regeneration research

The severity of peripheral nerve injuries is classified depending on whether demyelination occurs and on the extent of axonal and connective tissue damage [8]. The mildest form called neurapraxia is characterized by local demyelination without axon or connective tissue lesion. Axonal lesion in addition to demyelination but with preserved connective tissue is a more severe injury called axonotmesis, while in the most severe form of injury, which is called neurotmesis, axons and the connective tissue are fully transected [9]. A more detailed classification with five different severity degrees followed Seddon’s classification [10]. SC plasticity and regeneration potential have been extensively studied both in vivo and in vitro. In vivo studies involved the use of wild type and transgenic animals. Among the methods used to study regeneration, nerve transection and crush injury models are the most commonly employed. The crush injury model offers some particular advantages. It is typically performed through an acute traumatic compression of the nerve and it interrupts all the axons but preserves the SC basal lamina. This allows for an optimal regeneration and to investigate the SC regeneration potential. Rats and mice are often used in research for sciatic nerve lesions to model human PNS lesions [11, 12]. In vitro studies mainly involve the use of cell lines, primary or organotypic ex vivo cell cultures. These models present great ethical advantages and they allow to investigate signalling pathways specifically induced by different molecules and drugs on SCs and/or neurons [13, 14]. Although they are unable to fully predict what happens at the whole organ level, primary SC/neuron cocultures have been useful PNS models to investigate the molecular mechanisms involved in axonal regrowth fostered by SCs and remyelination. Moreover, microfluidic devices, which allow the compartmentalization of neuronal cell bodies, axons and myelinating cells, have demonstrated to be useful in in vitro research [15, 16].

Schwann cell reprogramming

SC reprogramming following peripheral nerve injury can be divided in two main partially overlap** processes: SC demyelination and conversion or transdifferentiation into repair SCs [5, 17]. The demyelination process is characterized by the repression of pro-myelinating genes such as Early growth response 2 (Erg2 or Krox20) and of myelin genes including Myelin basic protein (Mbp), Myelin protein zero (Mpz or P0), Peripheral myelin protein 22 (Pmp22) and Myelin associated glycoprotein (Mag). Moreover, genes typically expressed during development including genes coding for L1, neural cell adhesion molecule (NCAM), p75 neurotrophin receptor (p75NTR) and glial fibrillary acidic protein (GFAP), are re-expressed or upregulated [17, 18]. However, repair SCs differ from immature SCs during development by several aspects. Indeed, they are characterized by de-novo expression of genes that include Olig1 and Shh and by the upregulation of many proteins involved in the regeneration process. Among those, (1) c-Jun, the main driver of the SC-dependent repair program, glial cell-derived neurotrophic factor (GDNF), brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT3), artemin, nerve growth factor (NGF), vascular endothelial growth factor (VEGF), and VEGF receptor 1 support the survival of injured neurons and promote the regrowth of proximal axons [1, 16, 18,19,20]; (2) leukemia inhibitory factor (LIF), interleukin-1α (IL-1α) and -1β (IL-1β), tumor necrosis factor-α (TNF-α) and monocyte chemotactic protein 1 (MCP-1) initiate the immune response, promote macrophage invasion and activation, blood vessel formation and myelin breakdown [1, 21,22,23,24]; (3) c-Jun, SRY-box 2 (Sox2) and neuregulin 1 (NRG1) are involved in SC morphological changes and axoglial interactions, in the formation of a nerve bridge in case of nerve transection and of the regeneration tracks along which axons regrow; (4) zinc finger E-box-binding homeobox 2 (Zeb2), nuclear factor-kappa B (NF-kB) and histone deacetylases 1 and 2 (HDAC1/2) are involved in the remyelination of the regenerated axon [25,26,27,28,29]. A summary of the main factors involved in the regeneration process is illustrated in Fig. 2, while a more detailed description of the main factors and signaling pathways with the most recent findings is discussed below.

Fig. 2
figure 2

Summary of the main factors regulated after peripheral nerve injury. Overview of factors downregulated during demyelination (blue area), factors expressed during the repair program (green area), and factors involved in re-myelination post-injury (red area). Uphill areas indicate a high expression, whereas downhill areas indicate a low expression. Regulation of factors is illustrated throughout the different steps of the regeneration process

Full size image

Epigenetic regulation and other chromatin-remodeling enzyme functions

The complexity of the repair program requires a strict orchestration of gene expression and signalling pathways. Among the many levels of regulation, epigenetic regulation and chromatin-remodeling enzymes have been related to many aspects of SC development, maintenance and plasticity after peripheral nerve injury [2]. In SCs of uninjured nerves, Polycomb repressive complex 2 (PCR2) adds trimethyl marks to histone H3 lysine 27 (H3K27me3) on promoter regions of several genes to repress the expression of these genes and PRC2 inactivation results in the induction of genes that are normally upregulated following peripheral nerve injury [30]. Upon injury, H3K27 demethylation together with H3K4 methylation at promoter regions and H3K27 acetylation at enhancer regions promote the activation of injury-induced genes expressed during the repair program such as Shh and Gdnf [2, 30,31,32]. Recently, much research has focused on the functions of HDACs. By deacetylating and controlling the activity of non-histone targets including transcription factors, these enzymes have been shown to play key roles in the regulation of SC behavior following peripheral nerve lesion. HDAC1 and HDAC2, which belong to class 1 HDACs, have crucial functions in SC development including myelination and in PNS maintenance during adulthood [1, 24, 33]. Moreover, they are strongly upregulated following axonal injury [29]. Recent studies revealed that by interacting with Sox10, HDAC2 recruits the H3K9 demethylases KDM3A and JMJD2C to form a complex that de-represses and activates Oct6 and Krox20 genes during the repair program. In turn, Oct6, which is upregulated too early after lesion, negatively regulates c-Jun expression and thereby delays the conversion of SCs into repair SCs. Genetic inactivation of HDAC1/2 prevents Oct6 and Krox20 upregulation after lesion, which results in faster axonal regrowth but impairs the remyelination process. Interestingly, short-term treatment early after lesion with Mocetinostat, a pharmacological inhibitor of HDAC1/2, accelerates the regeneration process without impairing the remyelination process [29]. These findings are of particular interest for the development of potential treatments to improve peripheral nerve regeneration when lesions have led to large gaps between axons and their targets. Indeed, functional recovery of peripheral nerves after lesion is critically dependent on the speed of axon reconnection to their former target [34]. HDAC3, another class 1 HDAC, has been shown instead to limit myelination [119]. Notch inhibits myelination and its expression is downregulated at the beginning of the myelination process [121]. Moreover, Notch antagonistic activity to Krox20 classifies it as a negative regulator of myelination [17]. In adult nerves, Notch dysregulation results in demyelination, which suggests an involvement in the signaling pathway that induces myelin breakdown in vivo. Indeed, inhibition of Notch signaling in adult mice decelerates myelin breakdown that occurs after a nerve lesion [120]. In addition, Wang et al. [122] showed how the addition of Jagged1, a Notch activator, in rat injured nerves enhances functional nerve repair, suggesting that Notch stimulation in SCs could represent an interesting therapeutic strategy promoting nerve repair.

Conclusion

The PNS displays a remarkable ability to regenerate following injury. This process involves the coordinated action of multiple cell types and signaling pathways. Early after injury, damaged axons respond by generating a distress signal detected by SCs, which initiate the repair program. SCs respond by undergoing dynamic reprogramming and assuming an alternative differentiation state suited to meet the specific requirements arising from the injured condition. Although peripheral nerves display an impressive regenerative capacity as compared to the CNS, recovery for patients suffering from traumatic injuries and others peripheral neuropathies is often incomplete. This is mainly a result of the slow regeneration rate, which can reach approximately 1 mm per day, depending on the lesion site, and on the absence of a long-lasting repair-supportive environment. Moreover, the PNS repair ability decreases over time. SCs slowly lose their plasticity in an age-dependent way and the PNS environment becomes unsupportive to regeneration [45, 123]. Therefore, a greater understanding of the mechanisms driving SC plasticity is of utmost interest. Extensive research has been devoted to highlight the molecular mechanisms involved in SC reprogramming to provide mechanistic insights and novel therapeutic strategies to treat peripheral nerve injuries as well as to identify possible correlations with CNS mechanisms and strategies to improve CNS regeneration. Recent work demonstrates the involvement of morphogenetic transformations, epigenetic mechanisms and highlight many transcription factors and signaling pathways critical in this process. However, their induction and the temporal/quantitative activation as well as their mutual interactions have not been completely elucidated and future work should focus on learning how to manipulate repair cells, how to increase their repair-supportive functions, and how to extend their actions to meet the long periods required for axonal regeneration in clinical environment.