Schwann cells degrade myelin after injury by a novel form of selective autophagy, myelinophagy, which is positively regulated by the JNK/c-Jun pathway and is defective in the injured central nervous system.

SummaryRepair Schwann cells play a critical role in orchestrating nerve repair after injury, but the cellular and molecular processes that generate them are poorly understood. Here, we perform a combined whole-genome, coding and non-coding RNA and CpG methylation study following nerve injury. We show that genes involved in the epithelial-mesenchymal transition are enriched in repair cells, and we identify several long non-coding RNAs in Schwann cells. We demonstrate that the AP-1 transcription factor C-JUN regulates the expression of certain micro RNAs in repair Schwann cells, in particular miR-21 and miR-34. Surprisingly, unlike during development, changes in CpG methylation are limited in injury, restricted to specific locations, such as enhancer regions of Schwann cell-specific genes (e.g., Nedd4l), and close to local enrichment of AP-1 motifs. These genetic and epigenomic changes broaden our mechanistic understanding of the formation of repair Schwann cell during peripheral nervous system tissue repair.

There is consensus that, distal to peripheral nerve injury, myelin and Remak cells reorganize to form cellular columns, Bungner's bands, which are indispensable for regeneration. However, knowledge of the structure of these regeneration tracks has not advanced for decades and the structure of the cells that form them, denervated or repair Schwann cells, remains obscure. Furthermore, the origin of these cells from myelin and Remak cells and their ability to give rise to myelin cells after regeneration has not been demonstrated directly, although these conversions are believed to be central to nerve repair. Using genetic lineage-tracing and scanning-block face electron microscopy, we show that injury of sciatic nerves from mice of either sex triggers extensive and unexpected Schwann cell elongation and branching to form long, parallel processes. Repair cells are 2- to 3-fold longer than myelin and Remak cells and 7- to 10-fold longer than immature Schwann cells. Remarkably, when repair cells transit back to myelinating cells, they shorten ∼7-fold to generate the typically short internodes of regenerated nerves. The present experiments define novel morphological transitions in injured nerves and show that repair Schwann cells have a cell-type-specific structure that differentiates them from other cells in the Schwann cell lineage. They also provide the first direct evidence using genetic lineage tracing for two basic assumptions in Schwann cell biology: that myelin and Remak cells generate the elongated cells that build Bungner bands in injured nerves and that such cells can transform to myelin cells after regeneration.SIGNIFICANCE STATEMENT After injury to peripheral nerves, the myelin and Remak Schwann cells distal to the injury site reorganize and modify their properties to form cells that support the survival of injured neurons, promote axon growth, remove myelin-associated growth inhibitors, and guide regenerating axons to their targets. We show that the generation of these repair-supportive Schwann cells involves an extensive cellular elongation and branching, often to form long, parallel processes. This generates a distinctive repair cell morphology that is favorable for the formation of the regeneration tracks that are essential for nerve repair. Remyelination, conversely, involves a striking cell shortening to form the typical short myelin cells of regenerated nerves. We also provide evidence for direct lineage relationships between: (1) repair cells and myelin and Remak cells of uninjured nerves and (2) remyelinating cells in regenerated nerves.

After nerve injury, myelin and Remak Schwann cells reprogram to repair cells specialized for regeneration. Normally providing strong regenerative support, these cells fail in aging animals, and during chronic denervation that results from slow axon growth. This impairs axonal regeneration and causes significant clinical problems. In mice, we find that repair cells express reduced c-Jun protein as regenerative support provided by these cells declines during aging and chronic denervation. In both cases, genetically restoring Schwann cell c-Jun levels restores regeneration to control levels. We identify potential gene candidates mediating this effect and implicate Shh in the control of Schwann cell c-Jun levels. This establishes that a common mechanism, reduced c-Jun in Schwann cells, regulates success and failure of nerve repair both during aging and chronic denervation. This provides a molecular framework for addressing important clinical problems, suggesting molecular pathways that can be targeted to promote repair in the PNS.

Schwann cell c-Jun is implicated in adaptive and maladaptive functions in peripheral nerves. In injured nerves, this transcription factor promotes the repair Schwann cell phenotype and regeneration and promotes Schwann-cell-mediated neurotrophic support in models of peripheral neuropathies. However, c-Jun is associated with tumor formation in some systems, potentially suppresses myelin genes, and has been implicated in demyelinating neuropathies. To clarify these issues and to determine how c-Jun levels determine its function, we have generated c-Jun OE/+ and c-Jun OE/OE mice with graded expression of c-Jun in Schwann cells and examined these lines during development, in adulthood, and after injury using RNA sequencing analysis, quantitative electron microscopic morphometry, Western blotting, and functional tests. Schwann cells are remarkably tolerant of elevated c-Jun because the nerves of c-Jun OE/+ mice, in which c-Jun is elevated ∼6-fold, are normal with the exception of modestly reduced myelin thickness. The stronger elevation of c-Jun in c-Jun OE/OE mice is, however, sufficient to induce significant hypomyelination pathology, implicating c-Jun as a potential player in demyelinating neuropathies. The tumor suppressor P19ARF is strongly activated in the nerves of these mice and, even in aged c-Jun OE/OE mice, there is no evidence of tumors. This is consistent with the fact that tumors do not form in injured nerves, although they contain proliferating Schwann cells with strikingly elevated c-Jun. Furthermore, in crushed nerves of c-Jun OE/+ mice, where c-Jun levels are overexpressed sufficiently to accelerate axonal regeneration, myelination and function are restored after injury.SIGNIFICANCE STATEMENT In injured and diseased nerves, the transcription factor c-Jun in Schwann cells is elevated and variously implicated in controlling beneficial or adverse functions, including trophic Schwann cell support for neurons, promotion of regeneration, tumorigenesis, and suppression of myelination. To analyze the functions of c-Jun, we have used transgenic mice with graded elevation of Schwann cell c-Jun. We show that high c-Jun elevation is a potential pathogenic mechanism because it inhibits myelination. Conversely, we did not find a link between c-Jun elevation and tumorigenesis. Modest c-Jun elevation, which is beneficial for regeneration, is well tolerated during Schwann cell development and in the adult and is compatible with restoration of myelination and nerve function after injury.