Schwann cells support neuronal signalling. The discovery that these cells become dramatically reprogrammed after nerve injury, adopting migratory characteristics that promote repair, highlights the plasticity of mature cell types.
Cells surrounding injured or diseased tissue often undergo changes that promote repair. A striking example of this is the neuron-supporting Schwann cells of the peripheral nervous system (PNS). In adult nerves there are two types of Schwann cell, distinguished by whether or not they produce a fatty myelin sheath that wraps around neuronal projections called axons to support rapid signalling1,2,3. It has long been thought that, following injury, myelinating Schwann cells rapidly dedifferentiate to an unmyelinating progenitor state that facilitates axonal regeneration and repair. But writing in Neuron, Clements et al.4 challenge this simple notion.
When a nerve in the PNS is cut, the section still connected to the PNS remains intact, whereas the section below (distal to) the cut degenerates. Schwann cells from the cut ends spread across the gap, forming a bridge along which axons can regrow, re-forming the lower part of the nerve. Clements et al. cut the sciatic nerve in mice, and analysed gene expression in myelinating Schwann cells isolated from the intact nerve, the distal nerve and the repairing bridge.
The authors found extensive differences in gene expression between Schwann cells at the intact and distal nerve sites (Fig. 1). The comparison revealed the expected reductions in the expression of genes associated with myelin production and cell–cell junctional communication in distal cells — changes that might reflect dedifferentiation. But, in addition, the researchers found increased expression of unexpected gene sets in distal cells.
These included genes involved in communication with the extracellular matrix that surrounds cells, and genes such as Sox10, Nanog, Oct4 and Myc that are associated with stem cells and the epithelial-to-mesenchymal transition (EMT) — a cellular change in which the adhesive traits that characterize epithelial cells such as myelinating Schwann cells are lost, and migratory 'mesenchymal' traits are gained. Finally, the distal cells did not express genes characteristic of the embryonic tissue from which they arose during development. Together, these data suggest that, rather than simply dedifferentiating, distal Schwann cells that were previously myelinating assume a state intermediate between epithelial and mesenchymal, which shares several fundamental characteristics with embryonic stem cells.
The EMT is a graded response5 that can be mediated by changes in the cellular environment. Clements et al. next compared distal and bridge Schwann cells, and found that the bridge cells underwent rapid, much more extensive EMT-associated changes that resulted in invasive, mesenchymal-like characteristics.
What aspects of the bridge environment might drive these changes? Through a series of experiments in mice and in vitro, Clements and colleagues convincingly demonstrated that a major mediator of EMT in bridge cells is the local upregulation of the transforming growth factor β (TGF-β) protein in the tissue immediately around the cut. Elevated levels of TGF-β also stimulated the targeted migration of Schwann cells across the bridge by activating the Ephrin signalling pathway, which helps to sort the cells into migratory clusters. Finally, the authors demonstrated that bridge Schwann cells are more proliferative than distal cells, but that this effect is TGF-β-independent, suggesting that several different signals mediate Schwann-cell responses to injury. Together, these profound changes in bridge Schwann cells provide a conducive substrate for successful nerve regeneration.
Clements and colleagues' findings offer compelling insights into changes in the characteristics of Schwann cells following an injury. But there are caveats that should be considered, for example in the way in which the authors isolated Schwann cells for analysis. They engineered mice in which expression of a gene encoding a fluorescent protein was regulated by the promoter sequence that drives expression of myelin protein zero (MPZ), which is expressed by myelinating Schwann cells. They then used a process called fluorescence-activated cell sorting (FACS) to isolate fluorescing cells. But FACS favours cells that have high levels of fluorescence. The researchers showed that MPZ expression is drastically reduced following nerve damage, raising the possibility that FACS might isolate only a subset of cells — those that retain sufficient MPZ expression to express the fluorescent protein at high levels.
In future, it would be informative to assay other cell types in the bridge that do not express detectable MPZ (such as non-myelinating Schwann cells and connective-tissue cells called fibroblasts), to gain a full understanding of environmental modulators of Schwann-cell behaviour. In addition, because TGF-β is emerging as a major regulator of EMT (ref. 6), it will be crucial to understand how to effectively modulate and harness the control of its expression if the protein is to be used to help treat various injuries.
The involvement of EMT in development and repair is widespread. In wound healing, EMT is rapidly activated and results in wound closure7 in a manner conceptually similar to that described by Clements and colleagues. It has also emerged as a key component of tumour progression6. The role of EMT in injury responses in the central nervous system has yet to be fully assessed, but tumour studies suggest that it might be important there, too. For example, the formation of cancer stem cells that promote the progression of certain brain tumours8,9 depends on EMT.
EMT is a clearly a key phenomenon in multiple settings, but it is not always stable. For example, Schwann cells in intact nerves are epithelial in nature. Because damaged nerves in the PNS ultimately recover, bridge Schwann cells must either die or differentiate back to a myelinating cell type through a mesenchymal-to-epithelial transition — a change that is much less well understood than EMT. Perhaps it should not come as a surprise that there is plasticity between adult mesenchymal and epithelial states, given that many identity changes occur during development, and that differentiated epithelial cells can be experimentally reprogrammed to a mesenchymal-like stem-cell state10. However, it is somewhat unsettling to this developmental biologist to consider the precarious nature of adult cellular differentiation, because the identity of mature cell types has largely been considered to be irreversible.Footnote 1
Mirsky, R. et al. J. Periph. Nerv. Syst. 13, 122–135 (2008).
Jessen, K. R. & Mirsky, R. J. Anat. 191, 501–505 (1997).
Mirsky, R. & Jessen, K. R. Curr. Opin. Neurobiol. 6, 89–96 (1996).
Clements, M. P. et al. Neuron 96, 98–114 (2017).
Grunert, S., Jechlinger, M. & Beug, H. Nature Rev. Mol. Cell Biol. 4, 657–665 (2003).
Ye, X. & Weinberg, R. A. Trends Cell Biol. 25, 675–686 (2015).
Savagner, P. et al. J. Cell. Physiol. 202, 858–866 (2005).
Chow, K. H. et al. Cancer Res. 77, 5360–5373 (2017).
Kühnöl, C. D., Würfel, C., Staege, M. S. & Kramm, C. Oncol. Lett. 13, 3882–3888 (2017).
Takahashi, K. & Yamanaka, S. Cell 126, 663–676 (2006).