The regulation of the homeostasis and regeneration of peripheral nerve is distinct from the CNS and independent of a stem cell population

ABSTRACT Peripheral nerves are highly regenerative, in contrast to the poor regenerative capabilities of the central nervous system (CNS). Here, we show that adult peripheral nerve is a more quiescent tissue than the CNS, yet all cell types within a peripheral nerve proliferate efficiently following injury. Moreover, whereas oligodendrocytes are produced throughout life from a precursor pool, we find that the corresponding cell of the peripheral nervous system, the myelinating Schwann cell (mSC), does not turn over in the adult. However, following injury, all mSCs can dedifferentiate to the proliferating progenitor-like Schwann cells (SCs) that orchestrate the regenerative response. Lineage analysis shows that these newly migratory, progenitor-like cells redifferentiate to form new tissue at the injury site and maintain their lineage, but can switch to become a non-myelinating SC. In contrast, increased plasticity is observed during tumourigenesis. These findings show that peripheral nerves have a distinct mechanism for maintaining homeostasis and can regenerate without the need for an additional stem cell population. This article has an associated ‘The people behind the papers’ interview.

. We were not confident distinguishing macrophages (Mϕ) from NG2+/ PDGFRβ/ P75+ cells, labelled as P-LCs (pericyte-like cells) in the EM quantifications, and so these cells were grouped together (n=4-5 mice, mean±SEM, two-way ANNOVA was used). C) Quantification of immunofluorescent images showing the proportion of each cell type, identified by the indicated markers, in the uncut and regenerated nerves, 6 months following injury (n=4-9 mice, mean±SEM).

Immunostaining
After fixation in PFA, nerves were cryoprotected in 30% sucrose (w/v) in PBS overnight at 4ºC. The nerves were then transferred to a 1:1 mixture of 30% sucrose with O.C.T compound (TissueTek, Sakura) for 2 hours at RT and finally embedded in O.C.T before being snapfrozen in liquid nitrogen. 10-25µm longitudinal or transverse cryosections (Leica) were permeabilised in 0.3% Triton-X100 in PBS for 30 min, blocked in 10% goat serum/PBS for ≥1 hour and incubated in primary antibodies diluted in blocking buffer overnight (O/N) at 4°C. Sections were washed 3 times with PBS and the appropriate fluorescent secondary antibodies were used with Hoechst to counterstain the nuclei for 1h at RT. Samples were mounted in Fluoromount G (Southern Biotechnology) before imaging. Sciatic nerves from P0-CreER TR :Confetti mice were fixed in Antigenfix (DiaPath) to preserve the endogenous fluorescence and RedDot (Biotium) or To-Pro-3 (Thermo Fisher Cat# T3605) was used as a nuclear counterstain. For better preservation of axonal proteins, PBS containing 1mM CaCl 2 , 0.5mM MgCl 2 was used when needed. For P0 staining, harvested sciatic nerves were instead immediately snap frozen in liquid nitrogen. After cutting 10-25µm transverse cryosections, the sections were postfixed for 10min using 4% PFA at RT and washed thoroughly with PBS before blocking with 10% goat serum/PBS. The rest of the immunostaining protocol was as the staining protocol for the prefixed nerve.

Correlative light and electron microscopy (CLEM)
200µm vibrotome sections were screened using a widefield fluorescence microscope to identify sections containing a large number of fluorescently-labelled cells. These sections were imaged with a 40x lens using a SP8 confocal microscope (Leica). YFP, GFP and RFP was acquired (CFP excitation is not possible on this microscope. After image acquisition, samples were fixed in 2% (wt/vol) PFA, 1.5% (wt/vol) glutaraldehyde (both EM grade from TAAB) in 0.1M sodium cacodylate buffer for 30 mins at RT. Samples were then secondarily fixed in 1% (wt/vol) osmium tetroxide, 1.5% (wt/vol) potassium ferricyanide for 1h at 4 °C. After washes in 0.1M sodium cacodylate, samples were incubated in 1% (wt/vol) tannic acid in 0.5M sodium cacodylate at room temperature for 45 min. Further washes in 0.5M sodium cacodylate were followed by a final wash in distilled water, before the samples underwent dehydration by sequential short incubations in 70% (vol/vol) and 90% (vol/vol) ethanol and then two longer incubations in 100% ethanol. Samples were transferred to a 1:1 mix of propylene oxide and Epon resin (TAAB) for 90 min, then 100% Epon for two more incubations, one of several hours and one O/N. Finally samples were polymerised by baking at 60°C O/N. It is important that the side of the sample that was imaged by confocal microscopy faces the top of the Development: doi:10.1242/dev.170316: Supplementary information resulting resin block to ensure correlation between LM images and EM section images.
Ultrathin sections were collected and imaged as above.

Cell composition:
For the cell type quantifications, z-stack projections with an equal number of z-stacks were used. 4 or more non-overlapping fields of each section were imaged using a 63x objective on a SPE microscope. Three different sections were quantified per mouse (≥4 animals per group).
The area of each quantified field was 0.0135mm 2 . Confocal images were counted manually using Fiji software. Within TEM images, each cell type was identified and quantified based on their morphology and the presence of a nucleus. mSCs, nmSCs, pericytes and endothelial cells are morphologically very different, however, we were unable to differentiate between the pericyte-like cells and macrophages. Therefore these cell types were quantified as a single category.

Determination of proliferation rates
Turnover of each cell-type was determined by using measurements taken following 30 days of continuous EdU administration. The calculation used was: proportion of each celltype/(proportion of each proliferating cell-type x the total proliferation rate at 30 days).

ECM analysis:
For area measurements, 4 or more different fields of each section were imaged, with three sections counted for each mouse (≥4 mice per group). Images were converted to 8-bit grey scale TIFF images using Fiji software. Each image was thresholded and made binary. The thresholded area was outlined using the "Create Selection" function and the immunostained area quantified using the measurement function. For intensity measurements, projections used an equal number of z-stacks. The intensity of 9 different fields per image was measured and averaged using Fiji software (3 images were acquired for each section, 3 sections per animal, 6 animals per group).

Axon quantification:
The diameter of individual axons measured from 6 images per mouse and 8 mice per group was binned to assess distribution. All measurements were done with Photoshop to draw the axons and Fiji software was used to measure their diameter. Total number of myelinated axons were counted from 3 imaged sections of the entire nerve per mouse (n= 5 mice per group). Development: doi:10.1242/dev.170316: Supplementary information