Sustained MAPK/ERK Activation in Adult Schwann Cells Impairs Nerve Repair

Abstract

The MAPK/ERK pathway has a critical role in PNS development. It is required for Schwann cell (SC) differentiation and myelination; sustained embryonic MAPK/ERK activation in SCs enhances myelin growth overcoming signals that normally end myelination. Excess activation of this pathway can be maladaptive as in adulthood acute strong activation of MAPK/ERK has been shown to cause SC dedifferentiation and demyelination. We used a mouse model (including male and female animals) in which the gain-of-function MEK1DD allele produces sustained MAPK/ERK activation in adult SCs, and we determined the impact of such activation on nerve repair. In the uninjured nerve, MAPK/ERK activation neither impaired myelin nor reactivated myelination. However, in the injured nerve it was detrimental and resulted in delayed repair and functional recovery. In the early phase of injury, the rate of myelin clearance was faster. Four weeks following injury, when nerve repair is normally advanced, myelinated axons of MEK1DD mutants demonstrated higher rates of myelin decompaction, a reduced number of Cajal bands. and decreased internodal length. We noted the presence of abnormal Remak bundles with long SCs processes and reduced numbers of C-fibers/Remak bundle. Both the total number of regenerating axons and the intraepidermal nerve fiber density in the skin were reduced. Sustained activation of MAPK/ERK in adult SCs is therefore deleterious to successful nerve repair, emphasizing the differences in the signaling processes coordinating nerve development and repair. Our results also underline the key role of SCs in axon regeneration and successful target reinnervation.SIGNIFICANCE STATEMENT The MAPK/ERK pathway promotes developmental myelination and its sustained activation in SCs induced continuous myelin growth, compensating for the absence of essential myelination signals. However, the strength of activation is fundamental because acute strong induction of MAPK/ERK in adulthood induces demyelination. What has been unknown is the effect of a mild but sustained MAPK/ERK activation in SCs on nerve repair in adulthood. This promoted myelin clearance but led to abnormalities in nonmyelinating and myelinating SCs in the later phases of nerve repair, resulting in slowed axon regeneration, cutaneous reinnervation, and functional recovery. Our results emphasize the distinct role of the MAPK/ERK pathway in developmental myelination versus remyelination and the importance of signaling between SCs and axons for successful axon regeneration.

Keywords: MAPK/ERK; Schwann cells; nerve injury; nerve repair; regeneration; remyelination.

Figure 1.

MAPK/ERK activation in adult Schwann cells does not affect myelin and axons stability. A, Immunoblotting of p-ERK, p-Akt, and ERK1/2 proteins in contralateral nerves 6 weeks after tamoxifen dosing in sciatic nerves of MEK1DD mutant mice (MEK) compared with controls (Ctr) mice. Data were normalized vs Calnexin. B, TEM micrographs of uninjured sciatic nerve 6 weeks post-tamoxifen dosing. C, Summary of the main parameters analyzed to study myelination and axon survival. Data are reported as percentages of myelinated and nonmyelinated axons normalized for the area in square millimeters. The g-ratio of individual fibers in relation to their axon diameter is shown in the scatter plot in D. E, Quantification of IENFD/millimeter from uninjured skin biopsy specimens. F, Internodal distance measured in teased fibers of sciatic nerve. G, Percentage of innervated NMJs in uninjured nerves. For all analyses, three to four mice per group were analyzed. Scale bar, 5 μm. *p < 0.05, two-tailed unpaired t test. Error bars indicate SEM.

Figure 2.

Wallerian degeneration is faster in mutant mice after p-ERK induction. A, Electron micrographs of ipsilateral sciatic nerve sections 5 d postinjury. Arrows indicate intact myelin sheaths that are quantified in B. C, Macrophage count in TEM pictures reported as the number of macrophages per square millimeter. D, Western blot analysis of MBP expression in MEK1DD mutants (MEK) compared with controls (Ctr). Calnexin was used as a control. n = 3 mice/group. Scale bar, 10 μm. *p < 0.05, ***p < 0.001, two-tailed unpaired t test. Error bars indicate SEM.

Figure 3.

Slower recovery in MEK1DD mutant mice vs controls, 1 month postinjury. A, Beam test was used to assess balance and motor coordination after injury. B, C, Rotarod test (B) and toe-spreading reflex test (C) confirmed deficits in motor function in MEK1DD mutant mice. C, Representative photographs of the toe spreading from day 13 postinjury are shown. D, Pin-prick test was used to analyze recovery in sensory functions. n = 11 mice. All tests were analyzed with two-way repeated-measures ANOVA; Sidak's multiple-comparison test apart from the rotarod test in which two-tailed unpaired t test was used. *p < 0.05, **p < 0.005, ***p < 0.001. Error bars indicate SEM.

Figure 4.

MAPK/ERK activation in Schwann cells does not affect myelin thickness, but does affect myelin stability. A–C, Morphological analysis of MEK1DD mutant vs control sciatic nerve at 4 weeks postinjury was undertaken to determine the following: frequency distribution of g-ratio (A); frequency distribution of axon diameter (B); distribution graph of g-ratio in relation to axon diameter (C). D, TEM pictures of ipsilateral sciatic nerves from control and MEK1DD mutant mice, which illustrate that although myelin thickness does not change there are myelin defects shown in E: asterisks illustrate axons with myelin decompaction; arrows point to invaginating recurrent loops; arrowheads indicate supernumerary SC processes surrounding a myelinated axon with poorly compacted myelin that have the appearance of an early “onion bulb” structure. Error bars indicate SEM. n = 3 mice/group. Scale bars: D, 10 μm; E, top left, 5 μm; E, all other panels, 2 μm. Kolmogorov–Smirnov test performed.

Figure 5.

Cajal band disruption and DRP2 impairment in mutant ipsilateral nerves after injury. A, TEM pictures showing a lack of appositions between SC plasma membrane and the abaxonal layer of the myelin sheath in mutant mice. Scale bars: 2 μm; insert, 1 μm. Asterisk shows Cajal bands location; arrow points to the apposition. B, Quantification. C, Periaxin and DRP2 were quantified by Western blot 1 month after injury in sciatic nerves of MEK1DD mutant mice (MEK) compared with control (Ctr) mice; calnexin was used as a loading control. D, Teased fibers from ipsilateral sciatic nerves showed decreased internodal distance in mutant mice. n = 3 mice/group. *p < 0.05, ***p < 0.001, two-tailed unpaired t test. Holm–Sidak test was performed in B. Error bars indicate SEM.

Figure 6.

MAPK/ERK activation does not affect reinnervation of the gastrocnemius muscle at 4 weeks postinjury. Muscle innervation has been calculated as the number of reinnervated neuromuscular junctions. Gastrocnemius muscle sections of 100 μm were stained for a-bungarotoxin (α-BTX) as a marker of the postsynaptic NMJ, and for synaptic vesicle protein 2 (SV2) and neurofilament (2H3) antibodies for presynaptic terminal markers. n = 3 mice/group. No significant difference in two-tailed unpaired t test. Scale bar, 10 μm. Error bars indicate SEM.

Figure 7.

MAPK/ERK induction disrupts Remak bundle formation and small-fiber reinnervation after injury. A, Electron micrograph showing disruption of Remak bundle formation with decreased C-fiber numbers per Remak bundle. Arrows show Remak bundles in controls and mutants. Visible formation of elongated and supernumerary SCs failing to assemble into normal Remak bundles (arrowhead). Scale bar, 5 μm. B, Photomicrographs of transverse sections of ipsilateral skin biopsy specimens stained with PGP9.5. Quantification of epidermal reinnervation, in IENF per millimeter, showed a decrease in mutant mice. n = 3 mice/group. Arrows point to nerve fibers crossing the epidermis. *p < 0.05, two-tailed unpaired t test. Scale bar, 15 μm. Error bars indicate SEM.