Wistar rats (postnatal days 1–3, embryonic day 15 and 200 g adult; MF, Oriental Yeast, Osaka, Japan) were used. Animals were housed under a 12/12 h light/dark cycle (lights on, 08:00–20:00 h). All animals had free access to food (MF, Oriental Yeast, Osaka, Japan) and tap water. All experiments were performed in conformity to the guidelines of the Animal Care Committee of the Graduate School for Medicine, Osaka University. We made a maximum effort to minimize the number of animals used and to limit any suffering.

Primary rat SCs were isolated and cultured as previously described (Ogata et al., 2004). SCs were collected from the sciatic nerves of postnatal days 1–3 Wistar rats and cultured in Dulbecco’s Modified Eagle’s Medium (DMEM; GIBCO/BRL Life Technologies, Grand Island, NY, USA; not including vitamin B12) containing 10% fetal bovine serum (FBS; Sigma–Aldrich, St. Louis, MO, USA) and 1% penicillin and streptomycin. The following day, 10 μM cytosine arabinoside (Sigma–Aldrich) was added to the medium to eliminate contaminating fibroblasts. After 48 h, the medium was replaced with DMEM containing 3% FBS with 3 μM forskolin (Merck, Darmstadt, Germany) and 20 ng/mL of neuregulin (R&D Systems, Minneapolis, MN; growth medium) to expand the cells. Cells were then detached from the dishes using 0.25% trypsin (GIBCO/BRL Life Technologies) treatment and subculturing by replating at a 1:2 ratio onto poly-L-lysine-coated (Sigma–Aldrich) plastic dishes before confluence. We obtained a SC culture of >99% purity using these procedures. In all the experiments, cells were used between passages 3 and 8.

Schwann cells were plated at a density of 1.4 × 10⁴ cells in 6-cm plates and maintained in the growth medium for 24 h prior to stimulation with MeCbl (100 μM; Sigma–Aldrich). On days 1, 3, and 5 after the stimulation, cells were trypsinized and resuspended. Cell counting was performed in triplicate on separate 10 μl aliquots using a hemacytometer.

Cultured SCs were collected and homogenized with 100-μL Kaplan buffer [150 mM NaCl, 50 mM Tris-HCl (pH 7.4), 1% NP-40, 10% glycerol, and a protease inhibitor cocktail] and clarified by centrifugation. Each sample included 18 μg of protein, was separated by SDS–PAGE, and transferred onto polyvinylidene difluoride membranes. After blocking non-specific binding sites with a blocking buffer [5% skimmed milk/1% Tween 20 in 20 mM TBS (pH 7.6)] for 1 h, the membranes were incubated overnight at 4°C with primary antibodies against p44/42 MAPK (1:1000; Cell Signaling Technology 4695, Beverly, MA, USA), phospho-p44/42 MAPK (1:1000; Cell Signaling Technology 9101), Akt (1:1000; Cell Signaling Technology 4691), phosphor-Akt (1:1000; Cell Signaling Technology 4056), GAPDH (1:1000; Cell Signaling Technology 2118), cleaved caspase-3 (Asp175; 1:1000; Cell Signaling Technology 9661), cleaved caspase-9 (Asp353; 1:1000; Cell Signaling Technology 9507), MAG (1:1000; Chemicon MAB1567), P0 (1:1000; Abcam ab31851, Cambridge, UK), MBP (1:1000; Sigma–Aldrich M3821) and Acly (1:1000; Abcam ab40793). Subsequently, the membranes were incubated with an anti-rabbit IgG, horseradish peroxidase linked whole antibody from donkey (1:1000; GE Healthcare Life Sciences NA934, Little Chalfont, UK) and subjected to ECL reagent treatment. Protein expression levels were determined using the MF-ChemiBIS 3.2 imaging system (Berthold Technologies, Bad Wildbad, Germany). The integrated optical densities of immunoreactive protein bands were measured using ImageJ 1.45 s, which is a public-domain image analysis program that was developed at the U.S. National Institutes of Health.

Schwann cells were cultured on a poly-L-lysine-coated 35-mm plastic dish until they reached subconfluency. For the determination of apoptosis, cells were incubated with recombinant rat TNF-α, (100 ng/mL; Sigma–Aldrich) in the presence or absence of 100 μM MeCbl at 37°C for 24 h (Yuan et al., 2012). The effects of TNF-α on SCs apoptosis after treatment were determined by western blotting using an anti-cleaved caspase-3 antibody and anti-cleaved caspase-9 antibody.

For differentiation experiments, purified cells were cultured in the growth medium for 24 h. They were induced to differentiate by the addition of db-cAMP (1 mM; Sigma–Aldrich; differentiation medium; (Yamauchi et al., 2011; Varela-Rey et al., 2014) and cultured for 72 h in the differentiation medium in the presence or absence of 100 μM MeCbl. The effects of the differentiation of SC after treatment were determined by western blotting using anti-MBP antibody and anti-Acly antibody.

Rat DRG neurons were prepared as previously described (Callizot et al., 2011), with some modifications. DRG neurons from the spinal cord of 15-day-old Wistar Rat embryos (E15) were sterilely extracted and treated with a trypsin solution (0.25% Trypsin-EDTA, GIBCO) at 37°C for 20 min; the reaction was stopped by the addition of DMEM containing DNase I grade 2 (0.1 mg/mL, Roche) and 10% FBS. Cells were mechanically dissociated by three forced passages through the tip of a 10-mL pipette. They were then centrifuged at 500 × g for 10 min. The supernatant was discarded and the pellet was resuspended in a defined culture medium containing neurobasal medium (Invitrogen, Carlsbad, CA, USA) supplemented with 2% B27 nutrient supplement (Invitrogen), L-glutamine (0.2 mmol/L; Invitrogen), 1% penicillin and streptomycin (10 mg/mL), and nerve growth factor (50 ng/mL; Millipore, Billerica, MA, USA). Viable cells were counted in an Invitrogen Countess^®Automated Cell Counter. They were seeded at a density of 31,250 cells/well in an 8-well slide chamber that was precoated with poly-L-lysine and coated with laminin (1.4 g/cm²; Sigma–Aldrich) for 2 h at 37°C. Half of the medium was then changed every day. Cultures were maintained for 1 week to allow SCs to proliferate and unsheathe the axons of DRG neurons. At day 7, the cocultures were induced to form myelin via the addition of L-ascorbic acid (50 g/mL; Sigma–Aldrich) with or without MeCbl (100 μM), and half of the medium was then changed every day.

Dorsal root ganglions/SCs cocultures on 8-well slide chambers were fixed with 4% paraformaldehyde for 10 min and permeabilized with 100% methanol for 30 min at -20°C. After blocking with PBS + 0.2% Triton X + 5% bovine serum albumin (Sigma–Aldrich), the samples were incubated with primary antibodies against MBP (1:1000; Calbiochem NE1018, San Diego, CA, USA) and NF200 (1:1000; Sigma–Aldrich N4142) overnight at 4°C, followed by incubation with the appropriate secondary antibodies including Alexa Fluor 488 goat anti-rabbit IgG (Molecular Probes A11034, Eugene, OR, USA) and Alexa Fluor 568 goat anti-mouse IgG (Molecular Probes A-11004). DAPI (Wako Pure Chemical Industries, Osaka, Japan) was included in the Permafluor (Thermo Fisher Scientific, Waltham, MA, USA) mounting solution to visualize nuclei. The number of MBP-positive segments was counted using the NIS Elements BR software (Laboratory Imaging, Nikon).

All animal experiments were approved by the Ethics Review Committee for Animal Experimentation of the Osaka University. Thirty-six male Wistar rats weighing 180–220 g were used in this study. For all experimental procedures, animals were deeply anesthetized using an intraperitoneal injection of a mixture of midazolam (2 mg/kg), butorphanol (2.5 mg/kg), and medetomidine (0.15 mg/kg). Under sterile conditions, the left sciatic nerve was exposed at the level of the sciatic notch. Using a Hamilton syringe, 5 μL of saline or 2% LPC (Sigma–Aldrich) in saline was injected into the proximal sciatic nerve. Seven days later, an osmotic minipump (Alzet, Cupertino, CA, USA) was placed subcutaneously in the back (Okada et al., 2010) to deliver continuous saline or MeCbl (1 mg/kg/day) for 1 week. All surgeries were performed by the same surgeon.

For histological evaluation of the lesions, animals were sacrificed 1 or 2 weeks after the LPC injection. They were deeply anesthetized, and the sciatic nerve containing the area of LPC application was excised for the evaluation of the extent of demyelination and remyelination. Sciatic nerves were fixed in 4% paraformaldehyde for 24 h at room temperature and then stored in 20% sucrose in 0.01 MPBS. The tissues were embedded in Tissue Tek (Sakura Finetek Japan), snap frozen on liquid nitrogen, sectioned axially at 5 μm, and mounted on a glass slide. They were permeabilized with 100% methanol for 30 min at -20°C. After blocking with PBS + 0.2% TritonX + 5% bovine serum albumin, they were incubated with primary antibodies against MBP (1:1000; Calbiochem NE1018) and NF200 (1:1000; Sigma–Aldrich N4142) overnight at 4°C inside a wet chamber, followed by incubation with the appropriate secondary antibodies including Alexa Fluor 488 goat anti-rabbit IgG, and Alexa Fluor 568 goat anti-mouse IgG. DAPI was included in the Permafluor mounting solution to visualize the nuclei. The number of myelinated axons was counted using the NIS Elements BR software.

Motor function was evaluated with the sciatic functional index as previously described (Varejao et al., 2001; Pereira Lopes et al., 2013; Goulart et al., 2014). Sciatic functional index was calculated at 2 weeks after the LPC injection. Rats were made to walk across a narrow track. The hind feet were dipped in block ink and changes in footprints were recorded on white papers. For normal footprints, a sciatic functional index value is near 0, whereas a sciatic functional index value of approximately -100 reflects complete loss of function (Bain et al., 1989).

Under anesthesia, the sciatic nerve was exposed. A bipolar stimulating electrode was placed in the nerve trunk at its proximal portion and a recording electrode was placed in the anterior tibial muscle to record the compound muscle action potential. Nerve conduction velocity values were calculated by stimulating two different points of the sciatic nerve. Compound muscle action potential was detected and measured using the PowerLab devices and software (AD Instruments, Bella Vista, NSW, Australia).

Responses to thermal stimuli were assessed using a hot plate (Ugo basile, Varese, Italy), at a temperature of 52.5°C. The reaction time (in seconds) until the first signs of a painful response (hindpaw licking or escape) was recorded and the cut-off time was 45 s (Guevara et al., 2015). Withdrawal thresholds of the hind paw to mechanical stimulation were determined using von Frey monofilaments (0.008–26 g; TouchTest, North Coast Medical Inc, Gilroy, CA, USA). Each filament was applied from beneath the mesh floor to the middle part of the plantar surface of each hind paw until the individual filaments used started to bend (Pitcher et al., 1999; Noda et al., 2014). Values are normalized to the unaffected side.

The JMP software, version 11 (SAS Institute, Cary, NC, USA) was used to analyze the results and the data were expressed as the mean ± SEM. The data were first analyzed for normality with Shapiro–Wilk test. Those qualified were then analyzed subsequently with one-way ANOVA followed by a post hoc Student’s t-test, Dunnett test, and Tukey–Kramer HSD test. Wilcoxon test and Steel test was only used when normality test failed.

After peripheral nerve injury, SCs located in the distal nerve begin to dedifferentiate and proliferate; this reaction is a prerequisite process for the regeneration of damaged peripheral nerves. We focused on the effect of MeCbl on SC proliferation. SCs were cultured with or without 100 μM MeCbl for 5 days and the total cell number was then counted. The results indicate that there was no significant difference between the control and MeCbl groups on days 1, 3, and 5 (Figure 1). This finding suggests that MeCbl does not stimulate SC proliferation.

Some studies identified the activation of the Erk1/2 pathway for SC proliferation (Ogata et al., 2004; Monje et al., 2006). To determine whether MeCbl influences Erk1/2 activity in SCs, we detected the activity of Erk1/2 in SCs cultured for 1 h with MeCbl at concentrations of 1 nM to 100 μM in the growth medium. We observed that MeCbl suppressed Erk1/2 activities in SCs at a concentration ≥10 μM (Figure 2A). Subsequently, SCs were cultured with 100 μM MeCbl for 3 h in the growth medium. MeCbl temporarily led to a 0.38 ± 0.04-fold weaker activation of Erk1/2 than that observed in the control at 1 h after the addition of the compound (Figure 2B). We next examined the activation of the Akt pathway which plays an important role in SCs differentiation (Ogata et al., 2004). We detected the activity of Akt in SCs cultured for 1 h with MeCbl at concentrations of 1 nM–100 μM (Figure 2C) and with 100 μM MeCbl for 3 h (Figure 2D) in the growth medium. The Akt activation was not detected in neither conditions (Figures 2C,D). These results clearly demonstrate that MeCbl suppresses the Erk1/2 activity in SCs, albeit temporarily.

The activation of Erk1/2 has been reported to protect against apoptosis in several cell types including neurons (Kolch, 2000; Xifro et al., 2014). To determine the effect of MeCbl on SC apoptosis, we cultured SCs with MeCbl (100 μM) in the presence or absence of TNF-α in the growth medium. We added TNF-α to the medium at a concentration of 100 ng/mL because TNF-α at a low concentration of 0.001 ng/mL induces SC proliferation, whereas TNF-α at a high concentration of 100 ng/mL induces SC apoptosis (Yuan et al., 2012). We examined the activation of cleaved caspase-3 and cleaved caspase-9, which were markers of apoptosis. MeCbl did not cause SC apoptosis, and TNF-α-induced apoptosis was not rescued by the administration of MeCbl (Figures 3A,B).

Schwann cells are the main glial cells in the peripheral nervous system. After peripheral nerve injury, SCs first dedifferentiate and proliferate, and then redifferentiate and remyelinate newly grown axons in response to axon-derived signals, thus triggering a process of nerve regeneration (Harrisingh et al., 2004). cAMP is one of the signals that can mimic axonal contact with SCs and promotes the expression of the myelin marker galactocerebroside (Sobue and Pleasure, 1984) and a myelin protein (Ji et al., 2010; Yang et al., 2012) in SCs. First, we detected the expression of P0 and MAG, markers of SCs in the promyelinating state (Ogata et al., 2004). SCs were treated in the growth or the differentiation medium containing db-cAMP. MeCbl did not promote the expression of P0 (Figure 4A) and MAG (Figure 4B) in SCs in neither conditions. In SCs culture, P0 is clearly detectable even in the growth medium and the expression of MAG peaks at 48 h under the differentiation condition (Leitman et al., 2011). We therefore assumed that it might be difficult to detect the increasing expression of the markers in the promyelinating SCs by MeCbl and focused on the markers in the myelinating state. MBP is vital to the myelination process and is essential for the appropriate formation of myelin thickness and compactness in the central nervous system and peripheral nervous system (Zhang et al., 2007; Ryu et al., 2008). Therefore, we focused on the expression of MBP to evaluate the effect of MeCbl on SC differentiation. In the growth medium, MeCbl did not promote the expression of MBP in SCs (Figure 4C). On the other hand, the expression of MBP in SCs cultured in the differentiation medium was increased by the administration of MeCbl (Figure 4C).

Abundant lipid and cholesterol biosynthesis is necessary for the SCs myelination of peripheral nerves. Recent experiments using microarray analyses of the myelination process during development and the remyelination process after nerve injury have revealed that cholesterol/lipid metabolism in peripheral nerve myelination is also important (Nagarajan et al., 2002; Verheijen et al., 2003). Next, we examined the activation of Acly, which is a marker of lipid-synthesis enzymes. Together with the expression of MBP, MeCbl promoted the expression of Acly in SCs in the differentiation medium but not in the growth medium (Figure 4D). In addition, to clarify the increasing expression of MBP by MeCbl, we applied the following experiment.

Early studies have shown that after the proliferation stage SCs differentiate into myelination process and its initiation is the stimulation by direct physical contact with axonal membranes (Wood and Bunge, 1975; Salzer et al., 1980). To date, the culture consisting SCs and primary DRG neurons is the unique and stable in vitro system and it enables to support both active proliferation and myelination by purified isolated populations of postnatal rodent SCs (Eldridge et al., 1987). Therefore, we used a DRGs/SCs coculture system to estimate MeCbl-mediated myelination. DRGs/SCs cocultures were maintained for 21 days to examine the effect of MeCbl on the differentiation of axon-related SCs. The extent of myelination was quantified by measuring the number of myelinated sections. No significant differences in the number of myelin segments were observed on day 7 (Figures 5A–F,S). On day 14 after the induction of differentiation, MeCbl increased the number of MBP-positive segments by approximately twofold compared with that observed by the control (Figures 5G–L,S). In both groups, the differentiation of SC progressed for 21 days and the number of MBP-positive segments in the MeCbl group was approximately 1.5-fold compared with that observed in the control group (Figures 5M–S). Collectively, these results provide strong evidence that MeCbl promotes the differentiation and myelination of SCs in vitro.

To make the focal demyelination model of the sciatic nerve, we used LPC that dissolves myelin sheaths and leads to pure demyelination lesions which spontaneously remyelinate over time (Smith and Hall, 1980; Stoll et al., 1993; Wessig et al., 2007; Zhang et al., 2010). To confirm the effect of LPC, cross-sections of sciatic nerves were prepared and demyelination was assessed by immunofluorescence for MBP (Figures 6A–D). On 7 days postoperatively, MBP positive axons were clearly visible in saline injected LPC (-) group (Figures 6C,D), whereas it was rarely observed in the LPC (+) group (Figures 6B,D). Therefore, we confirmed that damages by needle insertion itself were absent in the sciatic nerves (Figures 6A–D). Furthermore, we confirmed that this demyelination model did not cause axonal damages because many neurofilament positive axons in LPC (+) group were detected (Figure 6B). Seven days after the injection to the sciatic nerve, saline or MeCbl was systemically administered with an osmotic pump. In the LPC (-) groups with or without MeCbl, the numbers of MBP positive axons are similar (Figures 6E–H,M) and same as that in normal sciatic nerve (Figure 6A) 7 days after systemic administration (14 days after the injection to the sciatic nerve). These results indicate that MeCbl did not affect the myelin of the normal sciatic nerve. The number of MBP positive axons decreased to 7% compared with the normal nerve 7 days after the LPC injection (Figures 6B,D), but it spontaneously increased to 41% 14 days after the LPC injection (Figures 6I,J,M). An administration of MeCbl for 7 days accelerated the recovery of remyelination and number of MBP positive axons was approximately twofold compared with that in LPC (+) group without MeCbl (Figures 6I–M). These results demonstrate that MeCbl promotes the remyelination of LPC-induced demyelination rat model.

To evaluate motor functional recovery, we administered a sciatic functional index test to the rats. At 2 weeks after the LPC injection, sciatic functional index values in the MeCbl (+) group were significantly higher than those recorded in the MeCbl (-) group (Figure 7A). For electrophysiological evaluation, compound muscle action potential and nerve conduction velocity were obtained by stimulating sites proximal and distal to the injection. Regarding compound muscle action potential, the amplitude was not lowered by the LPC injection and MeCbl did not affect it (Figure 7B). This was presumably because compound muscle action potential is affected by the number of axons and LPC does not lower the number of axons (Figure 6B). On the other hand, MeCbl treatment kept nerve conduction velocity at a normal level, whereas the LPC injection decreased nerve conduction velocity to a value <20 m/s (Figure 7C). These functional evaluations show that MeCbl promotes functional recovery in rat sciatic nerve demyelination models. The response latency on hot plate significantly increased by approximately 3.8 and 3.5-fold (Figure 7D) in LPC (+) MeCbl (-) group compared to those in LPC (-) MeCbl (-) and LPC (-) MeCbl (+) groups. Treatment with administration of MeCbl for LPC injection significantly improved of response latency in administration of saline for LPC injection (Figure 7D). The von-Frey monofilament test was employed to determine withdrawal thresholds to mechanical stimuli applied on the hind paw. MeCbl treatment for LPC injection significantly reduced compared with saline treatment for LPC injection and was improved to control level (Figure 7E). These results demonstrate that MeCbl promotes motor and sensory functional recovery in a rat demyelination model.