kts

M

okaSuita

inudiemaoutbol

methylcobalamin increases Erk1/2 and Akt activities through the methylation cycle. In a rat sciatic nervedministration of high doses of methylcobalamin improves nerve regeneration

organs.

tant focauseon of

Experimental Neurology 222 (2010) 191203

Contents lists available at ScienceDirect

Experimental

.e ladenosylcobalamin (AdoCbl). In mammalian cells, CNCbl and OHCblare inactive forms, AdoCbl acts as a coenzyme of methylmaronyl Co-A mutase in mitochondria, and MeCbl acts as a coenzyme ofmethionine synthase (MS), which is required for the formation ofmethionine from homocysteine in the methylation cycle thatinvolves methylation of DNA or proteins (Banerjee and Ragsdale,2003; Ghosh et al., 1991; Pfohl-Leszkowicz et al., 1991; Toohey,2006). There are reports that vitamin B12, including MeCbl, has abenecial effect on the nervous system. In vitro, the vitamin Bcomplex, including vitamin B12, promotes neurite outgrowth andvitamin B12-enriched medium produces the greatest mean neurite

on neurons, the most effective analog and its mechanism of action onneurons remains to be claried.

In the nervous system, the protein kinases Erk1/2 and Akt areactivated by certain neurotrophins, such as nerve growth factor (NGF)and brain-derived neurotrophic factor (BDNF) (Segal, 2003). Activa-tion of Erk1/2 promotes neurite outgrowth and Akt mediatesbranching of neurites in dorsal root ganglion (DRG) neurons (Markuset al., 2002). The survival of cerebellar granule neurons (CGN) ispromoted by the activation of Erk1/2 (Bonni et al., 1999) and Akt(Bhave et al., 1999). The activation of Erk1/2 and Akt induced by theFGL peptide (a broblast growth factor receptor agonist) results inoutgrowth (Fujii et al., 1996). MeCbl proteretinal cell cultures against glutamate cytotoKikuchi et al., 1997). In in vivo studies, high d

Methylcobalamin promotes nerve regeneration. Corresponding author. Department of Orthopaedic

School ofMedicine, 2-2 Yamadaoka, Suita, Osaka 565-087E-mail address: tanahiro-osk@umin.ac.jp (H. Tanaka

0014-4886/$ see front matter 2009 Elsevier Inc. Adoi:10.1016/j.expneurol.2009.12.017the spinal cord (Scalab-such as cyanocobalamincobalamin (OHCbl), and

were degenerating in the anterior gracile muscle of the gracile axonaldystrophy mutant mouse (Yamazaki et al., 1994). Despite theseprevious reports about the effects of vitamin B12, including MeCbl,rino et al., 1990). Vitamin B12 has analogs(CNCbl), methylcobalamin (MeCbl), hydroxoCerebellar granule neuronDorsal root ganglionNeurite outgrowthNeuronal survival

Introduction

Vitamin B12 (cobalamin) is importhe nervous system and its deciencycalled subacute combined degenerati 2009 Elsevier Inc. All rights reserved.

r normal functioning ofs a systemic neuropathy

nerve conduction and regeneration in a rat sciatic nerve injury model(Yamatsu et al., 1976a,b), streptozotocin-diabetic rats (Sonobe et al.,1988), and experimental acrylamide neuropathy (Watanabe et al.,1994). MeCbl promoted regeneration of motor nerve terminals thatcts cortical neuron andxity (Akaike et al., 1993;oses of MeCbl improved

neurite outgrow(Neiiendam et a

In this study,several vitaminneuronal survivaeffects are broushow that MeCrecovery in a rat

s, Osaka University Graduate1, Japan. Fax:+81 668793559.).

ll rights reserved.Sciatic nerve and functional recovery. Therefore, methylcobalamin may provide the basis for better treatments ofnervous disorders through effective systemic or local delivery of high doses of methylcobalamin to targetKeywords:Methylcobalamin injury model, continuous aMethylcobalamin increases Erk1/2 and Aand promotes nerve regeneration in a rat

Kiyoshi Okada a, Hiroyuki Tanaka a,b,, Ko Temporin a,Tsuyoshi Murase a, Hideki Yoshikawa a

a Department of Orthopaedics, Osaka University Graduate School of Medicine, 2-2 Yamadab Medical Center for Translational Research, Osaka University Hospital, 2-15 Yamadaoka,

a b s t r a c ta r t i c l e i n f o

Article history:Received 28 August 2009Revised 24 November 2009Accepted 17 December 2009Available online 4 January 2010

Methylcobalamin is a vitamAlthough some previous stmechanism of this effect r100 nM promotes neuritemethylation cycle, a meta

j ou rna l homepage: wwwactivities through the methylation cycleciatic nerve injury model

ichio Okamoto a, Yusuke Kuroda a, Hisao Moritomo a,

, Suita, Osaka 565-0871, Japan, Osaka 565-0871, Japan

B12 analog and is necessary for the maintenance of the nervous system.es have referred to the effects of methylcobalamin on neurons, the preciseins obscure. Here we show that methylcobalamin at concentrations abovegrowth and neuronal survival and that these effects are mediated by theic pathway involving methylation reactions. We also demonstrate that

Neurology

sev ie r.com/ locate /yexnrth and neuronal survival in primary rat neuronsl., 2004).we demonstrate that MeCbl is the most effective ofB12 analogs in promoting neurite outgrowth andl with activation of Erk1/2 and Akt and that theseght about through the methylation cycle. We alsobl promotes nerve regeneration and functionalmodel of sciatic nerve injury.

192 K. Okada et al. / Experimental Neurology 222 (2010) 191203Materials and methods

Cell culture

DRG neurons were cultured as described previously (Higuchi et al.,2003). DRG were obtained fromWistar rats at postnatal day (P9) anddissociated by incubation with 0.25% trypsin, 0.1% collagenase, and200 U/ml DNase I for 30 min at 37 C. The enzymatic reaction wasblocked by adding Dulbecco's modied Eagle's medium (DMEM, notincluding vitamin B12) containing 10% fetal bovine serum, and aftertrituration and centrifugation, the cells were resuspended in modiedSato medium (DMEM containing 5 g/ml insulin, 20 nM progester-one, 100 Mputrescine, 30 nM sodium selenite, 0.1 g/ml L-thyroxine,0.08 g/ml triiodo-L-thyronine, and 4 mg/ml bovine serum albumin)and plated on poly-L-lysine-coated four-well chamber slides. Prolif-eration of non-neuronal cells was prevented by addition of 10 M 5-uoro-2-deoxyuridine.

For CGN culture, the cerebellum was removed from Wistar rats atP9 as described previously (Mimura et al., 2006) and dissociated byincubation with 0.25% trypsin and 200 U/ml DNase I for 30 min at37 C. The enzymatic reaction was blocked by adding DMEMcontaining 10% fetal bovine serum, and after trituration andcentrifugation, the cells were resuspended in modied Sato medium.The cells were centrifuged and resuspended in modied Sato mediumagain, and then plated on poly-L-lysine-coated culture dishes or four-well chamber slides.

Immunocytochemistry

The four-well chamber or glass slides were xed in 4% parafor-maldehyde (PFA) for 30 min, blocked for 1 h, and incubated overnightat 4 C with primary antibody, followed by incubation for 1 h at roomtemperature with secondary antibody as described previously(Temporin et al., 2008a,b), and nuclei were labeled with DAPI(Wako Pure Chemical Industries, Osaka, Japan). For the terminaltransferase dUTP nick-end-labeling (TUNEL) assay, the slides werestained with the DeadEnd Fluorometric TUNEL System (PromegaCorporation, Madison, WI) after incubation with secondary antibody.The primary antibodywas anti-neural class III -tubulin (TuJ1) mousemonoclonal antibody (1:1000; Covance, Berkeley, CA), and thesecondary antibodies were Alexa 488-labeled goat anti-mouse IgGantibody (1:1000; Molecular Probes, Eugene, OR) or Alexa 568-labeled goat anti-mouse IgG antibody (1:1000; Molecular Probes).

Neurite outgrowth assay

Neurons were cultured with CNCbl (10 M; Sigma-Aldrich, St.Louis, MO), MeCbl (1 nM100 M; Sigma-Aldrich), OHCbl (10 M;Sigma-Aldrich), AdoCbl (10 M; Sigma-Aldrich), S-adenosylmethio-nine (SAM, 10 M; Sigma-Aldrich), 4-nitro-2,1,3-benzothiadiazole(Nbtd, 10 M; Sigma-Aldrich), U0126 (10 M; Cell SignalingTechnology, Beverly, MA), LY294002 (50 M; Cell Signaling Technol-ogy), or caspase inhibitor Z-VAD-FMK (Z-VAD, 20 M; PromegaCorporation) for 72 h and then immunostained with anti-TuJ1antibody. Total neurite length per neuron and the axonal length (thelength of the longest neurite per TuJ1-positive neuron)weremeasuredusing an image analyzer (Lumina Vision; Mitani Co., Fukui, Japan) asdescribed previously (Tanaka et al., 2002). Only neurites longer than20 m (about the diameter of a soma) and not in contact with othercells were measured. The mean total neurite length and axonal lengthwere calculated from at least 30 neurons in each experiment.

Neuronal survival assay

Neurons were cultured with CNCbl (10 M), MeCbl (1 nM

100 M), AdoCbl (10 M), OHCbl (10 M), SAM (10 M), Nbtd (1 M),U0126 (10 M), or LY294002 (50 M) for 72 h. Kainic acid (100 M;Calbiochem, San Diego, CA) was added to themedium of CGN cultures48 h after starting the culture, and thereafter, the neurons werecultured for 24 h. The neurons were visualized by staining with anti-TuJ1 antibody and apoptotic cells were visualized in the TUNEL assay.Normal and apoptotic cells were counted in ten randomized 200elds using an image analyzer (Lumina Vision). The apoptotic rateswere calculated by dividing the number of apoptotic cells by the totalnumber of cells. More than 100 cells were counted in eachexperiment.

Western blotting

CGN were cultured with BDNF (100 ng/ml; Sigma-Aldrich) orMeCbl (10 M). At time points for 5 min to 72 h, cultured cells werehomogenized with Kaplan buffer (50 mM Tris, pH 7.4, 150 mM NaCl,10% glycerol, 1% NP40, and protease inhibitor cocktail) and claried bycentrifugation. Theywere separated by 12% SDS-PAGE and transferredto polyvinylidene diuoride membranes. After blocking with 5% skimmilk for 1 h, the membranes were incubated with anti-p44/42 MAPKrabbit polyclonal antibody (1:1000; Cell Signaling Technology), anti-phospho-p44/42 MAPK rabbit monoclonal antibody (1:1000; CellSignaling Technology), anti-Akt rabbit polyclonal antibody (1:1000;Cell Signaling Technology), or anti-phospho-PKB (phospho-Akt)(pSer473) rabbit polyclonal antibody (1:1000; Sigma-Aldrich) at4 C overnight, followed by incubation with horseradish peroxidase-conjugated secondary antibody (1:1000; GE Healthcare, LittleChalfont, UK) and ECL reagents (GE Healthcare). The densities ofErk1/2, phospho-Erk1/2, Akt, and phospho-Akt were measured usingScion Image software (Scion Corporation, Frederick, MD). To calculatethe normalized density, the density of phospho-Erk1/2 or phospho-Akt was divided by the density of Erk1/2 or Akt, respectively, in thesame membrane.

Surgical procedures

All animal experiments were approved by the Ethics ReviewCommittee for Animal Experimentation of Osaka University.

Eight-week-old femaleWistar rats (180220 g) were anesthetizedby intraperitoneal injection of sodium pentobarbital (50 mg/kg). Theleft sciatic nerve was exposed at mid-thigh level and cut withmicroscissors. The dissection site was sutured end to end with 10-0nylon, and the skin was then sutured with 3-0 nylon. Continuousadministration of phosphate-buffered saline (PBS) or MeCbl (1 mg/kg/day) in PBS for 12 weeks was performed using an osmoticminipump (Model 2ML4; Alzet, Cupertino, CA), which was placedsubcutaneously in the right side of the back. All surgeries wereperformed by the same surgeon.

Serum concentration of vitamin B12

Blood samples (0.5 ml) were collected from the jugular vein or theright ventricle of rats under sodium pentobarbital anesthesia 3 daysafter administration. Samples were centrifuged for 20 min at 800 gand the supernatant was collected. Serum concentrations of vitaminB12 were measured from four rats from the MeCbl and PBS groupeach, by chemiluminescent immunoassay performed at Bio MedicalLaboratories, Tokyo, Japan.

Motor and sensory functional analysis

To evaluate sensory function, the toe pinch test was performedevery week, in which the reaction to pinching the third, fourth, andfth toes with ne forcepswas graded based on thewithdrawal reex,escape behavior, and vocalization (0, normal, brisk withdrawal reex,

escape behavior, and strong vocalization; 1, mildly impaired; 2,

193K. Okada et al. / Experimental Neurology 222 (2010) 191203moderately impaired; 3, totally impaired) (Kau et al., 2006). Toevaluate motor function, the toe spreading test was performed everyweek and the sciatic functional index (SFI) was evaluated every 2weeks (Temporin et al., 2008a,b; Tomita et al., 2007). In the toespreading test, the appearance of a toe-spreading reex when the tailwas lifted was estimated (2, normal spreading; 1, intermediatespreading; 0, no spreading) (Azzouz et al., 1996). For SFI measure-ment, the rats were allowed to walk a 60-cm-long track after pressingtheir hind paws onto an ink pad every 2 weeks and SFI was calculatedfrom their footprints using a formula developed previously (deMedinaceli et al., 1984). All functional studies were performed by thesame blinded observer who did not knowwhich rats were included incontrol group or in MeCbl group, respectively.

Electrophysiological analysis

Twelve weeks post operation, rats were anesthetized withsodium pentobarbital and placed on a smooth table. The sciaticnerve and tibialis anterior muscle were exposed. Terminal latenciesand compound muscle action potentials (cMAP) were obtained bystimulating the sciatic nerve with bipolar electrodes. A pair ofstimulating electrodes was gently placed on the proximal side 2 mmfrom the sutured site. The cMAP was recorded with a pair ofrecording needle electrodes; one of them was inserted in themiddle of the tibialis anterior muscle and another was placed in thedistal musculotendinous junction of the tibialis anterior muscle.Terminal latency and cMAP were detected and measured usingPowerLab devices and software (AD Instruments, Bella Vista, NSW,Australia).

Morphological and histological analysis

The rats were anesthetized with sodium pentobarbital andtranscardially perfused with PBS, followed by 4% PFA at 6 and 12weeks post operation. The sciatic nerve was then removed,embedded in parafn, sectioned axially at 5 m, and mounted on aglass slide. The slides were stained with 0.1% toluidine blue solution,and the number of myelinated axons was counted using NIS-Elements BR 3.00, SP3 software (Laboratory Imaging, Nikon, Tokyo,Japan). The other slides were xed in 4% PFA for 30 min, blocked for 1h, and incubated overnight at 4 C with primary antibody, followedby incubation for 1 h at room temperature with secondary antibody.The primary antibodies used were anti-neurolament 200 (NF)rabbit antiserum (1:1000; Sigma-Aldrich) and anti-myelin basicprotein (MBP) mouse antibody (1:100; Covance). The secondaryantibodies were Alexa 488-labeled goat anti-rabbit IgG antibody(1:1000; Molecular Probes) and Alexa 568-labeled goat anti-mouseIgG antibody (1:1000; Molecular Probes). The gross area of cross-sectional slices 2 and 10 mm distal to the sutured site werevisualized by staining for NF or MBP, which was measured andcalculated using NIS-Elements BR 3.00, SP3 software.

Statistics

Data are expressed as meanSEM. Statistical evaluation wasperformed by one-way ANOVA and post hoc Student's t-test usingStatView software, version 5.0 (SAS Institute, Cary, NC).

Results

MeCbl promotes neurite outgrowth in CGN and DRG neurons

Analogs of vitamin B12 include CNCbl, MeCbl, OHCbl, and AdoCbl,with each possessing different metabolic characteristics. We per-formed neurite outgrowth assays with these analogs in CGN to

determine the analog with the greatest effect on neurite outgrowth.Axonal length and total neurite length in the control group were124.16.0 m and 191.77.8 m, respectively; values were higherin the MeCbl group: 206.79.6 m and 291.812.5 m, respec-tively (Figs. 1a, c, f, and g). Furthermore, values in the MeCbl groupwere greater than those in the CNCbl group (176.67.8 m and255.210.1 m, respectively), the OHCbl group (168.08.0 m and236.19.7 m), and the AdoCbl group (137.34.7 m and 185.35.6 m) (Figs. 1bg). These results show that MeCbl was the analogwith the greatest effect on neurite outgrowth. We then investigatedwhether MeCbl promoted neurite outgrowth in DRG neurons aswell. The axonal length in MeCbl (266.315.1 m) was greater thanthat in the control group (179.710.1 m) (Figs. 1ik). On the basisof a previous report, vitamin B12 was used in this experiment at theconcentration of 10 M (Akaike et al., 1993). However, it remains tobe determined which concentration is most appropriate for neuriteoutgrowth. To examine the effect of MeCbl concentration on neuriteoutgrowth, CGN were cultured with MeCbl at concentrations 1 nM100 M. MeCbl promoted neurite outgrowth in CGN in a concen-tration-dependent manner and a signicant difference compared tocontrol was revealed at concentrations 100 nM (Fig. 1h). Thelongest axonal length was observed at 10 M concentration.

MeCbl promotes neurite outgrowth through the methylation cycle

MeCbl is a coenzyme of MS, which forms methionine fromhomocysteine in mammalian cells, and it is also known that MeCblacts as a methyl donor in certain methylation reactions through themethylation cycle (Toohey, 2006). To determine whether the effectof MeCbl on neurite outgrowth was mediated through this cycle, wecultured CGN and DRG neurons with MeCbl in the presence orabsence of Nbtd, an inhibitor of MS (Banks et al., 2007). Nbtd wasadded at 1 M concentration, which neither diminished the axonallength (Figs. 2k and l) nor increased the rate of apoptosis of neurons(Figs. 4k and l). However, after addition of Nbtd, the axonal length inthe MeCbl group was decreased to the level same as that in thecontrol group in CGN (Figs. 2ac and k) and DRG neurons (Figs. 2fh and l). Next, we evaluated the effects of SAM, the downstreammetabolite of methionine in the methylation cycle, in the presenceor absence of Nbtd. SAM increased the axonal length to 209.111.5 m in CGN (Figs. 2d and k) and to 231.613.5 m in DRGneurons (Figs. 2i and l). Unlike what was observed when Nbtd wasadded in the MeCbl group, addition of Nbtd in the SAM group didnot reduce the axonal length (Figs. 2e and jl). Then, to determinewhether MeCbl and SAM promoted neurite outgrowth through thesame metabolic pathway, neurons were cultured with both MeCbland SAM in the presence or absence of Nbtd. In the MeCbl + SAMgroup, the axonal length was increased to 218.111.6 m in CGNand to 238.411.7 m in DRG neurons, and these effects were notdiminished by the addition of Nbtd (Figs. 2k and l). Furthermore, theaxonal length in the MeCbl + SAM group showed no signicantdifference from that in the MeCbl group or the SAM group. Insummary, we showed that MeCbl and SAM had similar effects onneurite outgrowth but SAM did not increase the effect of MeCbl, andNbtd only inhibited the effect of MeCbl. These results led us toconsider that MeCbl promotes neurite outgrowth through themethylation cycle, which is a metabolic pathway related to bothMeCbl and SAM.

MeCbl promotes neuronal survival in CGN and DRG neurons

In cortical neuron culture and retinal neuron culture, MeCblpromotes neuronal survival (Akaike et al., 1993; Kikuchi et al., 1997).However, the analog of vitamin B12 that is most effective inpromoting neuronal survival remains unelucidated. To determinewhether MeCbl or other vitamin B12 analogs promoted neuronal

survival, we performed a neuronal survival assay in CGN and DRG

194 K. Okada et al. / Experimental Neurology 222 (2010) 191203neurons with four vitamin B12 analogs. In this experiment, kainicacid, which increased the apoptosis of neurons (Liao et al., 2002), wasadded to CGN culture for 24 h because the apoptotic rate of CGNwithout kainic acid was too small (9.31.5%) (Fig. 6i) to evaluate thedifference in neuronal survival in each group. In the control group,the apoptotic rate was 42.72.7%, and MeCbl decreased the rate to15.20.3%, which was less than that in the CNCbl group (25.51.6%), the OHCbl group (28.63.8%), and the AdoCbl group (45.42.7%) (Figs. 3af). Thus, it was shown that MeCbl was the mosteffective analog for neuronal survival as well as neurite outgrowth. InDRG neurons, MeCbl also decreased the apoptotic rate from 32.44.7% to 15.8 3.6% (Figs. 3hj). Then, we examined the effect ofMeCbl concentration on neuronal survival. CGN were cultured in thepresence of kainic acid with MeCbl at several concentrations. MeCbldiminished the apoptotic rate in a concentration-dependent mannerand there were signicant differences compared to control atconcentrations 100 nM (Fig. 3g).

Fig. 1. Methylcobalamin promotes neurite outgrowth in CGN and DRG neurons. (ae) CGmicrographs of CGN in each vitamin analog group are shown. Neurons and neurites were vistheMeCbl group than in the control group. (f and g) Axonal length (f) and total neurite lengthCGN culturedwithMeCbl at concentrations 1 nM100 M.MeCbl increased axonal length in Ch in the presence or absence of MeCbl. Neurons and neurites were visualized by staining wiwas greater in the MeCbl group than in the control group. Signicance was determined by onare representative of three independent experiments. pb0.05, pb0.01.MeCbl promotes neuronal survival through the methylation cycle

As shown in Fig. 2, MeCbl promotes neurite outgrowth through themethylation cycle. Therefore, we hypothesized that the effect ofMeCbl on neuronal survival also occurred through the methylationcycle. To examine this hypothesis, we performed a neuronal survivalassay in a manner similar to the neurite outgrowth assay. Addition ofNbtd raised the apoptotic rate in the MeCbl group to the level same asthat in the control group in CGN (Figs. 4ac and k) and DRG neurons(Figs. 4fh, and l). SAM reduced the apoptotic rate to 23.24.7% inCGN (Fig. 4k) and 16.73.7% (Fig. 4l) in DRG neurons, which was notinhibited by addition of Nbtd. Furthermore, the apoptotic rate in theMeCbl + SAM group was 20.44.6% in CGN (Fig. 4k) and 11.91.8%in DRG neurons (Fig. 4l), not signicantly different from that in theMeCbl or SAM group (Figs. 4k and l). These results are similar to theresults of the neurite outgrowth assay and also suggest that MeCblpromotes neuronal survival through the methylation cycle.

N from P9 rats were cultured with four vitamin B12 analogs for 72 h. Fluorescenceualized by staining with anti-TuJ1 antibody. The increase in axonal length was greater in(g) of CGN after culture with each of the four vitamin B12 analogs. (h) Axonal length ofGN at concentrations100 nM. (i and j) DRG neurons from P9 rats were cultured for 72th anti-TuJ1 antibody. (k) Axonal length in DRG neurons. The increase in axonal lengthe-way ANOVA and the post hoc Student's t-test. Error bars indicate meanSEM. Results

195K. Okada et al. / Experimental Neurology 222 (2010) 191203MeCbl increases Erk and Akt activities through the methylation cycle

In CGN and DRG neurons, it is known that neurotrophic factorssuch as NGF and BDNF increase neurite outgrowth or neuronalsurvival by activation of Erk1/2 or Akt (Bhave et al., 1999; Bonni et al.,1999;Markus et al., 2002; Neiiendam et al., 2004; Schmid et al., 2000).In this study, we showed that MeCbl promoted neurite outgrowth(Fig. 2) and neuronal survival (Fig. 4) through the methylation cycle.Therefore, we hypothesized that MeCbl affected Erk1/2 and Aktactivities through themethylation cycle. First, we detected the activityof Erk1/2 and Akt in CGN cultured for 72 h with MeCbl. MeCblproduced a 1.80.3-fold stronger activation of Erk1/2 and a 1.70.1-fold stronger activation of Akt than that observed in the controlgroup (Figs. 5ac). Then, to investigate the relationship between themethylation cycle and phosphorylation of Erk1/2 and Akt, CGN werecultured in the presence or absence of MeCbl, SAM, or Nbtd. Addition

Fig. 2.Methylcobalamin promotes neurite outgrowth through the methylation cycle. (aj) Cstaining with anti-TuJ1 antibody. The increase in neurite outgrowth was greater in the MeCmicrographs of neurons in the MeCbl (c and h) and SAM groups (e and j) affected by additionthe presence or absence of MeCbl, SAM, and Nbtd. Signicance was determined by one-wayrepresentative of three independent experiments. pb0.01.of Nbtd inhibited the activation of Erk1/2 and Akt promoted byMeCbl(Figs. 5dg). SAM increased Erk1/2 activity 2.50.2-fold and Aktactivity 2.60.2-fold compared with the control group, and theseeffects were not inhibited by addition of Nbtd. These ndings suggestthat MeCbl increases Erk1/2 and Akt activities through the methyl-ation cycle. In addition, to investigate the mechanism by which MeCblaffected the phosphorylation of Erk1/2 and Akt, we compared theeffects of MeCbl and BDNF on Erk1/2 and Akt activities. CGN werecultured with BDNF or MeCbl for 5 min to 72 h. BDNF led to a 7.22.8-fold stronger activation of Erk1/2 (Figs. 5h and i) and a 6.71.5-fold stronger activation of Akt (Figs. 5j and k) than that observed inthe control group at 5 min after addition, and activation of Erk1/2 andAkt was maintained for 1 h (Figs. 5i and k). MeCbl produced 3.90.3-fold stronger activation of Erk1/2 (Figs. 5h and i) and 4.00.1-foldstronger activation of Akt (Figs. 5j and k) 10 min after addition, andactivation was maintained for 72 h (Figs. 5i and k). Addition of both

GN (ae) and DRG neurons (fj) from P9 rats were cultured for 72 h and visualized bybl (b and g) and SAM groups (d and i) than in the control group (a and f). Fluorescenceof Nbtd are shown. (k and l) Axonal length in CGN (k) and DRG neurons (l) cultured inANOVA and the post hoc Student's t-test. Error bars indicate meanSEM. Results are

196 K. Okada et al. / Experimental Neurology 222 (2010) 191203BDNF and MeCbl activated Erk1/2 and Akt to the similar level to thataffected only by BDNF over the rst hour of application, but theactivation was more increased than that affected by BDNF or MeCbl at24 and 72 h from application (Figs. 5h-k). These results suggest thatMeCbl has a means of activating Erk1/2 and Akt that differs from thatof BDNF.

MeCbl promotes neurite outgrowth and neuronal survival with Erk andAkt activities

To reveal whether neurite outgrowth and neuronal survival werepromoted by the activation of Erk1/2 and Akt, we used U0126 andLY294002 to inhibit the phosphorylation of Erk1/2 and Akt,respectively, in a neurite outgrowth and neuronal survival assay.Different from the previous assay, herewe demonstrated the neuronalsurvival assay without kainic acid. Apoptotic rates were increased inthe U0126 group (47.35.7%), LY294002 group (45.11.8%), andU0126 + LY294002 group (82.33.4%) (Fig. 6i). Addition of MeCbldid not reduce the increased apoptotic rate of neurons induced byU0126 and/or LY294002 in a neuronal survival assay. Under theinuence of U0126 and/or LY294002, the neurites of most CGN weretoo short for proper evaluation because half or more than half of

Fig. 3. Methylcobalamin promotes neuronal survival of CGN and DRG neurons. (ae) CGNstaining with anti-TuJ-1 antibody (red) and the TUNEL assay (green). There were fewer TUNfour vitamin B12 analog groups. (g) Apoptotic rate in CGN with MeCbl at concentrations frFluorescence micrographs of DRG neurons from P9 rats cultured with or without MeCbl fo(green). (j) Apoptotic rate in DRG neurons. MeCbl decreased apoptosis in DRG neurons. Signiindicate meanSEM. Results are representative of three independent experiments. pb0.0neurons were apoptotic. Therefore, to avoid the inuence of apoptosisin a neurite outgrowth assay, we applied Z-VAD, which decreases theapoptosis by inhibiting the activity of caspase (Boise and Thompson,1997, Robertson et al., 2000). Under the application of Z-VAD,apoptotic rates were decreased to the control level in U0126 group(6.02.8%), in LY294002 group (4.31.7%), and in U0126 +LY294002 group (7.41.3%), and the neurites became appropriatefor evaluation. In the neurite outgrowth assay with Z-VAD, MeCblcould not increase axonal length with U0126 or LY294002 (Figs. 6afand j). In U0126 + LY294002 group, axonal length was shorter thanthat in control group (Figs. 6g, h, and j), which was not elongated byaddition of MeCbl. These results suggest that MeCbl increases neuriteoutgrowth and neuronal survival with Erk and Akt activities.

MeCbl promotes the recovery of injured sciatic nerve and its function

Weshowed thatMeCbl promotedneurite outgrowth (Figs. 1 and2)and neuronal survival (Figs. 3 and 4) in CGN and DRG neurons invitro. Then, we examined the effects of MeCbl in vivo in a rat model ofsciatic nerve injury. We cut the left sciatic nerve of rats at the mid-thigh level. The nerve was sutured end to end and we continuouslyadministered MeCbl (1 mg/kg/day) or PBS for 12 weeks. Because we

from P9 rats were cultured with four vitamin B12 analogs for 72 h and visualized byEL-positive cells in the MeCbl group than in the other groups. (f) Apoptotic rate in theom 1 nM to 100 M. MeCbl decreased apoptosis at concentrations 100 nM. (h and i)r 72 h and visualized by staining with anti-TuJ-1 antibody (red) and the TUNEL assaycance was determined by one-way ANOVA and the post hoc Student's t-test. Error bars5, pb0.01.

197K. Okada et al. / Experimental Neurology 222 (2010) 191203showed in our in vitro study that MeCbl did not affect neurons atconcentrations below 10 nM, we checked the serum concentration ofvitamin B12 in rats before functional and histological evaluation.Serum was obtained 3 days after surgery because the concentrationof vitamin B12 in serum was increased immediately and maintainedby its repeated administration (Nava-Ocampo et al., 2005). Theconcentration of vitamin B12 was 130.96.9 nM in the MeCbl groupand 669.266.8 pM in the control group. We considered that theconcentration in the MeCbl group was appropriate because MeCblpromoted neurite outgrowth and neuronal survival at concentrationsabove 100 nM (Figs. 1 and 3). Then, we observed functional recoveryfor 12 weeks. To evaluate the sensory function, we performed pinchtests of the third, fourth, and fth toes. The sensory function of thethird toe in the MeCbl group was not signicantly different from thatin the control group except at 8 weeks post operation (Fig. 7a). Onthe other hand, the sensory function of the fourth and fth toes in theMeCbl group improved more than that in the control group from 10

Fig. 4.Methylcobalamin promotes neuronal survival through the methylation cycle. (aj) CGMeCbl, SAM, and Nbtd for 72 h and visualized by staining with anti-TuJ1 antibody (red) andand SAM (d and i) groups. Micrographs of neurons in the MeCbl group (c and h) and the SAM(k) and DRG neurons (l) cultured in the presence or absence of MeCbl, SAM, and Nbtd. Signiindicate meanSEM. Results are representative of three independent experiments. pb0.0weeks post operation (Figs. 7b and c). Motor functional recovery wasbetter in the MeCbl group than in the control group from 11 weekspost operation in the toe-spreading test (Fig. 7d). Furthermore, thecourse of SFI recovery in the MeCbl group was signicantly differentfrom that in the control group at 5 weeks post operation, and the nalvalue in the MeCbl group (46.75.1) was better than that in thecontrol group (67.71.7) (Fig. 7e). For electrophysiologicalevaluation, cMAP and terminal latency were obtained by stimulatingthe side proximal to the sutured site. Though we did not detect asignicant difference in cMAP between the control and MeCbl groups(data not shown), recovery of terminal latency was greater in theMeCbl group than in the control group (Fig. 7f). These functionalevaluations showed that MeCbl promoted motor and functionalrecovery in rat sciatic nerve injury models.

Next, we examined the morphological and histological recovery ofaxons and myelin to determine whether these ndings resulted fromthe regeneration of axons and myelin. For anatomical and

N (ae) and DRG neurons (fj) from P9 rats were cultured in the presence or absence ofthe TUNEL assay (green). Numbers of apoptotic cells were lower in the MeCbl (b and g)group (e and j) affected by addition of Nbtd are shown. (k and l) Apoptotic rate in CGNcance was determined by one-way ANOVA and the post hoc Student's t-test. Error bars5.

198 K. Okada et al. / Experimental Neurology 222 (2010) 191203morphological evaluation, the axial slices of sciatic nerve werestained with toluidine blue and observed. The number of myelinatedaxons was increased both at 2 mm (Figs. 8a, b, e, f and n) and 10 mm(Figs. 8c, d, g, h and n) distal to the sutured site at 6 weeks and 12

Fig. 5. Methylcobalamin increases Erk1/2 and Akt activities through the methylation cycleactivities were detected by Western blotting. Erk1/2 and Akt activities were increased in the(dg) Erk1/2 (d) and Akt (e) activity in the presence or absence of MeCbl, SAM, or Nbtd for 10and Akt activities, an effect that was inhibited by addition of Nbtd. SAM also increased Erk1/2(h) and Akt (j) was observed for 5 min to 72 h after addition of BDNF (100 ng/ml) and/or Megroup than in the BDNF group. Quantication of normalized density of Erk1/2 (i) and Akt (k)h in the BDNF group, but for 72 h in the MeCbl group. Signicance was determined by one wrepresentative of three independent experiments. pb0.05.weeks post operation. It was also demonstrated that the formation ofmyelin was promoted by MeCbl, and especially the bers weresurrounded with thick myelin in MeCbl group at 12 weeks postoperation (Figs. 8f and h). For histological evaluation, the gross area

. (ac) CGN from P9 rats were cultured with MeCbl (10 M) for 72 h. Erk1/2 and AktMeCbl group. Quantication of normalized density of Erk1/2 (b) and Akt (c) is shown.min. Quantication of Erk1/2 (f) and Akt (g) activity is shown.MeCbl increased Erk1/2and Akt activities, which was not inhibited by addition of Nbtd. (hk) Activity of Erk1/2Cbl (10 M). Activation of Erk1/2 and Akt was observed for a longer period in the MeCblin the BDNF and/orMeCbl groups, respectively. Erk1/2 and Akt activities increased for 1ay ANOVA and the post hoc Student's t-test. Error bars indicate meanSEM. Results are

199K. Okada et al. / Experimental Neurology 222 (2010) 191203of sciatic nerves labeled for NF or MBP was measured at points 2 and10 mm distal to the sutured site. At 6 weeks post operation, the grossareas of NF-positive tissues at 2 mm distal and 10 mm distal sites inthe MeCbl group (101,4689872 m2 and 59,0028298 m2,respectively) were larger than those in the control group (46,5548457 m2 and 27,9696321 m2) (Fig. 8o), and the gross areas ofMBP-positive tissues at 2 and 10 mm distal sites in the MeCbl group(129,98414,337 m2 and 58,8508387 m2, respectively) werealso larger than those in the control group (46,41611,782 m2 and15,7145244 m2) (Fig. 8p). The gross areas of NF-positive tissuesand MBP-positive tissues were enlarged at 12 weeks post operation,which were still larger in the MeCbl group than those in the controlgroup (Fig. 8il, o and p). A uorescencemicrographmade bymergingthe gure labeled for NF with that labeled for MBP showed that theregenerated axon was correctly ensheathed by myelin (Fig. 8m). Theresults of these experiments suggest that MeCbl promotes functionalrecovery and increases the number and gross area of axons andmyelin in parallel to the functional recovery and indicate that MeCblpromotes nerve regeneration in a rat model of sciatic nerve injury.

Discussion

MeCbl is an analog of vitamin B12, and vitamin B12 has variouseffects in the nervous system. Recent reports show that vitamin B12 isinvolved in some neurotrophic actions. When the central nervoussystem is decient in vitamin B12, levels of epidermal growth factor,

Fig. 6.Methylcobalamin promotes neurite outgrowth and neuronal survival with Erk and Akh (a and b) under the application of U0126 (c and d), LY294002 (e and f), and U0126+ LY294LY294002 increased the apoptotic rates, which was not reduced by addition of MeCbl. (j) Axonot increase the axonal length under application of U0126 and/or LY294002. Signicance wmeanSEM. Results are representative of three independent experiments. pb0.05.glial brillary acidic protein, and p75 neurotrophin receptor aredecreased, and tumor necrosis factor is increased (Magnaghi et al.,2002; Scalabrino et al., 2003, 1999, 2000). The mechanisms of thisphenomenon remain to be claried, but it has been suggested thatvitamin B12 modulates the synthesis of at least some cytokines andgrowth factors (Scalabrino and Peracchi, 2006). The well-knownprotein kinases Erk1/2 and Akt are activated by certain neurotro-phins, such as NGF and BDNF, and this activation in turn promotesneurite outgrowth and neuronal survival (Bhave et al., 1999; Bonni etal., 1999; Markus et al., 2002; Neiiendam et al., 2004; Schmid et al.,2000). In our study, we showed that MeCbl promotes neuriteoutgrowth (Fig. 1) and neuronal survival (Fig. 3) with activatingErk1/2 and Akt (Fig. 5). We also show that MeCbl activates Erk1/2and Akt for 72 h, whereas BDNF activates them for 1 h (Fig. 5),suggesting that MeCbl and BDNF differ in the mechanism by whichthey activate these kinases. This discrepancy of the time course foractivation may be caused by the difference of time required for thephosphorylation and the methylation. BDNF activates Erk1/2 and Aktthrough TrkB, which is the one of the receptor kinases (Segal, 2003),and MeCbl may increase the activity of Erk1/2 and Akt by themethylation of some protein kinase (Toohey, 2006). We consider thisis the reason why the time courses for activation of Erk1/2 and Akt byMeCbl and BDNF are different. Furthermore, MeCbl did not affect theactivation of Erk1/2 and Akt for 1 h after the application of BDNF, butfrom 24 to 72 h, addition of MeCbl increased the activation more thanthat affected only by BDNF (Fig. 5). We consider that the effects of

t activities. (ah) CGN from P9 rats were cultured with or without MeCbl (10 M) for 72002 (g and h). (i) The apoptotic rate of CGNwithout kainic acid is presented. U0126 andnal length of CGN under application of U0126 and/or LY294002, and Z-VAD. MeCbl didas determined by one-way ANOVA and the post hoc Student's t-test. Error bars indicate

200 K. Okada et al. / Experimental Neurology 222 (2010) 191203MeCbl and BDNF may be additive but the effect of BDNF is so strongand covers the effect of MeCbl for the rst hour of application. In thefollowing experiments using U0126 or LY294002, we demonstratedthat the inhibition of Erk1/2 and Akt erased the effects of MeCbl onneurite outgrowth and neuronal survival (Fig. 6). Therefore, wesuggest that MeCbl promotes neurite outgrowth and neuronalsurvival with the activation of Erk1/2 and Akt.

In this study, we showed that MeCbl was the most effective of thevitamin B12 analogs tested in promoting neurite outgrowth andneuronal survival in CGN and DRG neurons. However, AdoCbl, one ofthe two active forms of vitamin B12, did not promote neuriteoutgrowth and neuronal survival. These results suggest that themetabolic pathway of MeCbl, but not AdoCbl, is associated withneurite outgrowth and neuronal survival. CNCbl and OHCbl, which areinactive forms of vitamin B12, promote neurite outgrowth andneuronal survival more effectively than AdoCbl, and we considerthat the reason for this is that they may be altered to MeCbl aftercoming into neurons, whereas AdoCbl is already the active form andmay not be changed to MeCbl. MeCbl works as a coenzyme of MS inmammalian cells (Banerjee and Ragsdale, 2003). MS is necessary forthe formation of methionine from homocysteine, and SAM, which isinvolved in some methylation reactions, is formed from methionine.This metabolic process is called the methylation cycle and MeCbl is amethyl donor in this cycle (Toohey, 2006). On the basis of our results(summarized in Fig. 9), we suggest that certain methylation reactionsare involved in neurite outgrowth, neuronal survival, and phosphor-ylation of Erk1/2 and Akt.

Fig. 7. Rat sciatic nerve injury models in 8-week-old female Wistar rats were observed for 12fth (c) toes was performed everyweek. In the third toe, sensory function showed no signicthe fourth and fth toes, sensory function improved in the MeCbl group from 10 weeks afterMeCbl group, toe spreading improved from 11 weeks after surgery. (e) SFI was examined elatency of the sciatic nerve was examined 12weeks after surgery. In theMeCbl group, the termway ANOVA and the post hoc Student's t-test. Error bars indicate meanSEM. There wereThe methylation cycle includes two major pathways: DNAmethylation and protein methylation. DNA methylation is animportant cellular phenomenon that is used to repress geneexpression and promote gene stability in various species (Nelson etal., 2008). DNA methylation plays major roles in the developmentalstage, but after cellular differentiation, changes in DNA methylationare less numerous and are thought to maintain the identity of cells(Ehrlich, 2003). Thus, it is hard to accept that MeCbl activates Erk1/2and Akt in neurons by DNA methylation. Protein methylation in themethylation cycle involves protein carboxyl methylation. Onecarboxyl methyltransferase, named isoprenylcysteine carboxylmethyltransferase (Icmt), acts on proteins formed by their precursorssynthesized with a C-terminal CAAX tetrapeptide motif (C is cystein;A is generally an aliphatic residue; X is a variable residue) (Grillo andColombatto, 2005). Methylation of the cysteine residue of the C-terminal CAAX motif of the protein has been shown to be importantfor localization, stability, and functions of the protein (Magee andSeabra, 2005). Ras protein, which is upstream of Erk1/2 and Akt intheir signaling pathway (Markus et al., 2002), is a CAAX protein andits methylation by Icmt is very important for its function (Bergo et al.,2004). Furthermore, methylation of Ras protein regulates Akt and themammalian target of rapamycin signaling pathway (mTOR) (Wang etal., 2008), and Erk1/2 also regulates mTOR (Tsokas et al., 2007). ThemTOR signaling pathway is crucial for neuronal development andlong-term modication of synaptic strength (Jaworski and Sheng,2006). Therefore, on the basis of these previous reports, we proposethe hypothesis that MeCbl increases methylation of Ras and promotes

weeks. (ac) To evaluate sensory function, a pinch test of the third (a), fourth (b), andant difference between the control andMeCbl groups, except at 8weeks after surgery. Insurgery. (d) To evaluate motor function, toe spreading was examined every week. In thevery 2 weeks. It improved in the MeCbl group from 5 weeks after surgery. (f) Terminalinal latency of the injured sciatic nerve improved. Signicancewas determined by one-10 animals in each of the control and MeCbl groups. pb0.05.

201K. Okada et al. / Experimental Neurology 222 (2010) 191203its function, and that Ras activates mTOR through activation of Erk1/2and Akt, which facilitates neurite outgrowth and neuronal survival.There is little evidence to prove such a hypothesis, but in this study,

Fig. 8.Methylcobalamin improves anatomical regeneration of sciatic nerve. (ah) Axial sectiopost operation. The slices at 2 mm distal (a, b, e and f) and 10 mm distal (c, d, g and h)photomicrographs at 60 magnication. (il) Fluorescence micrographs of cross-sectionaldistal to the sutured site were obtained from a rat sciatic nerve injury model at 12 weeksshows that NF is ensheathed by MBP. (n) The number of myelinated axons at 6 weeks andNF (o) and MBP (p). The area of NF and MBP was measured at 6 weeks and 12 weeks postindicate meanSEM. Four animals at 6weeks and 10 animals at 12 weeks post operation w(insets in ah), 100 m (il), 50 m (m). pb0.05.we show the possibility that MeCbl promotes neurite outgrowth andneuronal survival by specic protein methylation, such as methyla-tion of Ras.

ns of the sciatic nerve stained with toluidine blue at 6 weeks (ad) and 12 weeks (eh)to the sutured site are presented. Insets located at the lower left side in a-h displayslices of sciatic nerves labeled for NF (green; i and j) and MBP (red; k and l) at 10 mmpost operation. (m) White boxes in j and l enlarged and merged are presented, and it12 weeks post operation is calculated. (o and p) Mean gross area of tissues labeled foroperation respectively. Signicance was determined by the Student's t-test. Error barsere evaluated in each of the control and MeCbl groups. Scale bars, 200 m (ah), 20 m

202 K. Okada et al. / Experimental Neurology 222 (2010) 191203In previous reports on the effects of MeCbl on neurons in vivo, highlevels of MeCbl improved nerve conduction and regeneration instreptozotocin-diabetic rats (Sonobe et al., 1988), and experimentalacrylamide neuropathy (Watanabe et al., 1994). We used a rat sciaticnerve injury model in the present study to evaluate the effects ofMeCbl in vivo. However, we used continuous administration of a doseof MeCbl (1 mg/kg/day) higher than that used in previous reports(500 g/kg/day). The half-life of vitamin B12 in serum ranges fromapproximately 20 to 50min and one administration cannotmaintain a

Fig. 9. A summary of this study. MeCbl activates Erk1/2 and Akt through themethylation cycle and promotes neurite outgrowth and neuronal survival.high concentration of vitamin B12 (Nava-Ocampo et al., 2005).Therefore, to maintain the serum concentration of vitamin B12 above100 nM, we continuously administered a higher dose. MeCblimproved the recovery of motor and sensory function, and terminallatency of the sciatic nerve, but did not increase cMAP to an extentgreater than that observed in the control group 12 weeks postoperation. Our morphological and histological evaluation offers apossible explanation of this phenomenon. MeCbl increased not onlythe regeneration of axons but also the regeneration of myelin. InMeCbl group, nerve bers were surrounded by thick myelin and thegross areas of MBP-positive tissues were enlarged (Fig. 8). Weconsider that the recovery of terminal latency in our experiments wasrelated to the regeneration of myelin because a previous reportshowed that nerve conduction velocity and terminal latency arenearly proportional to the diameter of myelinated bers (Waxman,1980). cMAP is also thought to be related to the nerve diameter, but itis also known that cMAP decreases at the time of axonal maturation,which leads to the loss of the signal-conducting capability of axonsthat do not reach a proper target organ (Kuypers et al., 1999). Weconsider that axons mature 12 weeks post operation and cMAP mightshow a signicant difference between the MeCbl and control groups ifwe were to evaluate them at an earlier time.

MeCbl has already been used to treat peripheral nervous disorders,such as diabetic neuropathy, Bell's palsy, and carpal tunnel syndrome(Ide et al., 1987; Jalaludin, 1995; Kuwabara et al., 1999; Sato et al.,2005; Yaqub et al., 1992). However, some studies show that vitaminB12 has no effects on nervous diseases. For example, vitamin Bsupplementation by oral medication showed no benecial effect onthe cognitive decline in individuals with mild to moderate Alzheimerdisease (Aisen et al., 2008).We consider that this discrepancy is due tothe differences in the dose of vitamin B12 and the method ofadministration. In our study, MeCbl did not affect neurons atconcentrations below 10 nM, and it is difcult to maintain highconcentrations, especially with low-dose administration or oralmedication, because the concentration of vitamin B12 declinesimmediately (Nava-Ocampo et al., 2005) and the absorption ofvitamin B12 from the ileum and entire intestine is regulated bylimitation of binding to gastric intrinsic factor and passive diffusion(Andres et al., 2009; Kuzminski et al., 1998;Wolters et al., 2004). Fromthe results of our study, we consider that it is necessary to devise newclinical methods, such as high-dose systemic or local injection, todeliver high doses of MeCbl to target organs to treat nervous disordersmore effectively. Furthermore, our in vitro study suggests that MeCblactivates Erk1/2 and Akt by methylation reactions, such as Rasmethylation by Icmt. Recent reports showed that axon regeneration ispromoted in the adult central nervous system by modulation of theAkt/mTOR pathway (Park et al., 2008) and that Icmt is downregulatedin spinocerebellar ataxia type 1 mice (Lin et al., 2000). Thus, the high-dose MeCbl treatment might have the potential to treat not onlyperipheral nerve injury but also central nervous injury and spinocer-ebellar ataxia.

MeCbl shows great potential for the treatment of nervousdisorders. We have revealed a role for signaling pathways in neuronsaffected by MeCbl, but further understanding of this role anddevelopment of an effective delivery system for MeCbl may enableus to treat several nervous disorders and to obtain new insights intonerve regeneration.

Acknowledgment

This work was supported by Grant-in-Aid for ExploratoryResearch.

References

Aisen, P.S., Schneider, L.S., Sano, M., Diaz-Arrastia, R., van Dyck, C.H., Weiner, M.F.,Bottiglieri, T., Jin, S., Stokes, K.T., Thomas, R.G., Thal, L.J., 2008. High-dose B vitaminsupplementation and cognitive decline in Alzheimer disease: a randomizedcontrolled trial. JAMA 300, 17741783.

Akaike, A., Tamura, Y., Sato, Y., Yokota, T., 1993. Protective effects of a vitamin B12analog, methylcobalamin, against glutamate cytotoxicity in cultured corticalneurons. Eur. J. Pharmacol. 241, 16.

Andres, E., Dali-Youcef, N., Vogel, T., Serraj, K., Zimmer, J., 2009. Oral cobalamin (vitaminB(12)) treatment. An update. Int. J. Lab. Hematol. 31, 18.

Azzouz, M., Kenel, P.F., Warter, J.M., Poindron, P., Borg, J., 1996. Enhancement of mousesciatic nerve regeneration by the long chain fatty alcohol, N-Hexacosanol. Exp.Neurol. 138, 189197.

Banerjee, R., Ragsdale, S.W., 2003. The many faces of vitamin B12: catalysis bycobalamin-dependent enzymes. Annu. Rev. Biochem. 72, 209247.

Banks, E.C., Doughty, S.W., Toms, S.M., Wheelhouse, R.T., Nicolaou, A., 2007. Inhibitionof cobalamin-dependent methionine synthase by substituted benzo-fused hetero-cycles. FEBS J. 274, 287299.

Bergo, M.O., Gavino, B.J., Hong, C., Beigneux, A.P., McMahon, M., Casey, P.J., Young, S.G.,2004. Inactivation of Icmt inhibits transformation by oncogenic K-Ras and B-Raf.J. Clin. Invest. 113, 539550.

Bhave, S.V., Ghoda, L., Hoffman, P.L., 1999. Brain-derived neurotrophic factor mediatesthe anti-apoptotic effect of NMDA in cerebellar granule neurons: signaltransduction cascades and site of ethanol action. J. Neurosci. 19, 32773286.

Boise, L.H., Thompson, C.B., 1997. Bcl-x(L) can inhibit apoptosis in cells that haveundergone Fas-induced protease activation. Proc. Natl. Acad. Sci. U. S. A. 94,37593764.

Bonni, A., Brunet, A., West, A.E., Datta, S.R., Takasu, M.A., Greenberg, M.E., 1999. Cellsurvival promoted by the Ras-MAPK signaling pathway by transcription-dependent and -independent mechanisms. Science 286, 13581362.

de Medinaceli, L., DeRenzo, E., Wyatt, R.J., 1984. Rat sciatic functional index datamanagement system with digitized input. Comput. Biomed. Res. 17, 185192.

Ehrlich, M., 2003. Expression of various genes is controlled by DNA methylation duringmammalian development. J. Cell. Biochem. 88, 899910.

Fujii, A., Matsumoto, H., Yamamoto, H., 1996. Effect of vitamin B complex onneurotransmission and neurite outgrowth. Gen. Pharmacol. 27, 9951000.

Ghosh, S.K., Rawal, N., Syed, S.K., Paik, W.K., Kim, S.D., 1991. Enzymic methylation ofmyelin basic protein in myelin. Biochem. J. 275, 381387.

Grillo, M.A., Colombatto, S., 2005. S-adenosylmethionine and protein methylation.

Amino Acids 28, 357362.

Higuchi, H., Yamashita, T., Yoshikawa, H., Tohyama, M., 2003. Functional inhibition ofthe p75 receptor using a small interfering RNA. Biochem. Biophys. Res. Commun.301, 804809.

Ide, H., Fujiya, S., Asanuma, Y., Tsuji, M., Sakai, H., Agishi, Y., 1987. Clinical usefulness ofintrathecal injection of methylcobalamin in patients with diabetic neuropathy. Clin.Ther. 9, 183192.

Jalaludin, M.A., 1995. Methylcobalamin treatment of Bell's palsy. Methods Find. Exp.Clin. Pharmacol. 17, 539544.

Jaworski, J., Sheng, M., 2006. The growing role of mTOR in neuronal development andplasticity. Mol. Neurobiol. 34, 205219.

Kau, Y.C., Hung, Y.C., Zizza, A.M., Zurakowski, D., Greco, W.R., Wang, G.K., Gerner, P.,2006. Efcacy of lidocaine or bupivacaine combined with ephedrine in rat sciaticnerve block. Reg. Anesth. Pain Med. 31, 1418.

Kikuchi, M., Kashii, S., Honda, Y., Tamura, Y., Kaneda, K., Akaike, A., 1997. Protectiveeffects of methylcobalamin, a vitamin B12 analog, against glutamate-inducedneurotoxicity in retinal cell culture. Invest. Ophthalmol. Vis. Sci. 38, 848854.

Kuwabara, S., Nakazawa, R., Azuma, N., Suzuki, M., Miyajima, K., Fukutake, T., Hattori, T.,1999. Intravenous methylcobalamin treatment for uremic and diabetic neuropathyin chronic hemodialysis patients. Intern. Med. 38, 472475.

Candiani, R., 1990. Subacute combined degeneration and induction of ornithinedecarboxylase in spinal cords of totally gastrectomized rats. Lab. Invest. 62,297304.

Scalabrino, G., Nicolini, G., Buccellato, F.R., Peracchi, M., Tredici, G., Manfridi, A.,Pravettoni, G., 1999. Epidermal growth factor as a local mediator of theneurotrophic action of vitamin B(12) (cobalamin) in the rat central nervoussystem. FASEB J. 13, 20832090.

Scalabrino, G., Tredici, G., Buccellato, F.R., Manfridi, A., 2000. Further evidence for theinvolvement of epidermal growth factor in the signaling pathway of vitamin B12(cobalamin) in the rat central nervous system. J. Neuropathol. Exp. Neurol. 59,808814.

Scalabrino, G., Buccellato, F.R., Veber, D., Mutti, E., 2003. New basis of the neurotrophicaction of vitamin B12. Clin. Chem. Lab. Med. 41, 14351437.

Schmid, R.S., Pruitt, W.M.,Maness, P.F., 2000. AMAP kinase-signaling pathwaymediatesneurite outgrowth on L1 and requires Src-dependent endocytosis. J. Neurosci. 20,41774188.

Segal, R.A., 2003. Selectivity in neurotrophin signaling: theme and variations. Annu.Rev. Neurosci. 26, 299330.

Sonobe, M., Yasuda, H., Hatanaka, I., Terada, M., Yamashita, M., Kikkawa, R., Shigeta, Y.,1988. Methylcobalamin improves nerve conduction in streptozotocin-diabetic rats

203K. Okada et al. / Experimental Neurology 222 (2010) 191203the compound action current amplitudes in relation to the conduction velocity andfunctional recovery in the reconstructed peripheral nerve. Muscle Nerve 22,10871093.

Kuzminski, A.M., Del Giacco, E.J., Allen, R.H., Stabler, S.P., Lindenbaum, J., 1998. Effectivetreatment of cobalamin deciency with oral cobalamin. Blood 92, 11911198.

Liao, B., Newmark, H., Zhou, R., 2002. Neuroprotective effects of ginseng total saponinand ginsenosides Rb1 and Rg1 on spinal cord neurons in vitro. Exp. Neurol. 173,224234.

Lin, X., Antalffy, B., Kang, D., Orr, H.T., Zoghbi, H.Y., 2000. Polyglutamine expansiondown-regulates specic neuronal genes before pathologic changes in SCA1. Nat.Neurosci. 3, 157163.

Magee, T., Seabra, M.C., 2005. Fatty acylation and prenylation of proteins: what's hot infat. Curr. Opin. Cell Biol. 17, 190196.

Magnaghi, V., Veber, D., Morabito, A., Buccellato, F.R., Melcangi, R.C., Scalabrino, G.,2002. Decreased GFAP-mRNA expression in spinal cord of cobalamin-decient rats.FASEB J. 16, 18201822.

Markus, A., Zhong, J., Snider, W.D., 2002. Raf and akt mediate distinct aspects of sensoryaxon growth. Neuron 35, 6576.

Mimura, F., Yamagishi, S., Arimura, N., Fujitani, M., Kubo, T., Kaibuchi, K., Yamashita, T.,2006. Myelin-associated glycoprotein inhibits microtubule assembly by a Rho-kinase-dependent mechanism. J. Biol. Chem. 281, 1597015979.

Nava-Ocampo, A.A., Pastrak, A., Cruz, T., Koren, G., 2005. Pharmacokinetics of high dosesof cyanocobalamin administered by intravenous injection for 26 weeks in rats. Clin.Exp. Pharmacol. Physiol. 32, 1318.

Neiiendam, J.L., Kohler, L.B., Christensen, C., Li, S., Pedersen, M.V., Ditlevsen, D.K.,Kornum, M.K., Kiselyov, V.V., Berezin, V., Bock, E., 2004. An NCAM-derived FGF-receptor agonist, the FGL-peptide, induces neurite outgrowth and neuronalsurvival in primary rat neurons. J. Neurochem. 91, 920935.

Nelson, E.D., Kavalali, E.T., Monteggia, L.M., 2008. Activity-dependent suppressionof miniature neurotransmission through the regulation of DNA methylation.J. Neurosci. 28, 395406.

Park, K.K., Liu, K., Hu, Y., Smith, P.D., Wang, C., Cai, B., Xu, B., Connolly, L., Kramvis, I.,Sahin, M., He, Z., 2008. Promoting axon regeneration in the adult CNS bymodulation of the PTEN/mTOR pathway. Science 322, 963966.

Pfohl-Leszkowicz, A., Keith, G., Dirheimer, G., 1991. Effect of cobalamin derivatives on invitro enzymatic DNA methylation: methylcobalamin can act as a methyl donor.Biochemistry 30, 80458051.

Robertson, G.S., Crocker, S.J., Nicholson, D.W., Schulz, J.B., 2000. Neuroprotection by theinhibition of apoptosis. Brain Pathol. 10, 283292.

Sato, Y., Honda, Y., Iwamoto, J., Kanoko, T., Satoh, K., 2005. Amelioration bymecobalamin of subclinical carpal tunnel syndrome involving unaffected limbs instroke patients. J. Neurol. Sci. 231, 1318.

Scalabrino, G., Peracchi, M., 2006. New insights into the pathophysiology of cobalamindeciency. Trends Mol. Med. 12, 247254.

Scalabrino, G., Monzio-Compagnoni, B., Ferioli, M.E., Lorenzini, E.C., Chiodini, E.,without affecting sorbitol and myo-inositol contents of sciatic nerve. Horm. Metab.Res. 20, 717718.

Tanaka, H., Yamashita, T., Asada, M., Mizutani, S., Yoshikawa, H., Tohyama, M., 2002.Cytoplasmic p21(Cip1/WAF1) regulates neurite remodeling by inhibiting Rho-kinase activity. J. Cell Biol. 158, 321329.

Temporin, K., Tanaka, H., Kuroda, Y., Okada, K., Yachi, K., Moritomo, H., Murase, T.,Yoshikawa, H., 2008a. IL-1beta promotes neurite outgrowth by deactivating RhoAvia p38 MAPK pathway. Biochem. Biophys. Res. Commun. 365, 375380.

Temporin, K., Tanaka, H., Kuroda, Y., Okada, K., Yachi, K., Moritomo, H., Murase, T.,Yoshikawa, H., 2008b. Interleukin-1 beta promotes sensory nerve regenerationafter sciatic nerve injury. Neurosci. Lett. 440, 130133.

Tomita, K., Kubo, T., Matsuda, K., Yano, K., Tohyama, M., Hosokawa, K., 2007. Myelin-associated glycoprotein reduces axonal branching and enhances functionalrecovery after sciatic nerve transection in rats. Glia 55, 14981507.

Toohey, J.I., 2006. Vitamin B12 and methionine synthesis: a critical review. Is nature'smost beautiful cofactor misunderstood. Biofactors 26, 4557.

Tsokas, P., Ma, T., Iyengar, R., Landau, E.M., Blitzer, R.D., 2007. Mitogen-activated proteinkinase upregulates the dendritic translation machinery in long-term potentiationby controlling the mammalian target of rapamycin pathway. J. Neurosci. 27,58855894.

Wang, M., Tan, W., Zhou, J., Leow, J., Go, M., Lee, H.S., Casey, P.J., 2008. A small moleculeinhibitor of isoprenylcysteine carboxymethyltransferase induces autophagic celldeath in PC3 prostate cancer cells. J. Biol. Chem. 283, 1867818684.

Watanabe, T., Kaji, R., Oka, N., Bara, W., Kimura, J., 1994. Ultra-high dosemethylcobalamin promotes nerve regeneration in experimental acrylamideneuropathy. J. Neurol. Sci. 122, 140143.

Waxman, S.G., 1980. Determinants of conduction velocity in myelinated nerve bers.Muscle Nerve 3, 141150.

Wolters, M., Strohle, A., Hahn, A., 2004. Cobalamin: a critical vitamin in the elderly. Prev.Med. 39, 12561266.

Yamatsu, K., Kaneko, T., Kitahara, A., Ohkawa, I., 1976a. [Pharmacological studies ondegeneration and regeneration of peripheral nerves. (1) Effects of methylcobala-min and cobamide on EMG patterns and loss of muscle weight in rats with crushedsciatic nerve]. Nippon Yakurigaku Zasshi 72, 259268.

Yamatsu, K., Yamanishi, Y., Kaneko, T., Ohkawa, I., 1976b. [Pharmacological studies ondegeneration and regeneration of the peripheral nerves. (2) Effects of methylco-balamin on mitosis of Schwann cells and incorporation of labeled amino acid intoprotein fractions of crushed sciatic nerve in rats]. Nippon Yakurigaku Zasshi 72,269278.

Yamazaki, K., Oda, K., Endo, C., Kikuchi, T., Wakabayashi, T., 1994. Methylcobalamin(methyl-B12) promotes regeneration of motor nerve terminals degenerating inanterior gracile muscle of gracile axonal dystrophy (GAD) mutant mouse. Neurosci.Lett. 170, 195197.

Yaqub, B.A., Siddique, A., Sulimani, R., 1992. Effects of methylcobalamin on diabeticneuropathy. Clin. Neurol. Neurosurg. 94, 105111.Kuypers, P.D., Walbeehm, E.T., Heel, M.D., Godschalk, M., Hovius, S.E., 1999. Changes in

Methylcobalamin increases Erk1/2 and Akt activities through the methylation cycle and promotes .....IntroductionMaterials and methodsCell cultureImmunocytochemistryNeurite outgrowth assayNeuronal survival assayWestern blottingSurgical proceduresSerum concentration of vitamin B12Motor and sensory functional analysisElectrophysiological analysisMorphological and histological analysisStatistics

ResultsMeCbl promotes neurite outgrowth in CGN and DRG neuronsMeCbl promotes neurite outgrowth through the methylation cycleMeCbl promotes neuronal survival in CGN and DRG neuronsMeCbl promotes neuronal survival through the methylation cycleMeCbl increases Erk and Akt activities through the methylation cycleMeCbl promotes neurite outgrowth and neuronal survival with Erk and Akt activitiesMeCbl promotes the recovery of injured sciatic nerve and its function

DiscussionAcknowledgmentReferences