JOURNAL OF NEUROTRAUMAVolume 23, Number 12, 2006© Mary Ann Liebert, Inc.Pp. 1883–1894

Metabosensitive Afferent Fiber Responses after PeripheralNerve Injury and Transplantation of an Acellular Muscle Graft

in Association with Schwann Cells

OLIVIER ALLUIN,1 FRANÇOIS FERON,2 CHRISTOPHE DESOUCHES,3ERICK DOUSSET,1 JEAN-FRANÇOIS PELLISSIER,4 GUY MAGALON,3

and PATRICK DECHERCHI1

ABSTRACT

Studies dedicated to the repair of peripheral nerve focused almost exclusively on motor ormechanosensitive fiber regeneration. Poor attention has been paid to the metabosensitive fibers fromgroup III and IV (also called ergoreceptor). Previously, we demonstrated that the metabosensitiveresponse from the tibialis anterior muscle was partially restored when the transected nerve was im-mediately sutured. In the present study, we assessed motor and metabosensitive responses of the re-generated axons in a rat model in which 1 cm segment of the peroneal nerve was removed and im-mediately replaced by an autologous nerve graft or an acellular muscle graft. Four groups of animalswere included: control animals (C, no graft), transected animals grafted with either an autologousnerve graft (Gold Standard-GS) or an acellular muscle filled with Schwann Cells (MSC) or CultureMedium (MCM). We observed that (1) the tibialis anterior muscle was atrophied in GS, MSC andMCM groups, with no significant difference between grafted groups; (2) the contractile propertiesof the reinnervated muscles after nerve stimulation were similar in all groups; (3) the metabosen-sitive afferent responses to electrically induced fatigue was smaller in MSC and MCM groups; and(4) the metabosensitive afferent responses to two chemical agents (KCl and lactic acid) was decreasedin GS, MSC and MCM groups. Altogether, these data indicate a motor axonal regeneration and animmature metabosensitive afferent fiber regrowth through acellular muscle grafts. Similarities be-tween the two groups grafted with acellular muscles suggest that, in our conditions, implantedSchwann cells do not improve nerve regeneration. Future studies could include engineered conduitsthat mimic as closely as possible the internal organization of uninjured nerve.

Key words: afferent fiber; electrophysiology; metabosensitivity; nerve repair

1883

1Laboratoire des Déterminants Physiologiques de l’Activité Physique (UPRES EA 3285), Institut Fédératif de Recherche (IFR)107, Faculté des Sciences du Sport, Université de la Méditerranée, Marseille, France.

2Neurobiologie de Interactions Cellulaires et Neurophysiopathologie (CNRS UMR 6184), Institut Fédératif de Recherche (IFR)11, Faculté de Médecine, Université de la Méditerranée, Marseille, France.

3Services de Chirurgie de la Main, Chirurgie Plastique et Réparatrice des Membres, Assistance Publique–Hôpitaux de Marseille, Ho--pital de la Conception, Marseille, France.

4Biopathologie Nerveuse et Musculaire (JE 2053), Assistance Publique—Hôpitaux de Marseille, Faculté de Médecine, Université de la Méditerranée, Marseille, France.

INTRODUCTION

REPAIRING AN INJURED NERVE remains a challenge forthe clinician. When a peripheral nerve is severed,

Wallerian degeneration occurs in the distal part, followedby proliferation of Schwann cells (Büngner band) andmacrophages within the remaining intact basal laminatubes. Schwann cells synthesize neurotrophic factors(NGF, BDNF, Neurotrophin 3–4/5, EGF, IGF and GDNF)(Acheson et al., 1991; Bosch et al., 1989; Friedman et al.,1992; Heumann, 1987; Ikegami, 1990; Meyer et al., 1992;Rende et al., 1992a,b; Springer et al., 1990; Windebankand Poduslo, 1986) as well as extracellular matrix and ad-hesion molecules (L1 glycoprotein, N-CAM, N-cadherin)(Ard et al., 1987; Bixby and Harris, 1991; Bixby et al.,1988; Bunge and Bunge, 1986; Bunge, 1994; Kleitman etal., 1988a; Kleitman et al., 1988b), which favor guidanceand myelination of the regenerating axons. After nervecrush injury (axonotmesis), that is, when the connectivetissue scaffold of the nerve remains intact, the regenerat-ing axons grow in their original endoneural tubes and arethereby guided back to their original target sites (Bakerand Body, 1980; Barker et al., 1985; Brown and Butler,1976; Hyde and Scott, 1983). In this case, reinnervation isessentially topographic. By contrast, nerve transection(neurotmesis) results in a much greater disruption with ahigher probability of muscle reinnervation by inappropri-ate axons (Collins et al., 1986).

The long held dogma that growing axons are onlyguided by attractive molecules (neurotropism) is not cor-rect. It is now thought that axonal regeneration is based aswell on physical contacts between filipodia and their en-vironment (directional guidance). These two combinedmechanisms induce an axonal repair, coined as “preferen-tial” (Al-Majed et al., 2000; Brushart et al., 1998). As forthe motor fibers, the sensory nerve fibers tend to grow inthe original direction (Burgess and Horch, 1973; Horch,1979), but along the way they often diverge and reinner-vate inappropriate peripheral targets (Banks and Barker,1989; Koerber et al., 1989). As a consequence, a sensorydisorganization is observed (Ford and Woodhall, 1938;Hawkins, 1948), even if later pruning of inappropriatesprouts may refine to a limited degree the anomalous rein-nervation patterns (Pockett and Slack, 1982).

Although peripheral nerves have the ability to regen-erate, the recovery of sensory and motor functions ishardly ever completed. In humans, when the nerve de-fect is wide, surgeons use autologous nerve graft, har-vested from another nervous area in the body, to span theinjury site. However, this procedure generates a loss offunction at the donor site and requires multiple surgeries.Alternatively, biological and/or synthetic nerve guides,including conjunctive tissue (Lewin-Kowalik et al.,

2003), skin bundles or muscle fibers (Calder and Norris,1993; Fields et al., 1989; Lundborg, 2000b; Norris et al.,1988), or tube (Archibald et al., 1995; Dahlin and Lund-borg, 2001; Lundborg, 2000a; Lundborg et al., 2004)have been used to overcome these hurdles. Several stud-ies have shown that some of these conduits improve func-tional recovery in animal models of nerve injury (Des-ouches et al., 2005; Dvali and Mackinnon, 2003;Rummler and Gupta, 2004; Zhang et al., 2005). Further-more, a phase I clinical trial based on the implantationof collagen NeuraGen® guides (Integra NeurosciencesLtd. www.integra-Is.com/) is currently carried out. How-ever, none of these strategies has proved to be as effi-cient as autologous nerve grafting.

Interestingly, basal membrane of skeletal striated mus-cle is similar to the nerve membrane. It contains lamininand collagen IV, two extracellular matrix proteins thatshape conduits, similar to endoneural tubes, providingphysical guidance for regenerating axons (Bertelli et al.,2005; Glasby et al., 1986a,b,c, 1990; Hall, 1997; Lund-borg, 2000a; Meek et al., 1996). Moreover, it has beenobserved that Schwann cells proliferate in the proximalstump and participate actively in nerve regeneration(Geller and Fawcett, 2002; Jones et al., 2001; Son et al.,1996). Several convergent studies demonstrated thatSchwann cells implanted either in biodegradable tube orin intestinal mucosa, increased the functional recovery inanimals whose sciatic nerve segment has been previouslyremoved [for review (Belkas et al., 2004; Desouches etal., 2005; Dvali and Mackinnon, 2003; Rummler andGupta, 2004; Strauch, 2000; Suematsu, 1989; Taras etal., 2005; Zhang et al., 2005)]. However, the therapeuticbenefit of Schwann cell transplants has been assessed us-ing mainly motor function tests. To the best of our knowl-edge, no denatured skeletal muscle associated withSchwann cells have been used to repair peripheral nerve.In addition, sensitive recovery after grafting denaturedskeletal muscle has never been studied. In order to ap-praise the recovery of ascending fibers, we designed astudy in which acellular skeletal striated muscle, im-planted or not with Schwann cells, were sutured betweenthe two free nerve stumps of a peroneal nerve after sec-tion of a 1 cm segment. Ten weeks later, the electro-physiological responses of afferent fibers were assessedas previously described (Decherchi et al., 1998, 2001;Dousset et al., 2001, 2003).

MATERIALS AND METHODS

Animals

Four- to five-month-old Lewis rats, weighing 250–350g (Charles River®, Les Oncins, France), were housed in

ALLUIN ET AL.

1884

smooth-bottomed plastic cages at 22°C with a 12-hlight/dark cycle. Food (Purina, rat chow) and water wereavailable ad libitum. Anesthesia and surgical procedureswere performed according to the French law on AnimalCare Guidelines and the Animal Care Committee of Uni-versity Aix-Marseille II approved our protocols. All an-imals were weighed before each experiment.

Schwann Cell Culture

Sciatic nerves were excised from 12-week-old femaleLewis inbred rats. The nerves were dissected free fromtheir surrounding membrane (epineurium) and placed inserum-free DMEM-F12 culture medium (Invitrogen) be-fore being cut into small pieces using a McIlwain tissuechopper (Brinkmann). After several washes in HBSS, thetissue was placed in 1 mL of trypsin-EDTA (0.25%Trypsin, 1 mM EDTA; Invitrogen) and incubated for 10min at 37°C/5% CO2 with mechanical dissociation at 2-min intervals. Trypsin activity was stopped using 9 mLof DMEM-F12 containing 10% fetal calf serum. The re-sulting cell suspension was centrifuged at 400g for 5 min,and the cell pellet was resuspended in DMEM-F12 con-taining 10% FCS. The cells were plated on plastic wellscoated with poly-L-lysine (2 �g/cm2). After 2 h, the un-attached cells were collected and replated onto poly-L-lysine coated plastic wells for a further 8 h, after whichthe remaining floating cells were again replated ontopoly-L-lysine coated plastic wells and maintained for afurther 16 h at 37°C/5% CO2. Following the final incu-bation, the remaining unattached cells were collected,centrifuged at 400g for 5 min, and the cell pellet was re-suspended in serum-free DMEM-F12 containing 50ng/mL TGF�, plated on poly-L-lysine (2 �g/cm2)–coatedplastic dishes, and maintained for 5 days at 37°C/5%CO2. Cells were grown until they reached the requiredquantity for grafting (n � 2 million per animal).

Surgical Protocol and Experimental Groups

Male inbred host rats (n � 60) were randomized intofour groups: (1) control group (C, n � 23) without anysurgery, (2) gold standard group (GS, n � 19) in whicha 1-cm segment of the left peroneal nerve was cut outand immediately autografted in inverted position, (3)muscle Schwann cell group (MSC, n � 7) in which a 1-cm segment of peroneal nerve was removed and imme-diately replaced by an acellular muscle graft containingSchwann cells, and (4) muscle medium culture group(MCM, n � 11) in which a 1-cm segment of peronealnerve was replaced by an acellular muscle graft contain-ing culture medium.

Female inbred donor rats (n � 3) were sacrificed byintra-peritoneal overdose (1 mL) of sodium pentobarbi-

tal (Pentobarbital Sodique, Sanofi Santé Animale). Mus-cles of lower limbs (i.e., gastrocnemius, tibialis anterior,and quadriceps) were harvested, and bundles were lon-gitudinally cut. In order to clear the cells off, muscle bun-dles were immersed into liquid nitrogen (�196°C). Af-ter thermal equilibrium was achieved, the muscles weretransferred into distilled water (room temperature) andallowed to thaw for 10 min. This procedure was repeatedtwo times (Lassner et al., 1994). Then, acellular musclebundles were stored in a �80°C freezer until further use.

Before grafting, pieces of acellular muscles were slowlythawed. Animals were deeply anesthetized (PentobarbitalSodique, Sanofi Santé Animale, 60 mg/kg). The peronealnerve from the left limb was dissected free from sur-rounding tissues on a length of 3–4 cm, and a 1-cm seg-ment was removed. The nerve segment was immediatelyreplaced either by muscle bundle or by the nerve segmentin inverted position, and sutured (Ethilon 9-0, Ethicon) atthe two free nerve stumps. Then, Schwann cells or cul-ture medium were injected into muscle grafts. Musclesand skin were stitched (Flexocrin 3-0, B. Braun). No clin-ical signs of pain or unpleasant sensation (i.e., screech,prostration, hyperactivity, anorexia) and no paw-eatingbehavior were observed during the study.

Recording Protocol

At 10 weeks post-grafting, rats were re-anesthetizedby an intra-peritoneal injection of solution containingsodium pentobarbital (Pentobarbital Sodique, SanofiSanté Animale, 60 mg/kg). A tracheotomy was per-formed, and rats were artificially ventilated (Harvard vol-umetric pump: rate 40–60 min�1, tidal volume 2–4 ml;Southmatick, MA). A catheter was inserted into the rightfemoral artery and pushed up to the fork of the abdomi-nal aorta in order to transport the chemicals (i.e., potas-sium chloride [KCl] and lactic acid [LA]) to the contro-lateral muscle. This catheter was positioned in order tolet the blood flow freely to the left lower limb muscles.Animals were paralyzed by an intra-arterial injection ofpancuronium bromide (Pavulon, 10 mg � kg�1; Sanofi,Fresnes, France). The left peroneal nerve was dissectedfree from surrounding tissue on a length of 3–4 cm, andits proximal portion was cut. In order to record the af-ferent activity from the tibialis anterior muscle, the freeend of the distal nerve was divided into several filamentbundles, using an operating microscope (�40, MZ75, Leica, Heerbrugg, Switzerland). Each bundle was posi-tioned on a monopolar tungsten electrode and immersedin paraffin oil. The nerve activity was referred to a nearbyground electrode implanted in a close muscle, amplified(50–100 K), and filtered (30 Hz to 10 kHz) by a differ-ential amplifier (P2MP SARL, Marseille, France). The

METABOSENSITIVE FIBERS REGROWTH

1885

afferent discharge was recorded (Biopac MP150 and AcqKnowledge software) and fed into pulse window dis-criminators (P2MP SARL, Marseille, France), which si-multaneously analyzed afferent populations. The outputof these discriminators provided noise-free tracings (dis-criminated units), which were counted by data analysissystem (Biopac AcqKnowledge software) at 1-sec inter-vals (in Hz) and then displayed on a computer. The dis-criminated units were counted and recorded on separatetracings. Due to the small size of action potentials of thethin afferent fibers in each bundle, the window discrim-inators allowed us to select one or two units in each af-ferent population (i.e., 2–4 units per filament bundle) andto study activities of the afferent populations. More de-tails of the recording protocol may be found in our pre-vious manuscripts (Decherchi et al., 1998, 2001; Dous-set et al., 2001, 2003).

Muscular Stimulation

Two stimulation electrodes (inter-electrode distance:4–5 mm) were placed onto the surface of the tibialis an-terior muscle. Contractions were produced by a GrassS88K neurostimulator delivering trains of rectangularpulses through an isolation unit. The intensity of pulsetrains was determined in order to be supramaximal aftera threshold eliciting a twitch. The voltage was 25% morethan those evoking a “relative” maximal force. The du-ration of stimulus trains was 500 msec, and trains wererepeated each second to produce a series of contractions.The pulse duration was 2 msec, and five single stimula-tions were delivered in each 500-msec train (10 Hz). Fa-tigue was assessed from the decay of force throughouttrials. Duration of fatigue trial was 3 min. In every case,the intensity of muscle stimulation was adjusted to elicitthe same maximal increase in muscle force at the begin-ning of fatigue trials.

Twitch Measurement

The contractile response of the tibialis anterior to mus-cle stimulation (twitch, which is a reflection of the ten-sion generated in the muscle) was performed with a neu-rostimulator (Grass S88K delivered single rectangularshocks, duration: 0.1 msec, frequency 0.5 Hz) and mea-sured with an electromanometer (Micromanometer 7001,Ugo Basile, Biological Research apparatus). Twitcheswere analyzed in terms of peak amplitude (A), contrac-tion time (CT: time interval between the beginning of thecontraction curve and peak twitch tension), maximum re-laxation rate (MRR), defined as the slope of a tangentdrawn to the steepest portion of the relaxation curve.MRR was normalized to the total twitch amplitude (A;MRR/A � mean relaxation rate constant, msec�1), as

suggested by Esau et al. (1983), who have shown thatMRR values are linearly related to A. Twitch wasrecorded with Biopac MP150 system (sampled at 2000Hz, filtered with low pass at 150 Hz) and analyzed withAcqKnowledge 3.7.3 software.

Identification of Afferent Fibers

As previously described (Decherchi et al., 1998, 2001;Dousset et al., 2001, 2003), the following tests were per-formed for each selected filament bundle: (1) determina-tion of the receptive field in order to confirm that therecorded afferent activity was initiated from the muscleswhen pressure was exerted on the belly with a blunt rod;(2) measurements of the conduction velocity of discrim-inated units; (3) response of afferents after a 3-min lowfrequency (10 Hz) electrical stimulation of tibialis ante-rior muscle (simulated fatigue); (4) response of afferentunits to intra-arterial bolus injection of potassium chlo-ride (KCl) solution (1, 5, 10, and 20 mM in 0.5 mL ofsaline) or lactic acid (LA) solutions (0.5, 1, 2, and 3 mMin 0.1 mL of saline). A 15-min delay between each in-jection was performed in order to let the afferent activ-ity going back to its baseline. Numerous figures, includ-ing examples of recordings of afferent activity afterfatigue and chemical stimuli application, may be foundin our previous manuscripts (Decherchi et al., 1998, 2001;Dousset et al., 2001, 2003).

Muscular Atrophy Measurement

After the electrophysiological study, rats were sacri-ficed by an intra-arterial overdose (1 mL) of sodium pen-tobarbital solution (Pentobarbital Sodique, Sanofi SantéAnimale, 60 mg/mL). Left tibialis anterior muscle washarvested and immediately weighed on a precision scale(OHAUS, Navigator™, N30330 model). Comparisons ofmuscle mass atrophy were performed using a muscleweight/body weight ratio.

Histology and Microscopy

Peroneal nerve or denatured muscle grafts were har-vested free from surrounding tissues, rapidly washed inphosphate buffer (PBS; Gibco) and immediately im-mersed in a 4% glutaraldehyde-containing PBS solutionduring 24 h. Fixed peroneal nerves were stored in sodiumazide solution (0.1%) at 4°C. In each group, half of thesamples were stained with p-phenylenediamine (PPD)and the other half with hematoxylin and eosin solutions.

p-Phenylenediamine staining. Samples were washedthree times (5 min) in PBS and immersed into osmiumtetroxide (2%) solution for 1 h. After 3 washes (5 min)in PBS, samples were immersed in increasingly concen-trated acetone solutions (50%, 35 min; 70%, 1 h; 95%,

ALLUIN ET AL.

1886

1 h; 100%, 2 h, respectively). Samples were immersedin increasingly concentrated araldite solutions (50%, 3 h;80%, 5 h, respectively). Specimens were immersed inaraldite 100% and placed in heat chamber (80°C) during12 h for resin polymerization.

After inclusion, semithin sections (0.8 �m) were cutusing an ultramicrotome (Ultracut R, Leica) and collectedon coated slides (SuperFrost Plus, Menzel-Glaser). Afterbeing dried for 12 h on a hot plate, sections were stainedwith a p-phenylenediamine-ethanol (70°) solution,washed in distilled water, dried for 5 h on a hot plate andmounted with glycerol-containing medium (Glycergel,DakoCytomation).

Hematoxylin-eosin staining. Samples were washedthree times (5 min) in PBS and immediately immersedin a PBS-sucrose (30%) solution until specimens sank tothe bottom of the container. Samples were included in acryostat embedding glue (Tissutek, Durham, NC) at�25°C. After inclusion, sections (7 �m) were cut on acryotome (Cryostat CM3050s, Leica) and collected oncoated slides (SuperFrost Plus, Menzel-Glaser). Afterdrying, slides were sequentially immersed in hematoxylinsolution (3 min), distilled water, eosin solution (3 min),and distilled water. Finally, sections were mounted usingglass coverslips and mounting medium (Glycergel, Dako-Cytomation).

Slides were examined using an optical microscope(Eclipse E800, Nikon), which was associated with high-resolution color digital camera (Eclipse DXM 1200,Nikon).

Statistical Analysis

Afferent fiber baseline discharge (Fimpulses � sec�1)was averaged at time zero, irrespective of the stimulusapplied later. Significant changes in afferent activity in-duced by each test agent were determined with respectto the corresponding averaged baseline value. Frequen-cies, twitch parameters, and muscular atrophy ratio aregiven as mean � SEM and correspond to measured rawvalues. Data processing was performed using a softwareprogram (Instat 3.0, GraphPad software, San Diego, CA).Analysis of variance allowed us to assess significantmodifications of the afferent activity, muscular contrac-tion, and muscular atrophy, followed by a Tukey’s post-hoc test to indicate the direction and magnitude of dif-ferences between different conditions. Histology wasdescribed only with qualitative evaluation; no statisticalanalysis was performed.

RESULTS

Nerve Afferent Responses

Response to electrical induced fatigue. As shown inFigure 1, a significant increase of afferent discharge fre-quency (p � 0.05) was observed in group C and GS af-ter a 3-min stimulation of the tibialis anterior (C ��35.9 � 0.17%; GS � �41% � 0.2%). No significantchange was observed in the MSC and MCM groups(MSC � �7.1 � 0.4%; MCM � �1.5 � 0.3%). Whencompared to C and GS groups, animals grafted with mus-

METABOSENSITIVE FIBERS REGROWTH

1887

FIG. 1. Afferent response (in Fimpulses � sec�1 expressed in % of baseline activity) recorded in control (C), gold standard (GS),Muscle � Schwann cells (MSC), and muscle � culture medium (MCM) groups. At 10 weeks post-surgery, tibialis anterior mus-cle was stimulated (10 Hz) and afferent response was recorded (*p � 0.05 in comparison to baseline activity of the same group;†p � 0.05 in comparison to control group).

ALLUIN ET AL.

1888

FIG. 2. Afferent response (in Fimpulses � sec�1 expressed in % of baseline activity) recorded in control (C), gold standard (GS),muscle � Schwann cells (MSC), and muscle � culture medium (MCM) groups. At 10 weeks post-surgery, tibialis anterior mus-cle was stimulated using injections of increasingly concentrated solutions of potassium chloride (KCl) and afferent response wasrecorded (*p � 0.05 and ***p � 0.0001 in comparison to baseline activity of the same group; ††p � 0.01 in comparison to con-trol group).

FIG. 3. Afferent response (in Fimpulses � sec�1 expressed in % of baseline activity) recorded in control (C), gold standard (GS),muscle � Schwann cells (MSC), and muscle � culture medium (MCM) groups. At 10 weeks post-surgery, tibialis anterior mus-cle was stimulated using injections of increasingly concentrated solutions of lactic acid (LA) and afferent response was recorded(*p � 0.05 and **p � 0.01 in comparison to baseline activity of the same group).

cle conduits exhibit a dramatic loss of afferent responseto electrical induced fatigue.

Responses to chemical injections. (a) Potassium chlo-ride (KCl): Figure 2 shows that significant increases ofafferent discharge frequency were observed in the Cgroup whatever the concentration of KCl solution used:C � �26.61 � 10.73% (1 mM, p � 0.05), �29.57 �10.62% (5 mM, p � 0.05), �38.52 � 14.97% (10 mM,p � 0.0001), �78.17 � 22.97% (20 mM, p � 0.0001).For the GS group, a small increase of afferent dischargewas observed only at the concentration of 20 mM: GS ��9.92 � 4.33% (20 mM, p � 0.05). No significantchange was observed in the MSC and MCM groups. Whencompared to C group, animals grafted with either an in-verted nerve or a muscle conduit exhibit a dramatic lossof afferent response to KCl. (b) Lactic acid (LA): Figure3 shows that significant increases of afferent dischargefrequency was observed in C group after lactic acid in-jections: C � �47.35 � 15.82% (1 mM, p � 0.01),�59.76 � 26.95% (2 mM, p � 0.01), �24.63 � 5.94%(3 mM, p � 0.01). For the GS group, a small increase ofafferent discharge was observed only at the concentra-tion of 2 mM: GS � �10.4 � 4.1% (2 mM, p � 0.05).No significant change was observed in the MSC and MCM

groups. When compared to C group, animals grafted witheither an inverted nerve or a muscle conduit exhibit a dra-matic loss of afferent response to lactic acid.

Muscular Atrophy Evaluation

Figure 4 shows that the weight of tibialis anterior mus-cle in GS (�0.088 � 0.007%), MSC (�0.106 � 0.01%),

MCM (�0.117 � 0.016%) groups is significantly de-creased (p � 0.001) when compared to the C group(�0.23 � 0.006%). There is no significant difference be-tween grafted groups.

Twitch Response

Every tibialis anterior muscle from C and GS ani-mals, but only half of them (57.14% [n � 4] and45.45% [n � 5] for MSC group and MCM group, re-spectively) have produced muscular contractions whenthe proximal stump of peroneal nerve was electricallystimulated. Table 1 shows no difference in twitch re-sponse between groups.

Histology

Ten weeks post-grafting, animals from the graftedgroups (Fig. 5b,c,d) exhibit numerous small diameteraxons. When compared to other grafted groups, ani-mals from the GS group display specific features suchas an epineurial structure, a weak connective tissuedensity and a robust vascularization with blood vesselslocated along axon bundles (Fig. 5B). Rare blood ves-sels were observed in the connective tissue of muscle-grafted groups (MCM and MSC; Fig. 5C,D). Withinmuscle graft groups, we observed a cluster distributionof unmyelinated and myelinated regenerating axons(Fig. 5c,d) and neuroma-like structures composed ofaxons bundles lacking fascicular organization andmembrane (i.e., epineurium, perineurium, and en-doneurium; Fig. 5C,D). No anatomical difference wasobserved between MCM and MSC groups.

METABOSENSITIVE FIBERS REGROWTH

1889

FIG. 4. Muscle weight/body weight ratio (expressed in % of body mass). At 10 weeks post-surgery, tibialis anterior muscleand total body were weighted in every animal in control (C), gold standard (GS), muscle � Schwann cells (MSC), and muscle �culture medium (MCM) groups (***p � 0.001 in comparison to control group).

DISCUSSION

The goal of our study was to assess for the first timethe metabosensitive recovery of a transected peronealnerve, after transplantation of an acellular muscle con-duit filled or not with a Schwann cell suspension. Theseanimals were compared to rats sutured with a one cen-timeter inverted nerve graft (GS group). We show herethat chemical and electrical stimulation of the tibialis an-terior muscle did not elicit any change in the response ofthe afferent fibers in muscle-grafted animals while a sim-ilar stimulation in the GS group induced a weak and un-conventional activity. Previously, metabosensitive activ-ities have never been shown in this type of nerveautograft. Only transected nerve, acutely sutured, wereshown to include functional metabosensitive fibers, 2.5and 7 months post-trauma (Decherchi et al., 2001; Mar-queste et al., 2002).

In the present study, the reduced atrophy of the tibialisanterior muscle and its contraction when motor fiberswere stimulated in half of the muscle-grafted animals in-dicate that some motor fibers have regenerated after nerveresection and muscle transplantation. It is known thatmuscle afferents regenerate more slowly than efferents.Myelinated regenerating fibers were found in musclegrafts but their neuroma-like organisation prevented toprovide an efficient regeneration. Finally, functional re-covery was not improved by injecting Schwann cellswithin acellular grafted muscles.

After nerve injury and loss of tissue, transplantation ofan acellular matrix between the two stumps is a key com-ponent for a successful regeneration. This matrix pro-vides a guide for the migration of fibroblasts, Schwanncells and endothelial cells (Williams et al., 1983, 1987).However, acellular muscles used in the current studywere not totally suitable to provide such a permissive en-vironment. In spite of the partially motor responserestoration, afferent response in muscle graft groupfailed. This finding is concordant with two studies show-ing an absence of axon guidance (Ide et al., 1983; Leni-han et al., 1998) but discordant with others (Glasby et

al., 1986a,b,c, 1990; Hall, 1997; Lundborg, 2000a; Meeket al., 1996). One possible explanation for this discrep-ancy is the use of an envelope around the muscle. Whenmuscles are grafted with a coating matrix (i.e., biologicor synthetic tube), the number of regenerating axons andthe speed of recovery is increased (Meek et al., 1996,1999, 2001). Conversely, within an unsheathed acellularmuscle, regenerating axons can sprout out of the guideand reinnervate inappropriate targets. This is probablywhy we found growing axons within grafted muscles butno metabosensitive afferent response during electro-physiological recordings. Furthermore, an unenclosedmuscle remains open to inflammatory molecules or scar-producing cells which can restrict axonal regeneration(Lutz, 2004a,b; Teare et al., 2004). Our results also in-dicate that the response of the metabosensitive fibers af-ter a 3-min electrical induced fatigue is more consistentthan the responses obtained after chemical (KCl or lac-tic acid) arterial injections, the kinetic of response tochemical being different to the control group. Regener-ated metabosensitive fibers are more sensible to themetabolites released during fatigue than the circulatingchemicals found in the muscular interstitium. The vas-cular changes occurring after denervation may explainthis difference (Borisov et al., 2000).

A very weak metabosensitive afferent response was ob-served in nerves grafted with an acellular muscle contain-ing culture medium or Schwann cells. Nevertheless, theweight loss of the normally innervated muscle was not asstrong as it would be in absence of activation. This findingas well as our data concerning the motor pathway indicatethat a partial motor reinnervation of the tibialis anterior hasbeen achieved in half of muscle-grafted animals. As demon-strated previously, a 10-week delay post-injury is enoughto get a motor recovery in mice (Keilhoff et al., 2002; Verduet al., 1995; Young et al., 2001) and rats (Young et al.,2001). However, it may take longer to re-establish efficientmetabosensitive circuitry, as suggested by one of our pre-vious studies (Decherchi et al., 2001).

Compared to motor fibers, sensory fibers exhibit smalldiameter axons. In each animal of every grafted group,

ALLUIN ET AL.

1890

TABLE 1. TWITCH RESPONSE OF TIBIALIS ANTERIOR MUSCLE FROM C, GS, MCM, AND MSC GROUPS

Groups

Parameter C GS MCM MSC

A (g) 12.37 � 1.75 23.73 � 4.28 17.62 � 6.27 13.42 � 6.26MRR/A (sec�1) �20.62 � 2.7 �14.72 � 1.04 �22.75 � 1.66 �22.21 � 7.65CT (msec) 37.25 � 4.26 36.08 � 1.56 33.84 � 1.87 40.58 � 6.09

Values are means � SEM of muscular response to nerve stimulation. A, twitch amplitude; MRR/A, ratio of maximal relaxationrate to A; CT, contraction time.

we observe regenerating unmyelinated fibers suggestingthat sensory fibers have regenerated. However, they werenot functional. As mentioned before, one explanationcould be the axon rerouting toward inappropriate targets.

The risk of getting such diverted innervation is increasedwhen the diameter of the grafted muscles exceeds the di-ameter of the transected nerve. When acellular musclevolume is increased, axons loose their way and do not

METABOSENSITIVE FIBERS REGROWTH

1891

FIG. 5. Histology of normal and repaired nerves at 10 weeks post-surgery. (A,a) Control peroneal nerve. (B,b) Autologousnerve graft (gold standard). (C,c) Nerve grafted with an acellular muscle containing culture medium (MCM). (D,d) Nerve graftedwith an acellular muscle containing Schwann cells (MSC). Sections were stained with p-phenylenediamine (PPD). Scale bar �15 �m (A–D), 30 �m (a–d).

find their distal original muscle targets. Alternatively, itcan be surmised that regenerating sensory axons are notfully mature. The absence of epineurial membrane andthe neuroma-like organisation observed in muscle-grafted animals are supportive of this second hypothesis.

The data of the current study suggest that metabosen-sitivity recovery in nerve transected animal model shouldbe assessed in future experiments using engineered con-duits which mimic as closely as possible the molecularand spatial organisation found in unsevered nerve.

ACKNOWLEDGMENTS

We are grateful to Elisabeth Pasquale for her techni-cal assistance. We thank La Fondation de l’Avenir, AFM(Association Française contre les Myopathies),ALARME (Association Libre d’Aide à la Recherche surla Moelle Epinière), AMDT (Association Méditer-ranéenne pour le Développement des Transplantations),and La Fondation NRJ (sous l’égide de l’Institut deFrance) for their financial support.

REFERENCES

ACHESON, A., BARKER, P.A., ALDERSON, R.F., MILLER,F.D., and MURPHY, R.A. (1991). Detection of brain-derivedneurotrophic factor-like activity in fibroblasts and Schwanncells: inhibition by antibodies to NGF. Neuron 7, 265–275.

AL-MAJED, A.A., NEUMANN, C.M., BRUSHART, T.M.,and GORDON, T. (2000). Brief electrical stimulation pro-motes the speed and accuracy of motor axonal regeneration.J. Neurosci. 20, 2602–2608.

ARCHIBALD, S.J., SHEFNER, J., KRARUP, C., and MADI-SON, R.D. (1995). Monkey median nerve repaired by nervegraft or collagen nerve guide tube. J. Neurosci. 15,4109–4123.

ARD, M.D., BUNGE, R.P., and BUNGE, M.B. (1987). Com-parison of the Schwann cell surface and Schwann cell ex-tracellular matrix as promoters of neurite growth. J. Neuro-cytol. 16, 539–555.

BAKER, D., and BODY, A. (1980). reinnervation of stretch re-ceptors in cat muscle after nerve crush, in: Ontogenesis andFunctional Mechanisms of Peripheral Synapses. J. Taxis(ed), Elsevier: Amsterdam, pps. 251–253.

BANKS, R.W., and BARKER, D. (1989). Specificities of af-ferents reinnervating cat muscle spindles after nerve section.J. Physiol. 408, 345–372.

BARKER, D., SCOTT, J.J., and STACEY, M.J. (1985). Sen-sory reinnervation of cat peroneus brevis muscle spindles af-ter nerve crush. Brain Res. 333, 131–138.

BELKAS, J.S., SHOICHET, M.S., and MIDHA, R. (2004). Pe-ripheral nerve regeneration through guidance tubes. Neurol.Res. 26, 151–160.

BERTELLI, J.A., TALEB, M., MIRA, J.C., and GHIZONI,M.F. (2005). The course of aberrant reinnervation followingnerve repair with fresh or denatured muscle autografts. J. Pe-ripher. Nerv. Syst. 10, 359–368.

BIXBY, J.L., and HARRIS, W.A. (1991). Molecular mecha-nisms of axon growth and guidance. Annu. Rev. Cell Biol.7, 117–159.

BIXBY, J.L., LILIEN, J., and REICHARDT, L.F. (1988). Identi-fication of the major proteins that promote neuronal process out-growth on Schwann cells in vitro. J. Cell. Biol. 107, 353–361.

BORISOV, A.B., HUANG, S.K., and CARLSON, B.M. (2000).Remodeling of the vascular bed and progressive loss of capil-laries in denervated skeletal muscle. Anat. Rec. 258, 292–304.

BOSCH, E.P., ZHONG, W., and LIM, R. (1989). Axonal sig-nals regulate expression of glia maturation factor–beta inSchwann cells: an immunohistochemical study of injured sci-atic nerves and cultured Schwann cells. J. Neurosci. 9,3690–3698.

BROWN, M.C., and BUTLER, R.G. (1976). Regeneration ofafferent and efferent fibres to muscle spindles after nerve in-jury in adults cats. J. Physiol. 260, 253–266.

BRUSHART, T.M., GERBER, J., KESSENS, P., CHEN, Y.G.,and ROYALL, R.M. (1998). Contributions of pathway andneuron to preferential motor reinnervation. J. Neurosci. 18,8674–8681.

BUNGE, M.B., and BUNGE, R.P. (1986). Linkage betweenSchwann cell extracellular matrix production and ensheath-ment function. Ann. N.Y. Acad. Sci. 486, 241–247.

BUNGE, R.P. (1994). The role of the Schwann cell in trophicsupport and regeneration. J. Neurol. 242, S19–S21.

BURGESS, P.R., and HORCH, K.W. (1973). Specific regen-eration of cutaneous fibers in the cat. J. Neurophysiol. 36,101–114.

CALDER, J.S., and NORRIS, R.W. (1993). Repair of mixedperipheral nerves using muscle autografts: a preliminarycommunication. Br. J. Plast. Surg. 46, 557–564.

COLLINS, W.F., 3RD, MENDELL, L.M., and MUNSON, J.B.(1986). On the specificity of sensory reinnervation of catskeletal muscle. J. Physiol. 375, 587–609.

DAHLIN, L.B., and LUNDBORG, G. (2001). Use of tubes inperipheral nerve repair. Neurosurg. Clin. North Am. 12,341–352.

DECHERCHI, P., DARQUES, J.L., and JAMMES, Y. (1998).Modifications of afferent activities from tibialis anterior mus-cle in rat by tendon vibrations, increase of interstitial potas-sium or lactate concentration and electrically-induced fa-tigue. J. Peripher. Nerv. Syst. 3, 267–276.

DECHERCHI, P., VUILLON-CACCIUTOLO, G., DAR-QUES, J.L., and JAMMES, Y. (2001). Changes in afferentactivities from tibialis anterior muscle after nerve repair byself-anastomosis. Muscle Nerve 24, 59–68.

ALLUIN ET AL.

1892

DESOUCHES, C., ALLUIN, O., MUTAFTSCHIEV, N., et al.(2005). [Peripheral nerve repair: 30 centuries of scientific re-search] [in French]. Rev. Neurol. (Paris) 161, 1045–1059.

DOUSSET, E., DECHERCHI, P., GRELOT, L., and JAMMES,Y. (2001). Effects of chronic hypoxemia on the afferent nerveactivities from skeletal muscle. Am. J. Respir. Crit. CareMed. 164, 1476–1480.

DOUSSET, E., DECHERCHI, P., GRELOT, L., and JAMMES,Y. (2003). Comparison between the effects of chronic andacute hypoxemia on muscle afferent activities from the tib-ialis anterior muscle. Exp. Brain Res. 148, 320–327.

DVALI, L., and MACKINNON, S. (2003). Nerve repair, graft-ing, and nerve transfers. Clin. Plast. Surg. 30, 203–221.

ESAU, S.A., BELLEMARE, F., GRASSINO, A., PERMUTT,S., ROUSSOS, C., and PARDY, R.L. (1983). Changes in re-laxation rate with diaphragmatic fatigue in humans. J. Appl.Physiol. 54, 1353–1360.

FIELDS, R.D., LE BEAU, J.M., LONGO, F.M., and ELLIS-MAN, M.H. (1989). Nerve regeneration through artificialtubular implants. Prog. Neurobiol. 33, 87–134.

FORD, F.R., and WOODHALL, B. (1938). Phenomena due tomisdirection of regenerating fibers of cranial spinal and au-tonomic nerves. Arch. Surg. 36, 480–499.

FRIEDMAN, W.J., ALTIOK, N., FREDHOLM, B.B., andPERSSON, H. (1992). Mechanisms of nerve growth factormRNA regulation by interleukin-1 beta in hippocampal cul-tures: role of second messengers. J. Neurosci. Res 33, 37–46.

GELLER, H.M., and FAWCETT, J.W. (2002). Building a bridge:engineering spinal cord repair. Exp. Neurol. 174, 125–136.

GLASBY, M.A., GILMOUR, J.A., GSCHMEISSNER, S.E.,HEMS, T.E., and MYLES, L.M. (1990). The repair of largeperipheral nerves using skeletal muscle autografts: a com-parison with cable grafts in the sheep femoral nerve. Br. J.Plast. Surg. 43, 169–178.

GLASBY, M.A., GSCHMEISSNER, S., HITCHCOCK, R.J.,and HUANG, C.L. (1986a). Regeneration of the sciatic nervein rats. The effect of muscle basement membrane. J. BoneJoint Surg. Br. 68, 829–833.

GLASBY, M.A., GSCHMEISSNER, S.E., HUANG, C.L., andDE SOUZA, B.A. (1986b). Degenerated muscle grafts usedfor peripheral nerve repair in primates. J. Hand Surg. Br. 11,347–351.

GLASBY, M.A., GSCHMEISSNER, S.G., HITCHCOCK, R.J.,and HUANG, C.L. (1986c). The dependence of nerve regener-ation through muscle grafts in the rat on the availability and ori-entation of basement membrane. J. Neurocytol. 15, 497–510.

HALL, S. (1997). Axonal regeneration through acellular mus-cle grafts. J. Anat. 190, 57–71.

HAWKINS, G.L. (1948). Faultly localization in nerve regen-eration. An index of functional recovery following suture. J.Neurosurg. 5, 11–18.

HEUMANN, R. (1987). Regulation of the synthesis of nervegrowth factor. J. Exp. Biol. 132, 133–150.

HORCH, K. (1979). Guidance of regrowing sensory axons af-ter cutaneous nerve lesions in the cat. J. Neurophysiol. 42,1437–1449.

HYDE, D., and SCOTT, J.J. (1983). Responses of cat peroneusbrevis muscle spindle afferents during recovery from nerve-crush injury. J. Neurophysiol. 50, 344–357.

IDE, C., TOHYAMA, K., YOKOTA, R., NITATORI, T., andONODERA, S. (1983). Schwann cell basal lamina and nerveregeneration. Brain Res. 288, 61–75.

IKEGAMI, R. (1990). [Changes of nerve growth factor (NGF)content in injured peripheral nerve during regeneration: lo-cal synthesis of NGF by Schwann cells] [in Japanese]. Nip-pon Seikeigeka Gakkai Zasshi 64, 612–622.

JONES, L.L., OUDEGA, M., BUNGE, M.B., and TUSZYN-SKI, M.H. (2001). Neurotrophic factors, cellular bridges andgene therapy for spinal cord injury. J. Physiol. 533, 83–89.

KEILHOFF, G., FANSA, H., and WOLF, G. (2002). Differ-ences in peripheral nerve degeneration/regeneration betweenwild-type and neuronal nitric oxide synthase knockout mice.J. Neurosci. Res. 68, 432–441.

KLEITMAN, N., SIMON, D.K., SCHACHNER, M., andBUNGE, R.P. (1988a). Growth of embryonic retinal neuriteselicited by contact with Schwann cell surfaces is blocked byantibodies to L1. Exp. Neurol. 102, 298–306.

KLEITMAN, N., WOOD, P., JOHNSON, M.I., and BUNGE,R.P. (1988b). Schwann cell surfaces but not extracellular ma-trix organized by Schwann cells support neurite outgrowthfrom embryonic rat retina. J. Neurosci. 8, 653–663.

KOERBER, H.R., SEYMOUR, A.W., and MENDELL, L.M.(1989). Mismatches between peripheral receptor type andcentral projections after peripheral nerve regeneration. Neu-rosci. Lett. 99, 67–72.

LASSNER, F., DI BENEDETTO, G., PUGNALONI, A., et al.(1994). [Regeneration of peripheral nerves after interpositionof acellular muscle into larger defects]. Handchir. Mikrochir.Plast. Chir. 26, 258–261.

LENIHAN, D.V., SOJITRA, N.M., and GLASBY, M.A.(1998). Stimulated jitter measurement in the assessment ofrecovery after different methods of peripheral nerve repair.J. Hand Surg. Br. 23, 12–16.

LEWIN-KOWALIK, J., MARCOL, W., KOTULSKA-WOL-WENDER, K., et al. (2003). Dead-ended autologous con-nective tissue chambers in peripheral nerve repair—early ob-servations. Acta Physiol. Hung. 90, 157–166.

LUNDBORG, G. (2000a). A 25-year perspective of peripheralnerve surgery: evolving neuroscientific concepts and clinicalsignificance. J. Hand Surg. Am. 25, 391–414.

LUNDBORG, G. (2000b). Brain plasticity and hand surgery:an overview. J. Hand Surg. Br. 25, 242–252.

METABOSENSITIVE FIBERS REGROWTH

1893

LUNDBORG, G., ROSEN, B., DAHLIN, L., HOLMBERG, J.,and ROSEN, I. (2004). Tubular repair of the median or ul-nar nerve in the human forearm: a 5-year follow-up. J. HandSurg. Br. 29, 100–107.

LUTZ, B.S. (2004a). The role of a barrier between two nervefascicles in adjacency after transection and repair of a pe-ripheral nerve trunk. Neurol. Res. 26, 363–370.

LUTZ, B.S. (2004b). Structural and functional regeneration ofmuscle-related axons after transection and repair of the ratsciatic nerve using nonvascularized autologous fascia as abarrier between tibial and peroneal nerve fascicles. J. Re-constr. Microsurg. 20, 637–644.

MARQUESTE, T., DECHERCHI, P., DOUSSET, E.,BERTHELIN, F., and JAMMES, Y. (2002). Effect of mus-cle electrostimulation on afferent activities from tibialis an-terior muscle after nerve repair by self-anastomosis. Neuro-science 113, 257–271.

MEEK, M.F., DEN DUNNEN, W.F., SCHAKENRAAD, J.M.,and ROBINSON, P.H. (1996). Evaluation of functional nerverecovery after reconstruction with a poly (DL-lactide-epsilon-caprolactone) nerve guide, filled with modified denaturedmuscle tissue. Microsurgery 17, 555–561.

MEEK, M.F., DEN DUNNEN, W.F., SCHAKENRAAD, J.M.,and ROBINSON, P.H. (1999). Evaluation of several techniquesto modify denatured muscle tissue to obtain a scaffold for pe-ripheral nerve regeneration. Biomaterials 20, 401–408.

MEEK, M.F., ROBINSON, P.H., STOKROOS, I., BLAAUW,E.H., KORS, G., and DEN DUNNEN, W.F. (2001). Elec-tronmicroscopical evaluation of short-term nerve regenera-tion through a thin-walled biodegradable poly(DLLA-ep-silon-CL) nerve guide filled with modified denatured muscletissue. Biomaterials 22, 1177–1185.

MEYER, M., MATSUOKA, I., WETMORE, C., OLSON, L.,and THOENEN, H. (1992). Enhanced synthesis of brain-de-rived neurotrophic factor in the lesioned peripheral nerve:different mechanisms are responsible for the regulation ofBDNF and NGF mRNA. J. Cell. Biol. 119, 45–54.

NORRIS, R.W., GLASBY, M.A., GATTUSO, J.M., and BOW-DEN, R.E. (1988). Peripheral nerve repair in humans usingmuscle autografts. A new technique. J. Bone Joint Surg. Br.70, 530–533.

POCKETT, S., and SLACK, J.R. (1982). Pruning of axonaltrees results in increased efficacy of surviving nerve termi-nals. Brain Res. 243, 350–353.

RENDE, M., HAGG, T., MANTHORPE, M., and VARON, S.(1992a). Nerve growth factor receptor immunoreactivity in neu-rons of the normal adult rat spinal cord and its modulation af-ter peripheral nerve lesions. J. Comp. Neurol. 319, 285–298.

RENDE, M., MUIR, D., RUOSLAHTI, E., HAGG, T.,VARON, S., and MANTHORPE, M. (1992b). Immunolo-calization of ciliary neuronotrophic factor in adult rat sciaticnerve. Glia 5, 25–32.

RUMMLER, L.S., and GUPTA, R. (2004). Peripheral nerve re-pair: a review. Curr. Opin. Orthop. 15, 215–219.

SON, Y.J., TRACHTENBERG, J.T., and THOMPSON, W.J.(1996). Schwann cells induce and guide sprouting and rein-nervation of neuromuscular junctions. Trends Neurosci. 19,280–285.

SPRINGER, J.E., LACOVITTI, L., MAGUIRE, B.A., andCOLLIER, T.J. (1990). A dopaminergic neurotrophic factor(DNTF) id secreted by Schwann cells grown in culture. Soc.Neurosci. abst 822.

STRAUCH, B. (2000). Use of nerve conduits in peripheralnerve repair. Hand Clin. 16, 123–130.

SUEMATSU, N. (1989). Tubulation for peripheral nerve gap:its history and possibility. Microsurgery 10, 71–74.

TARAS, J.S., NANAVATI, V., and STEELMAN, P. (2005).Nerve conduits. J. Hand Ther. 18, 191–197.

TEARE, K.A., PEARSON, R.G., SHAKESHEFF, K.M., andHAYCOCK, J.W. (2004). Alpha-MSH inhibits inflammatorysignalling in Schwann cells. Neuroreport 15, 493–498.

VERDU, E., BUTI, M., and NAVARRO, X. (1995). The ef-fect of aging on efferent nerve fibers regeneration in mice.Brain Res. 696, 76–82.

WILLIAMS, L.R., DANIELSEN, N., MULLER, H., andVARON, S. (1987). Exogenous matrix precursors promotefunctional nerve regeneration across a 15-mm gap within asilicone chamber in the rat. J. Comp. Neurol. 264, 284–290.

WILLIAMS, L.R., LONGO, F.M., POWELL, H.C., LUND-BORG, G., and VARON, S. (1983). Spatial-temporalprogress of peripheral nerve regeneration within a siliconechamber: parameters for a bioassay. J. Comp. Neurol. 218,460–470.

WINDEBANK, A.J., and PODUSLO, J.F. (1986). Neuronalgrowth factors produced by adult peripheral nerve after in-jury. Brain Res. 385, 197–200.

YOUNG, C., MILLER, E., NICKLOUS, D.M., and HOFF-MAN, J.R. (2001). Nerve growth factor and neurotrophin-3affect functional recovery following peripheral nerve injurydifferently. Restor. Neurol. Neurosci. 18, 167–175.

ZHANG, N., YAN, H., and WEN, X. (2005). Tissue-engi-neering approaches for axonal guidance. Brain Res. BrainRes. Rev. 49, 48–64.

Address reprint requests to:Patrick Decherchi, Ph.D.

Laboratoire des Déterminants Physiologiques de l’Activité Physique (UPRES EA 3285)

Université de la Méditerranée (Aix-Marseille II)Institut Fédératif de Recherche Etienne-Jules Marey (IFR107)

Faculté des Sciences du Sport de Marseille-LuminyCase Postale 910

163, Avenue de Luminy13288 Marseille Cedex 09, France

E-mail: patrick.decherchi@univmed.fr

ALLUIN ET AL.

1894