ROTH VISAL UNG

EFFET DE L'ENTRAINEMENT LOCOMOTEUR SUR LA RÉCUPÉRATION DES FONCTIONS

LOCOMOTRICES CHEZ LA SOURIS PARAPLÉGIQUE

Thèse présentée à la Faculté des études supérieures de l'Université Laval dans le cadre du programme de doctorat en neurobiologie pour l'obtention du grade de Philosophiae Doctor (PhD)

FACULTE DE MEDECINE UNIVERSITÉ LAVAL

QUÉBEC

2011

Roth-VisalUng,2011

II

RÉSUMÉ

Une blessure à la moelle épinière (BME) est un traumatisme qui endommage les

fibres nerveuses permettant la communication entre le cerveau et le reste du corps. La

prévalence d'une BME est d'environ 1.3 million de Nord-Américains. Il n'existe

malheureusement aucune cure pour réparer la moelle épinière lésée. La principale

conséquence d'une BME est une perte des fonctions sensorielles et motrices volontaires,

sous le niveau de la lésion (ex, lésion thoracique entraîne paraplégie). Les patients souffrent

également de problèmes de santé qui se développent progressivement. Des problèmes

immunitaires, métaboliques, hormonaux, cardiovasculaires, musculaires, osseux et mentaux

apparaîtront chez une majorité de patients. Le manque de connaissance lié au

développement de ces troubles de santé constitue la problématique de recherche au cœur de

cette thèse.

Le but de cette thèse est de mieux comprendre l'étendue de ces problèmes de santé, et

de concevoir un traitement novateur pour diminuer, voire prévenir, certains de ces

problèmes. En se sens, les objectifs sont : 1) De terminer la caractérisation des ces

problèmes chez notre modèle animal. 2) De bien établir les conséquences fonctionnelles de

la plasticité neuronale sous-lésionnelle sur le système moteur et locomoteur. 3) D'établir le

rôle précis des récepteurs 5-HT2 dans l'activation pharmacologique des circuits spinaux

locomoteurs in vivo. 4) De déterminer les effets de substances aux propriétés anaboliques

sur le système musculaire et locomoteur. 5) D'évaluer les effets de d'un entraînement seul,

puis d'une approche multidisciplinaire combinant l'entraînement locomoteur,

l'administration d'agents aux propriétés anaboliques et d'activateurs des réseaux

locomoteurs spinaux, sur les dérèglements des systèmes locomoteur, musculaire et osseux.

I l l

ABSTRACT

Spinal cord injury (SCI) leads to a loss of nerve fibres that allow communication

between supraspinal centers and the rest of the body. In North America, the prevalence of

SCI is around 1.3 million of persons. Unfortunately, there is currently no cure to repair SCI.

The major consequence of SCI is a loss of sensory and voluntary motor functions below the

level of injury. The majority of people living with a SCI will also experience other health

problems including, immune, metabolic, hormonal, cardiovascular, muscular, bone and

mental problems. The lack of knowledge surrounding the development of such

complications is at the basis of the problematic of the current thesis.

The goal of this thesis is to better understand the extent of the health problems

following SCI and to develop a novel treatment that can prevent some of the complications.

In order to reach our goal, the objectives are: 1) To complete the characterization of the

problems in our animal model. 2) To establish the functional consequences of sublesional

neuronal plasticity at the motor and locomotor system level. 3) To establish the role of 5-

HT2 receptors in the pharmacological activation of spinal locomotor circuitry in vivo. 4) To

determine the effects of anabolic treatments on the muscular and locomotor systems. 5) To

evaluate the effects of training without any form of assistance, and then of a combined

strategy including pharmacological activation of spinal locomotor circuitry, locomotor

training and anabolic treatments.

IV

REMERCIEMENTS

Je tiens avant tout à remercier le Dr Pierre Guertin, qui m'a accueilli dans son

laboratoire. Merci pour ton enseignement, pour ta confiance et pour m'avoir permis

d'élargir mes horizons de recherche. Les dernières années dans ton laboratoire ont été

déterminantes dans mon choix de carrière et les connaissances acquises au cours de mon

doctorat sauront certainement me servir ultérieurement.

Sincères remerciements à tous les membres passés et présents du laboratoire pour

leur soutient au cours de mon cheminement. Merci Nicolas pour ton aide et les nombreux

échanges d'idées. Felicitation pour tes travaux (Tu l'as eu ton Ph.D!!). Bonne chance pour

tes projets futurs. Merci Pascal, Inge et Éric pour votre présence et soutient technique

grandement apprécié. Merci aux amis de l'unité de Neurosciences, Pédiatrie, du CRULRG

ainsi qu'à mes amis de longues dates, qui ont su me soutenir moralement et me divertir lors

de moments plus difficiles.

Je tiens à remercier spécialement mes parents et Béatrice, je ne sais pas ce que je

ferais sans vous, votre présence est tellement précieuse. Béatrice, Nora, Estelle et bébé (s)

futur (s), vous êtes une source de motivation et d'amour intarissable.

AVANT-PROPOS

Les six articles présentés dans cette thèse sont directement reliés à l'obtention du

grade de Ph.D. La contribution de l'auteur principal à cette thèse et aux six articles joints à

celle-ci a été significative, et ce, de la conception des protocoles de recherche jusqu'à la

rédaction des articles et de la thèse. Le premier article est une revue de littérature et

constitue le chapitre 2 de l'introduction, les cinq autres études sont des articles de

recherche.

Article 1 : Ung RV, Lapointe NP. Guertin PA. Early adaptive changes in chronic

paraplegic mice: a model to study rapid health degradation after spinal cord injury. Spinal

Cord. 2008 Mar;46(3): 176-80.

Dans cet article, l'étudiant est l'auteur principal, sous la supervision de son directeur, le Dr

Pierre Guertin. Nicolas Lapointe, étudiant gradué du Dr Guertin a contribué à la rédaction

de l'article de revue. L'ensemble des travaux a été effectué dans le laboratoire du Dr

Guertin.

Article 2 : Ung RV. Lapointe NP, Tremblay C, Larouche A, Guertin PA. Spontaneous

recovery of hindlimb movement in completely spinal cord transected mice: a comparison of

assessment methods and conditions. Spinal Cord. 2007 May;45(5):367-79.

Dans cet article, l'étudiant est l'auteur principal, sous la supervision de son directeur, le Dr

Pierre Guertin. Nicolas Lapointe, autre étudiant gradué du Dr Guertin, Christina Tremblay

et Amélie Larouche, étudiantes d'été, ont participé à la collecte des résultats

comportementaux, l'analyse des résultats et à la rédaction de l'article. L'ensemble des

travaux a été effectué dans le laboratoire du Dr Guertin.

Article 3 : Ung RV. Landry ES. Rouleau P. Lapointe NP, Rouillard C. Guertin PA. Role of

spinal 5-HT2 receptor subtypes in quipazine-induced hindlimb movements after a low-

thoracic spinal cord transection.Eur J Neurosci. 2008 Dec;28(l l):2231-42.

VI

Dans cet article, l'étudiant est l'auteur principal, sous la supervision de son directeur, le Dr

Pierre Guertin. Éric Landry, Nicolas Lapointe, étudiants gradués du Dr Guertin, Pascal

Rouleau, assistant de recherche du Dr Guertin et Claude Rouillard, chercheur collaborateur

ont participé à la collecte des données, l'analyse des résultats et à la rédaction de l'article.

L'ensemble des travaux a été effectué dans le laboratoire du Dr Guertin.

Article 4 : Ung RV, Rouleau P. Guertin PA. Effects of Co-Administration of Clenbuterol

and Testosterone Propionate on Skeletal Muscle in Paraplegic Mice. J Neurotrauma, 2010

27(6): 1129-42.

Dans cet article, l'étudiant est l'auteur principal, sous la supervision de son directeur, le Dr

Pierre Guertin. Pascal Rouleau, assistant de recherche du Dr Guertin, a participé à la

collecte des résultats. L'ensemble des travaux a été effectué dans le laboratoire du Dr

Guertin.

Article 5 : Ung RV. Lapointe NP, Rouleau P. Guertin PA. Non-assisted treadmill training

does not improve motor recovery and body composition in spinal cord-transected mice.

Spinal Cord. 2010 Epub 23 Fév.

Dans cet article, l'étudiant est l'auteur principal, sous la supervision de son directeur, le Dr

Pierre Guertin. Nicolas Lapointe, étudiant gradué et Pascal Rouleau, assistant de recherche

du Dr Guertin, ont participé à la collecte des résultats. L'ensemble des travaux a été effectué

dans le laboratoire du Dr Guertin.

Article 6

Ung Rv, Rouleau P, Guertin PA. Functional and physiological effects induced by spinal

locomotor network-activating tritherapy and clenbuterol in paraplegic mice. J Neurotrauma

2010, soumis le 11 septembre, en revision.

VII

Dans cet article, l'étudiant est l'auteur principal, sous la supervision de son directeur, le Dr

Pierre Guertin. Pascal Rouleau, assistant de recherche du Dr Guertin, a participé à la

collecte des résultats. L'ensemble des travaux a été effectué dans le laboratoire du Dr

Guertin.

VIII

TABLE DES MATIÈRES

RÉSUMÉ n

ABSTRACT m

REMERCIEMENT IV

AVANT-PROPOS V

TABLE DES MATIÈRES VIII

LISTE DES ABRÉVIATIONS XI

PRÉAMBULE 1

CHAPITRE I

Moelle épinière et locomotion 3

1.1 Terminologie 3

1.2 Anatomie de la moelle épinière 4

1.3 Contrôle supraspinale de la locomotion 6

1.4 Générateur central de patrons locomoteurs (CPG) 8

1.4.1 Évidences de CPG, localisation et neurones candidats 13

1.4.2 Évidences de CPG chez l'humain 16

1.5 Rôle des afférences sensorielles durant la locomotion 17

CHAPITRE II

Early adaptative changes in chronic paraplegic mice: A model to study rapid health degradation after spinal cord injury 21

CHAPITRE II (SUITE)

Conséquences d'une lésion de la moelle épinière sur le muscle et les fibres musculaires 35

2.1 Atrophie musculaire et conversion des fibres 35

2.2 Substances anaboliques pour contrer l'atrophie musculaire 31

CHAPITRE III

Potentiel de plasticité de la moelle épinière suite à une lésion 40

3.1 Le choc spinal, évidences de plasticité sous-lésionnelle 40

3.2 Réorganisation fonctionnelle de la moelle épinière 42

IX

3.2.1 Récupération motrice spontanée 42

3.2.2 Entraînement locomoteur chez l'animal 43

3.2.3 Entraînement locomoteur chez l'humain 45

3.3 Stimulation pharmacologique de la moelle épinière 47

3.3.1 L-Dopa, récepteurs dopaminergiques et adrénergiques 48

3.3.2 Sérotonine et récepteurs sérotoninergiques 52

CHAPITRE IV

Problématique 59

1.1 Problématique

1.2 Hypothèse de travail 60

1.3 Approche méthodologique 65

CHAPITRE V

Spontaneous recovery of hindlimb movement in completely spinal cord transected mice: a comparison of assessment methods and conditions 66

CHAPITRE VI

Role of spinal 5-HT2 receptor subtypes in quipazine-induced hindlimb movements after a low-thoracic spinal cord transection 98

CHAPITRE VII

Effects of co-administration of clenbuterol and testosterone propionate on skeletal muscle in paraplegic mice 132

CHAPITRE VIII

tdmill training does not improve locomo I mice 169

CHAPITRE V11I

Non-assisted treadmill training does not improve locomotor recovery and body composition in spinal cord- transected mic

CHAPITRE IX

Functional and physiological effects induced by spinal locomotor network-activating tri therapy and clenbuterol in paraplegic mice 190

CHAPITRE X

Discussion et conclusion 226

10.1 Résumé des études de la thèse 226

10.2 Limitation des études 227

10.2.1 Différence de rétablissement moteur entre

souris mâles et femelles 227

10.2.2 Évaluation de la récupération locomotrice 229

10.2.3 Pharmaco- et hormono-thérapies 231

10.3 Perspectives futures pour notre approche multidisciplinaires 235

10.4 Vers une transition de ces approches chez le patient 240

Conclusion 241

BIBLIOGRAPHIE 243

XI

LISTE DES ABRÉVIATIONS

5-HT : Sérotonine, sérotoninergique

ACOS : Grille d'évaluation locomotrice Average Combined Score

AOB : Grille d'évaluation locomotrice développée par Antri-Orsal-Barthe

BME : Blessure à la moelle épinière

BWSTT : Entraînement locomoteur sur tapis roulant avec support de poids (Body weight

supported treadmill training)

CMO : Contenu minéral osseux

CPG : Générateur central de patrons (Central pattern generator)

C, T, L, S : Segment cervical, thoracique, lombaire et sacral de la moelle épinière

DMO : Densité minérale osseuse

EDL : Extenseur digitorum longus

EMG : Électromyogramme, électromyographique

ENG : Électroneurogramme, électroneurographique

FES : Stimulations électriques fonctionnelles (Functional electrical stimulation)

FIM-L : Mesure d'indépendance fonctionnelle pour la locomotion (Functional

Indépendance Measure for Locomotion)

FRA : Afférences de réflexe de flexion (Flexor reflex afférents)

IEG : Gènes à expression précoce (Immediate early genes)

KO : Knock-out

LM : Mouvement de type locomoteur (locomotor movement)

MLR : Région locomotrice mésencéphalique (Mesencephalic locomotor region)

NLM : Mouvement de type non-locomoteur (non locomotor movement)

SNA : Système nerveux autonome

SNC : Système nerveux central

SNP : Système nerveux périphérique

UBG : Unité génératrice d'activité rythmique (Unit burst generator)

WISCI : Indice de marche pour blessés médullaires (Walking Index for Spinal Cord

Injury)

PRÉAMBULE

Les blessures à la moelle épinière (BME) affectent environ 1.3 million de Nord-

Américains (www.christopherreeve.org). Au Québec, annuellement, 200 nouvelles

victimes s'ajoutent aux quelques 7000 cas déjà répertoriés. Dans 60% des cas, les BME

surviennent avant l'âge de 30 ans et touchent majoritairement des hommes, de l'ordre de

80%. Les principales causes de BME sont les accidents de la route, les accidents lors

d'activités récréatives et sportives (plongeon, sport alpin, sport motorisé, etc.), les accidents

de travail ainsi que les actes violents (tentative de meurtre, suicide). Elles peuvent

également faire suite à des problèmes de santé, tels que des infections, des tumeurs et des

ischémies qui touchent la moelle épinière. Étant donné que les connections avec les centres

supraspinaux sont rompues, le cerveau ne peut plus contrôler volontairement les

mouvements et analyser adéquatement l'information provenant de la périphérie, ce qui se

traduit par une paralysie. Selon le niveau (cervical, thoracique, lombaire ou sacral) et

l'étendue de la BME, les blessés médullaires ont une perte partielle ou totale de certaines

fonctions motrices et sensorielles qui affectent les régions localisées sous le niveau de

lésion. Si seules les jambes sont affectées, il s'agit d'une paraplégie alors que si les

membres supérieurs et inférieurs sont touchés nous parlons de tétraplégie.

Pour une majorité de patients, d'importants problèmes de santé accompagnent ces

déficits sensorimoteurs. Ceux-ci subissent les conséquences de leur sédentarité, des

dérèglements des systèmes immunitaire, hormonal et vasculaire ainsi que de la

réorganisation des réseaux de neurones du système nerveux central (SNC), périphérique

(SNP) et autonome (SNA). Ces changements se traduisent, entre autre, par une perte de

masse corporelle associée à une fonte musculaire, une détérioration de la qualité de l'os, de

l'hypertonicité et des spasmes musculaires, de l'hyper-réflexie, des anormalités

cardiovasculaires, des troubles érectiles et urinaires. En dépit des avancements en soin de

santé, l'espérance de vie des blessés médullaires est plus courte que la population générale.

Ces derniers sont davantage sujets à des complications de santé pouvant entraîner des

hospitalisations de longue durée et des décès. Par exemple, les blessés médullaires sont

souvent sujets à des embolies pulmonaires, des pneumonies, des septicémies, des

problèmes cardiaques, etc.

Malheureusement, la moelle épinière lésée ne peut se régénérer par elle-même et,

malgré les importantes avancées de la science, il est pour l'instant impossible de

reconnecter la moelle épinière au cerveau. Il est donc primordial de développer des

stratégies pouvant pallier aux nombreuses conséquences qui découlent d'une BME.

S'inscrivant dans ce domaine d'expertise, notre laboratoire cherche à comprendre le

fonctionnement du réseau locomoteur de la moelle épinière, de caractériser les

changements anatomiques et métaboliques suite à une BME, de trouver des avenues

thérapeutiques afin de permettre une récupération de la motricité et de corriger certains des

problèmes secondaires associés à la BME.

Dans les prochaines sections, une revue des différents concepts et connaissances

reliés à nos travaux seront présentés. Par la suite, nos recherches portant sur le

rétablissement moteur, l'administration de substances anaboliques et pharmacologiques et

l'entraînement locomoteur sur tapis roulant seront étayées.

CHAPITRE I - MOELLE ÉPINIÈRE ET LOCOMOTION

1.1 Terminologie

Tout au long de cette thèse les termes reliés à la motricité et à la locomotion seront

utilisés. Cette section fournira des explications quant à l'utilisation de ces termes, ce qui

facilitera la compréhension de la thèse. Le terme « moteur » est employé au sens large,

c'est-à-dire tout ce qui est relié à la motricité peu importe le type de mouvements. Les

termes « locomoteur, locomotion et fonctions locomotrices » (sans l'utilisation du terme

rétablissement), font référence au fait de pouvoir se propulser vers l'avant dans le but de se

déplacer. Ceci implique donc une coordination des pattes avant-arrière, l'équilibre et une

posture adéquate. Le terme « locomotion fictive » est employé pour certains modèles

animaux d'isolation complète de la moelle épinière (in vivo ou in vitro) ou de

décérébration. Pour les modèles in vivo, l'animal est maintenu en position et ne peut pas se

déplacer. La locomotion fictive se définit donc comme étant une activité alternée entre les

fléchisseurs et extenseurs ipsi- et contra-latéral enregistrée dans les muscles, les nerfs ou les

racines ventrales. La locomotion fictive est induite pharmacologiquement ou par

stimulation électrique. Pour sa part, la « locomotion spinale » se définit par une capacité à

se propulser chez l'animal ayant une lésion complète de la moelle épinière. Nous

définissons le « rétablissement moteur » comme étant le retour du mouvement en général,

peu importe son aspect, alors que le « rétablissement locomoteur, de la locomotion ou des

fonctions locomotrices » se définit par l'amélioration des capacités ambulatoires. Pour

notre modèle animal de souris complètement lésée à la moelle épinière, nous utilisons

souvent les termes « mouvement de type locomoteur » (LM) et « de type non-locomoteur »

(NLM). Les LMs se définissent comme un mouvement de flexion-extension observé en

alternance dans les pattes arrière gauche et droite. Les NLMs constituent tous mouvements

qui ne sont pas générés en alternance dans les pattes arrière. Les NLMs incluent les

mouvements de flexion-extension unilatéral et les mouvements spastiques.

1.2 Anatomie de la moelle épinière.

La moelle épinière fait partie du SNC. Cet organe est souvent décrit comme étant un

relais entre le cerveau et le reste du corps (description qui demeure toutefois imprécise et

incomplète). Chez l'humain, la longueur de la moelle épinière varie entre 43 et 45 cm et

possède un diamètre entre 1 et 1.5 cm. Elle est située dans la colonne vertébrale, ce qui lui

confère une première barrière de protection contre les chocs mécaniques, et est entourée de

3 couches tissulaires soit: la dure-mère, la pie-mère et l'arachnoïde. Elle s'étend du tronc

cérébral jusqu'à la première vertèbre lombaire. La moelle épinière est divisée en 31

segments auxquels sont reliés 2 paires de nerf, sensoriel et moteur. On retrouve ainsi, 8

segments (et paires de nerfs) cervicaux, 12 thoraciques, 5 lombaires, 5 sacraux et 1

coccygien.

Une coupe transversale de la moelle épinière montre que celle-ci est constituée de

matière blanche en périphérie, de matière grise au centre et d'un canal central, dans lequel

on retrouve le liquide cérébrospinal. La matière blanche est divisée en colonne ascendante

(dorsale et latérale), relayant l'information sensitive au cerveau et descendante (ventrale),

relayant l'information motrice au reste du corps. Les colonnes sont subdivisées en

différentes voies, fascicules ou funicules. Nous n'entrerons toutefois pas dans une

description détaillée du rôle des différentes voies qui constituent les colonnes ascendantes

et descendantes.

Quant à la matière grise, elle est principalement constituée des corps cellulaires des

neurones. Des 100 milliard de neurones qui forment le système nerveux, selon Kandel,

Schwartz et Jessel, 10 milliard se retrouveraient probablement dans la moelle épinière

comme tel. En se basant sur les caractéristiques cytologiques et l'arrangement de ces

neurones, Rexed a subdivisé la matière grise de la moelle épinière en différentes régions ou

laminae (Fig. 1.1) (Rexed, 1952). Les laminae I-II se situent dans la zone dorsale de la

moelle épinière. Cette zone constitue les neurones principalement impliqués dans le relais

des inputs sensoriels concernant la température, la douleur et la sensation en général. Les

neurones des laminae III-V sont impliqués dans le relais des inputs sensoriels, pour la

plupart, non-nociceptifs, tels que la proprioception et le touché léger. Plusieurs neurones de

la laminae V reçoivent de l'information sensorielle nociceptive provenant des structures

viscérales. Les neurones de la lamina VI reçoivent principalement des inputs sensoriels

concernant la proprioception et la vibration provenant des muscles et des articulations. La

zone intermédiaire de la moelle épinière est constituée des laminae VII et X. La lamina VII

reçoit la plupart de ses afférences des laminae II à VI et des structures viscérales. Alors que

la majorité des neurones des couches superficielles de la corne dorsale reçoivent des inputs

uniquement ipsilatéraux, la lamina VII reçoit également de l'information provenant du côté

contralateral. Les laminae VIII-EX constituent la corne ventrale de la moelle épinière. Ils

contiennent les interneurones et neurones moteurs, impliqués dans l'exécution du

mouvement. De ce fait, leurs axones projettent vers les différents muscles. La laminae X

contient les neurones de la commissure de la matière grise de la moelle épinière. Ceux-ci

pourraient être impliqués dans la coordination ipsi- et controlatérale durant la locomotion.

Fig 1.1 Représentation schématique d'une coupe transversale de moelle épinière de chat au niveau lombaire.

La matière blanche et les différentes voies ascendantes et descendantes se retrouvent en périphérie (non-

illustrées) alors que la matière grise est localisée plus centralement. Les différentes laminae de Rexed y sont

représentées (figure reproduite de Rexed, 1952).

1.3 Contrôle supraspinal de la locomotion

En situation où la moelle épinière est intacte, les centres supraspinaux contrôlent les

mouvements volontaires et sont intimement liés à la locomotion. Es l'initient, la modulent,

aident au maintient de l'équilibre du corps, permettent la coordination de la locomotion

avec d'autres fonctions motrices et permettent son adaptation à certaines perturbations de

l'environnement (Orlovsky, 1991). Bien entendu, tout ceci est orchestré en intégrant les

inputs provenant de la périphérie. Donc, durant la locomotion, le rôle des centres

supraspinaux peut être divisé en 3 systèmes fonctionnels qui: 1) initie et contrôle la vitesse

de locomotion, 2) guide la trajectoire des membres en fonction des afférences visuelles et

3) ajuste de façon précise les patrons moteurs en intégrant les afférences proprioceptives

(Fig. 1.2).

Les travaux datant d'une cinquantaine d'années ont démontré l'existence d'une

région locomotrice mésencéphalique (MLR, mesenphalic locomotor region) (Shik et al.,

1966). Une stimulation électrique tonique de cette région déclenchait la locomotion. De

plus, l'augmentation de l'intensité de stimulation était positivement corrélée à la vitesse de

locomotion. Les axones de ces neurones ne font toutefois pas synapses directement avec le

centre locomoteur de la moelle épinière, mais avec les neurones de la formation réticulée

qui intègrent et transmettent les inputs à la moelle épinière, via la voie réticulospinale

(Deliagina et al., 2008; Brocard et al., 2010). D'autres régions ont également montré leur

capacité à initier la locomotion, telle la région locomotrice subthalamique et la région

locomotrice pontomédullaire (Mori et al., 1989; Noga et al., 1988).

Cu « s * «t tronc

Van les musclas du tronc et de la cutoM

Fig 1.2 Principales voies descendantes qui contrôlent le mouvement de façon volontaire. Pour la locomotion,

les centres supraspinaux initie et contrôle la vitesse de locomotion, guide la trajectoire des membres et ajuste

de façon précise les patrons moteurs en intégrant les afférences proprioceptives. Suite à une blessure à la

moelle épinière, ce contrôle est fortement limité ou aboli, (figure reproduite de lecerveau.mcgill.ca)

Plusieurs régions du cerveau ont montré leur importance afin d'intégrer

l'information visuelle et ajuster la locomotion en conséquence. Des expériences chez le chat

ont montré que le cortex moteur contribuait à cette tâche. À l'aide d'enregistrements de

neurones du cortex moteur, l'auteur a noté une augmentation des décharges lorsque les

animaux franchissaient des obstacles qui se trouvaient sur leur chemin (Drew, 1988). Plus

8

récemment, avec un paradigme similaire, une étude provenant de ce même laboratoire a

montré le rôle de la population de neurones du cortex pariétal postérieur dans le contrôle de

la coordination entre les membres durant la locomotion. Ces neurones seraient impliqués

dans l'intégration de l'information visuelle en estimant les attributs spatio-temporels des

obstacles par rapport à l'individu. Une lésion de cette région n'empêchait pas la locomotion,

mais celle-ci était sévèrement perturbée lorsque des obstacles se trouvaient sur le trajet

(Lajoie et al., 2010).

Le cervelet ajusterait, de façon précise, le patron locomoteur en régulant l'intensité

et les paramètres temporels. L'information proprioceptive provenant de membres ainsi que

les inputs du générateur central de patrons locomoteurs (CPG) vers les centre supraspinaux

se transmettraient au cervelet via les voies spino-cérébelleuses dorsale et ventrale. Celui-ci

intégrerait cette information avec les afférences vestibulaires et visuelles afin d'ajuster la

posture et de contrôler l'équilibre (pour revue, voir Morton et Bastian, 2004).

Après une lésion de la moelle épinière, les centres supraspinaux ne peuvent plus, ou

difficilement, initier la locomotion ou tout autre mouvement volontaire. Le contrôle de

l'équilibre et de la posture est perdu car l'intégration de l'information sensorielle est

perturbée. Toutefois les neurones de la moelle épinière demeurent activables, et une

locomotion peut toujours être induite, si la blessure ne touche pas le centre locomoteur de

la moelle épinière.

1.4 Générateur central de patrons locomoteurs (CPG)

La conception qui veut que la moelle épinière soit considérée comme un simple

centre de relais a été mise à jour. En plus de servir de liaison entre le SNC et le SNP, on y

retrouve des réseaux neuronaux impliqués dans la génération d'activité rythmique (CPG).

Un CPG est définit comme étant un circuit neuronal capable de générer une activité

rythmique indépendamment des inputs périphériques et supraspinaux. Ceux-ci sont, entre

autre, sollicités pour les fonctions de mastication, de déglutition, de vomissement, de

respiration, d'éjaculation et de locomotion (Kinkead, 2009; Nakamura et al., 1999;

Giuliano et Clément, 2005; Guertin et Steuer, 2009).

Les études qui ont mené au concept de CPG pour la locomotion - que nous référerons

uniquement à CPG pour la suite - découlent des travaux datant du début du 20e siècle

effectués par Maurice Philippson, Charles Sherrington et Thomas Graham Brown.

Philippson a montré que des chiens qui ont la moelle épinière sectionnée pouvaient générer

des mouvements rythmiques au niveau des pattes arrière (Philippson, 1905; revue dans

Clarac, 2008). Quelques années plus tard, dans son ouvrage « Flexion-reflex of the limb,

cross extension-reflex, and reflex stepping and standing » (Sherrington, 1910), l'auteur y

comparait les réflexes de flexion et d'extension croisée sur trois préparations animales

différentes : le chat décérébré, spinal et intact. Il a également montré les similitudes entre le

réflexe de flexion et la phase de balancement, ainsi qu'entre le réflexe d'extension croisé et

la phase d'appui lors de la locomotion. De plus ces expériences nous ont montré qu'il était

possible de générer un patron locomoteur en l'absence des centres supraspinaux. À la suite

de ses travaux, Sherrington émettait l'hypothèse que la locomotion était un enchaînement

de réflexes dont les paramètres étaient contrôlés par les afférences sensorielles, provenant

de la périphérie.

Thomas Graham Brown, en coupant les racines dorsales suivi d'une lésion complète

de la moelle, a montré que les afférences sensorielles n'étaient pas nécessaires à la

génération et au maintient d'un patron locomoteur. Ceci suggérait donc qu'il existerait,

dans la moelle épinière, un réseau capable d'induire et de moduler de tels mouvements

rythmiques. Par la suite, il soumettait l'hypothèse des demi-centres qui stipulait que la

locomotion était contrôlée par deux systèmes de neurones complémentaires. Chaque demi-

centre serait relié respectivement aux muscles fléchisseurs et extenseurs. Les demi-centres

s'inhiberaient mutuellement, ce qui assurerait l'activation d'un seul demi-centre à la fois.

L'action alternée des demi-centres serait expliquée par la fatigue des connections

inhibitrices (Fig. 1.3) (Graham Brown, 1911; 1914). Le concept d'une organisation d'un

réseau locomoteur constitué de demi-centres a été appuyé par les travaux de Jankowska.

Chez le chat spinal, suite à l'administration de L-Dopa, une stimulation des afférences du

10

réflexe de flexion (FRA) induit des bouffés d'activité de longue latence et de longue durée

dans les motoneurones fléchisseurs tout en inhibant les motoneurones extenseurs. Le même

patron d'activation et d'inhibition était observé dans les groupes d'interneurones reliés aux

fléchisseurs et extenseurs (Jankowska et al., 1967a; 1967b).

B

x 2 level

Rhythm Generator

f jtcnvCK Ftoot

c ^ î p ^ D r t^vn

ExW*o> Fl**»

2 * level Z.f*-vy Fl»«o»

3 level

Rhythm Generator Pattern Formation Last Order Interneurons

Sensory

Fig 1.3 Différents modèles de CPG basés sur le concept des demi-centres pour expliquer la génération et la

modulation de la locomotion. Les cercles représentent la population des interneurones spinaux et les losanges

représentent les motoneurones. Les connections excitatrices et inhibitrices sont représentées par les lignes se

terminant par des pointe de flèches et des petits cercles, respectivement. (A) Modélisation « classique » du

CPG à 1 niveau, tel que proposé par Graham Brown. (B) et (C) Modélisation du CPG à 2 niveau et (D) 3

niveaux possédant deux circuiteries distinctes pour la génération d'un rythme et la formation d'un patron

(Modifiée de McCrea et Rybak, 2008).

Chez les invertébrés les premières évidences de CPG ont été montrées chez

l'écrevisse (Hughes and Wiersma, 1960) et le criquet (Wilson, 1961). Ces chercheurs ont

enregistré de l'activité rythmique au niveau des pattes natatoires et des ailes sur des

préparations de système nerveux isolé. Suite à ces travaux, de nombreux modèles

11

d'invertébrés ont permis d'approfondir les connaissances sur le CPG (pour une revue, voir

Clarac et Pearlstein 2007). Les évidences de l'existence de CPGs chez les mammifères sont

apparues autour de la même période. En plus de leurs recherches portant sur le MLR, Shik

Severin et Orlovsky ont également contribué à montrer l'existence d'un CPG chez le chat.

Ils ont remarqué que les extenseurs des pattes arrière s'activaient de 5 à 10 ms avant que les

pattes touchent au sol (Shik et al., 1966) ce qui fût plus tard corroboré par d'autres

laboratoires (Halbertsma, 1983; Gorassini et al, 1994). Ceci suggère donc que l'activation

des extenseurs est d'origine centrale et non pas due à un traitement de l'information

périphérique.

Une démonstration convaincante de l'existence d'un CPG, pouvant générer un patron

d'activité complexe en l'absence de toute influence supraspinal et périphérique, a été

effectuée chez le chat (Grillner et Zangger, 1979). Suite à un isolement complet de la

moelle épinière (i.e. coupée des centres supraspinaux et des afférences périphériques), les

auteurs ont montré que l'administration de L-Dopa induisait encore une activité de type

locomoteur. En enregistrant dans divers muscles et nerfs, ils ont montré un patron

d'activation plus complexe qu'une simple activation et inhibition entre tous les fléchisseurs

et extenseurs des membres (Fig 1.4). Il a alors été proposé que le CPG soit plutôt constitué

d'unités génératrices d'activité rythmique (Unit burst generator, UBG), contrôlant

différentes populations de motoneurones et présentes à chaque articulation de chaque

membre (Grillner, 1981). Les UBGs permettaient d'expliquer la co-activation musculaire

entre fléchisseurs et extenseurs durant les phases de flexion et d'extension. Toutefois ce

modèle n'a pas encore trouvé de solution pour expliquer les patrons d'activation musculaire

plus complexes.

Plus récemment, des nouveaux modèles d'organisation du CPG à 2 ou 3 niveaux,

ayant pour base l'organisation en demi-centre, ont été proposés (Fig 1.3) (Perret et

Cabelguen, 1980; Rybak et al., 2006 a, b; Burke et al., 2001). L'organisation d'un CPG à 2

niveaux stipule que le CPG est constitué de 2 circuiteries distinctivement responsables de la

génération d'un rythme et de la formation d'un patron. La circuiterie de formation de

12

patrons projetterait ces connections vers les populations de motoneurones fléchisseurs et

extenseurs, lesquelles recevraient également les projections d'interneurones véhiculant

l'information sensorielle. Pour l'organisation d'un CPG à 3 niveaux, la circuiterie

responsable de la formation d'un patron projetterait ses connections vers un groupe

d'interneurones (interneurones de dernier ordre ou last order interneurons) qui projetterait,

à son tour, vers les motoneurones (Fig 1.3) (pour une revue, voir McCrea et Rybak, 2008;

Guertin, 2009).

HC Stance £ MC,

Swing

Gluteus maximus

iliopsoas

Quadriceps

Hamstrings

Triceps surae

Tibialis anterior

Normalized step cycle (%)

_ Extensors

| Flexors

Hip extensor

Hip flexors

Knee extensors

Knee flexors

Ankle extensors

Ankle flexor

TRENDS m Nauratcttno*

Fig 1.4 Patron d'activation musculaire normalisé lors de la marche chez l'humain. À noter que tous les

extenseurs ou fléchisseurs des différentes articulations ne sont pas activés en phase et que l'activation des

deux groupes de muscles n'est pas tout-à-fait réciproque, tel que stipulé dans le concept des demi-centres,

originalement proposé par Graham Brown. L'activation musculaire est séquentielle et suit un patron

déterminé. HC (heel contact) TO (Toe off). (Figure tirée de Capaday, 2002).

13

1.4.1 Évidences de CPG, localisation et neurones candidats

Tel que mentionné précédemment, la première démonstration complète de

l'existence de CPG chez les mammifères a été effectuée chez le chat, à la fin des années 70

par Grillner et Zangger. Malgré la démonstration et les évidences de l'existence d'un CPG

locomoteur chez plusieurs espèces animales, incluant l'humain, sa localisation précise et les

éléments constituant sa circuiterie restent à élucider. Les récentes avancées ont toutefois

permis d'en connaître un peu plus à ce sujet.

En utilisant une préparation de moelle épinière de rat néonatal qui était

compartimentée de façon à permettre l'activation de segments spécifiques par NMDA et 5-

HT et en enregistrant la locomotion fictive induite au niveau de racines dorsales, les

chercheurs ont conclu que les segments LI et L2 contenaient la circuiterie responsable

d'induire un rythme locomoteur et d'organiser les patrons d'alternance. Lorsque les

segments en aval de LI et L2 étaient stimulés, seule une activation tonique était enregistrée

(Cazalets et al., 1995). Chez la souris néonatale, les mêmes segments ont montré un rôle

primordial dans l'activation du CPG. L'administration de 5-HT pouvait induire un rythme

locomoteur (ipsi- et contralateral) dans les segments L2 et L5 après une transsection de la

moelle épinière en Tl 1-12. De plus, une alternance ipsilatérale, entre les segments L2 et L5

était perçue, suggérant l'alternance entre fléchisseurs et extenseurs. Lorsque la transsection

était effectuée entre les segments L3 et L4, la locomotion fictive était observée au niveau

L2, mais avait été abolie en L5 (Nishimaru et al., 2000). D'autres études ont montré que

l'induction de la locomotion fictive n'était pas restreinte aux segments LI et L2, mais que la

circuiterie serait plus étendue à travers la moelle épinière, avec un rôle plus dominant pour

les segments supra-lombaires (Kjaerulff et Kiehn, 1996; Cowley & Schmidt, 1997).

In vivo, chez la souris adulte, l'administration de drogues activatrice du CPG peut

induire des mouvements lorsque la transsection de la moelle épinière est effectuée entre les

segments T9 et T10. Cependant, lorsque la lésion est effectuée proche ou au niveau des

segments lombaires (T12-L1) ces mêmes drogues n'ont plus d'effets pro-locomoteurs,

14

même que les réflexes de flexion suite à une stimulation plantaire sont abolis (données non-

publiées).

Chez le chat adulte spinalisé en T13 et stimulé au niveau perineal, une locomotion

peut être induite lorsque de la clonidine est administrée restrictivement en L3-L4, ou en L5-

L7. Cette locomotion pouvait être inhibée lorsque de la yohimbine était administrée au

niveau de ces mêmes segments. Lorsque des lésions subséquentes étaient effectuées au

niveau des segments L3 et L4, la locomotion était complètement abolie (Marcoux et

Rossignol, 2000). De façon similaire, le chat spinal entraîné peut ré-exprimer ses fonctions

locomotrices et, suite au rétablissement locomoteur, une seconde lésion dans les segments

L2 ou rostral L3 n'affectait pas la locomotion. Cependant lorsque la seconde lésion était

effectuée dans la portion caudale de L3 ou en L4, la locomotion est complètement abolie et

les entraînements subséquents ne pouvaient permettre un « second rétablissement » des

fonctions locomotrices (Langlet et al., 2005).

Il semblerait donc que le réseau locomoteur serait étendu le long des segments

lombaires, mais que l'intégrité des segments L3 et L4 chez le chat et LI et L2 chez la

souris, le rat et l'humain (voir section 1.3.2) soit cruciale pour soutenir les fonctions

locomotrices (revue dans Guertin, 2009).

Plus récemment d'autres données ont apporté davantage de précision sur l'étendue

du réseau locomoteur chez les rats néonataux. Les stimulations électriques des afférences

sensorielles provenant des segments caudales, évoquaient une activité rythmique dans les

segments lombaires (Whelan et al., 2000; Delvolve et al., 2001; Strauss et Lev-Tov, 2003).

Lorsque les segments sacraux étaient complètement isolés par transsection, il était encore

possible de produire une activité rythmique dans ces derniers. Toutefois lorsque la moelle

épinière était intacte, l'activité rythmique était contrôlée par les segments rostro-lombaires.

D existerait donc un couplage entre les segments lombaires et sacraux pouvant être modulé

par les afférences sensorielles (Cazalets et Bertrand, 2000). Ceci expliquerait pourquoi des

stimulations sensorielles périnéales ou caudales chez le chat spinal améliorent les

performances locomotrices (Lovely et al., 1986; Barbeau et Rossignol, 1987). Chez le rat

15

paraplégique des stimulations épidurales de la moelle épinière au niveau lombaire (L2) et

sacral (SI) permet d'induire une meilleure locomotion que lorsque seulement un des

segments est stimulé (Courtine et al., 2009).

Quant à la distribution transversale de la circuiterie locomotrice, les études

électrophysiologiques et de marquages cellulaires ont montré que celle-ci serait

principalement localisée au niveau ventro-intermédiaire de la moelle épinière dans les

laminae VII-VIII et X (Kjaerulff et al., 1994; Tresch et Kiehn, 1999; Cina et Hochman,

2000; Dai et al., 2005).

Les outils moléculaires et génétiques ont permis d'approfondir nos connaissances

sur les interneurones constituant la circuiterie du CPG (pour une revue voir, Jessel et al.,

2000; Goulding et al., 2002; Kiehn et al., 2008; Guertin, 2009). Malgré le fait que leurs

connections ne sont pas encore toutes élucidées, il est possible d'associer un certain type

d'interneurones à un comportement lié à la locomotion. À l'aide de marqueurs

moléculaires, quatre classes de neurones ont été identifiées dans les portions intermédiaires

et ventrale de la moelle épinière : V0, VI, V2 (V2a glutamatergique et V2b gabaergique) et

V3. Les connections des interneurones V0 sont strictement controlatérales et sont associées

à l'alternance ipsi-controlatérale des membres. Les souris knock-out (KO) pour lesquelles

les interneurones V0 sont absents montrent des mouvements synchronisés bilatéraux ou des

sautillements durant la locomotion (Lanuza et al., 2004). Les interneurones VI seraient

inhibiteurs et associés au rythme locomoteur. Lorsque ces derniers sont abolis, les souris

montrent des patrons locomoteurs ralentis, par contre l'alternance entre les pattes arrière

gauche et droite demeure intact (Gosgnach et al., 2006). Les projections des interneurones

V2a seraient exclusivement ipsilatérales (Lundfald et al., 2007). Ceux-ci auraient un rôle

pour l'alternance gauche-droite et pour le maintient de la fréquence et de l'amplitude des

bouffées locomotrices (Crone et al., 2008). Les interneurones V2b seraient inhibiteurs, mais

leur rôle plus spécifique durant la locomotion reste à déterminer (Lundfald et al., 2007).

Les interneurones V3 seraient requis pour établir un rythme locomoteur robuste et balancé

au niveau des bouffées d'activité locomotrice en projetant leurs connections directement sur

les motoneurones contra-latéraux et sur des interneurones inhibiteurs (Zhang et al., 2008).

16

D'autres interneurones candidats pouvant former la circuiterie du CPG ont été

identifiés par l'expression du récepteur EphA4 ou du facteur de transcription HB9. Les

molécules EphA4 et leurs récepteurs sont requis pour le guidage axonal durant le

développement. Tout comme pour les interneurones VO, les interneurones EphA4 sont

reliés à l'alternance ipsi-controlatérale des membres puisque des souris KO pour le

récepteur ont une locomotion en sautillement (Coonan et al., 2001; Kullander et al., 2003).

En se basant sur les caractéristiques physiologiques et anatomiques des connections des

interneurones HB9, il a été proposé que ceux-ci auraient un rôle pour la génération du

rythme locomoteur (Wilson et al., 2005; Hinckley et al., 2005).

1.4.2 Évidences de CPG chez l'humain

Alors que l'existence d'un CPG a été démontrée chez les invertébrés et quelques

mammifères tels que le chat (Grillner et Zanger, 1979) et le marmoset (Fedirchuk et al.,

1998), la démonstration, hors de tout doute, de son existence chez l'humain demeure

problématique, mais des évidences indirectes de l'existence d'un centre locomoteur chez

l'humain ont été rapportées.

À la fin des années 1980, des chercheurs ont montré que l'activité rythmique causée

par le myoclonus chez un patient paraplégique pouvait être modulée par les FRAs. Les

chercheurs ont par la suite rapporté les similitudes dans l'organisation du réseau réflexe

entre ce patient et le chat spinal injecté à la L-Dopa, ce qui leur permettait de suggérer

l'existence d'un réseau locomoteur chez l'humain s'apparentant à celui du chat (Bussel et

al., 1988). Une étude de cas a rapporté une première évidence de l'existence d'un tel réseau

chez un blessé médullaire au niveau cervical (Calancie et al., 1994). Des mouvements

rythmiques involontaires de ses jambes, s'apparentant à un patron de locomotion, étaient

observés lorsque le sujet était étendu sur le dos, les hanches en extension. Les mouvements

se décrivaient comme une alternance entre flexions et extension des hanches, genoux et

chevilles. Cette étude ne peut être considérée comme étant une évidence convaincante d'un

17

CPG chez l'humain car le sujet était paraplégique incomplet et qu'il avait retrouvé une

certaine sensibilité au niveau de ses membres paralysés (touché léger, vibration, douleur et

température). Quelques années plus tard, une autre étude a montré des évidences de CPG

chez l'humain (Dimitrijevic et al., 1998). Par stimulations épidurales au niveau L2 chez six

patients paraplégiques complets, classifies ASIA A, les chercheurs ont induit des

mouvements rythmiques de type locomoteur dans les jambes. Lorsque les stimulations

étaient données en aval ou en amont du segment L2, une réponse tonique ou des

mouvements rythmiques, mais non-locomoteurs étaient induits. Plus récemment, des

mouvements rythmiques alternés entre des muscles antagonistes ont été rapportés chez un

patient paraplégique classifié ASIA A, affligé d'une lésion anatomiquement complète. Le

rythme de ces mouvements pouvait être modulé par les manipulations des hanches ou des

pincements de la peau (Nadeau et al., 2010). Les auteurs nous rappellent que des

observations similaires ont été effectuées 60 ans auparavant chez un patient ayant une

lésion complète de la moelle épinière (Kuhn, 1950). De plus en plus d'évidences de

l'existence d'un tel réseau chez l'humain nous proviennent des études portant sur

l'entraînement locomoteur suite à une blessure à la moelle épinière (voir section 3.2.3).

1.5 Rôle des afférences sensorielles durant la locomotion

Tel que discuté précédemment, le CPG dans la moelle épinière est capable de

générer la locomotion indépendamment des afférences sensorielles, mais celles-ci

interagissent dynamiquement avec le CPG pour moduler, voire contrôler (proposé par

Edgerton et al., 2008) les caractéristiques de la locomotion (Barbeau et Rossignol, 1987,

1994; Bélanger et al., 1996; Rossignol et al., 2006). En situation où la moelle épinière est

intacte toutes les informations provenant de la périphérie, qu'elles soient cutanées,

proprioceptives, visuelles, auditives ou vestibulaires sont perçues et intégrées dans le SNC

(moelle épinière et centre supraspinaux). Celui-ci sélectionne alors le patron moteur

approprié pour optimiser la locomotion à son environnement. Lorsqu'il y a blessure à la

moelle épinière, l'information cutanée et proprioceptive, qui est intégrée en majeure partie

au niveau spinal, augmente en importance puisque les afférences et efférences des centres

18

supraspinaux sont limitées ou complètement éliminées. Cette section mettra l'accent sur le

rôle des afférences sensorielles, intégrées dans la moelle épinière, durant la locomotion.

Parmi les principales fonctions qui sont associées aux afférences sensorielles

notons: 1) l'initiation de la locomotion et le contrôle de la durée des différentes phases du

cycle locomoteur, 2) la modulation du patron d'activité musculaire via différentes voies

réflexes et 3) l'adaptation de la locomotion à son environnement (Rossignol et al., 2006;

Hultborn et Nielsen, 2007; Frigon et Rossignol, 2008). Toutes ces fonctions sont

évidemment interconnectées durant la locomotion.

Les propriocepteurs de la hanche jouent un rôle important pour l'initiation. Chez le

chat spinal, Sherrington avait noté qu'une extension de la hanche induisait des mouvements

de type locomoteur (Sherrington, 1910). Pareillement, chez les patients paraplégiques chez

lesquels des évidences de CPG ont été rapportées, le degré d'extension ou de flexion de la

hanche initiait ou terminait les bouffées d'activité rythmique observées dans les jambes

(Calancie et al., 1994; Nadeau et al., 2010). De plus, le contrôle de la durée des différentes

phases du cycle locomoteur (phase de balancement ou phase d'appui) est, entre autre,

modulé par les propriocepteurs de la hanche. Lorsque la hanche et ses fléchisseurs

atteignent un certain seuil d'extension et d'étirement, la phase d'appui se termine et la phase

de balancement est initiée (McVea et al., 2005). La transition entre ces 2 phases est

interrompue si l'extension de la hanche est bloquée (Grillner et Rossignol, 1978). Les

afférences lb et la des extenseurs de la cheville influencent également les phases du cycle

locomoteur. Chez le chat décérébré, il a été montré que l'augmentation de la charge (load)

sur les muscles extenseurs prévenait l'initiation de la phase de balancement et à l'inverse, le

délestage (unloading) initiait la transition vers la phase de balancement (Duysens et

Pearson, 1980). La stimulation spécifique de ces afférences a permis de corroborer ces

observations (Conway et al., 1987; Guertin et al., 1995; Frigon et al., 2010). Durant la

locomotion fictive, la stimulation électrique de ces afférences durant la phase d'appui

prolongeait la durée des électroneurogrammes (ENGs) des extenseurs de la hanche, du

genou et de la cheville. Par conséquence, ceci prolongeait la phase d'appui. À l'opposé, la

stimulation de ces afférences durant la phase de balancement réinitialise la phase d'appui.

19

Les voies réflexes, particulièrement celle du réflexe d'étirement, pourrait informer la

moelle épinière et exercer un contrôle sur l'activité musculaire dépendamment de la tâche

(task-dependent) ou de la phase (phase-dependent). Chez l'humain, dans le soleus, le

réflexe de Hoffman ou réflexe-H (analogue électrique du réflexe d'étirement) est modulé

différemment selon la tâche exécutée. Comparativement à des sujets qui se maintiennent

debout, l'amplitude du réflexe diminue lorsque ceux-ci marchent et diminue davantage

lorsqu'ils courent (Capaday et Stein, 1986, 1987). Des résultats similaires ont été noté chez

le chat en locomotion fictive (Bennett et al., 1996; Gosgnach et al., 2000). L'amplitude du

réflexe d'étirement est aussi modulé selon la phase du cycle de marche i.e. l'excitabilité du

réflexe augmente graduellement du début de la phase d'appui jusqu'à la transition vers la

phase de balancement, où il est fortement diminué (Capaday et Stein, 1986). fl a été estimé

que de 30% à 60% de l'activité musculaire du soleus durant la phase d'appui serait

attribuable à l'étirement (réflexe d'étirement) de ce muscle (Yang et al., 1991).

Les afférences cutanées, associées aux voies réflexes, agissent sur l'activité

musculaire durant la locomotion dans le but de contrer les perturbations ou d'adapter la

locomotion aux exigences de l'environnement. Des expériences effectuées chez des chats

spinaux et intacts ont montré que le contact de la partie dorsale de la patte avec un obstacle

durant la phase de balancement induisait une flexion du genou suivit d'une flexion de la

cheville et de la hanche afin de passer par dessus l'obstacle (Forssberg et al., 1975, 1977;

Forssberg, 1979). Lorsque la même stimulation est donnée durant la phase d'appui, des

augmentations de courte latence dans l'amplitude du réflexe des extenseurs du genou et de

la cheville sont observées. Une réponse identique a été obtenue chez le chat en locomotion

fictive et pour lequel le nerf fibulaire superficiel a été stimulé (Quevedo et al., 2005). De

façon similaire, la stimulation du nerf tibial chez l'humain induit une flexion de la cheville à

la transition de la phase d'appui à la phase de balancement et une extension de la cheville à

la fin de la phase de balancement (Zehr et al., 1997).

Récemment il a été montré que suite à une lésion de la moelle épinière, les

afférences cutanées provenant de la patte étaient essentielles pour réaliser des tâches

20

requérant une certaine précision et pour le rétablissement locomoteur (Bouyer et Rossignol,

2003a, b). Chez les animaux intacts (non-spinalisés), la dénervation des pattes arrière

affectait peu la locomotion sur une surface nivelée, mais d'importants déficits étaient

observés lorsque les animaux devaient marcher sur les barreaux d'une échelle. Dans les

jours suivant la dénervation, les animaux étaient incapables de traverser l'échelle. Avec le

temps, les chats avaient développé une stratégie compensatoire afin d'effectuer la tâche.

Après spinalisation, la dénervation complète des pattes arrière empêchait l'animal de

supporter son poids durant la locomotion et de positionner ses pattes adéquatement sur le

tapis roulant. Par contre les animaux qui avaient une dénervation partielle pouvaient

retrouver leurs fonctions locomotrices.

21

C H A P I T R E I I - E A R L Y A D A P T A T I V E C H A N G E S I N C H R O N I C P A R A P L E G I C

MICE: A MODEL TO STUDY RAPID HEALTH DEGRADATION AFTER SPINAL CORD INJURY.

Ce chapitre passe en revue les changements au niveau de l'anatomie générale, des

propriétés musculaires et du profil lipidique, sanguin et hormonal observés suite à une

blessure à la moelle épinière. Nous y présentons principalement les observations faites dans

notre laboratoire, chez la souris paraplégique. Cet article a été publié dans Spinal Cord,

2008, 46 (3) : 176-80.

Abstract

Study design: Literature review

Objective: To describe quantitatively some of most important anatomic, systemic and

metabolic changes occurring soon (one month) after spinal cord trauma in mice.

Setting: University Laval Medical Center

Results: Significant changes in weight, mechanical and contractile muscle properties, bone

histomorphometry and biomechanics, deep vein morphology, complete blood count,

immune cell count, lipid metabolism, and anabolic hormone levels were found occurring

within one month in completely spinal cord transected (Th9/10) mice.

Conclusion: These data reveal that many changes in mice and humans are comparable

suggesting, in turn, that this model may be a valuable tool for neuroscientists to investigate

the specific mechanisms associated with rapid health degradation post-SCI.

Keywords: Health degradation; secondary consequences, paraplegic mouse; SCI; bone,

muscle.

22

Introduction

Spinal cord injury (SCI) leads generally to an irreversible loss of motor control and

sensations below the level of trauma. In recent years, the secondary consequences and

complications associated with chronic SCI have received increasing attention. Indeed, it is

now well-recognized that SCI patients, in particular those with complete or near-complete

lesions, develop important and often life-threatening complications after their accident. For

instance, muscle wasting, osteopenia or osteoporosis, hormone dysregulation,

cardiovascular problems and immune deficiency are among the problems typically

encountered by chronic SCI individuals (Giangregorio et al., 2006; Bauman et al., 1999;

Bauman et al., 2000; Cruse et al., 2000). Although many of these complications occur soon

after trauma, little is known about the detailed mechanisms underlying their development

and progression. Furthermore, no animal model had been characterized to specifically study

these health problems. Since mice are increasingly recognized as offering clear molecular

and genetic advantages over other species, we chose paraplegic mice as an animal model to

study the many anatomic and metabolic changes associated with health degradation after

SCI. In this article, we summarize recently published data mainly from our laboratory

reporting changes in weight, mechanical and contractile muscle properties, bone

histomorphometry and biomechanics, deep vein morphology, complete blood count,

immune cell count, lipid metabolism, and anabolic hormone levels.

Animal model

All data reported in this review, that are from our laboratory, were obtained using a

completely spinal cord transected (Tx) mouse model (Landry et al., 2004; Picard et al.,

2007; Rouleau et al., 2007 a, b). All experimental procedures were conducted in accordance

with the Canadian Council for Animal Care guidelines and accepted by the Laval

University Animal Care, Use, and Ethics Committee. In brief, adult male CD1 mice

(Charles River Canada, St-Constant, Quebec) weighing 30-40 g were used. Pre-operative

cares included subcutaneous injection of lactate-Ringer's solution (1 ml), analgesic (0.1

23

mg/kg, Buprenorphine), and antibiotic (5 mg/kg, Baytril). All surgical procedures were

performed under aseptic conditions in deeply anesthetised animals (2.5% isoflurane). After

exposing the area between the 6 and the 12th thoracic level, the spinal cord was completely

transected using microscissors inserted between the 9th and 10th thoracic vertebrae. The

opened skin area was then sutured and animals were placed for a few hours on heating

pads. Post-operative cares included subcutaneous injection of lactate-Ringers's solution (2

x 1 ml/day), buprenorphine (0.2 mg/kg/day) and Baytril (5 mg/kg/day). Complete spinal

cord Tx was confirmed by 1) full paralysis of the hindlimbs initially, 2) post-mortem visual

and microscopic examination of the spinal cord lesions, and 3) histological examination of

coronal or mid-sagittal spinal cord sections stained with luxol fast blue/cresyl violet for

myelinated axons and Nissl substance, respectively. Only data from complete spinal cord

Tx animals were use for analyses.

Results and Discussion

1. Body weight

After monitoring weekly body weights for one month, we found a rapid and

significant reduction in body weight in complete Tx mice (N = 29, 1 animal died upon

surgery). A sudden loss that reached 24% (P < 0.001) after only 7 days was monitored (see

Table l).5 Body weight values did not return to normal levels after 4 weeks. Post-mortem

measurements of individual limb mass and volume revealed greater losses beneath lesion

level compared with above lesion level. Indeed, at 7 days post-Tx, hindlimb mass and

volume decreased by 28% whereas, a 21% reduction was found in the forelimbs ( N = S , P <

0.001 and P < 0.05, respectively). A partial return to near normal values was found in the

forelimbs but not in the hindlimbs at 4 weeks post-Tx. These relatively rapid adaptive

changes are somewhat comparable to what has been observed generally in patients. A

significant weight loss has indeed been reported in patients soon after their accident (Cox et

al., 1985). Reasons underlying this initial weight loss are not fully understood. It is unclear,

for instance, the extent to which reduced physical activity, metabolic changes or hormone

24

dysregulations may play a role in this phenomenon. The decrease in hindlimb size may be

partially explained by muscular atrophy due to reduced muscle activity and paralysis per se

(see Section 2. Hindlimb muscles). However, the forelimb atrophy, which was somehow

unsuspected in paraplegic animals, suggests that factors other than reduced muscular

activity (i.e., forelimbs typically remained active to move around and to reach for food and

water) contribute also to induce these rapid changes (see also Section 6. Hormone

dysregulation). However, after a few years, this early weight loss is often transformed into

a weight gain. In fact, SCI patients generally tend to progressively undergo an increase in

weight (mainly fat tissue increase) leading to overweight and obesity problems (Cox et al.,

1985).

2. Hindlimb muscles

As mentioned above, the hindlimbs are specifically affected in complete paraplegic

mice. Results from our laboratory have demonstrated that part of this weight loss is due to a

specific decrease in muscle mass. Soleus muscles were dissected out (Af = 8), weighed and

tested in vitro for muscle properties. At 1 week post-Tx, a 32% (P < 0.001) loss in mass

was already detected in soleus muscles (Table 1, Landry et al., 2004). Similar values were

found several weeks post-Tx. This decrease in mass corresponded also to a proportional

decrease in strength. Indeed, using a set-up specifically designed for electrical muscle

stimulation (using an electromagnetic field) and force-generating measurements in vitro,

we found at 1 week post-Tx, a 33% (P < 0.001) decrease (43% at 2 week post-Tx) in

absolute maximal tetanic force (P0), combined with a 21% (26% at 2 week post-Tx) and

48% (P < 0.05) increase in time-to-peak tension (TPT) and half-time relaxation (Vi RT),

respectively (Landry et al., 2004). Studies in rats have reported that fibre-type conversion

(slow oxidative to fast-oxidative) may be induced later on after Tx (Lieber et al., 1986;

Talmadge et al., 2002). Slow-twitch fibres (type I) in the Soleus were found to

progressively acquire some of the biochemical profile and contractile properties of fast-

twitch fibres (type Ha or lib) after 3 months post-SCI. In turn, P0 values in late chronic Tx

rats were similar to those seen in early Tx mice. Since our results showed increased TPT

and Vi RT values at 1 and 2 weeks but a return towards control values at 4 week post-Tx,

25

data from late chronic rats and early chronic mice together provide evidence of bi-phasic

changes in muscle property after Tx (i.e., early transient increase in TPT and Yi RT values

followed by a sustained decrease). While the early transient changes in contraction and

relaxation times are likely due to rapid alterations in Ca2+-induced-Ca2+-release mechanism,

ryanodine and dihydropyridine receptors expression and free cytosolic Ca2+ concentration

(see Discussion, Landry et al., 2004), longer-term changes are most probably caused by

slower mechanisms including fibre-type conversion and protein degradation (i.e., triggered

by increased calpain, lysosomal and ubiquitin-mediated proteolysis). In chronic SCI

patients, biopsies and physiological tests using leg muscles have revealed comparable

changes (i.e., decreased muscle strength and size, conversion to fast-twitch properties, etc.)

to those in chronic Tx animals (Grimby et al., 1976; Burnham et al., 1997; Gerrits et al.,

1999; Scott et al., 2006). However, additional data from early SCI patients (less than 1

month) would be required before concluding that bi-phasic contractile property changes

(i.e., TPT and Vi RT), as seen in mice, exist also in humans.

3. Femoral bones

It is well-documented that SCI is associated with increasing risks of fracture. In fact,

nearly all SCI individuals experience a significant loss of bone mineral tissue (up to 30% in

the femora) leading to a marked increase of fracture incidence (Ragnarsson et al., 1981;

Zehnder et al., 2004). In Tx mice (Af = 12), histomorphometric data from our laboratory

revealed a drastic decrease in trabecular bone volume (-22%, P = 0.02), thickness (-11%, P

= 0.04), and number (-15%, P = 0.09) within 10-30 days post-trauma (Table 1, Picard et al.,

2007). Densitometric measurements using dual-energy X-ray absorptiometry on the

femoral bones of Tx mice (N=14) reported no change in bone mineral density (BMD) but a

14% reduction (P < 0.001) in bone mineral content (BMC). Other models of disuse and

immobilization have also reported comparable bone losses. For instance, a 10-30%

decrease in femoral cancellous tissue was found within a few weeks (up to 50% after 18

weeks) of unilateral hindlimb immobilization in adult female rats (Li et al., 1990). It

remains unclear to what extent, rapid bone loss in SCI patients (most of which are young

adults) shares similar mechanisms with hormone-related osteopenia or osteoporosis seen in

26

elderly people. However, given that clear differences in bone loss progression has been

observed between animal models of disuse and age/hormone-related models, this suggests

that differences in bone remodelling mechanisms may exist between young immobilized

patients and elderly people. In fact, a murine model of disuse (hindlimb immobilization

with a cast) provided evidence suggesting that bone loss occurring within a few days to a

few weeks post-immobilization involves both a sharp decrease of osteoblastic activity and a

rapid increase of osteoclastic activity (i.e., low osteocalcin and high acid phosphatase

levels, Rantakokko et al., 1999). Further studies in mice (e.g., SCI, cast-immobilization,

tail-suspension) are likely to provide new insights into the molecular mechanisms of rapid

bone loss after paralysis or immobilization.

4. Deep vein size and blood lipid profile

SCI is also associated with the development of deep venous thrombosis (DVT) in the

lower limbs and, hence, with rapidly increasing risks of cardiovascular and pulmonary

complications soon after trauma (Waring et al., 1991). However, the specific mechanisms

underlying DVT formation following SCI remain poorly understood. Using in vivo

confocal microscopy, we recently established that deep vein changes in size can be found

as soon as at 1 week post-Tx in mice (Rouleau et al., 2007b). In fact, the femoral and

saphenous veins were found to undergo a large increase (> 1.5-fold) in diameter (P < 0.01).

This change in venous diameter remained similar for the entire period studied (4 weeks) (N

= 20). In this same study, we also analyzed the blood lipid profile using a clinical chemistry

analyser (Olympus AU400, Melville, NY). We found also during the same period

decreased concentrations of cholestérols (-25%), triglycerides (up to -45%), low-density

lipoproteins (LDL, up to -55%), high-density lipoproteins (HDL, up to -14%) but not

platelets (N = 40). These results may appear surprising since high LDL-triglyceride levels

are generally associated with DVT formation in humans. However, comparable data were

found in patients, since acute chronic SCI subjects were found also to display low LDL-

triglyceride levels (Apstein et al., 1998). Results in acute SCI patients and early Tx mice

suggest that LDL-triglyceride changes are unlikely to contribute to DVT formation soon

after trauma. If changes in deep vein size, as seen in Tx mice, were to be found also in

27

patients, it would strongly suggest that deep vein enlargement is a leading factor in DVT

formation after SCI. We know, indeed, that venous stasis (reduced blood flow) is a key

factor in DVT formation after immobilization (Waring et al., 1991) and, interestingly, it has

been associated with blood vessel enlargement in pregnant women (Macklon et al., 1997).

5. Blood and bone marrow cell counts

Immune deficiency may lead to life-threatening complications after SCI. To examine

possible factors that may be associated with this pathological condition, we characterized

using a CELL-DYN 3700 automatic blood analyzer (Abbott Laboratories, North Chicago,

IL) changes in red and white blood cells after Tx in adult CD1 mice (N = 40). A complete

blood count revealed unchanged or moderately decreased erythrocyte, platelet, hemoglobin

and hematocrit levels (Rouleau et al., 2007a). In contrast, leukocyte counts were greatly

reduced in Tx mice compared with controls. Total leukocyte numbers decreased by 35% (P

= 0.002) at 1 week post-Tx and remained low at 2, 3, and 4 weeks (P < 0.05). A detailed

analysis of leukocyte subtypes including lymphocytes, monocytes, neurotrophils and

eosinophils, revealed the existence of differential modulatory changes. Lymphocyte

numbers were reduced by 47% (P < 0.001) on average (as much as 53% at 1 week post-

Tx). Monocytes and neutrophils generally remained unchanged whereas eosinophil counts

gradually decreased by 81% (P = 0.027) after 4 weeks. Analyses from bone marrow

samples revealed comparable changes. We found a general decrease in lymphocytes and

mixed changes in neutrophils, monocytes, and megakaryocytes after Tx (Rouleau et al.,

2007a). These results can be compared, to some extent, with those from SCI patients where

reduced lymphocyte levels (specifically lymphocytes-T et NK cells) were found at 3

months post-SCI (Cruse et al., 1992) which may perhaps contribute to the state of immune

deficiency found generally in SCI patients (Cruse et al., 2000; Nash et al., 2000).

6. Hormone dysregulation

Serum levels of testosterone, GH, DHEA, PTH, and insulin were examined using

ELISA in control and Tx mice at 7, 14, 21 or 28 days post-Tx (N = 40). We found early

28

transient changes in testosterone (decreased) and GH (increased) levels during the first 2

weeks post-Tx (Rouleau et al., 2007a). In contrast, DHEA, PTH, and insulin levels were

reduced throughout the time period studied. Specifically, levels of testosterone in Tx mice

were reduced by 40% and 50% at 1 and 2 weeks post-Tx, respectively. However, at 3 and

4 weeks post-Tx, testosterone concentration returned to near normal values with average

serum levels ranging from 12.08 ng/ml to 12.83 ng/ml. In contrast, GH serum levels

drastically increased at 1 week post-Tx with an average level 3 times greater than control

animals (357.5 ng/ml). However, at 2, 3, and 4 weeks post-Tx, GH concentration returned

to near normal levels with values ranging from 363.9 ng/ml to 417.6 ng/ml. On the other

hand, while testosterone and GH levels were only transiently changed, those of PTH,

DHEA, and insulin were diminished for the entire time period studied. Specifically, insulin

levels were reduced by 84.5% at 1 week post-Tx and remained low. DHEA serum levels

were reduced by as much as 75% a few weeks after Tx. Similar reductions were found with

PTH levels. Comparable changes have been found in patients. Acutely injured men were

found to display decreased serum testosterone during the first few weeks post-injury

(Naftchi et al., 1980). Increased GH and decreased PTH levels have also been reported in

chronic SCI patients (Mechanick et al., 1997). It remains unclear what role these hormonal

changes may play in health degradation post-SCI. However, results in mice suggest that

some of these changes may participate to immune deficiency since high correlations were

found between specific anabolic hormone and immune cell type changes. For instance, the

transient increase of GH was strongly correlated with changes in blood monocyte and

megakaryocyte levels. This is also supported by data from the literature showing that GH

receptors are expressed in peripheral mononuclear cell types and that GH can increase

macrophageal activity and stimulate progenitor cell hematopoiesis (Kiess et al., 1985;

Edwards et al., 1992; Blazar et al., 1995). Regarding insulin and PTH, their sustained

decrease post-Tx was highly correlated with the decrease in total blood leukocytes and

lymphocytes. This finding is supported by results showing that 1) their receptors are

expressed in leukocytes including lymphocytes and, 2) these hormones can stimulate

lymphocyte synthesis in vivo (Helderman et al., 1978; Atkinson et al., 1987; Walrand et al.,

2005; Perry et al., 1984).

29

Conclusion

This review article has reported essentially recent data from a mouse model of SCI.

Anatomic, systemic, and metabolic changes were found to rapidly occur in adult mice after

a transection of the spinal cord at the low-thoracic level (complete paraplegia). To some

extent, the early changes found in paraplegic mice where comparable to those reported in

SCI patients. This supports the idea that a detailed characterization of health degradation in

this animal model may provide the basis for additional studies and, hence, contribute to

understand further the molecular and cellular changes underlying health degradation in

patients. This work may eventually contribute to the development of new therapeutic

approaches aimed at preventing these changes and life-threatening complications after SCI.

References

Apstein MD, George BC. Serum lipids during the first year following acute spinal cord

injury. Metabolism 1998; 47: 367-370.

Atkinson MJ, Hesch RD, Cade C, Wadwah M, Perils AD. Parathyroid hormone stimulation

of mitosis in rat thymic lymphocytes is independent of cyclic amp. J Bone Miner Res 1987;

2: 303-309.

Bauman WA, Kahn NN, Grimm DR, Spungen AM. Risk factors for atherogenesis and

cardiovascular autonomic function in persons with spinal cord injury. Spinal Cord 1999;

37: 601-616.

Bauman WA, Spungen AM. Metabolic changes in persons after spinal cord injury. Phys

MedRehabil Clin N Am 2000; 11: 109-140.

Blazar BR, Brennan CA, Broxmeyer HE, Shultz LD, Vallera DA. Transgenic mice

expressing either bovine growth hormone (bgh) or human gh releasing hormone (hgrh)

30

have increased splenic progenitor cell colony formation and DNA synthesis in vitro and in

vivo. Exp Hematol 1995; 23: 1397-1406.

Burnham R, Martin T, Stein R, Bell G, MacLean I, Steadward R. Skeletal muscle fibre type

transformation following spinal cord injury. Spinal Cord 1997; 35: 86-91.

Cox SA, Weiss SM, Posuniak EA, Worthington P, Prioleau M, Heffley G. Energy

expenditure after spinal cord injury: An evaluation of stable rehabilitating patients. J

Trauma 1985; 25: 419-423.

Cruse JM, Lewis RE, Bishop GR, Kliesch WF, Gaitan E. Neuroendocrine-immune

interactions associated with loss and restoration of immune system function in spinal cord

injury and stroke patients. Immunol Res 1992; 11: 104-116.

Cruse JM, Lewis RE, Dilioglou S, Roe DL, Wallace WF, Chen RS. Review of immune

function, healing of pressure ulcers, and nutritional status in patients with spinal cord

injury. J Spinal Cord Med 2000; 23: 129-135.

Edwards CK, 3rd, Arkins S, Yunger LM, Blum A, Dantzer R, Kelley KW. The

macrophage-activating properties of growth hormone. Cell Mol Neurobiol 1992; 12: 499-

510.

Gerrits HL, De Haan A, Hopman MT, van Der Woude LH, Jones DA, Sargeant AJ.

Contractile properties of the quadriceps muscle in individuals with spinal cord injury.

Muscle Nerve 1999; 22: 1249-1256.

Giangregorio L, McCartney N. Bone loss and muscle atrophy in spinal cord injury:

Epidemiology, fracture prediction, and rehabilitation strategies. J Spinal Cord Med 2006;

29: 489-500.

31

Grimby G, Broberg C, Krotkiewska I, Krotkiewski M. Muscle fiber composition in patients

with traumatic cord lesion. Scand J Rehabil Med 1976; 8: 37-42.

Helderman JH, Strom TB. Specific insulin binding site on t and b lymphocytes as a marker

of cell activation. Nature 1978; 274: 62-63.

Kiess W, Butenandt O. Specific growth hormone receptors on human peripheral

mononuclear cells: Reexpression, identification, and characterization. J Clin Endocrinol

Metab 1985; 60: 740-746.

Landry E, Frenette J, Guertin PA. Body weight, limb size, and muscular properties of early

paraplegic mice. J Neurotrauma 2004; 21: 1008-1016.

Li XJ, Jee WS, Chow SY, Woodbury DM. Adaptation of cancellous bone to aging and

immobilization in the rat: A single photon absorptiometry and histomorphometry study.

AnatRec 1990; 227: 12-24.

Lieber RL, Johansson CB, Vahlsing HL, Hargens AR, Feringa ER. Long-term effects of

spinal cord transection on fast and slow rat skeletal muscle. I. Contractile properties. Exp

Neurol 1986; 91: 423-434.

Macklon NS, Greer IA, Bowman AW. An ultrasound study of gestational and postural

changes in the deep venous system of the leg in pregnancy. Br J Obstet Gynaecol 1997;

104: 191-197.

Mechanick JI, Pomerantz F, Flanagan S, Stein A, Gordon WA, Ragnarsson KT.

Parathyroid hormone suppression in spinal cord injury patients is associated with the

degree of neurologic impairment and not the level of injury. Arch Phys Med Rehabil 1997;

78: 692-696.

32

Naftchi NE, Viau AT, Sell GH, Lowman EW. Pituitary-testicular axis dysfunction in spinal

cord injury. Arch Phys Med Rehabil 1980; 61: 402-405.

Nash MS. Known and plausible modulators of depressed immune functions following

spinal cord injuries. J Spinal Cord Med 2000; 23: 111-120.

Picard S, Lapointe NP, Brown JP, Guertin PA. Histomorphometric and densitometric

changes in the femora of spinal cord transected mice. Spinal Cord 2007 (submitted).

Perry HM, 3rd, Chappel JC, Bellorin-Font E, Tamao J, Martin KJ, Teitelbaum SL.

Parathyroid hormone receptors in circulating human mononuclear leukocytes. J Biol Chem

1984;259:5531-5535

Ragnarsson KT, Sell GH. Lower extremity fractures after spinal cord injury: A

retrospective study. Arch Phys Med Rehabil 1981; 62: 418-423.

Rantakokko J, Uusitalo H, Jamsa T, Tuukkanen J, Aro HT, Vuorio E. Expression profiles

of mrnas for osteoblast and osteoclast proteins as indicators of bone loss in mouse

immobilization osteopenia model. J Bone Miner Res 1999; 14: 1934-1942.

Rouleau P, Ung RV, Lapointe NP, Guertin PA. Hormonal and immunological changes in

mice after spinal cord injury. J Neurotrauma 2007a; 24: 367-378.

Rouleau P, Guertin PA. Early changes in deep vein diameter and biochemical markers

associated with thrombi formation after spinal cord injury in mice. J Neurotrauma 2007b;

24(8): xx-xx (in press).

Scott WB, Lee SC, Johnston TE, Binkley J, Binder-Macleod SA. Contractile properties and

the force-frequency relationship of the paralyzed human quadriceps femoris muscle. Phys

Ther 2006; 86: 788-799.

33

Talmadge RJ, Roy RR, Caiozzo VJ, Edgerton VR. Mechanical properties of rat soleus after

long-term spinal cord transection. JAppl Physiol 2002; 93: 1487-1497.

Walrand S et al. Insulin regulates protein synthesis rate in leukocytes from young and

elderly healthy humans. Clin Nutr 2005; 24: 1089-1098.

Waring WP, Karunas RS. Acute spinal cord injuries and the incidence of clinically

occurring thromboembolic disease. Paraplegia 1991; 29: 8-16.

Zehnder Y et al. Long-term changes in bone metabolism, bone mineral density, quantitative

ultrasound parameters, and fracture incidence after spinal cord injury: A cross-sectional

observational study in 100 paraplegic men. Osteoporos Int 2004; 15: 180-189.

34

Table 1. Summary of the anatomic, metabolic and systemic changes in the complete paraplegic mouse model.

General anatomy body weight y 24% Lipid profile HDL I n * forelimb y 21% LDL | 55%

hindlimb ^ 28% cholesterol y 25% triglyceride ^f 45%

Muscle (soleus)

weight y 32% TPT ^ 26% VS RT ^ 48% Absolute Po T 43%

ref 22

Blood cells erythrocyte «-» hemoglobin «-» hematocrit • * platelet «-»

ref 15

Bone (femora)

BMC V 14% BMD «-»

ref 22

Blood immune cells leukocytes y 35% lymphocyte ^r 53%

volume y 22% monocyte «-»

thickness y 10% neutrophil «-»

number y 15% eosinophil y 81%

separation ^ 24%

ref. 19 ref. 22

Anabolic hormone testosterone ^ 50% Blood vessel (diameter)

femoral f 162%

saphene ^ 155%

growth hormone ^ 300% DHEA I 75% insuline y 85% PTH | 70%

y decrease A increase «-» no change

35

CHAPITRE n (SUITE) - CONSÉQUENCES D'UNE LÉSION DE LA MOELLE

ÉPINIÈRE SUR LE MUCLE ET LES FIBRES MUSCULAIRES.

2.1 Atrophie musculaire et conversion des fibres

Étant donné que des analyses plus approfondies ont été effectuées pour évaluer les

effets de l'entraînement et de l'administration de substances anaboliques sur les muscles des

pattes arrière de souris paraplégiques, cette section se veut un complément d'information

concernant l'atrophie des fibres musculaires discutée dans la section précédente.

Les raisons menant à une atrophie musculaire sont généralement séparées en 3

groupes : 1) induite par l'inactivité, 2) la malnutrition ou la maladie (i.e. cachexie) et 3) le

vieillissement (i.e. sarcopénie). Malgré la différence des événements déclencheurs, toutes

ces formes d'atrophie musculaire se caractérisent par une diminution du diamètre des fibres

musculaires, de la force pouvant être générée ainsi qu'une augmentation de la fatigue

musculaire. Ces conséquences sont le résultat de changements au niveau de la synthèse et

de la dégradation protéique (Jackman et Kandarian, 2004; Kandarian et Jackman, 2006).

Lorsque les voies de signalisation intracellulaire de dégradation protéique sont activées,

plusieurs systèmes de protéolyse, fonctionnant de concert, sont enclenchés afin d'induire

l'atrophie musculaire.

Le système calpains calcium-dépendant et caspases-3 serait impliqué dans les stades

initiaux de l'atrophie. Il participerait au désassemblage myofibrillaire en ciblant les

protéines telles la connectine, vinculine, protéine-C, et la nebuline (Huang & Forsberg,

1998). Toutefois, il est connu que ce système ne peut dégrader l'actine et la myosine. Les

autres systèmes de protéolyse concernent l'ubiquitination, les lysosomes et les protéasomes.

La façon dont les protéines sont marquées par l'ubiquitination détermine le système de

dégradation de la protéine. Lorsque les protéines sont mono- ou bi-ubiquitinisées, elles sont

reconnues par les lysosomes. Ce système ciblerait davantage les protéines membranaires,

incluant les récepteurs, les ligands, les canaux et les transporteurs (Hicke & Dunn, 2003).

36

D'autre part, lorsque celles-ci sont marquées d'une chaîne d'ubiquitines, les protéasomes

sont responsables de leur dégradation (Fig 2.1) (Jagoe & Goldberg, 2001).

W Uhinui t inat inn Ubiquitination

Repeated

26S Proteasome

Fig. 2.1 Les 3 systèmes de protéolyse impliqués dans la dégradation protéique lors de l'atrophie musculaire.

Processus associés à A) calpains (et caspase-3) B) lysosomes et C) protéasomes (Figure tirée de Jackman et

Kandarian, 2004).

En plus d'induire de l'atrophie musculaire, l'inactivité engendre une transformation

du phénotype des fibres musculaires. Chez l'animal adulte, il existe 4 types de fibres

musculaires. Les fibres à contraction lente, ou fibres de type I, et les fibres à contraction

rapide ou de type Ha, IIx et Hb. En générale, la vitesse de contraction et la force produite

sont déterminées selon l'ordre suivant: lib > IIx > Ha > I, et ceci est inversement

proportionnel à la résistance à la fatigue. Toutefois, l'ordre pourrait ne pas correspondre à

la réalité chez les blessés médullaires.

37

Les analyses sur la conversion des fibres musculaires sont généralement effectuées

dans le soleus, extenseur de la cheville, muscle antigravitaire et principalement constitué de

fibres de type I. Après une lésion de la moelle épinière (ou en état d'inactivité prolongée) on

observe une conversion du phénotype de celles-ci, i.e. les fibres de type I se modifient en

type de fibre II. Cette transition a pour conséquence de modifier les propriétés contractiles

(Lieber et al., 1986 a, b, Landry et al., 2004, tel que discuté dans la première partie du

chapitre II). D'autres expériences ont montré que cette conversion se faisait

progressivement, alors que des fibres montraient un type « hybride », i.e. que les fibres

possédaient 2 phénotypes ou plus à la fois soit : I + Ha, I + IIx, I + Ha + IIx, ou Ha + IIx. Il

a même été suggéré que les fibres hybrides pourraient constituer un phénotype stable

puisqu'un an après la spinalisation, certains types hybrides étaient encore décelés

(Talmadge et al., 1999). Chez le rat ou la souris, très peu, voire aucune fibre de type lib ne

sont observées dans les soleus d'animaux intacts ou paraplégiques (Talmadge et al., 1999).

Les mécanismes sous-jacents à l'atrophie et à la conversion des fibres musculaires

ne sont pas tous élucidés (Zhang et al., 2007). Toutefois, certaines stratégies mises en place

dans le but d'augmenter l'activation neuromusculaire, comme les stimulations électriques

musculaires, la surcharge fonctionnelle sur les jambes et les exercices d'endurance

pourraient ralentir ou renverser ces effets secondaires.

2.2 Substances anaboliques pour contrer l'atrophie musculaire

Plusieurs substances aux propriétés anaboliques peuvent effectivement contrer ou

renverser l'atrophie musculaire. Parmi celles-ci nous allons élaborer sur le clenbuterol et la

testosterone, 2 substances aux propriétés anaboliques.

Le clenbuterol a été développé pour traiter les troubles respiratoires reliés à l'asthme

(Anderson et Wilkins, 1977). Des études subséquentes ont montré son efficacité à stimuler

la croissance musculaire (MacLennan et Edwards, 1989; Carter et al., 1991; Choo et al.,

38

1992; Lynch et al., 1999). Son efficacité pour renverser l'atrophie ou causer l'hypertrophie

musculaire tient du fait qu'il agit au niveau des récepteurs 62-adrénergiques. L'activation de

ces derniers enclenche les cascades cellulaires impliquées dans la synthèse protéique et

inhibe les mécanismes de dégradation protéique (Lynch et Ryall, 2008). Le clenbuterol a

montré ses propriétés anaboliques dans diverses situations de pertes musculaires. À titre

d'exemple, l'administration de clenbuterol induisait une augmentation de la masse

musculaire et de la force produite par le soleus et l'extenseur digitorum longus (EDL) sur

des modèles de souris présentant les symptômes de dystrophic musculaire, lors de

dénervation musculaire et dans les cas d'inactivité induite par suspension (hindlimb

unloading) (Hayes et William, 1994; Maltin et al., 1986; Zeman et al., 1987; Dodd et

Koesterer, 2002).

Les effets du clenbuterol ont également été étudiés lors de lésion à la moelle

épinière mais, étonnamment, les effets de cette substance n'ont pas été évalués sur le

muscle. Il s'est avéré que cette substance améliorait le rétablissement locomoteur et

réduisait les dommages anatomiques de la contusion et les risques de scoliose suite à une

blessure à la moelle épinière (Zeman et al., 1999; Zeman et al., 1997).

La testosterone, hormone stéroïdienne principalement sécrétée par les testicules

chez l'homme, est bien connue pour ses propriétés anaboliques (Bhasin et al., 1996;

Graham et al., 2008; Choong et al., 2008). Les hormones stéroïdiennes sont généralement

utilisées pour augmenter la masse musculaire, diminuer la masse adipeuse et augmenter les

performances athlétiques. Les effets hypertrophiques de la testosterone passent par

l'activation des récepteurs androgéniques, ce qui entraîne une augmentation de la synthèse

protéique. Les voies de dégradation protéique ne sembleraient pas affectées (Ferrando et

al., 1998). Parmi les mécanismes menant à l'hypertrophie musculaire, notons la

prolifération des cellules satellites, l'augmentation de l'accrétion myonucléaire, la

différenciation des cellules pluripotentes en lignées cellulaires myogéniques et l'inhibition

de la differentiation adipogénique (Herbst te Bhasin, 2004; Kadi, 2008). La testosterone

favorise l'hypertrophie des fibres de type I et II mais il semblerait que les fibres de types I

seraient plus sensibles aux traitements (Hartgens et al., 1996; Sinha-Hikim et al., 2002).

39

Chez la souris nous avons montré que la testosterone en circulation dans le système

diminuait rapidement suite à une transsection de la moelle épinière, ce qui pourrait

contribuer à l'importante atrophie musculaire observée (Rouleau et al., 2007).

L'administration de testosterone a montré des effets anaboliques sur une diversité de

modèles d'atrophie musculaire, tels que la sarcopénie, l'amyotrophie spinale, l'atrophie

induite par administration de glucocorticoïdes et lors de la gonadectomie (Kovacheva et al.,

2010; Johansen et al., 2009; Zhao et al., 2008b; Axell et al., 2006). Chez l'homme,

l'administration supra-physiologique de testosterone, avec ou sans entraînement, augmente

la masse et la force musculaire. De plus, ces augmentations sont corrélées positivement à la

dose administrée (Bhasin et al., 1996; Bhasin et al., 2001). Cette hormone est également

utilisée lors de thérapie de remplacement de testosterone, entre autre, pour contrer les

diverses conséquences du vieillissement et de l'hypogonadisme (Gooren et Bunck, 2004;

Zitzmann et Nieschlag, 2000).

Par ailleurs, il a été suggéré que la testosterone pouvait également avoir un rôle

neuroprotecteur suite à des lésions de nerfs en facilitant le rétablissement moteur (Kujawa

et al., 1989; Brown et al., 1999). D'autres évidences montrent que la testosterone peut être

synthétisée dans le CNS, au niveau de l'hippocampe et que ces metabolites influençaient

rapidement la formation de nouvelles synapses. De plus, la supplementation de testosterone

chez des rats castrés augmente la densité des boutons dendritiques en 48 heures dans l'aire

Cal de l'hippocampe. Par le fait même, on retrouve des récepteurs androgènes dans cette

aire. Ceci suggère donc que, du moins dans l'hippocampe, la testosterone et ces metabolites

participent à la modulation de plasticité synaptique (Hirotaka et al., 2007). Ainsi,

l'administration exogène de testosterone et/ou de clenbuterol pourrait s'avérer bénéfique sur

notre modèle animal de paraplégie tant au niveau musculaire que du rétablissement moteur.

40

CHAPITRE III - POTENTIEL DE PLASTICITÉ DE LA MOELLE ÉPINIÈRE SUITE À UNE LÉSION

3.1 Le choc spinal, évidences de plasticité sous-lésionnelle

Le choc spinal se caractérisait originalement par une dépression ou inhibition

temporaire des réflexes spinaux sous-lésionnels. La durée de cet état varie selon l'espèce

animale mais, également selon l'interprétation qu'on fait du phénomène. Afin de mieux

comprendre et définir le choc spinal, un modèle en 4 phases a été élaboré (Ditunno et al.,

2004):

1) aréflexie ou hypo-réflexie (0-1 jour)

2) retour initial des réflexes (1-3 jours)

3) hyper-réflexie initiale (1-4 semaines)

4) hyper-réflexie finale et/ou spasticité (1-12 mois)

Selon ces auteurs, une description plus détaillée du choc spinal était nécessaire

puisque chez l'humain, les réflexes ne sont pas tous abolis. De plus, les blessés médullaires

se remettent de l'état d'aréflexie/hypo-réflexie de façon graduelle, sur plusieurs mois. De ce

fait, ils jugeaient important d'élaborer une description plus adaptée du phénomène. Malgré

une compréhension encore incomplète des mécanismes neuronaux qui expliquent les

différentes phases de ce nouveau modèle, les études et les observations effectuées nous font

constater tout le potentiel plastique et la réorganisation neuronale qui s'effectue dès les

premiers instants d'une BME.

Tout d'abord, l'aréflexie/hypo-réflexie serait probablement due à une diminution de

l'excitabilité de base, produite par une perte de la transmission de 5-HT et de noradrenaline

provenant des centres supraspinaux. Ceci engendrerait une hyperpolarisation des

motoneurones (Schadt et Barnes, 1980). Une perte d'excitabilité au niveau des

motoneurones-y a également été répertoriée (Weaver et al., 1963). Ces derniers contrôlent

la sensibilité à l'étirement musculaire, ce qui aide au maintient de la tension musculaire. Par

41

ailleurs, une augmentation de l'inhibition spinale a été observée (Ashby et al., 1974).

L'augmentation serait causée par une perte des influences inhibitrices provenant des centres

supraspinaux, qui font connections avec des neurones inhibiteurs de la moelle épinière. En

parallèle, des augmentations de l'expression des récepteurs glycinergiques et de l'enzyme

GAD-67, impliquée dans la synthèse de GABA, pourrait être en partie responsables de

l'augmentation d'inhibition (Edgerton et al., 2001; Tillakaratne et al., 2002). D'autre part,

une dégénérescence synaptique ainsi qu'une rétraction dendritique contribueraient

également à accentuer l'état réfractaire observé dans la moelle épinière (Llewellyn-Smith et

Weaver, 2001).

Le retour initial des réflexes ainsi que l'hyper-réflexie/spasticité seraient expliqués

par une augmentation en densité des récepteurs pour certains neurotransmetteurs

excitateurs, tels les récepteurs NMDA (Grossman et al., 2000, 2001), les récepteurs 5-HT2A

(Basura et al., 2001; Kong et al., 2010), a r et a2-adrénergiques (Roudet et al., 1993, 1994;

Giroux et al., 1999) et 5-HTIA/7 (Giroux et al., 1999). Ceci permettrait d'accroître la

sensibilité aux neurotransmetteurs. En plus de contribuer à l'état d'hyper-excitabilité, l'up-

régulation de ces récepteurs pourrait également jouer un rôle dans la récupération motrice

spontanée. L'hyper-réflexie et la spasticité seraient également expliquées par une

diminution de l'inhibition, entre autre par une diminution de l'inhibition réciproque, de

l'inhibition autogénique lb et de l'inhibition pré-synaptique (Nielsen et al., 2007).

D'autres évidences de changements sous-lésionnels nous proviennent des facteurs

neurotrophiques. En effet, des variations au niveau de l'expression de BDNF, NT-3,

synapsin 1 et NGF sont observées (Gomez-Pinilla et al., 2004; Li et al., 2007). Ces facteurs

neurotrophiques jouent un rôle important dans le développement et la plasticité du système

nerveux. Par ailleurs, les gènes à expression précoce ou immediate early genes (ŒGs) sont

également reconnus comme signes de plasticité. De ce fait, dans les heures et les jours

suivant une BME, au niveau des segments L1-L2, l'expression des TEGs est respectivement

augmentée et diminuée pour c-fos et Nor-1 dans la zone dorsale et intermédiaire de la

moelle épinière (Landry et al., 2006a).

42

3.2 Réorganisation fonctionnelle de la moelle épinière

Dans la section précédente, tous les changements et les mécanismes proposés pour

expliquer les phases d'aréflexie, de retour initial des réflexes et de l'hyper-réflexie/spasticité

caractérisant le choc spinal nous montrent l'étendue du potentiel plastique de la moelle

épinière. Depuis plusieurs années, il a été montré qu'il était possible de retrouver les

fonctions locomotrices suite à une BME car le réseau locomoteur de la moelle épinière

demeure fonctionnel (lorsque ce dernier n'est pas directement touché par la BME).

3.2.1 Récupération motrice spontanée

Il a été montré que sans entraînement ou stimulation pharmacologique, le chat

néonatal spinal récupérait ses fonctions locomotrices avec support de poids (Forssberg et

al., 1980 a, b). On observe aussi une récupération « spontanée » chez le chat adulte,

toutefois la queue ou le périnée de l'animal doit être stimulé (de Leon et al., 1998). Les rats

néonataux qui ont subit une transsection de la moelle épinière pouvaient aussi retrouver

leur fonctions locomotrices avec support de poids, alors que ceux transsectés lors du

sevrage montraient une motricité limitée, caractérisée par des mouvements qualifiés de

spastiques (Stelzner et al., 1975). Chez la souris adulte spinale, à l'exception d'une étude

(Leblond et al., 2003), le rétablissement locomoteur est limité à des mouvements de type

locomoteur ou non-locomoteur, sans support de poids (Fong et al., 2005; Guertin et al.,

2005; Lapointe et al., 2006, voir chapitre 5). Les mécanismes responsables de cette

récupération motrice spontanée ne sont pas encore élucidés, mais il est fort probable que

ceux-ci s'apparentent ou soient directement reliés au retour des réflexes et à l'hyper-réflexie,

tel que discuté dans la section précédente (voir section 3.1).

Étant donné ce recouvrement plutôt limité, et pour faciliter la réorganisation de la

moelle épinière en un état plus « fonctionnel », plusieurs stratégies ont été élaborées. Dans

le cadre de ce projet, nous consacrerons plus d'importance à l'entraînement locomoteur ainsi

qu'à l'utilisation d'agents pharmacologiques pouvant activer le CPG.

43

3.2.2 Entraînement locomoteur chez l'animal

Depuis maintenant plusieurs années, il a été montré que des chats spinaux pouvaient

retrouver une certaine capacité locomotrice lorsqu'ils étaient entraînés à marcher sur un

tapis roulant (Lovely et al., 1986; Barbeau et Rossignol, 1987). Même sans entraînement,

les chats adultes montraient un rétablissement de la locomotion. Toutefois, ce

rétablissement était plus limité lorsqu'on le comparait à des animaux entraînés. En effet, les

animaux non-entraînés effectuaient autour de 25 cycles locomoteurs sur un tapis roulant qui

atteignait une vitesse de 0.3 m/s alors que les chats entraînés s'exécutaient sur le tapis à une

vitesse atteignant 0.8 m/s et pouvaient effectuer autour de 100 cycles locomoteurs. Ces

mêmes auteurs ont également montré que les performances locomotrices déclinaient

suivant l'arrêt de l'entraînement. Par contre le retour des fonctions locomotrices était

accéléré lorsque ces animaux étaient ré-entraînés (de Leon et al., 1998a). D'autre part, ces

mêmes auteurs ont montré que la récupération motrice est dépendante de la tâche pour

laquelle les animaux sont entraînés. Des chats entraînés à marcher sur tapis roulant ont une

excellente récupération locomotrice, par contre ces mêmes animaux ont davantage de

difficultés à maintenir une posture adéquate lorsqu'ils doivent se tenir debout. Inversement,

des animaux entraînés à se maintenir debout réussissent bien à cette tâche, mais offrent une

performance locomotrice limitée sur tapis roulant (de Leon et al., 1998b).

En parallèle, des travaux effectués par le Dr Rossignol et son équipe, ont également

montré que suite à un entraînement sur tapis roulant, les chats paraplégiques récupéraient

progressivement leur capacité locomotrice (Barbeau et Rossignol, 1987, Rossignol et al.,

2002). Dans les premiers jours d'entraînement, les chats étaient entraînés à l'aide de

stimulations périnéales. Ces derniers n'exécutaient que de faibles mouvements alternés des

pattes arrière. Après plusieurs semaines d'entraînement régulier, les animaux contactaient le

tapis plus fréquemment avec le côté plantaire de leurs pattes et la fréquence de mouvements

locomoteurs avec support de poids était augmentée. De plus, les stimulations périnéales

n'étaient plus nécessaires afin de déclencher la locomotion, seules les stimulations

provenant du tapis roulant suffisaient (Rossignol et al., 2000).

44

Une autre étude, portant sur des comparaisons de certaines caractéristiques de la

locomotion a rapporté que les chats spinalisés et entraînés présentaient une cinématique et

des patrons d'activation musculaire des pattes arrières similaires à ce qu'ils montraient

avant la spinalisation (Bélanger et al., 1996). Toutefois des différences étaient observées.

Les chats spinalisés présentaient une diminution de la longueur des pas. Les articulations de

la hanche, du genou et de la cheville étaient plus fléchies alors qu'une augmentation de

l'amplitude des EMGs des muscles fléchisseurs était observée. De plus, au début de la

phase de balancement, les auteurs ont noté un retard dans l'activation du semi-tendineux

(fléchisseur du genou) et une avance dans l'activation du tibialis antérieur (fléchisseur de la

cheville).

Il est également possible de rétablir les fonctions locomotrices de rongeurs ayant

des lésions incomplètes de la moelle épinière. Par exemple, les souris s'entraînant d'elles-

mêmes sur une roue d'entraînement placée dans leur cage, montraient un rétablissement des

fonctions locomotrices (Engesser-Cesar et al., 2007). De façon similaire, des rats qu'on

obligeait à l'entraînement sur roue après une compression de la moelle épinière au niveau

cervical, montraient des capacités locomotrices accrues par rapport aux animaux non-

entraînés (Sandrow-Feinberg et al., 2009). L'entraînement locomoteur avec support de

poids sur tapis roulant a également montré des effets bénéfiques pour le rétablissement de

la locomotion (Multon et al., 2003).

Chez la souris spinale, une récupération locomotrice spontanée, mais limitée, a été

répertoriée. Des mouvements de type non-locomoteur (i.e. crampes, mouvements

rythmiques rapides et mouvements unilatéraux) ainsi que des mouvements de type

locomoteur (flexion-extension de faible amplitude des pattes arrière en alternance et sans

support de poids) ont été observés. Toutefois les souris spinales n'exécutaient pas de

mouvements avec support de poids et placement plantaire (Guertin, 2005; Lapointe et al.,

2006; Ung et al., 2007). Chez les rongeurs complètement spinalisés, l'entraînement seul ne

parvient pas à augmenter davantage les capacités locomotrices à des niveaux équivalents à

ceux du chat (de Leon et al., 2002b; chapitre 8). Ceci serait probablement dû à une plus

45

grande difficulté d'évoquer de façon répétitive et constante des mouvements locomoteurs.

Afin de manier plus adéquatement les pattes arrière des rats et souris, des chercheurs ont

développé un appareil robotisé, le « rat-stepper » (Timoszyk et al., 2002, de Leon et al.,

2002a; 2002b; Nessler et al. 2005). Cet appareil permet autant de manipuler les pattes des

rongeurs (principalement la cheville) dans une trajectoire donnée que d'enregistrer leurs

mouvements. De plus, l'appareil est doté d'un système pouvant déterminer et ajuster le

support de poids, donc la charge imposée sur les pattes arrière lors de l'entraînement.

Récemment, en utilisant cette approche robotisé, un nouveau paradigme d'entraînement a

été proposé: l'entraînement « assisté au besoin » (assist-as-needed), qui permet d'induire un

mouvement des pattes paralysées, non pas d'une manière rigide et répétitive, mais à

l'intérieur d'une limite prédéterminée (Edgerton et Roy, 2009b; Fong et al., 2009). Ainsi, en

permettant une certaine variabilité dans le mouvement, la récupération locomotrice

s'améliorerait davantage (Cai et al., 2006). Toutefois, pour les animaux complètement

spinalisés, même l'entraînement assisté par le robot ne parvient pas à améliorer la

locomotion de façon significative. Par conséquent, une approche pharmacologique est

généralement requise (Fong et al., 2005; Edgerton et Roy, 2009a).

Il est important de mentionner que d'autres méthodes d'entraînement existent, par

exemple, l'entraînement sur vélo robotisé (Nothias et al., 2005; Yoshihara, 2006) et

l'entrainement à la nage (Smith et al., 2006; Magnuson et al., 2009).

3.2.3 Entraînement locomoteur chez l'humain

En se basant sur les études effectuées chez l'animal, des protocoles d'entraînement

locomoteur ont été développés pour l'humain (Barbeau et al., 1987; Wernig et Muller,

1992; Wernig et al., 1995; Norman et al., 1995). La majorité de ces protocoles consistent à

entraîner les patients paraplégiques sur un tapis roulant, en supportant un certain

pourcentage de leur poids à l'aide d'un harnais. Les patients sont ainsi entraînés par des

thérapeutes qui induisent manuellement un mouvement de locomotion au niveau des

jambes paralysées. Cette méthode, à laquelle on réfère généralement à BWSTT ou « body

46

weight supported treadmill training » a permis d'importantes améliorations des fonctions

locomotrices (Wernig et Muller 1992; Wernig et al., 1995; Dietz et al, 1998; Behrman et

Harkema, 2000; Wirz et al., 2001; Behrman et al., 2005).

Avec ce type d'entraînement, des patients présentant un diagnostique de paralysie

complète peuvent supporter davantage de poids sur leurs jambes après plusieurs semaines

d'entraînement intensif (Dietz et al., 1995). D'autres chercheurs ont noté une augmentation

de l'activité électromyographique (EMG) des extenseurs suite à l'entraînement (Wirz et al.,

2001) ainsi qu'un patron d'activation musculaire rythmique dans leurs jambes (Maegele et

al., 2002). Cependant, ces patients n'amélioraient pas leur capacité locomotrice jusqu'à

marcher sans support de poids et par conséquent, ne pouvaient non plus se déplacer sur le

sol de façon autonome, sans assistance (Van de Crommert et al., 1998). Un récent article

portant sur l'entraînement par BWSTT d'un blessé médullaire diagnostiqué ASIA-B (Tab.

3.1) a montré des améliorations importantes au niveau de la cinématique des jambes, des

patrons d'activations et de l'amplitude des EMGs. Par contre, aucune amélioration au

niveau du diagnostique ASIA n'a été décelée (Forrest et al., 2008).

Tableau 3.1 Classification des types de blessés médullaires selon l'International Standards for Neurological Classification of Spinal Cord Injury (ASIA).

B

D

E

Complète : Aucune fonction sensorielle ni motrice n'est conservée dans les segments S4-S5

Incomplète : Aucune fonction motrice, mais des fonctions sensorielles sont préservées sous

le niveau de la blessure.

Incomplète : Fonctions motrices préservées sous le niveau de la blessure et la plupart des

muscles importants sous le niveau de la blessure on une cote de moins que 3*

Incomplète : Fonctions motrices préservées sous le niveau de la blessure et la plupart des

muscles importants sous le niveau de la blessure ont une cote égale ou supérieur à 3*

Normal : Fonctions motrices et sensorielles normales

♦concerne le mouvement actif des muscles. 0 : paralysie total, 1 : contraction palpable, 2 : mouvements actifs en l'absence de gravité, 3 : mouvements actifs contre la gravité, 4 : mouvements actifs contre une faible résistance, 5 : mouvements actifs contre une forte résistance

47

Une vaste étude, regroupant 6 centres, a exploré les différences de recouvrement

locomoteur entre le BWSTT et une approche d'entraînement dite plus traditionnelle :

entraînement sur le sol à l'aide de barres parallèles qui consiste à se tenir debout et/ou

marcher, en proposant l'hypothèse que le BWSTT permettrait un meilleur rétablissement

locomoteur que l'approche traditionnelle (Dobkin et al., 2006). Le recouvrement

locomoteur était évalué sur la base de 2 paramètres principaux, soit la grille d'évaluation

Functional Indépendance Measure for Locomotion ou FIM-L pour les patients

diagnostiqués ASIA-B et C et la vitesse de déplacement au sol pour les patients classifies

ASIA-C et D. D'autres paramètres secondaires ont également été utilisés pour l'évaluation

du rétablissement moteur et l'état de santé général: Distance parcourue en 6 minutes

(évaluant l'endurance et la condition physique), Berg balance scale (évaluant le contrôle du

tronc et des jambes), Walking Index for Spinal Cord Injury (WISCI, grille d'évaluation

similaire au FIM-L, en plus détaillée), Lower extremity motor score (évaluant la force

musculaire), Ashworth scale (évaluant l'hypertonicité) et la SF-54 (évaluant la qualité de

vie perçue par le patient). Contrairement à leur hypothèse de départ, aucune différence n'a

été observée entre les 2 types d'entraînement. Néanmoins, les 2 formes d'entraînement ont

permis une progression remarquable au niveau ambulatoire. Par conséquent, peu importe la

forme que prend l'entraînement, il faut à tout le moins s'entraîner.

À cet égard, chez l'humain, d'autres protocoles d'entraînement existent. Entre autre,

des chercheurs ont développé un vélo dont le mouvement cyclique des jambes est induit par

le mouvement des bras (Huang et Ferris, 2004). Ceci permet aux blessés médullaires de

s'entraîner avec une plus grande autonomie. Par ailleurs, des protocoles d'entraînement sur

vélo combinés à des stimulations électriques fonctionnelles (FES) des muscles des jambes

ont également été évalués (Johnston et al., 2009; Griffin et al., 2009).

3.3 Stimulation pharmacologique de la moelle épinière

Il est bien connu que des patrons de type locomoteur peuvent être induits et modulés

par stimulation pharmacologique. Dans cette section nous aborderons les récepteurs

48

monoaminergiques et substances pharmacologiques impliqués et utilisés dans l'activation et

la modulation du CPG locomoteur pour les études de cette thèse.

3.3.1 L-Dopa, récepteurs dopaminergiques et adrénergiques

L-Dopa

La L-Dopa, précurseur de la dopamine et de la noradrenaline, est surtout utilisée

pour traiter les patients atteints de la maladie de Parkinson. Les effets de cette dernière

peuvent également être notés dans la moelle épinière. Dans les années 60, une étude chez le

lapin a montré que la L-Dopa induisait une activité rythmique et alternée, enregistrée au

niveau des nerfs reliés au tibialis antérieur et au gastrocnemius medial (Viala et Buser,

1969). Vers la fin des années 60, le laboratoire du Dr Lundberg a produit une série d'études

concernant les effets de la L-Dopa sur la transmission synaptique dans la moelle épinière.

Plus particulièrement, les études effectuées par Jankowska ont permis de montrer que les

décharges réflexes provenant des FRAs survenaient après une longue latence et étaient de

longue durée (Jankowska et al., 1967a; 1967b). De plus ces FRAs étaient observés en

alternance, dans les motoneurones reliés aux muscles fléchisseurs ipsilatéraux et aux

muscles extenseurs controlatéraux. Ainsi, il a été suggéré que les réseaux neuronaux

responsables de ces décharges pouvaient être reliés à ceux de la locomotion.

Grillner et Zangger ont montré qu'il était possible d'évoquer une locomotion fictive,

ou un patron de type locomoteur au niveau des pattes arrière en administrant de la L-Dopa

avec ou sans stimulation de certains nerfs périphériques, de racines dorsales ou de la

colonne dorsale de la moelle épinière chez des chats spinalisés, déafférentés et/ou curarisés

(Fig. 3.1). Les ENGs ou EMGs montraient une activité rythmique et alternée entre les

fléchisseurs et les extenseurs. Lorsque des mouvements étaient induits, ils étaient décrits

comme étant de type locomoteur et impliquaient toutes les articulations des pattes, non pas

juste de simples flexions-extensions des chevilles et des genoux (Grillner et Zangger,

1979). D'autres études chez le chat spinal ont confirmé que ce précurseur

49

catecholaminergique pouvait effectivement induire une locomotion sur tapis roulant

(Barbeau et Rossignol, 1991; Barbeau et al., 1993).

bit. DR D. Col.

rSt

Fig 3.1 Electroneurogramme montrant une activité alternée entre fléchisseurs et extenseurs des pattes gauche

et droite, suite à l'administration de L-Dopa et de stimulation des racines dorsales (schéma de gauche) et de la

colonne dorsale (schéma de droite) (Tirée de Grillner et Zangger, 1979)

Chez les rongeurs, la L-Dopa administrée in vivo montre également des effets sur la

genèse et la modulation de mouvements rythmiques et alternés. Cependant, administrée

seule, la L-Dopa n'était pas en mesure d'induire un patron locomoteur soutenu, alors que

peu de mouvements sont observés en « air-stepping » (i.e. animal maintenu dans les airs

par un harnais où aucune stimulation sensorielle n'est appliquée). Combinée avec d'autres

agonistes sérotoninergiques, tel que la quipazine ou lorsque les injections sont combinées à

une stimulation sensorielle provenant d'un tapis roulant, les effets pro-locomoteurs de la L-

Dopa étaient amplifiés (McEwen et al., 1997; Guertin, 2004; Landry et Guertin, 2004).

Chez les blessés médullaires, l'administration de L-Dopa a montré son efficacité

pour diminuer la spasticité ou les mouvements épisodiques involontaires (syndrome de la

jambe sans repos) (Eriksson et al., 1996; Lee et al., 1996; De Mello et al., 2004). Chez des

patients qui ont une lésion incomplète de la moelle épinière, son utilisation combinée à un

entraînement locomoteur n'a toutefois pas permis d'améliorer les fonctions locomotrices

50

outre ce qui a été observé avec un placebo (Marie et al., 2008). Cependant, la combinaison

de cet agent pharmacologique avec d'autres agonistes sérotoninergiques pourrait s'avérer

avantageuse pour l'activation du réseau locomoteur, surtout auprès de patients ayant une

lésion complète. Des résultats préliminaires ont permis de constater que la combinaison de

L-Dopa et buspirone chez un patient monoplégique était sécuritaire, sans effets secondaires

significatifs (Guertin et Brochu, 2009).

Il est possible d'aller cibler de manière plus spécifique les différents sous-types de

récepteurs dopaminergiques et adrénergiques. De ce fait, plusieurs études ont montré

l'efficacité de certains sous-types à induire et moduler des mouvements de types

locomoteurs.

Récepteurs dopaminergiques

Les récepteurs dopaminergiques sont divisés en 2 classes, subdivisés en 5 sous-

types: Duike (Di et D5) et D2-Hke (D2, D3 et D4). Tous ces sous-types se retrouvent dans la

moelle épinière et leur distribution serait diffuse à travers la matière grise de toutes les

laminae, avec une prédominance des récepteurs D2 (Zhu et al., 2007).

En disséquant pharmacologiquement et génétiquement la contribution des différents

sous-types de récepteurs dopaminergiques pour activer le CPG de la locomotion, nous

avons montré que les récepteurs Di auraient un rôle plus important dans l'induction de

mouvements locomoteurs (LMs, pour locomotor movements) (Lapointe et al., 2009). Dans

les expériences effectuées, ni le quinpirole, le 7-OH-DPAT ou le PD168077 (agonistes D2/3,

D3 et D4, respectivement) n'ont pu induire de mouvement chez les souris spinales.

L'administration de SKF-81297, agoniste D]/5, induisait un nombre similaire LMs, autant

chez les souris sauvages que chez les souris KO pour le récepteur D5. À l'opposé, lorsque

ces animaux étaient prétraités avec un antagoniste spécifique pour les récepteurs D1/5, les

LMs étaient complètement éliminés. L'activation des récepteurs D1/5 ne suffit toutefois pas

à induire une locomotion proprement dite. Ce n'est qu'une fois combinée avec l'activation

51

des récepteurs 5-HTIA/7 que des LMs avec placements plantaires et support de poids sont

observés (Lapointe et Guertin, 2008).

Chez le chat, l'apomorphine, agoniste pour tous les sous-type de récepteurs

dopaminergiques, n'a pas permis d'activer le CPG locomoteur (Barbeau et Rossignol,

1991), tout comme chez la souris (Lapointe et al., 2009). Aucun test concernant l'utilisation

d'agonistes spécifiques aux récepteurs dopaminergiques n'a été répertorié chez les primates

ou les humains blessés à la moelle épinière.

Récepteurs adrénergiques

On retrouve 2 classes de récepteurs adrénergiques (a et 6), qui se subdivisent en 9

sous-types (aiA - CIIB - (XID, (X2A - «2B - «2c» 8i - 62 - 83)- Les récepteurs a sont ceux

principalement associés à l'induction de la locomotion, plus spécifiquement les récepteurs

v.2- Dans la moelle épinière, les récepteurs a sont distribués dans tous les segments et

régions (dorsale, intermédiaire et ventrale) de la moelle épinière, avec une prédominance

dans la corne dorsale pour les récepteurs 02 (Roudet et al., 1993; 1994). Nos connaissances

concernant l'implication des récepteurs adrénergiques pour la locomotion nous proviennent

principalement des expériences effectuées chez le chat spinal. Les données recueillies sont

pour la plupart associées à l'utilisation de la clonidine, agoniste des récepteurs (12-

adrénergiques. L'administration de cette molécule induit la locomotion lorsque l'animal est

stimulé au niveau de la queue ou du périnée et qu'il est placé sur un tapis roulant (Barbeau

et al., 1987; Barbeau et Rossignol, 1991; Giroux et al., 2001). À l'opposé, la yohimbine

(antagoniste des récepteurs ci2-adrénergiques) permettait de diminuer la locomotion chez le

chat intact, mais pas chez le spinal qui a récupéré ses fonctions locomotrices. Ceci suggère

que chez le chat spinal, les récepteurs (X2-adrénergiques ne sont pas essentiels pour une

locomotion spinale. Les agonistes ai-adrénergiques peuvent aussi initier la locomotion,

mais le nombre d'animaux qui répondent au traitement est plus faible. La clonidine, la

tizanidine et l'oxymétazoline, tous des agonistes des récepteurs (X2, initiaient une

locomotion sur tapis roulant chez tous les animaux testé 7 jours après la spinalisation, alors

52

qu'un seul animal spinal répondait au traitement à la méthoxamine, agoniste ai (Chau et al.,

1998).

Chez le rongeur la clonidine n'a aucun effet pro-locomoteur, (Lapointe et al., 2008a;

2008b). En fait, l'injection de cette molécule semblait plutôt supprimer les mouvements.

Alors que des pincements de la queue induisaient des mouvements des pattes paralysées,

aucun mouvement (LM ou NLM) n'était observé lorsque la clonidine était administrée. Par

contre, à long terme, son administration répétée, sans aucune stimulation, améliorait

quelque peu le recouvrement locomoteur spontané.

Les expériences sur l'administration d'agonistes adrénergiques chez les primates et

les humains sont plutôt rares. Une étude chez le marmoset spinalisé et déafférenté, a montré

que la clonidine induisait de l'activité rythmique et alternée dans les nerfs des fléchisseurs

ipsi- et controlatéraux, ainsi qu'entre les fléchisseurs et extenseurs ipsilatéraux (Fedirchuk

et al., 1998). Chez l'humain, l'administration de clonidine n'a pas ou peu d'effet pro­

locomoteur (Stewart et al., 1991; Dietz et al., 1995; Rémy-Néris et al., 1999). Toutefois une

amélioration dans l'exécution des mouvements était perçue, possiblement due aux

propriétés anti-spastiques de cette molécule.

3.3.2 Sérotonine et récepteurs sérotoninergiques

Les récepteurs 5-HT constituent une vaste famille de récepteurs monoaminergiques.

Dans le SNC, on retrouve 7 familles de récepteurs 5-HT (5-HTi à 5-HT7) qui se subdivisent

en 14 sous-types (Barnes et Sharp, 1999). Ceux-ci sont impliqués dans une multitude de

processus physiologiques. Les dérèglements du système 5-HT dans le cerveau sont

responsables de nombreux troubles psychosomatiques (Filip et Bader, 2009). Au niveau de

la moelle épinière, la 5-HT et ses récepteurs sont impliqués dans plusieurs fonctions

rythmiques, telles que la miction et l'élimination des selles (Birder et al., 2010; Tuladhar et

al., 2000), la déglutition (Hachim et Bieger, 1987), la respiration (Kinkead et al., 2002),

53

l'érection et l'éjaculation (Giuliano et Rampin, 2004; Giuliano et Clément, 2005) et, bien

entendu, la locomotion (Schmidt et Jordan, 2000; Guertin et Steuer, 2009).

La majorité des neurones sérotoninergiques originent du noyau raphé et de la

formation réticulée (Dahlstrôm et Fuxe, 1964). Leurs projections se terminent

principalement dans la corne dorsale où les afférences sensorielles font synapses, mais

aussi dans les régions intermédiaire et ventrale, où se retrouvent la circuiterie du CPG et les

motoneurones. Du fait que plusieurs sous-types de récepteurs 5-HT se retrouvent dans la

moelle épinière, différents agonistes ont été étudiés afin d'initier et de moduler des patrons

locomoteurs. Une diversité de préparation de moelle épinière isolée et de modèle animal

ont permis de mieux comprendre la locomotion. Les récentes avancées en stimulation

pharmacologique de la moelle épinière montrent que seulement quelques sous-types de

récepteurs 5-HT seraient impliqués dans le contrôle de la locomotion.

Le récepteur 5-HTIA

Dans la famille des récepteurs 5-HTi, le récepteur 5-HTiA est pour l'instant le seul

reconnu comme étant impliqué dans l'activation et la modulation du CPG. Ces derniers sont

principalement localisés dans la corne dorsale (Thor et al., 1992.) Des expériences

électrophysiologiques ont préalablement déterminé que l'activation des récepteurs 5-HTiA

induisait une hyperpolarisation neuronale par l'ouverture des canaux potassiques couplés à

une protéine-G (pour revue voir Nicoll et al., 1990). Dans la moelle épinière, ceci aurait

pour conséquence une dépression de la réponse sensitive. Cependant, d'autres études sur

des préparations de moelle épinière de tortues et de grenouilles montrent que l'activation de

ce récepteur pouvait induire une dépolarisation neuronale, ce qui aurait un effet excitateur

sur le motoneurone (Perrier et al., 2003; Holohean et al., 1992). Ainsi après l'ajout de 5-HT

ou de 8-OH-DPAT (agoniste 5-HTIA, mais également 5-HT7, voir section ultérieure), les

auteurs ont noté des augmentations dans l'amplitude et la duré des potentiels évoqués au

niveau des racines ventrales (Holohean et al., 1992).

54

Chez le chat partiellement lésé et entraîné, l'administration de 8-OH-DPAT a

montré un effet néfaste sur la locomotion (Brustein et Rossignol, 1999). À l'opposé, chez le

rat spinal, l'administration chronique de ce même agoniste 5-HT améliorait la récupération

locomotrice (Antri et al., 2003). Similairement, une étude provenant de notre laboratoire a

montré l'importance des récepteurs 5-HTiA dans l'induction de la locomotion chez la souris

spinale (Landry et al., 2006b). En effet, une réduction du nombre de LMs était observée

suite à l'administration de 8-OH-DPAT chez des souris paraplégiques, prétraitées avec des

antagonistes spécifiques pour les récepteurs 5-HTiA. De plus, chez des KO pour le

récepteur 5-HT7 et prétraitées avec ces mêmes antagonistes, les LMs étaient bloqués en

quasi-totalité.

D'autre part, des études portant sur les difficultés respiratoires après BME (fonction

associée également à un CPG) ont montré que la buspirone, agoniste plus spécifique pour

les 5-HTRIA, était bénéfique pour le rétablissement de la capacité respiratoire (Teng et al.,

2003; Choi et al., 2005). Cette molécule, combiné avec la L-Dopa, n'a que rarement été

utilisée afin d'étudier le rétablissement des fonctions locomotrices après une BME (Guertin

et Brochu, Guertin et al., accepté dans Neurorehabil Neural Repair; chapitre 9).

Les récepteur 5-HT2

On retrouve 3 sous-types de récepteurs 5-HT2, les 5-HT 2A, 2B et 20 Lors de leur

activation chacun de ces récepteurs se couple à une protéine G, ce qui active une

phospholipase C et entraîne des augmentations des concentrations intracellulaires d'inositol

phosphate et de Ca2+ (Baxter et al., 1995; Grotewiel et Sanders-Bush, 1999). Ces 3 sous-

types de récepteurs se retrouvent dans la moelle épinière, mais dépendamment des études et

des espèces animales utilisées, quelques différences quant à leur présence et leur

localisation sont observées (Helton et al., 1994; Pompeiano et al., 1994; Fonseca et al.,

2001; Maeshima et al., 1998; Cornea-Hébert et al., 1999; Doly et al., 2004; Lauder et al.,

2000; Molineaux et al., 1989; Mengod et al., 1990). En général, les récepteurs 5-HT2A se

retrouvent surtout au niveau des cornes ventrales, proche des régions ou l'on retrouve les

55

motoneurones. Ds sont également observés, en plus faible densité, dans la région

intermédiaire et dans les cornes dorsales. Chez les mammifères, les récepteurs 5-HT2B, sont

surtout retrouvés dans le cerveau et la moelle épinière en développement, près des cellules

neuroépithéliales. La localisation des récepteurs 5-HT2C est plutôt diffuse au niveau de la

matière grise de la moelle épinière. Les 3 sous-types de récepteurs 5-HT2 ont montré des

effets soit sur la dépolarisation des motoneurones, l'induction et la modulation d'activité

rythmique ENG au niveau des racines ventrales ou sur l'induction de mouvements et la

récupération des fonctions locomotrices.

Des expériences in vitro sur des moelles épinières de souris néonatales ont montré

que l'activité induite par la 5-HT était inhibée suite à l'addition de kétansérine (antagoniste

non-spécifique pour les récepteurs 5-HT2, mais ayant plus d'affinité pour le récepteur 5-

HT2A) (Madriaga et al., 2004). Dans cette même étude, l'activité rythmique était également

induite par de l'alpha-métyl-5-HT (agoniste non spécifique pour les récepteurs 5-HT2). De

façon plus spécifique, Liu et Jordan ont montré que l'activité rythmique enregistrée au

niveau des racines ventrales, induite par stimulation électrique ou pharmacologique des

neurones 5-HT de la région parapyramidale, était abolie avec l'ajout de kétansérine dans le

bain. De plus ils ont montré que des antagonistes spécifiques pour les récepteurs 5-HT7

pouvaient interrompre la locomotion fictive (voir section 5-HT7 plus bas). Ces résultats ont

permis aux auteurs de conclure que les récepteurs 5-HT2A et 5-HT7 étaient responsables de

l'activation et la modulation de la locomotion fictive (Liu et Jordan, 2005).

Plusieurs études in vivo viennent appuyer les résultats in vitro, à l'égard de la

contribution des récepteurs 5-HT2 lors de la locomotion. À titre d'exemple, chez le chat

spinal adulte, la quipazine et le DOI (agonistes à large spectre pour les récepteurs 5-HT2 et

5-HT2A/2C* respectivement) augmentent l'excitabilité des motoneurones alpha et l'activité

motrice au niveau des pattes paralysées (Barbeau et Rossignol, 1990; Jackson et White,

1990). D'autres études montrent que 24h après une transsection de la moelle épinière de rats

néonataux, l'administration de quipazine induisait des mouvements de type locomoteurs en

« air-stepping ». Une augmentation de la fréquence de mouvements est observée lorsque la

quipazine est combiné avec la L-DOPA (McEwen et al., 1997). Les études de notre

56

laboratoire abondent en ce sens. Chez des souris adultes spinalisées, l'administration de

quipazine seule, ou en combinaison avec de la L-DOPA est capable d'induire des

mouvements de type locomoteur (LM) et non locomoteur (NLM), soit en « air-stepping »

ou sur tapis roulant. À l'opposé, le mCPP et le TFMPP, agonistes des récepteurs 5-HT2B/2C

et 5-HTIB n'induisaient aucun LM, seuls des NLMs pouvaient être observés (Guertin,

2004a; Landry et Guertin, 2004). Ces dernières expériences suggèrent que les récepteurs 5-

HT2A auraient un rôle prépondérant durant la locomotion.

En administration chronique chez des rats spinaux, la quipazine améliorait

également les performances locomotrices à long terme, sans entraînement, mais avec des

stimulations sensorielles provenant du tapis roulant et du pincement de la queue (Antri et

al., 2002). Les auteurs ont noté des augmentations du score locomoteur et de l'amplitude

des EMG, une meilleure alternance entre l'activation des fléchisseurs et extenseurs, une

amélioration de la coordination entre les membres postérieurs, ainsi que des améliorations

au niveau de la cinématique.

Contrairement à ce qui a été suggéré par notre laboratoire (Landry et Guertin, 2004),

un autre groupe de recherche proposent plutôt que les récepteurs 5-HT2C seraient davantage

impliqués pour le recouvrement locomoteur après lésion de la moelle épinière (Kim et al.,

2001; Kao et al., 2006). En administrant quotidiennement du mCPP, les auteurs ont noté

que les rats spinaux, avec ou sans transplants de tissus embryonnaires dans la moelle

épinière et avec entraînement sur tapis roulant, généraient plus de LMs avec support de

poids. De plus, leur tests effectués avec la quipazine, le DOI et le 8-OH-DPAT n'ont pu

permettre le rétablissement des fonctions locomotrices à des niveaux équivalents à ceux

observés avec le mCPP (Kim et al., 1999; Kao et al., 2006). Les différences seraient

attribuables, en partie, au modèle animal. Alors que nous effectuons les lésions de la moelle

épinière à l'âge adulte, ce groupe réalisait leur transsection chez l'animal néonatal et testait

les drogues lorsque l'animal atteignait l'âge adulte. La plasticité des 2 systèmes (i.e.

néonatal vs adulte) pourrait expliquer les différences dans la réponse aux traitements

pharmacologiques.

57

Sur des tranches de moelle épinière de grenouille, Holohean et Hackman ont montré

que les récepteurs 5-HT2B contribuaient à faciliter la dépolarisation neuronale (Holohean et

Hackman, 2004). D n'a toutefois jamais été démontré que ce récepteur contribuait de façon

significative à locomotion chez l'animal adulte. En fait, l'activation de ce récepteur pourrait

plutôt être nuisible pour la locomotion (Landry et Guertin, 2004).

Les récepteurs 5-HT-/

Bien qu'ils ne soient pas spécifiquement étudiés pour cette thèse, il faut noter

l'importante contribution des récepteurs 5-HT7 pour la locomotion. Leur présence est

surtout répertoriée dans les laminae I et II de la corne dorsale (Meuser et al., 2002; Doly et

al., 2005) et dans la zone intermédiaire entre les segments LI et L5 de la moelle épinière

(Hochman et al., 2001).

La contribution des récepteurs 5-HT7 durant la locomotion a été suggérée suite à des

expériences effectuées sur des moelles épinières de rats et de souris in vitro. La locomotion

fictive induite électriquement ou pharmacologiquement était abolie suite à l'ajout de SB

269970, un antagoniste spécifique pour le récepteur 5-HT7 (Madriaga et al., 2004; Liu et

Jordan, 2005; Pearlstein et al., 2005). Des études in vivo sont venues corroborer ce qui avait

été préalablement suggéré sur des préparations de moelle épinière. Tout d'abord, Antri et

collaborateurs ont montré que le 8-OH-DPAT, administré de façon chronique, facilitait le

rétablissement des fonctions locomotrices chez le rat spinal (Antri et al., 2005). Des études

provenant de notre laboratoire ont également montré que le 8-OH-DPAT induisait des LMs

7 jours seulement après la spinalisation chez la souris adulte. En étudiant plus

spécifiquement la contribution des récepteurs 5-HTiA et 5-HT7 par l'utilisation de souris

sauvages et KO pour le récepteur 5-HT7, il a été montré que ces 2 sous-types de récepteurs

5-HT étaient importants pour le recouvrement locomoteur de souris paraplégiques adultes

(Landry et al., 2006b). Une autre étude in vivo a montré que le récepteur 5-HT7 était

nécessaire pour produire une locomotion coordonnée chez la souris adulte sauvage non-

transectée (Liu et al., 2009). Toutefois ce récepteur ne semblait pas indispensable à la

58

locomotion chez les souris KO, car celles-ci ne montraient pas de déficit locomoteur

(Hedlund et al., 2003; Liu et al., 2009). Ceci suggère donc l'existence de mécanismes

compensateurs, provenant des autres systèmes impliqués dans l'activation et la modulation

de la locomotion (Landry et al., 2006b).

Corticospinal Rubrospinal

NE, 5-HT, DA.GLU

Fig 3.2 Schéma résumé du contrôle locomoteur. En situation où la moelle épinière est intacte, les neurones du

CPG reçoivent de l'information des structures supraspinales et périphériques. Lorsqu'il y a blessure à la

moelle épinière la contribution des centres supraspinaux est limitée (lignes pointillée). L'activation du CPG,

et par le fait même l'induction et la modulation de la locomotion, se fait via plusieurs neurotransmetteurs,

entres autres, la norepinephrine/noradrenaline (NE), la sérotonine (5-HT), la dopamine (DA) et le glutamate

(GLU) (Modifiée de Rossignol et al., 2009).

59

CHAPITRE IV - PROBLÉMATIQUE

1.1 Problématique

Environ la moitié des BME se solde par une perte quasi-totale des fonctions

sensorielles et motrices. De ce nombre seulement 5% des blessés médullaires récupéreront

certaines des fonctions perdues (Dobkin et Havton, 2004). En se basant sur l'évaluation

motrice 1 semaine suivant une BME, 80-90% des blessés médullaires classifies ASIA-A

demeureront ASIA-A et 50% des patients classifies ASIA-B n'amélioreront pas

significativement leur capacité ambulatoire (Kim et al., 2004; Maynard et al., 1979; Waters

et al., 1994; Mehrholz et al., 2008). Les patients souffrent également de problèmes de santé

secondaires. Plus spécifiquement, les patients développent des problèmes immunitaires,

métaboliques, hormonaux, cardiovasculaires, d'atrophie musculaire, d'ostéoporose, etc.

Malheureusement, aucun traitement n'existe pour réparer la moelle épinière lésée. Afin de

pallier à certaines des conséquences de ce traumatisme, nous pensons que l'entraînement

locomoteur pourrait s'avérer bénéfique tant pour améliorer les performances locomotrices

que pour contrer les problèmes secondaires découlant de l'immobilisation chronique.

Toutefois, l'entraînement locomoteur seul ne peut améliorer les fonctions

locomotrices pour des lésions complètes de la moelle épinière. fl faut alors combiner

plusieurs approches thérapeutiques. À cet égard, les travaux du Dr Barbeau ont été

importants dans cette approche. Tant chez l'animal que chez l'humain, ses travaux ont

montré que la locomotion était facilitée en combinant des agonistes adrénergiques à

l'entraînement locomoteur. Chez le chat spinal, la combinaison de clonidine et

entraînement locomoteur accélérait le rétablissement des fonctions locomotrices (Barbeau

et al., 1993). Chez l'humain, une stratégie similaire a été adoptée. En administrant de la

clonidine et de la cyproheptadine (antagoniste sérotoninergique) à l'entraînement sur tapis

roulant avec support de poids, les patients paraplégiques ont pu retrouver une certaine

motricité fonctionnelle. Plus récemment, les travaux du laboratoire du Dr Edgerton ont

montré d'importantes améliorations locomotrices lorsque des rats spinaux étaient entraînés

sur tapis roulant avec support de poids et que cet entraînement était assisté par un appareil

60

robotisé, des stimulations pharmacologiques (quipazine et/ou 8-OH-DPAT) et des

stimulations électriques épidurales de la moelle épinière (Fong et al., 2005; Courtine et al.,

2009). D'autres ont combiné l'entraînement locomoteur avec l'ajout de facteur de

croissance ou des greffes de cellules embryonnaires (Nothias et al., 2005; Mitsui et al.,

2005).

Puisqu'une approche multidisciplinaire semble davantage bénéfique, il importe de

développer une combinaison thérapeutique adéquate, qui favoriserait autant le

rétablissement des fonctions locomotrices et certains paramètres de santé. En premier il faut

valider notre méthode d'évaluation du rétablissement locomoteur et évaluer le

recouvrement moteur spontané (i.e. sans traitement) chez l'animal spinal adulte. Par la

suite, il faut déterminer l'effet des différentes thérapies de façon individuelle, puis les

combiner afin d'évaluer leurs effets synergiques sur le rétablissement locomoteur. Notre

approche se distingue des autres laboratoires par notre modèle de transsection complète de

la moelle épinière (lésion complète intervertébrale). De plus, lors des sessions

d'entraînement, nous ne supportons pas le poids de nos animaux, nous n'utilisons pas

d'assistance robotique ni de stimulation électrique et notre pharmacologie est différente

(Guertin et al., 2010; Guertin et al., 2010 Epub; chapitre 9). Dans le but de mieux

comprendre les effets de nos traitements suite aux blessures à la moelle épinière, nous

avons décidé d'utiliser un modèle de souris ayant une lésion complète de la moelle

épinière. De cette façon l'influence des centres supra-spinaux ne pourra biaiser les résultats

observés. Les études présentées pour cette thèse suivront l'ordre suivant.

1.2 Hypothèse de travail

Etant donnée le recouvrement locomoteur spontané très limité des souris ayant une

lésion complète de la moelle épinière, la majorité des grilles d'évaluation qualitative de

rétablissement locomoteur ne semble pas suffisamment adaptée pour ce modèle animal. La

majorité d'entre elles ont été développées pour la souris ou le rat partiellement lésé. De ce

fait, elles évaluent des critères qui ne sont pas valables chez la souris complètement spinale

61

(par exemple, la coordination patte avant-arrière et mouvement de la queue, habileté à se

déplacer sur une barre horizontale etc.). En contre partie, la méthode d'évaluation ACOS a

été développée spécifiquement pour évaluer le rétablissement locomoteur très limité chez

l'animal ayant une lésion complète de la moelle épinière. Elle détermine le rétablissement

locomoteur en fonction de la fréquence de mouvements générés, ce que ne fait aucune autre

méthode d'évaluation qualitative. La première partie de l'étude consistait donc à

déterminer, à l'aide de différentes grilles d'évaluation, le niveau de rétablissement moteur

spontané chez des souris ayant une lésion complète de la moelle épinière.

Pour cette étude nous voulions également déterminer s'il existait une différence entre

des souris spinales mâles et femelles dans le rétablissement locomoteur spontané. La

raison derrière cette question est que certaines études montrent des effets neuroprotecteurs

pour les hormones mâles ou femelles. Par exemple, la testosterone permet un

rétablissement moteur fonctionnel plus rapide suite à une lésion du nerf facial ou sciatique

(Tanzer et Jones, 1997; Vita et al., 1983) alors que l'oestrogène limite l'étendue d'une

blessure à la moelle épinière et prévient la dégénération axonale (Sribnick et al., 2005;

2006). Une étude s'est consacrée plus spécifiquement aux différences de rétablissement

locomoteur entre les mâles et les femelles chez le rat et la souris partiellement lésé (Hauben

et al., 2002). Les auteurs ont montré que les femelles avaient un meilleur score locomoteur

que les mâles suite à des lésions partielles de moelle épinière. Toutefois, aucune étude ne

s'est intéressée à cette question chez l'animal ayant une lésion complète de la moelle

épinière. Nos hypothèses pour l'article 1 étaient :

1. L'ACOS serait la méthode d'évaluation la plus discriminative pour évaluer le

rétablissement locomoteur spontané limité chez des souris spinales. Il faudrait

cependant la combiner à d'autres méthodes d'évaluation pour davantage de précision

sur la qualité du mouvement.

2. D n'existerait pas de différence de récupération motrice entre les souris mâles et

femelles étant donné la sévérité de la lésion.

62

En second lieu, le système sérotoninergique joue un rôle prépondérant dans l'activation

et la modulation du CPG locomoteur, mais il apparaît que seuls quelques sous-types de

récepteurs 5-HT seraient impliqués dans ce processus. Suite à une lésion spinale, il a été

suggéré que les récepteurs 5-HT2A seraient davantage liés à l'induction de mouvements de

type locomoteur (Landry et Guertin, 2004). Cependant, d'autres laboratoires ont montré

que l'activation des récepteurs 5-HT2C étaient nécessaire pour l'induction ce type de

mouvement (Kao et al., 2006). La seconde étude avait donc pour objectif d'évaluer la

contribution des sous-types des récepteurs 5-HT2 (5-HT2A-2B-2C) dans l'induction

pharmacologique de mouvements locomoteurs. Afin de mieux comprendre le rôle de ce

récepteur dans le recouvrement locomoteur post-lésionnel, nous avons également étudié les

changements au niveau de sa distribution spatio-temporelle suite à une lésion de moelle

épinière. Implicitement, nous voulions déterminer si l'ajout d'une molécule activant les

récepteurs 5-HT2A s'avérerait efficace dans l'élaboration d'un traitement pharmacologique

visant à restaurer les fonctions locomotrices. Nos hypothèses pour l'article 2 étaient:

1. En réponse à une perte des afférences 5-HT provenant des centres supraspinaux, une

augmentation de l'expression des récepteurs 5-HT2A serait perçue dans les jours et

semaines suivant une transsection complète de la moelle épinière.

2. Les récepteurs 5-HT2A seraient davantage liés à la circuiterie de la locomotion, par

rapport aux autres sous-types de récepteurs 5-HT2.

Par la suite, nous avons poursuivit nos travaux en ciblant plus spécifiquement le

système musculaire au niveau des pattes arrière. Il est bien connu que la masse musculaire

au niveau des jambes (ou pattes arrière) montre une atrophie après une blessure de la

moelle épinière. De plus, cette atrophie est dépendante de la sévérité de la lésion; plus une

lésion est sévère, plus importante sera l'atrophie. La diminution de masse musculaire

combinée à des changements au niveau des propriétés contractiles musculaires ont pour

conséquences de diminuer la force pouvant être générée et la résistance à la fatigue

(Talmadge et al., 2002; Landry et al., 2004). Conséquemment, afin de favoriser le retour

63

des fonctions locomotrices, il faut être en mesure de préserver ou renverser la diminution

de masse musculaire après une blessure de la moelle épinière.

La testosterone et le clenbuterol ont été utilisés étant donné que ces 2 substances ont

montré des effets anaboliques sur la masse musculaire et des effets neuroprotecteurs suite à

des lésions du SNP ou du SNC. L'administration supra-physiologique de testosterone, avec

ou sans entraînement augmente la masse et la force musculaire (Bhasin et al., 1996). De

plus, suite à une lésion du nerf facial ou sciatique, qui induit une paralysie des muscles

faciaux ou des pattes arrière, il a été montré que l'administration de testosterone accélérait

le rétablissement moteur (Kujawa et al., 1989; Brown et al., 1999). Pour sa part, la

supplementation de clenbuterol montre des effets similaires. Par exemple, l'administration

de clenbuterol induisait une augmentation de la masse musculaire et de la force produite par

le soleus et l'extenseur digitorum longus (EDL) sur des modèles de souris présentant les

symptômes de dystrophic musculaire, lors de dénervation musculaire et dans les cas

d'inactivité induite par suspension (Hayes et William 1994; Maltin et al., 1986; Zeman et

al., 1987; Dodd et Koesterer 2002). Par ailleurs il a été montré que chez des animaux

blessés à la moelle épinière, l'administration de clenbuterol favorisait le rétablissement

locomoteur en réduisant les dommages secondaires observés dans la moelle épinière après

le traumatisme (Zeman et al 1999). Ainsi, nous avons évalué les effets de l'administration

de testosterone et clenbuterol tant au niveau de la composition corporelle que sur la

récupération des fonctions locomotrices. Nos hypothèses pour l'article 3 étaient:

1. L'administration de clenbuterol et/ou de testosterone, préviendrait ou renverseraient

l'atrophie musculaire causée par l'inactivité due à la spinalisation.

2. L'augmentation de la masse musculaire, combinée au rôle neuroprotecteur rapporté

dans la littérature pour la testosterone et le clenbuterol, pourraient améliorer la

récupération motrice spontanée des souris paraplégiques.

Par la suite, nous avons évalué les effets de l'entraînement sur la récupération

locomotrice des souris paraplégiques. Tout d'abord, nous voulions déterminer si un

64

entraînement sur tapis roulant sans aucune autre forme d'assistance autre que les

stimulations sensorielles provenant du frottement des pattes arrière sur le tapis roulant,

pourrait être efficace pour rétablir une certaine motricité. Le rationnel derrière cette forme

d'entraînement était que cette forme de stimulation semblait suffisante afin d'induire des

mouvements chez la souris ayant une lésion complète de la moelle épinière (Guertin 2005;

Lapointe et al., 2006; Ung et al., 2007). Nos hypothèses pour l'article 4 étaient :

1. Un meilleur rétablissement moteur et locomoteur serait observé chez les souris

entraînées comparativement aux souris non-entraînées.

2. La masse musculaire des souris entraînées serait plus importante que celles des souris

non-entraînées.

Finalement, nous avons combiné un traitement pharmacologique à l'entraînement

locomoteur dans le but de proposer une approche multidisciplinaire qui favoriserait le

rétablissement locomoteur et renverserait l'atrophie musculaire et la détérioration osseuse.

Pour cette approche thérapeutique, nous avons combiné un traitement pharmacologique (L-

Dopa + carbidopa + buspirone) qui a montré son efficacité à activer le CPG et à induire des

mouvements locomoteurs avec support de poids (Guertin et al., 2010). Avec cette

pharmacologie, nous avons entraîné nos souris spinales sur tapis roulant. À cette

combinaison, nous avons ajouté du clenbuterol pour ces propriétés anaboliques sur le

système musculaire (Ung et al., 2010a). Notre hypothèse pour l'article 5 était:

1. L'entraînement assisté par la stimulation pharmacologique de L-Dopa + carbidopa

+ buspirone combiné à une administration de clenbuterol permettrait un meilleur

rétablissement des fonctions locomotrices, une augmentation de la masse

musculaire et moins de détérioration au niveau des propriétés osseuses, que des

souris non-entraînées ou des souris entraînées seulement avec la L-Dopa +

carbidopa + buspirone.

65

1.3 Approche méthodologique

Afin de répondre aux questions de cette thèse, nous avons utilisé un modèle animal

de souris. La souris a été choisi comme modèle animal dans le laboratoire parce qu'au

départ, peu de travaux s'intéressaient à ce modèle pour l'étude des blessures à la moelle

épinière. Etant donné que la moelle épinière est un organe relativement bien conservé au

cours de l'évolution, les observations notées chez la souris, risquent de se retrouver chez les

autres mammifères. La possibilité de pouvoir manipuler génétiquement la souris devient

également attrayant pour ce modèle animal (voir Lapointe et al., 2009; Landry et al., 2006).

Nous avons décidé d'utiliser le modèle de lésion complète de la moelle épinière, de cette

façon on isole complètement la moelle épinière des centres supraspinaux qui pourraient

influencer le rétablissement moteur ou locomoteur. De plus, parce que la majorité de nos

études utilisent des substances pharmacologiques, nous voulons nous assurer que les

observations effectuées font suite à un effet direct de ces drogues sur la moelle épinière. La

lésion de la moelle épinière se fait sans laminectomie, entre la 9e et la 10e vertèbre

thoracique. La chirurgie sans laminectomie réduit considérablement les complications de

santé, notamment les risques de scoliose.

Au niveau des analyses du rétablissement moteur et locomoteur, nous avons utilisé

une méthode d'évaluation semi-quantitative, ACOS combinée à une grille d'évaluation

motrice qualitative, l'AOB (le rationnel de ces choix se retrouve dans le chapitre V). Nous

combinons également des analyses de la cinématique du mouvement en évaluant le

déplacement angulaire des articulations de la cheville du genou et de la hanche dans le but

d'avoir une analyse plus détaillée du rétablissement moteur ou locomoteur. Suite à la

chirurgie et pour les certaines études, le poids de l'animal ainsi que la masse et l'aire

musculaire sont rapportés pour nous donner une idée générale de la santé des animaux et

pour voir si les traitements ont des effets bénéfiques sur l'atrophie musculaire.

66

C H A P I T R E V - S P O N T A N E O U S R E C O V E R Y O F H I N D L I M B M O V E M E N T I N

COMPLETELY SPINAL CORD TRANSECTED MICE: A COMPARISON OF

ASSESSMENT METHODS AND CONDITIONS

La première étude consistait à déterminer, à l'aide de différentes grilles d'évaluation,

le niveau de rétablissement moteur spontané chez des souris ayant une lésion complète de

la moelle épinière. Nous voulions savoir 1) si une grille était mieux adaptée pour évaluer le

recouvrement moteur de souris complètement spinales 2) s'il existait des différences entre

différentes conditions d'évaluation, soit l'arène (open-field) et le tapis roulant et 3) s'il

existait des différences de récupération motrice entre des souris mâles et femelles. Cette

article a été publié dans Spinal Cord, 2007,45 (5) : 367-79.

Study design: To compare results obtained with a variety of locomotor rating scales in

Th9/10 spinal cord transected (Tx) mice.

Objectives: To assess spontaneous recovery in Tx mice with a variety of rating scales to

find the most sensitive methods for assessing recovery levels in Tx mice and differences

associated with gender (male vs. female) and condition (open-field vs. treadmill).

Setting: Laval University Medical Center, Neuroscience Unit & Laval University,

Department of Anatomy and Physiology, Quebec City, Quebec, Canada.

Methods: Scales including the Basso, Beattie & Bresnahan (BBB), the Basso Mouse Score

(BMS), the Antri, Orsal and Barthe (AOB), the Motor Function Score (MFS) and the

Averaged Combined Score (ACOS) were used to assess, in open-field and treadmill

conditions, spontaneous locomotor recovery in male and female CD1 mouse at 7, 14, 21,

28 and 35 days post-Tx.

Results: The ACOS rating scale revealed a significant progressive increase of spontaneous

(without intervention) hindlimb movements during 5-weeks post-Tx. The other methods

detected a progressive increase for the first 2-3 weeks post-Tx without any significant

67

progress in week 4 and 5. Generally, scores obtained with each method were non-

significantly different between males and females or between open-field and treadmill

conditions. None of the mice were found to display weight-bearing capabilities or plantar

foot placements during this time period.

Conclusion: These results further confirm the existence of a limited but significant increase

of locomotor function recovery, occurring without intervention, in completely spinalized

animals. Although each method could detect small levels of recovery, the ACOS method

was discriminative enough to detect progressive changes up to 5 weeks post-Tx. In

conclusion, the ACOS rating scale was the most discriminative method for assessing the

spontaneous return of hindlimb movements found in completely spinalized mice, both in

open-field and treadmill conditions.

Introduction

Animal models have been increasingly used in the last twenty years to investigate the

pathological changes induced by spinal cord injury (SCI). These models have allowed the

study of potential new treatments and approaches to reduce secondary cellular damage and

scar formation or to increase neuronal regeneration and reconnection.1 Recently, mice have

been used more frequently for SCI research due to the availability of genetically engineered

models and molecular tools.2 Murine models with different types of SCI such as contusion,

displacement, crush, clip compression, ischemia, and transection are commonly used for

investigations.3

A number of scales and methods are available to assess functional recovery levels in

SCI mice. One of the most commonly used methods is the Basso, Beattie and Bresnahan

locomotor rating scale or BBB.4 However, this scale has been designed specifically for

spinal cord contused rats in open-field conditions and its utilization in mice has been

reported as problematic.5'6 Consequently, efforts have been made to develop alternative

methods adapted to SCI mice - adapted BBB5, Motor Function Score (MFS)7, Basso Mouse

68

Scale (BMS)8, and Average Combined Score (ACOS).9 These methods take into account

the fact that the hindlimb main articulations (hip, knee, and ankle) are not all easily

detectable in SCI mice9, that progression of locomotor function recovery is different in

mice than in rats (e.g. progression of tail movements)6 and that mice do not exhibit visually

detectable differences in toe drag.5

In addition, concerns have been raised by some researchers that most currently used

methods are not appropriate or sensitive enough to evaluate severely SCI or Tx animals.9'10

For instance, forelimb vs. hindlimb coordination or fine foot placement, assessed by most

standard methods, constitutes irrelevant criteria for evaluation of Tx animals. In line with

this, most studies with Tx mice produced recovery levels that are considered non­

significant by their authors.7'11'12 However, it has been clearly shown, using alternative

methods, that some significant levels of spontaneous motor and locomotor recovery can be

found in completely spinal cord transected mammals. Indeed, average scores up to level 5

have been reported after one month with the 22 level-AOB (Antri, Orsal and Barthe) scale

in untreated spinalized rats.1 Spontaneous full weight-bearing steps at relatively low

treadmill speeds have been described in untrained but tail-stimulated spinal cats after 2-3

weeks post-Tx.14 Also, in spinal mice, weight-bearing steps and plantar foot placements

have been detected in a few cases with tests performed on a motor-driven treadmill at

relatively low-speeds with tail stimulation.15 Without intervention (i.e. no graft, drug

treatment, tail stimulation or body-weight support), small but significant levels of

spontaneous recovery have been reported in spinal mice. Indeed, rhythmic bilaterally

alternated movements of small amplitude (i.e. locomotor-like but with no weight-bearing

and plantar foot placement capabilities) have been found after 2-3 weeks in the hindlimbs

of Tx mice tested in open-field conditions.9 Taken together, these results have

demonstrated the existence of spontaneous locomotor function recovery in completely

spinalized and developmentally mature mammals. However, locomotor scales have never

been compared to determine, in completely spinalized mice, which better detect this type of

recovery.

69

Here, we examined a number of locomotor scales to assess spontaneous locomotor

recovery levels in the hindlimbs of low thoracic Tx mice. Differences between males and

females, as well as between open-field and treadmill conditions, were also examined. The

aim was to identify which of these methods are better suited to assess spontaneous recovery

in completely spinalized mice.

Methods

Animal model and surgical procedures

All experimental procedures were conducted in accordance with the Canadian Council

for Animal Care guidelines and accepted by the Laval University Animal Care and Use

Committee. A total of twenty-two mice (11 male and 11 female CD1 mice, Charles River

Canada, St-Constant, Quebec), approximately eight week-old and initially weighing 30-40g

were used for this study. All mice were spinal cord transected at the low-thoracic level.16"18

In brief, preoperative care included subcutaneous injection of 1 ml lactate-Ringer's

solution, an analgesic (0.1 mg/kg buprenorphine) and an antibiotic (5 mg/kg Baytril). A

complete transection of the spinal cord was performed intervertebrally using microscissors

inserted between the 9th and 10th thoracic vertebrae in mice under complete anesthesia with

2.5% isoflurane. To ensure that complete transection was achieved, the inner vertebral

walls were explored and entirely scraped several times with scissor tips in order to disrupt

any small fibers which had not been severed. The incision was then sutured and the animals

were placed on a heating pad for a few hours. Postoperative care, provided for four days,

included subcutaneous injection of lactate-Ringer's solution (2 x 1 ml/day), buprenorphine

(0.2 mg/kg/day) and Baytril (5 mg/kg/day). Bladders were emptied manually until a

spontaneous return of the micturition reflex. Animals were left in their cage with food and

water ad libitum. Complete spinal cord transection was confirmed by 1) initial full paralysis

of the hindlimbs, 2) post-mortem visual examination of the spinal cord lesion for evidence

of spared tissue, and 3) coronal or midsagittal spinal cord sections stained with luxol fast

blue/cresyl violet for myelinated descending axons and Nissl substance.

70

Experimental protocol and assessment methods

Mice were left resting in their cage for two days after surgery to allow recovery before

testing. Tests were performed at 3, 7, 14, 21, 28 and 35 days post-Tx in both open-field and

treadmill conditions in a randomized manner to avoid carried-over fatigue. Animals were

also allowed to rest for approximately 40 minutes between the two conditions. Tests at 3

days were considered as a control, given that essentially no sign of hindlimb movement

recovery is observed after only few days post-Tx.9 In open-field conditions, mice were

examined inside a closed circular arena (60 x 60 cm) entirely made of transparent plexiglas

to facilitate video camera monitoring and recording.9 In treadmill conditions, we used a

custom-made 10-track adjustable-speed treadmill running at 8-10 cm/sec.17"19 Mice were

filmed using a digital video camera system (Sony DCR-PC9, shutter speed: 1/1000;

acquisition: 30 frames/sec) fixed on a tripod and positioned at a 45° angle above (open-

field) or behind (treadmill) in order to observe most hindlimb movement characteristics

including bilateral alternation. Data were directly collected and stored on a computer,

before being displayed and analyzed off-line by two trained observers. Five different

methods for assessing locomotor recovery were chosen based on their complementarity.

Although some of these methods have not necessarily been designed for spinal cord Tx

mice to be tested both in open-field and treadmill conditions, they have been shown to

provide, in some cases, valuable information in both conditions.10'18 Since many of these

methods have already been used in very different conditions than originally designed for

(e.g. BBB used in mice, see Engesser-Cesar, C. et al. 2005; Ma, M. et al. 2001)812, it

became of interest to compare them, in the same study, in order to clearly establish whether

or not some methods are more sensitive than others for Tx mice.

Basso, Beattie & Bresnahan locomotor scale (BBB)

This locomotor rating scale4'20 has been used extensively for the last ten years to assess

locomotor performance in incompletely spinal cord injured rats. It consists of 21

discriminative levels with progressively increasing scores:

71

0 - no hindlimb movement

1 - slight movement of one or two joints, usually the hip and/or the knee

2 - extensive movement of one joint with or without slight movement of one other

joint

3 - extensive movement of two joints

4 - slight movement of all three joints

5 - slight movement of two joints and extensive movement of the third

6 - extensive movement of two joints and slight movement of the third

7 - extensive movement of all three joints

All scores above 7 include some additional levels of plantar foot placement and/or

weight support (for details, see Basso and colleagues).20 Slight or extensive amplitude were

defined as less than half or more than half the normal range of joint motion respectively.

Occasional, frequent, and consistent were defined as < 50%, 51% - 94%, and > 95% of

total number of observed movements. Scores were determined for each of the two

hindlimbs and then averaged. Note that a modified BBB scale has been developed recently

by Dergham and colleagues.5 However, only scoring levels in the upper range of the scale

were modified, which is why we did not use it for testing in addition to the BBB scale.

2.2. Antri, Or sal and Barthe motor scale (AOB)

This locomotor rating scale created by Antri, Orsal and Barthe10 was designed

specifically to assess hindlimb movements in Tx rodents. It does not assess forelimb vs.

hindlimb coordination which is considered irrelevant in assessing spontaneous recovery in

transected animals. This is because regenerative processes across the lesion in adults are

only possible in animals with partially injured spinal cords, following grafting

interventions, or through the use of specific regenerative treatments. The AOB scale

consists of 22 discriminative scores:

0 - no movement

1 - weak limb jerks

2 - weak rhythmic movements with no bilateral alternation

72

3 - large rhythmic movements with no bilateral alternation

4 - weak rhythmic movements with occasional bilateral alternation

5 - large rhythmic movements with occasional bilateral alternation

6 - weak rhythmic movements with frequent bilateral alternation

7 - large rhythmic movements with frequent bilateral alternation

8 - weak rhythmic movements with consistent bilateral alternation

9 - large rhythmic movements with consistent bilateral alternation

Additional levels include body-weight support and plantar foot placement capabilities

(for additional details, see Antri and colleagues).10 Conditions of observation and criteria

for evaluating amplitude and frequency were the same as for the BBB scale (see above).

Note that characteristics occurring only rarely (e.g., once or twice) were not considered

sufficient to fulfill the requirements for up-grades to higher corresponding levels.

Basso Mouse Scale (BMS)

This method was developed for incomplete SCI mice preferably in open-field

conditions. It takes into account that locomotor recovery progression is different in mice

than in rats.8 It is a 9-point scale divided as follows:

0 - No ankle movement

1 - Slight ankle movement

2 - Extensive ankle movement

3 - Plantar placing of the paw with or without weight support or occasional, frequent or

consistent dorsal stepping but no plantar stepping

4 - Occasional plantar stepping

J - Frequent or consistent plantar stepping, no coordination or frequent or consistent

plantar stepping, some coordination, paws rotated at initial contact and lift off

6 - Frequent or consistent plantar stepping, some coordination, paws parallel at initial

contact or frequent or consistent plantar stepping, mostly coordinated, paws rotated

at initial contact and lift off

73

7 - Frequent or consistent plantar stepping, mostly coordinated, paws parallel at initial

contact and rotated at lift off or frequent or consistent plantar stepping, mostly

coordinated, paws parallel at initial contact and lift off, and severe trunk instability

8 - Frequent or consistent plantar stepping, mostly coordinated, paws parallel at initial

contact and lift off, and mild trunk instability or frequent or consistent plantar

stepping, mostly coordinated, paws parallel at initial contact and lift off, normal

trunk stability, and tail down or up and down

9 - Frequent or consistent plantar stepping, mostly coordinated, paws parallel at initial

contact and lift off, normal trunk stability, and tail always up.

Hindlimb Motor Function Score (MFS)

This method was developed by Farooque specifically to evaluate SCI mice. It consists

of a 10-point scale:

0 - No movement of the hindlimbs

1 - Barely perceptible movement of any hindlimb joints (hip, knee or ankle)

2 - Brisk movements at one or more hindlimb joints in one or both limbs but no

coordination

3 - Alternate stepping and propulsive movements of hindlimbs but no weight bearing, 4

- Weight bearing and can walk with some deficit

5 - Normal walking

6 - Normal walking and can walk on a 2-cm-wide bar

7 - Can walk on a 1.5-cm-wide-bar

8 - Can walk on a 1-cm-wide-bar

9 - Can walk on a 0.7-cm-wide-bar

10 - Can walk on a 0.5-cm-wide-bar

Average combined Score (ACOS)

74

This method is used routinely in our laboratory.9'17'18 In addition to being partially

quantitative and therefore more objective, it is a useful method for distinguishing

locomotor-like movements (LM) from non-locomotor movements (NLM) in the hindlimbs

of spinalized mice. LM and NLM frequency, incidence, and amplitude were assessed

during a 4-min bout of video-recorded activity.9 To ease comparisons, a unique average

combined score (ACOS) is created by simple arithmetic combination of the collected

values - NLM and LM frequency (per min), amplitude, and incidence (see below for

details, ACOS = [NLM + (2 x LM)] x amplitude). One LM was defined as an entire step­

like cycle consisting of an extension phase or stance followed by a flexion phase or swing

occurring in both hindlimbs consecutively (i.e. bilaterally alternated or out-of-phase

relation). Extension began with foot contact onset (i.e. dorsal or plantar foot) until the lift

off or the end of foot contact with the ground or treadmill belt. Flexion began with foot

contact ending (i.e. lift off) until next foot contact or extension onset. In the case where the

foot never quite cleared the ground or was constantly rubbing against the treadmill belt (or

the ground), then extension was more generally defined as when the hindlimb was in a

relatively extended position, and flexion when it was not extended and generally flexed.

One NLM was defined as one non-bilaterally coordinated movement (i.e. not followed by a

flexion-extension on the other side). There included unilateral movements, jerks, brief

sequences of fast-paw shaking (typically lasting 1-2 sec/episode and counted as one NLM),

twitches, and kicks. Amplitude was characterized by assigning one of three values; 0 - if no

movement was observed; 1 - if the amplitude of most movements was less than half the

range of motion of normal steps; 2 - if the amplitude of most movements was at least more

than half the range of motion of normal steps. Note that amplitude was scored for LM and

NLM indistinctively. Incidence corresponded with the number of mice (out of all mice

tested in a group) in which NLMs or LMs were observed. Plantar foot placement and body-

weight support were reported as either present or not. Note that, in the equation, LM is

multiplied by a factor of '2' for very simple and logical reasons. We consider, from our

experience, that it is easier for most observers to count as ' 1 ' rather than '2', an event

defined as one bilaterally alternating movement (i.e. 1 LM). However, what are being

described really are two consecutive movements (one in each of the two hindlimbs). To

respect the 'linearity' of score progression of the ACOS method, then one LM (one

75

movement in each of the two hindlimbs) which is then twice more 'valuable' than only one

NLM (one single hindlimb movement) should therefore, in the end, be multiplied by a

factor of '2 ' to reflect the fact that two movements are really being described with one LM.

Statistics

Results were reported as means ± SE. For differences between days, a Friedman test

followed by a Dunn's Multiple Comparison test was used. In order to evaluate gender- and

condition-related differences, a two-way repeated measure ANOVA followed by a

Bonferonni post-hoc was used. Statistical differences between the linear regression slopes

were examined with ANCOVA. P values < 0.05 were considered statistically significant.

Results

Low-thoracic spinal cord transected (Tx) mice (males and females) were filmed over a

period of 4 minutes in open-field and in treadmill conditions at 3, 7, 14, 21, 28 and 35 days

post-surgery. Spontaneously occurring hindlimb movements were subsequently analyzed

(off-line) using five different locomotor scoring methods. Data from male and female mice

were generally pooled together except for Fig.4 where gender-related differences were

specifically examined.

Recovery levels with time

Figure IA shows that Tx mice developed spontaneous hindlimb movement under both

conditions, that corresponded to average scores lower than level 1 on the BBB scale during

the first two weeks post-Tx. Indeed, increasing scores, although considered as non­

significant (P = 0.066), were found at 7 days (0.40 ± 0.07) compared with controls (3 days,

0.16 ± 0.06). At 14 and 21 days, further increased scores that reached 0.84 ± 0.09, (P <

0.01) and 1.40 ± 0.16, (P < 0.001) respectively were found to be significantly greater than

controls (Fig. IA). No significant additional progress (P > 0.05) was detected between 21,

76

28 and 35 days showing that performances assessed with the BBB scale reached a plateau

level at 21 days ( 1.40 ± 0.15) until the end of the study period ( 1.62 ± 0.18 at 35 days).

Comparable results were found with the AOB scale. Indeed, Fig. IB shows increasing

average scores that reached 0.52 ± 0.08 and 0.93 ± 0.09 at 7 and 14 days post-Tx

respectively. Significant differences (P < 0.001) compared with controls (0.20 ± 0.09 at 3

days) were found at 14, 21, 28 and 35 days (Fig. IB).

As with the BBB scale, performances assessed with the AOB scale reached a plateau at

21 days after which no additional significant progress (P > 0.05) was found when

comparing 21 to 28 days and 28 to 35 days.

The BMS scale evaluated recovery levels to be comparable with those described above.

An increase, although non significant (P = 0.059), was found at 7 days post-Tx (0.38 ±

0.08) compared with controls (0.14 ± 0.07, Fig. 1C). Further significant increases (P <

0.001) were observed until a plateau level was reached at 14 days post-Tx. Similar results

were found with the MFS scale with significant (P < 0.001) increases in performance post-

Tx (i.e. at 14 days and later) until a plateau level was reached at 21 days (Fig. ID).

The ACOS method detected a progressive increase in performance up to 35 days post-

Tx (Fig. 2E). At 7 days, an increase (0.66 ± 0.18) close to the level of significance (P =

0.051) was found compared with controls (0.09 ± 0.04). Further significantly (P < 0.001)

higher scores were found at 14, 21, 28 and 35 days. In contrast with the other methods,

ACOS scores kept progressively increasing at 21 days compared with 14 days (8.21 ± 1.23

vs. 2.55 ± 0.69, P < 0.01) and at 35 days compared with 14 days (11.61 ± 2.03 vs. 2.55 ±

0.69, P < 0.001). The average score at 35 days, although representing a 45% increase

compared with the 28 days score (7.96 ± 1.97), did not reach statistical difference (P =

0.171, Fig. 2E). Although, an apparent plateau level was reached at 21 and 28 days,

detailed scores for each of the two testing conditions revealed that no plateau of

performance was reached with ACOS, at least when treadmill testing was employed (see

Fig.3E).

77

As mentioned in the Methods, the ACOS score is the result of a combination of four

distinct factors assessed separately - NLM & LM Frequency, Amplitude and Incidence. In

Fig. 2A-D, these values were found to progress differently throughout the time period

studied. NLM (frequency/min) was found to reach 0.64 ± 0.17 at 7 days, which is almost

significantly (P = 0.051) different than controls at 3 days (0.09 ± 0.04, Fig. 2A).

Significantly different values were found at 14 days and later. The highest average value

was observed at 35 days with 8.84 ± 1.34 NLMs. NLMs were found to progressively

increase with time up to 35 days, except for a small non-significant reduction at 28 days.

LMs were also observed to increase during the time period studied. While virtually no LM

was found at 3 and 7 days post-Tx (except in one mouse where 1 LM was found), LMs

began to be detected at 14 days with an average score of 0.17 ± 0.10 (Fig. 2B). Increases

were found reaching significantly different values at 21 (0.81 ± 0.26, P < 0.01), 28 (1.67 ±

0.81, P < 0.05) and 35 days (1.13 ± 0.32, P < 0.001) compared with controls (no LM).

However, no significantly progressive increase was found given the low average frequency

values reported. One reason for this is illustrated in Fig. 2D where LMs were found in less

than 60% of the mice tested at 21, 28 or 35 days (which considerably affected the average

values). In contrast, NLMs were found in nearly all mice tested (95%) at 21 days and later.

Fig. 2C shows that the assessed amplitudes of all movements (NLMs and LMs) values

remained in the lower range of the scale (i.e. 1- small amplitude and 2-large amplitude, see

Methods).

Comparison of methods

Side-by-side comparisons clearly show that scores differed mainly between the ACOS

system and the other methods. This is illustrated in Fig. 3, where the BBB, AOB, BMS and

MFS scales provided virtually identical scores. As mentioned earlier, scores with these four

methods significantly increased up to 21 days, after which a plateau was reached (see also

Fig. 1). Overall, these methods provided values lower than 2 - one of the highest scores was

2.13 ± 0.30 obtained with the AOB scale in open-field conditions at 21 days (see Fig. 3B,

Table 2). The ACOS method clearly provided significantly higher scores than those

obtained with the other methods. Figure 3E illustrates that in open-field conditions, ACOS

78

reached its highest average value at 35 days (8.97 ± 2.78). However, the highest ACOS

score was found in treadmill conditions at 35 days with 14.25 ± 2.93 (Fig. 3E).

Correspondingly, linear regression analyses revealed that significantly steeper slopes were

found with the ACOS method compared with the other scales tested (Table 1). In open-

field conditions, a near five-fold increase was found with the ACOS method (slope values

of 0.28 with ACOS vs. < 0.07 with any of the other methods, P = 0.002, Table 1 and Fig.

3A). In treadmill conditions, a ten-fold increase in slope was reported between ACOS and

the other methods (0.46 with ACOS vs. < 0.04 with the others, P < 0.001, Table 1 and Fig.

3B). Comparison of linear regression slopes clearly showed that ACOS scores increased

over a wider range of scoring levels than the other methods during the time period studied.

As said, ACOS offers a better discriminative power than the other methods for assessing

progressively increasing performances in this animal model, especially in treadmill

conditions.

Difference between open-field and treadmill conditions

In general, no difference was found between open-field and treadmill conditions, using

the BBB scale. However, a few non-significant differences were detected. For instance, at

21 days post-Tx, average BBB scores were greater (P = 0.119) in open-field (open circles)

than in treadmill (filled circles) conditions (Fig. 3A, 1.66 ± 0.25 vs. 1.14 ± 0.18). A similar

difference (P = 0.080) was found at 28 days (1.81 ± 0.20 vs. 1.35 ± 0.20 respectively, see

also Table 1). With the AOB scale no overall significant difference between conditions was

observed. However, significantly different scores (P = 0.014) were found at 21 days in

open-field vs. treadmill conditions (2.13 ± 0.30 vs. 1.27 ± 0.17, Fig. 3B). Comparable

differences with the MFS scale were found at 21 days between open-field and treadmill

conditions (Fig. 3D). Significant difference between the two tasks was only found with the

BMS scale (P = 0.009) where at 21 (P = 0.013) and 28 days (P = 0.001) difference reached

significant levels (Fig. 3C). Therefore, in general, scores were higher in open-field than in

treadmill conditions from 21 to 35 days post-Tx. In turn, average scores were lower in

open-field than in treadmill conditions at 3, 7 and 14 days post-Tx although no statistically

different levels were reached. In contrast, the ACOS method displayed average scores that

79

were consistently higher in treadmill than in open-field conditions - except at 21 days.

However, these differences reached significant (P = 0.014) levels only at 14 days post-Tx

(Fig. 3E). Methods giving highest average scores were ACOS (14.25 ± 2.93: treadmill at 35

days) followed by AOB (2.13 ± 0.30: open-field at 21 days) (Table 2). It is worth noting

also that progression of performance assessed with the ACOS method was clearly different

in open-field vs. treadmill conditions. As mentioned earlier, the linear regression slopes in

open-field and treadmill conditions were steeper than those of all the other methods tested

(Table 1). However, a close-to-perfect linear relationship (r2 = 0.99, slope = 0.46) reflecting

a steady increase of performance over the five-week period was found on a treadmill

whereas some plateau level was reached in open-field conditions since ACOS scores at 21,

28 and 35 days post-Tx were non-significantly different (P > 0.05).

Differences between male and female

No systematic difference was found between male and female animals. However a few

punctual differences were found at some time points. For example, at 21 days (P = 0.130)

higher scores were found in females with the AOB scale (Fig. 4B, filled squares) whereas

at 14 days higher (P = 0.029) scores were found in males with the ACOS method (Fig. 4E,

open triangles). Otherwise, no significant differences were found between males and

females as clearly illustrated by the graphs.

Discussion

The results showed that a limited but significant increase of locomotor function

recovery can occur without intervention, in completely spinalized mice, tested in open-field

or treadmill conditions. Generally, hindlimb movements, assessed with the BBB, AOB,

BMS and MFS scales, corresponded to scoring levels lower than 2 (i.e. no weight-bearing

stepping, propulsive or large movements or plantar foot placement) that reached a plateau

at 14 or 21 days post-Tx. Only the ACOS scoring system was discriminative enough to

detect progressive changes up to 35 days post-Tx, especially in treadmill conditions. None

80

of the methods provided significantly different scores between males and females or

between open-field and treadmill conditions, albeit few exceptions.

Possible mechanisms underlying spontaneous recovery without intervention

As summarized above, the results clearly demonstrated the existence of spontaneously

occurring hindlimb movement in completely spinal cord Tx mice. These data further

confirm the results of previous studies that have reported significant spontaneous recovery

in complete Tx rats, cats, and mice. Indeed, average scores up to level 5 were assessed after

a few weeks with the 22 level-AOB scale in untreated but tail-stimulated spinalized rats

examined on a motor-driven treadmill.13 Spontaneous full weight-bearing steps at relatively

low treadmill speeds have also been found in untrained and tail-stimulated spinal cats after

a few weeks post-Tx.14 In spinalized mice, weight-bearing steps and plantar foot

placements have been reported in a few animals. Tests were performed at relatively low-

speeds on a motor-driven treadmill, employing tail stimulation and using quantitative

kinematic analysis.15 Without intervention (i.e. no graft, drug treatment, tail stimulation or

body-weight support), small but significant levels of spontaneous recovery have been found

in all Tx mice tested in open-field conditions.9 Indeed, it was reported that rhythmic and

bilaterally alternated movements of small amplitude (i.e. locomotor-like but with no

weight-bearing and plantar foot placement capabilities) spontaneously occur after a few

weeks in these animals. The lower level of recovery reported by Guertin (2005)9 is in

contrast with that reported by Leblond and colleagues (2003).15 This discrepancy was first

attributed to different testing conditions (open-field vs. motor-driven treadmill, see

Discussion in Guertin, 20059; Guertin and Steuer, 2005.18 However, the present results

provide additional insights regarding this discrepancy. Indeed, the Tx mice tested on a

motor-driven treadmill (at speeds similar to those used by Leblond and colleagues)15 were

found, with several assessment methods, not to reach recovery levels that included large

amplitude movements, weight-bearing stepping and plantar foot placements. Therefore, the

relatively high level of recovery reported by Leblond and colleagues15 was most likely

attributable to afferent-induced activation of the central pattern generator (CPG), caused by 91

tail stimulation. We can not exclude the possibility that, in both studies, some of the

81

recovery was facilitated or partially caused by repeated testing sessions over time which

could constitute some form of training-induced effects upon CPG reorganization and

activation.14 However, arguing against this possibility is the fact that we tested the mice

only 2 x 4 min per week (i.e. 4 min in open-field and then in treadmill conditions). This

suggests that the spontaneous recovery described, at least in this study, is due to

spontaneous sublesional neuronal network changes (e.g. reorganization and plasticity) and

increased CPG excitability. The idea that plasticity at sublesional levels post-Tx may be

associated with recovery is supported by results from in vitro preparations showing signs of

increased CPG excitability and spontaneous fictive locomotor activity in rodent isolated

spinal cords. ' Although it would be beyond the scope of this article to discuss the

cellular mechanisms underlying such spontaneous recovery in Tx animals, recent results

suggest a role for specific immediate early genes such as c-fos, Nor-1 and Nur77 as early

genetic events that may lead to locomotor network reorganization and spontaneous

activation.24 Other trans-membranal changes such as increased expression of subsets of

serotonin, noradrenaline, glutamate and glycine receptors have also been proposed as

factors that may contribute to functional recovery in Tx animals.25"27

Why would performances 'level off' after 2-3 weeks with most assessment methods?

It is clear that small recovery levels such as those seen in this animal model, which are

characterized mainly by small amplitude movements, can rarely reach BBB scores above

level 1 or 2. This is because large amplitude movements at one or two joints are required to

reach levels 2 and 3, respectively.4 A similar limiting factor is found with the BMS scale,

given that small amplitude movements generally do not qualify for level 2 (i.e. extensive

ankle movement). If in a few animals, larger movements at the ankle joint were observed,

the performances would most certainly reach a plateau at level 3, characterized by either

plantar foot placement or dorsal stepping (see BMS scale in Engesser-Cesar and

colleagues). With the MFS scale, level 2 is relatively easy to reach with the type of

performance generally observed in the conditions of the present study. Indeed, the 2nd MFS

level is only characterized by any movement at one or more joints in one or both limbs.

However, a substantial increase of performance is necessary to qualify as level 3 (i.e.

82

alternate stepping and propulsive movement of hindlimbs)7, making it very difficult to

reach that level in the case of spontaneous recovery without intervention. Also, the word

propulsive in the MFS scale may be confusing since if no body-weight support is displayed,

then it is difficult to associate any spontaneous hindlimb movement with propulsion per se.

Regarding the AOB scale, higher scores were generally reported than with the above

methods. This may be explained by the fact that this scale was specifically designed for

completely spinal cord transected animals.10 Therefore, level 2, which is defined by weak

amplitude movements, can easily be reached by the type of performance found in this

study. Level 3 is also easier to reach than with the other methods, given that it requires

larger amplitude movements but not necessarely specifically at the ankle joint or at more

than one joint or that qualify as stepping (i.e. no bilateral alternation required). Therefore, it

is clear that the type of performance observed in Tx mice without intervention can not be

characterized generally by more than 2 scoring levels with the BBB, BMS, MFS scales and,

to some extent, the AOB scale.

However, the present results provide evidence suggesting that another factor also

contributed to the development of a plateau in the scoring level. The ACOS method and, in

particular, one of its factors, incidence, contributed to the initial progressive increase of the

assessed performances. Indeed, analysis of the incidence factor showed that only about

50% of the mice tested displayed some hindlimb movement (NLMs) at 7 days and that few

displayed coordinated alternating movements (LMs, Fig. 2E). Therefore, even if with BBB,

BMS, MFS, and AOB scales, level 1 can be reached at 7 days by some animals, the fact

that only 50% of them displayed some hindlimb movements or NLMs explains the report

of average scores closer to 0.5 than to 1.0 (Fig. 1A-D at 7 days). In fact, the apparent

progressive increase of performance at 14 days compared with 7 days can be almost

entirely explained by an increase of incidence (i.e. the number of mice in which NLMs

were observed). Indeed, incidence values for NLMs reached 80% of mice tested at 14 days

which, again, nicely fits with the increase of average scoring levels assessed with most of

these methods (i.e. just below 1.0, Fig. 1A-D at 14 days). This idea, that a progressive

increase of incidence during the first two weeks is a determinant factor in the increase of

average scoring levels reported with these methods, is supported by the fact that very

83

similar graphs are found both for incidence (Fig. 2E, NLMs) and for average BBB, BMS,

MFS and AOB scores (Fig. 1A-D). In other words, the initial increase of average scoring

levels assessed with the BBB, BMS, MFS and AOB methods are mainly associated with an

increase in incidence. This also strongly suggests that the plateau level reached at 14 or 21

days may be explained both by the fact that the incidence had reached near 100% (NLMs)

and that, as mentioned earlier, scores above level 2 with these methods are nearly

impossible to reach.

How and why is ACOS different?

The ACOS method is different than the other tested methods in many ways. For

instance, it is a semi-quantitative rather than an entirely qualitative method, given that

NLM and LM frequency (counts per min) and incidence are calculated. Not only does this

allow the assessment of a factor rarely examined by the other methods (i.e. frequency), but

it also allows performances to be plotted against a wider range of Y-axis values.

Consequently, this makes it a more sensitive scale to report the progression of locomotor

function recovery (i.e. movement frequency). Thus, performance regarding NLMs,

progressively increased from average scores of '0' (at 3 days) up to '9' (at 35 days, Fig.

2A). Another difference of the ACOS method is that bilaterally coordinated movements

(LMs) are reported separately to non-coordinated movements (NLMs). Although these two

factors are recognized by many of the other methods as distinctive characteristics of

functional recovery (e.g. scores < 2 or 3 with the MFS or AOB respectively correspond to

our definition of NLMs and not LMs),7'10 only the ACOS method allows their evaluation

and quantification separately and in parallel. This allowed the finding that NLMs

progressed differently than LMs during the time period studied (Fig. 2A vs. 2B).

Also, as mentioned earlier, the assessment of incidence both for NLMs and LMs

provided information not available with the other tested methods. The quantification of

movement amplitude, although done with arbitrary values (i.e. 1- small amplitude, 2-large

amplitude), revealed an important aspect of the performances found in this study. This was

that spontaneously occurring movements in Tx mice during this five-week period do not

84

qualify as large amplitude movements (Fig. 2C). All the characteristics or factors assessed

with the ACOS method (i.e., NLM & LM frequency, amplitude and incidence) when

combined to obtain the ACOS score, contributed to produce increasing average scores

reflecting generally a progressive increase of the performances during the entire time period

studied. This is true if average scores obtained in both testing conditions were averaged as

in Fig. 2D. Surprisingly however, if examined separately, the ACOS scores in open-field

conditions were found to differ considerably compared with those on a treadmill (Fig. 3E,

Table 1). In open-field conditions, the average scores were found to reach a plateau not

dissimilar to those reported with the other tested methods. In fact, it is only with tests on a

treadmill that scores and therefore the assessed performances were found to linearly

increase for five weeks post-Tx. The reasons for this are unclear. However, it is possible

that conditions associated specifically with the treadmill tests (i.e. the hindlimbs dragging

behind with the entire front part including the dorsal paws rubbing continuously against the

moving treadmill belt) led to some activation of cutaneous and proprioceptive receptors of

the hindlimbs. This could have facilitated the progressive increase of performance given

that such afferent inputs are well-known to modulate, excite and reset CPG activity. This

has been shown in decerebrate and paralyzed cats with Ia- and Ib-proprioceptor afferent or

cutaneous afferent stimulation during fictive locomotion.28'29

Concluding remarks

These results contributed to demonstrate that the ACOS scale, especially in treadmill

conditions, provides a more sensitive method for assessing the type of recovery and

performance occurring spontaneously without intervention in Tx mice. However, it does

not assess many of the detailed characteristics inherent to higher recovery levels such as

balance, fine placement of the foot and digits, agility, speed, etc. Therefore, to examine

higher levels of locomotor recovery, such as after regeneration and reconnection across the

lesion induced by grafts or drug treatments, it may be preferable to combine a number of

assessment methods for SCI mice and other species. The idea of combining a number of

assessment methods for a more complete evaluation of performance is supported by results

that have been published by others.30'31

85

References

1 Schwab ME. Repairing the injured spinal cord. Science 2002; 295: 1029-1031.

2 Steward O, Schauwecker PE, Guth L, Zhang Z, Fujiki M, Inman D, Wrathall J,

Kempermann G, Gage FH, Saatman KE, Raghupathi R, Mcintosh T. Genetic

approaches to neurotrauma research: opportunities and potential pitfalls of murine

models. Exp Neurol 1999 ; 157: 19-42.

3 Kwon BK, Oxland TR, Tetzlaff W. Animal models used in spinal cord regeneration

research. Spine 2002 ; 27:1504-1510.

4 Basso DM, Beattie MS, Bresnahan JC. A sensitive and reliable locomotor rating scale

for open field testing in rats. J Neurotrauma 1995; 12: 1-21.

5 Dergham P, Ellezam B, Essagian C, Avedissian H, Lubell WD, McKerracher L. Rho

signaling pathway targeted to promote spinal cord repair. J Neurosci 2002 ; 22 : 6570-

6577.

6 Basso DM. Behavioral testing after spinal cord injury: congruities, complexities, and

controversies. J Neurotrauma 2004; 21: 395-404.

7 Farooque M. Spinal cord compression injury in the mouse: presentation of a model

including assessment of motor dysfunction. Acta Neuropathol (Berl) 2000; 100: 13-22.

8 Engesser-Cesar C, Anderson AJ, Basso DM, Edgerton VR, Cotman CW. Voluntary

wheel running improves recovery from a moderate spinal cord injury. J Neurotrauma

2005; 22: 157-171.

9 Guertin PA. Semiquantitative assessment of hindlimb movement recovery without

intervention in adult paraplegic mice. Spinal Cord 2005 ; 43: 162-166.

86

10 Antri M, Orsal D, Barthe JY. Locomotor recovery in the chronic spinal rat: effects of

long-term treatment with a 5-HT2 agonist. Eur J Neurosci 2002; 16: 467-476.

11 Wamil AW, Wamil BD, Hellerqvist CG. CM 101-mediated recovery of walking ability

in adult mice paralyzed by spinal cord injury. Proc Natl Acad Sci USA 1998 ; 95:

13188-13193.

12 Ma M, Basso DM, Walters P, Stokes BT, Jake man LB. Behavioral and histological

outcomes following graded spinal cord contusion injury in the C57B1/6 mouse. Exp

Neurol 2001 ; 169: 239-254.

13 Antri M, Barthe JY, Mouffle C, Orsal D. Long-lasting recovery of locomotor function

in chronic spinal rat following chronic combined pharmacological stimulation of

serotonergic receptors with 8-OHDPAT and quipazine. Neurosci Lett 2005; 384: 162-

167.

14 de Leon RD, Hodgson JA, Roy RR, Edgerton VR. Locomotor capacity attributable to

step training versus spontaneous recovery after spinalization in adult cats. J

Neurophysiol 1998; 79: 1329-1340.

15 Leblond H, L'Espérance M, Orsal D, Rossignol S. Treadmill locomotion in the intact

and spinal mouse. J Neurosci 2003; 23: 11411-11419.

16 Guertin PA. Role of NMDA receptor activation in serotonin agonist-induced air-

stepping in paraplegic mice. Spinal Cord 2004a; 42: 185-190.

17 Landry ES, Guertin PA. Differential effects of 5-HT(l) and 5-HT(2) receptor agonists

on hindlimb movements in paraplegic mice. Prog Neuropsychopharmacol Biol

Psychiatry 2004 ; 28:1053-1060.

87

18 Guertin PA, Steuer I. Ionotropic 5-HT3 Receptor Agonist-Induced Motor Responses in

the Hindlimbs of Paraplegic Mice. J Neurophysiol 2005 ; 94: 3397-3405.

19 Guertin PA. Synergistic activation of the central pattern generator for locomotion by 1-

beta-3,4-dihydroxyphenylalanine and quipazine in adult paraplegic mice. Neurosci Lett

2004b; 358: 71-74.

20 Basso DM, Beattie MS, Bresnahan JC. Graded histological and locomotor outcomes

after spinal cord contusion using the NYU weight-drop device versus transection. Exp

Neurol 1996; 139: 244-256.

21 Strauss I, Lev-Tov A. Neural pathways between sacrocaudal afférents and lumbar

pattern generators in neonatal rats. J Neurophysiol 2003; 89: 773-784.

22 Bonnot A, Morin D, Viala D. Genesis of spontaneous rhythmic motor patterns in the

lumbosacral spinal cord of neonate mouse. Brain Res Dev Brain Res 1998; 108: 89-99.

23 Norreel JC, Pflieger JF, Pearlstein E, Simeoni-Alias J, Clarac F, Vinay L. Reversible

disorganization of the locomotor pattern after neonatal spinal cord transection in the rat.

J Neurosci 2003 ; 23:1924-1932.

24 Landry ES, Guertin PA. Profile of immediate early gene expression in the lumbar spinal

cord of paraplegic mice. Soc Neurosci Abstr 2005; 865.2.

25 Talmadge RJ, Roy RR, Edgerton VR. Alterations in the glycinergic neurotransmitter

system are associated with stepping behavior in neonatal spinal cord transected rats. Soc

Neurosci Abstr 1996; 380.

26 Giroux N, Rossignol S, Reader TA. Autoradiographic study of alphal- and alpha2-

noradrenergic and serotonin 1A receptors in the spinal cord of normal and chronically

transected cats. J Comp Neurol 1999; 406: 402-414.

88

27 Ung RV, Landry ES, Lapointe N, Rouillard C, Levesque D, Guertin P. Expression,

activation, and function of 5-HT2A receptor in the lumbar spinal cord of adult

paraplegic mice. Soc Neurosci Abstr 2005; 516.7.

28 Guertin P, Angel MJ, Perreault MC, McCrea DA. Ankle extensor group I afférents

excite extensors throughout the hindlimb during fictive locomotion in the cat. J Physiol

1995; 487: 197-209.

29 Perreault MC, Angel MJ, Guertin P, McCrea DA. Effects of stimulation of hindlimb

flexor group II afférents during fictive locomotion in the cat. J Physiol 1995; 487: 211-

220.

30 Kunkel-Bagden E, Dai HN, Bregman BS. Methods to assess the development and

recovery of locomotor function after spinal cord injury in rats. Exp Neurol 1993; 119:

153-164.

31 Metz GA, Merkler D, Dietz V, Schwab ME, Fouad K. Efficient testing of motor

function in spinal cord injured rats. Brain Res 2000 ; 883: 165-177.

89

Figures and Legends

Figure 1. Spontaneous recovery assessed with qualitative methods. Evaluation of

spontaneous recovery with 4 different qualitative scales A) BBB, B) AOB, C) BMS and D)

MFS . Three days after spinal cord transection, score levels close to zero. This time point

served as a control. For each method tested, a non-significant increase was observed at 7

days. Compared with day 3, significant levels of recovery were found at 14 days and

subsequently. After 21 days, all scales reached a plateau as no significant differences were

found between scores at 21, 28 and 35 days. * : P < 0.01, ** : P < 0.001.

Figure 2. Spontaneous recovery assessed with the ACOS method. Evaluation of

spontaneaous recovery with a semi-quantitative locomotor scale called ACOS. The method

is a combination of different assessed parameters: A) NLM, B) LM and C) Amplitude and

D) Incidence. NLMs and LMs are found to significantly increase at 14 days and 21 days,

respectively. Amplitude and incidence reached plateaus after 21 days. At this time point,

nearly all mice produced NLMs and 50% of them displayed LMs. Compared with scores at

day 3, ACOS scores showed significant differences at 14 days vs. 35 days (a two-way

repeated measure ANOVA followed with a Bonferonni post-hoc). Differences were found

also between 14 days and 21 days and between 28 days and 35 days. # : P < 0.05, * : P <

0.01, **:/>< 0.001.

Figure 3. Differences between open-field and treadmill conditions. Two-way repeated

measure ANOVA did not reveal task-related differences between open-field and treadmill

conditions except for C) BMS scale where in open-field, score levels are significantly

higher (P = 0.009) than on treadmill. However, when using a Student paired T-test,

differences between conditions were found at some time points for B) AOB, D) MFS and

E) ACOS. For all scales, except ACOS, before 14 days, scores tended to be lower in open-

field than on the treadmill. From 21 days up to 35 days, scores were higher in open-field

conditions. Mice evaluated with ACOS tended to have higher scores on the treadmill

throughout the testing period, except at 21 days where, in open-field conditions, scores

were higher compared with the treadmill. ** : P < 0.01, * : P < 0.001.

90

Figure 4. Differences between males and females. Gender related differences assessed with

5 locomotor rating scales. Two-way repeated measure ANOVA did not reveal any

significant difference between males and females. # : P < 0.05

Table 1 Summary of linear regression. Slopes and squared correlation coefficients (r2)

relating the progression in spontaneous motor recovery tested with all scales. All

regressions were significantly different from 0.

Table 2 Summary of data collected for each task. Average scores calculated for each

locomotor scale at each day of the testing period. Data are reported as mean ± SE.

Distinction is made between open-field (upper panel) and treadmill (lower panel).

Table 3 Summary of data collected for each gender. Average scores calculated for each

locomotor scale at each day of the testing period. Results from open-field and treadmill are

combined for each gender. Data are reported as mean ± SE. Distinction is made between

males (upper panel) and females (lower panel).

91

7 14 21 26 35 Days post-SCI

7 14 21 28 Days post-SCI

Figure 1

92

A) NLM

& c III

Q-tu

14 21 28 35

C) Amplitude D) Incidence n.z-

1.0- M p —f »

_ >^ ? 0.8- àf (U f \

G i j T

11) y "5 o.e- - r / « d À / Q . T <f 0.4- j

0.2- 0.0- i 1 i i — — i

14

14 21 Days post-SCI

8 c OD T3 D C

Days post-SCI

Figure 2

93

B) AOB

2.0

| 1.5 1 f

8 10

0.5

0.0 - * * — r —i 1 1— 14 21 28

E)

35

) ACOS 2 0 -

1 5 -

*77>

tt H r : : <D __ J => 1 0 -<u w\ l _ f r 8 À L / T (O * A ' . b \ / .

5 - Y Y y I / / 1 /

^&—-**> 0 ——w^—! 1 1 1 1

14 21 28 Days post-SCI

35

-& open-field ♦ treadmill

D)

7 14 21 28 Days post-SCI

Figure 3

94

A) BBB 25

1

20-

W

! « . h-—3 Gr 0 C / / : &

1 f l-

/J'SL 05-

00 -

T * * ' ^

B) AOB

14 21 28 35

25

20

15

10-

05-

oo

- ô " male - - female

• [A

•

r T /

itff

7 14 21 28 35

D) MFS

7 14 21 28 Days post-SCI

35

2 5 n

20-

1 5 -

10

05

00 7 14 21 28

Days post-SCI 35

Figures 4

95

Open field Treadmill

Methods slope r2 slope r2

BBB 0.068 0.96 0.027 0.96 AOB 0.069 0.85 0.041 0.97 BMS 0.050 0.79 0.037 0.85 MFS 0.050 0.76 0.033 0.86 ACOS 0.278 0.73 0.463 0.99

Table 1

96

Methods

Days post spinal cord injury

Methods 3 7 14 21 28 35

BBB 0.04 ±0.04 0.30 ± 0.09 0.75 ±0.11 1.66 ±0.25 1.81 ±0.20 1.76 ±0.27 AOB 0.05 ±0.05 0.43 ±0.11 0.86 ±0.10 2.13 ±0.30 1.70 ±0.24 2.00 ± 0.29 BMS 0.02 i 0.02 0.36 ±0.10 1.06±0.17 1.70 ±0.22 1.54±0.15 1.46±0.15

o MFS 0.05 ±0.05 0.69 ±0.19 1.24 ±0.15 1.80 ±0.08 1.83 ±0.08 1.80 ±0.11 (D NLM 0.01 ±0.01 0.64 ± 0.27 1.09 ±0.31 6.60 ±1.29 3.84 ± 0.76 6.10 ±1.44 CD Incidence NLM 1/21 10/21 17/21 20/20 20/20 19/20 O LM 0.00 ± 0.00 0.01 ±0.01 0.02 ± 0.02 1.00 ±0.46 0.33 ± 0.20 0.91 ±0.34

Incidence LM 0/21 1/21 12/21 10/20 10/20 12/20 Amplitude 0.05 ±0.05 0.48 ±0.11 0.81 ± 0.09 1.00 ±0.00 1.00 ±0.00 1.00 ±0.07 ACOS 0.01 ± 0.01 0.66 ± 0.29 1.17 ±0.33 8.70 ±1.06 4.69 ±1.06 8.97 ± 2.78

BBB 0.29 ±0.12 0.51 ±0.12 0.91 ±0.14 1.14±0.18 1.35 ±0.20 1.48 ±0.23 AOB 0.36 ±0.13 0.61 ±0.12 1.00 ±0.14 1.27 ±0.17 1.55 ±0.35 1.65 ±0.28 BMS 0.26 ±0.11 0.39 ±0.13 0.90 ±0.16 0.97 ±0.13 0.99 ±0.13 1.15±0.13 MFS 0.48 ±0.17 0.78 ±0.16 1.28 ±0.17 1.38 ±0.12 1.53 ±0.15 1.58 ±0.13

1 NLM 0.17 ±0.07 0.64 ±0.19 3.32 ±1.00 6.51 ±1.38 5.23 ± 0.96 11.58±2.15 03 Incidence NLM 7/21 12/18 16/20 19/20 18/20 19/20 r- LM 0.00 ± 0.00 0.01 ± 0.01 0.31 ± 0.20 0.61 ± 0.26 3.00 ±1.59 1.34 ±0.56

Incidence LM 0/21 1/18 5/20 13/20 9/20 9/20 Amplitude 0.31 ±0.10 0.61 ±0.11 0.80 ± 0.09 0.95 ± 0.05 0.90 ± 0.07 0.95 ± 0.05 ACOS 0.17 ±0.07 0.66 ± 0.20 3.94 ±1.29 7.73 ±1.49 11.23 ±3.69 14.25 ±2.93

Table 2

97

Methods

Days post spinal cord injury

Methods 3 7 14 21 28 35

BBB 0.25 ±0.12 0.43 ±0.13 0.75 ±0.14 1.40 ±0.20 1.58±0.18 1.58±0.19 AOB 0.28 ±0.12 0.47 ±0.12 0.92 ±0.15 1.39 ±0.18 1.64 ±0.30 2.06 ± 0.31 BMS 0.18 ±0.09 0.41 ±0.13 0.96 i 0.20 1.40 ±0.21 1.29±0.11 1.29 ±0.13 MFS 0.33 ±0.15 0.74 ± 0.20 1.16±0.19 1.61 ±0.10 1.69 ±0.11 1.78 ±0.09

<D NLM 0.13 ±0.07 0.90 ±0.31 3.40 ±1.07 7.10 ±1.69 4.56 ±0.91 7.98 ±1.88 ro Incidence NLM 5/20 8/17 14/19 18/18 18/18 18/18

LM 0.00 ±0.00 0.01 ± 0.01 0.33 ± 0.21 0.93 ± 0.48 1.81 ±1.17 1.49 ±0.66 Incidence LM 0/20 1/17 5/19 10/18 10/18 9/18 Amplitude 0.23 ± 0.09 0.53 ±0.12 0.76 ±0.11 1.00 ±0.00 1.00 ±0.00 1.00 ±0.00 ACOS 0.13 ±0.07 0.93 ± 0.33 4.10±1.36 8.96 ±2.16 8.18 ±2.90 10.95 ±3.10

BBB 0.08 ± 0.04 0.38 ± 0.09 0.90 ±0.11 1.40 ±0.24 1.58 ±0.22 1.65 ±0.28 AOB 0.14 ±0.07 0.55 ±0.11 0.93 ± 0.09 1.95 ±0.30 1.61 ±0.30 1.64 ±0.27 BMS 0.11 ±0.07 0.34 ±0.10 1.00 ±0.13 1.28 ±0.19 1.24 ±0.17 1.32 ±0.15 MFS 0.20 ±0.12 0.73 ±0.16 1.34 ±0.14 1.57 ±0.12 1.66 ±0.13 1.61 ±0.14

CD

ro NLM 0.06 ± 0.03 0.44 ±0.16 1.12 ±0.26 6.20 ±1.01 4.65 ± 0.85 9.64 ±1.93 E <D

Incidence NLM 3/22 13/22 19/22 21/22 20/22 20/22 Ll_ LM 0.0010.00 0.01 10.01 0.02 ± 0.02 0.70 ±0.28 1.57 ±1.16 0.68 ± 0.23

Incidence LM 0/22 1/22 2/22 13/22 10/22 12/22 Amplitude 0.14 ±0.07 0.55 ±0.10 0.84 ± 0.08 0.95 ± 0.05 0.91 ±0.06 0.95 ± 0.08 ACOS 0.06 ± 0.03 0.45 + 0.17 1.16 ±0.27 7.61 ±1.41 7.78 ±2.74 12.15 ±2.76

Table 3

98

C H A P I T R E V I - R O L E O F S P I N A L 5 - H T 2 R E C E P T O R S U B T Y P E S I N Q U I P A Z I N E -

INDUCED HINDLIMB MOVEMENTS AFTER A LOW-THORACIC SPINAL CORD

TRANSECTION.

Ce chapitre avait pour but d'étudier la contribution des récepteurs 5-HT2 dans la

genèse de mouvements locomoteurs. Nous y avons montré que la densité de l'ARNm des

récepteurs 5-HT2A augmentait dans la zone intermédiaire latérale de la moelle épinière

jusqu'à 1 semaine après une spinalisation. Nous avons également montré que

l'administration de quipazine, agoniste 5-HT2 à large spectre, générait des mouvements

rythmiques de type locomoteur. Ces mouvements dépendaient essentiellement des

récepteur 5-HT2A- Cet article a été publié dans European Journal of Neuroscience, 2008, 28

(11): 2231-42.

Abstract

A role of serotonin receptors (5-HTRs) in spinal rhythmogenesis has been proposed

several years ago based mainly upon data showing that bath-applied 5-HT could elicit

locomotor-like rhythms in in vitro isolated spinal cord preparations. Such a role was

partially confirmed in vivo after revealing that systemically administered 5-HTR2 agonists,

such as quipazine, could induce some locomotor-like movements (LM) in completely

spinal cord-transected (Tx) rodents. However, given the limited binding selectivity of

currently available 5-HTR2 agonists, it has remained difficult to determine clearly if one

receptor subtype is specifically associated with LM induction. In situ hybridization, data

using tissues from L1-L2 spinal cord segments, where critical locomotor network elements

have been identified in mice, revealed greater 5-HTR2A mRNA levels in low-thoracic Tx

than non-Tx animals. This expression level remained elevated for several days, specifically

in the lateral intermediate zone, where peak values were detected at one week post-Tx and

return to normal at 3 weeks post-Tx. Behavioural and kinematic analyses revealed

quipazine-induced LM in one-week-Tx mice either non-pretreated or pretreated with

99

selective 5-HTR2B and/or 5-HTR2c antagonists. In contrast, LM completely failed to be

induced by quipazine in animals pretreated with selective 5-HTR2A antagonists. Altogether,

these results provide strong evidence suggesting that 5-HTR2A are specifically associated

with spinal locomotor network activation and LM generation induced by quipazine in Tx

animals. These findings may contribute to design drug treatments aimed at promoting

locomotor function recovery in chronic spinal cord-injured patients.

Introduction

Clear evidence suggests that locomotion is controlled by networks located in the

spinal cord referred to as central pattern generators or CPGs (Grillner and Zangger, 1979)

that receive descending serotonergic inputs from the raphe nucleus and parapyramidal

region (Ballion et al., 2002; Jordan et a i , 2008). Moreover, serotonin (5-HT) and

corresponding precursors or agonists were found to trigger and modulate spinal network-

mediated locomotor-like activity (Barbeau and Rossignol, 1990; Cazalet et al., 1992;

Cowley and Schmidt, 1994; Kiehn and Kjaerulff, 1996).

Additional experiments have provided data suggesting that only a subset of spinal 5-

HT receptors (5-HTRs) may be specifically involved in 5-HT-induced locomotor

rhythmogenesis. For instance, in isolated spinal cord preparations from mice, 5-HTR2

agonists and antagonists were found respectively to either evoke or disrupt sustained fictive

locomotor rhythms (Madriaga et a l , 2004). Electrical or pharmacological stimulation of the

parapyramidal region was also shown to induce locomotor-like activity in neonatal rat

preparations that is blocked by 5-HTR2A antagonists (Liu and Jordan, 2005). In completely

Tx rodent models, administration of 5-HTR2 agonists such as quipazine or DOI was found

to induce some locomotor-like movements (LM) and to promote locomotor function

recovery (McEwen et a l , 1997; Feraboli-Lohnherr et a i , 1999; Kim et a l , 1999; Antri et

a l , 2002; Gerasimenko et a l , 2007; Guertin, 2004b; Fong et a l , 2005).

100

Although all subtypes of 5-HTR2 (i.e., 2A, 2B and 2C) are present in the spinal cord,

preliminary evidence suggests a subtype-specific distribution. For instance, greater 5-

HTR2A mRNA levels were reported in motor areas (lamina DC) of the spinal cord (Fonseca

et a l , 2001) whereas 5-HTR2C transcripts are widely distributed throughout the gray matter

(Pompeiano et al., 1994; Fonseca et a i , 2001). However, the distribution and pattern of

expression of 5-HTR2B in the adult spinal cord remain unclear.

Altogether, clear evidence demonstrates that spinal 5-HTR2 are involved in the

control of locomotion. However, it remains to determine whether one subtype of 5-HTR2 is

specifically associated with LM induction, although in vitro studies suggest a specific role

of 5-HTR2A in CPG activation (Madriaga et al., 2004; Liu and Jordan, 2005). Thus, this

study aimed at determining whether 5-HTR2A is associated with locomotor network

activation and LM induction in adult Tx mice. Although, the detailed localization and

spatial distribution of CPG neurons in most mammalian species remain unclear, selective

lesions have established the existence of critical CPG elements in L1-L2 segments in mice

(Nishimaru et a l , 2000). Since the expression pattern of some 5-HT receptors has been

reported to change post-Tx (Giroux et a l , 1999), we characterized spatiotemporal changes

of 5-HTR2A expression post-trauma, by measuring its mRNA levels in L1-L2 at fixed

delays post-Tx. We also quantitatively assessed and compared the acute effects of

quipazine in 5-HTR2A, 5-HTR2B, and/or 5-HTR2c antagonist-pretreated Tx animals in order

to pharmacologically dissect the contribution of each receptor subtype to CPG activation

and LM induction in vivo. Part of this work has been presented in abstract form (Ung et a l ,

2005).

Methods

Animal model

All experimental procedures were conducted in accordance with the Canadian

Council for Animal Care guidelines and were accepted by the Laval University Animal

101

Care and Use Committee. Male CD1 mice (N=10S) (Charles River Canada, St-Constant,

Quebec) initially weighing 30-35 g (approximately eight week-old) were used for this

study. Pre-operative cares included subcutaneous injections of an analgesic (0.1 mg/kg,

buprenorphine), an antibiotic (5 mg/kg, enrofloxacin) and lactate-Ringer's solution (1 ml).

All surgical procedures were performed under aseptic conditions. Mice were anesthetised

with 2.5% isoflurane. A small incision was made on their back in order to expose some

thoracic segments. The spinal cord was then completely transected intervertebrally with

microscissors inserted between the 9th and 10th thoracic vertebrae (Th9/10). To ensure that

complete Tx was achieved, the inner vertebral walls were explored and entirely scraped

several times with small scissor tips. The opened skin area was sutured and animals were

placed for a few hours on heating pads. Mice were left in their cage with food and water ad

libitum for 7 days to allow sufficient rest and recovery from surgery before testing. Post­

operative cares provided for 4 days included subcutaneous injections of buprenorphine (0.2

mg/kg/day), enrofloxacin (5 mg/kg/day) and lactate-Ringer's solution ( 2 x 1 ml/day).

Bladders were manually expressed twice daily for 4-7 days post-Tx. Complete low-thoracic

Tx was confirmed by 1 ) initial full paralysis of the hindlimbs, 2) post-mortem visual and

microscopic examination of the spinal cord lesion, and 3) histological assessment of

coronal or midsagittal spinal cord sections stained with luxol fast blue/cresyl violet for

myelinated axons and Nissl substance, respectively.

In situ hybridization

Immediately after sacrifice using CO2, the LI and L2 segments were taken from non-

Tx (N =7) and Tx animals either at 3 hours (N = 5), 1 day (N = 6), 3 days (N = 5), 7 days

(N = 5), 14 days (N = 5), or 28 days (N = 5) post-Tx. Collected tissues were immediately

put on dry ice for rapid freezing and then stored at -80°C. They were thawed at room

temperature and post-fixed in 0.1 M PBS solution containing 4% PFA for 15 minutes.

Cryostat-prepared sections (12 pm-thick) were mounted on Snowcoat X-TRA slides

(Surgipath, Winnipeg, MA, Canada) and stored at -80°C until use. A specific [35S]UTP-

labelled complementary RNA (cRNA) probe was used to assess tissue mRNA levels

(Pritchett et ah, 1988; Cyr et al., 1998). The 5-HTR2A probe was inserted into a pSP64

plasmid at the Sal 1/EcoRI sites of the polylinker as previously described by Ruat and

102

colleagues (Ruât et al., 1993). The single-stranded riboprobe, complementary to the

mRNA, was synthesized and labelled with the Promega riboprobes kit (Promega, Madison,

WI, USA), [35S]UTP (Perkin Elmer Inc, Canada) and SP6 RNA polymerase. In situ

hybridization (ISH) was performed as previously described elsewhere (Beaudry et a l ,

2000; Langlois et a l , 2001). Briefly, sections were fixed for 10 minutes in PFA solution

(0.1 M PBS containing 4% PFA, m/v) and rinsed twice in 0.1 M PBS for 5 minutes. They

were then incubated in triethanolamine (TEA 100 mM, pH: 8.0) and then 10 minutes

acetylation in 0.25% (v/v) acetic anhydride (TEA 100 mM). Sections were rinsed in

standard saline citrate (SSC) (NaCl 300 mM, sodium citrate 30 mM) and dehydrated using

increasing concentrations of ethanol. The [35S]UTP-labelled riboprobe was added at a

concentration of 2xl06cpm/100 ul in the hybridization solution (Denhart's solution, dextran

sulphate and 50% (v/v) deionized formamide) and heated at 65°C for 5 minutes. Each slide

was covered with 100 pi of hybridization solution and cover-slipped. The hybridization

took place overnight at 58°C. Slides were soaked in SSC (NaCl 600 mM, sodium citrate 30

mM) for 30 minutes, coverslips were removed, and slides were washed four times in SSC.

Samples were treated with Rnase A (2 mg/100 ml) for 30 minutes at 37°C in incubation

buffer (NaCL 5 M, Tris-Hcl 1 M, EDTA 500 mM, H20). A series of SSC baths, one of

which being at 55°C for 30 minutes were then performed. Subsequently, tissues were

dehydrated in ethanol, dried, and apposed to a Biomax MR film (Kodak). Autoradiograms

were developed 120 hours later. Levels of radiolabelling were quantified by computerized

densitometry. Digitized spinal cord images were captured with a CCD camera model XC-

77 (Sony) equipped with a 60 mm f/2.8D magnification lens (Nikon). Computerized

densitometry of mRNA spots was achieved using Scion Image software (B 4.0.2). Optical

densities of autoradiograms were translated in nanocuries per gram of tissue (nCi/g) using l4C radioactivity standards (ARC 146-14C standards, American Radiolabeled Chemicals

Inc., St-Louis, MO, USA). Radiolabelling levels were measured in medial and lateral

portions of the ventral horn, intermediate zone and dorsal horn areas (see diagram,

Fig.2Av/«). Average levels in each area were calculated bilaterally and from two adjacent

sections from the same animal. Background intensity was subtracted from every

measurement.

103

5-HTR2 ligands

Quipazine [2-(l-Piperazinyl) quinoline dimaleate] (Tocris, Ellisville, MO, USA), a

non-selective 5-HTR2 agonist (often referred to as a 5-HTR2A/2C agonist, Barnes and Sharp,

1999) was dissolved in sterile water. Quipazine was chosen over other 5-HTR2 agonists

such as DOI, m-CPP or TFMPP for its clear LM-inducing effects in untrained and non

sensory-stimulated Tx mice (Landry and Guertin, 2004, unpublished data for DOI). The

selective 5-HTR2A antagonist MDL-100,907 [fl-(+)-»-(2,3-dimethoxyphenyl)-l-[2-(4-

fluorophenylethyl)]-4-piperidine-methanol] (Sanofi-Aventis, Paris, France) was dissolved

in sterile water containing 0.005% acetic acid. The selective 5-HTR2B antagonist SB204741

[N-(l-Methyl-lH-5-indolyl)-N'-(3-methyl-5-isothiazolyl)urea] (Sigma-Aldrich, St. Louis,

MO, USA) was dissolved in 0.25% (v/v) DMSO and sterile water. The selective 5-HTR2C

antagonist SB242084 [6-chloro-5-methyl-l-[[2-(2-methylpyrid-3-yloxy) pyrid-5-yl]

carbamoyl] indoline dihydrochloride] (Sigma-Aldrich, St. Louis, MO, USA) was dissolved

in sterile water.

Corresponding acetic acid and DMSO vehicle solutions were tested alone and in

combination with quipazine on 20 mice. Administration alone had no effect and pre­

treatrnent with these vehicle controls did not alter quipazine-induced movements (Fig.l).

All drugs were administered intraperitoneally (i.p.). One (1) mg/kg quipazine was used for

all tests based upon preliminary data (dose-response data with 0.5 - 4 mg/kg) revealing this

dose to be threshold for inducing LM (e.g., 0.7 mg/kg is subthreshold for LM induction in

air-stepping-condition, Guertin, 2004a). Doses chosen for antagonists were determined

based upon the potency and respective binding affinity for the different 5-HTR2 subtypes

(pKi ± SEM of 5-HTR2A, 5-HTR2B, and 5-HTR2C, respectively, Knight et al. 2004). We

used 0.2 mg/kg of MDL-100,907 (8.73 ± 0.20, 5.99 ± 0.06, 7.52 ± 0.13), 10.0 mg/kg of

SB204741 (<5.00, 6.90 ± 0.27, 5.56 ± 0.07) and 2.0 mg/kg of SB242084 (6.07 ±0.18, 6.84

±0.28, 8.15 ±0.10).

Pharmacological testing

104

All tests were performed at 7 days post-Tx to allow enough recovery time from

surgery. Mice were either treated with quipazine alone or in combination (as a pre­

treatrnent) with one or several 5-HTR2 antagonists. Mice (N = 50) were thus divided into

five groups (10 mice/group) which were either treated with 1) quipazine alone, 2) MDL-

100,907-pretreated + quipazine, 3) SB204741-pretreated + quipazine, 4) SB242084-

pretreated + quipazine or 5) SB204741+SB242084-pretreated + quipazine. The first group

was used to assess separately the acute effects of quipazine on LM induction. The other

four groups were used to pharmacologically dissect the contribution of 5-HTR2A, 5-HTR2B,

5-HTR2C or 5-HTR2B/2C to quipazine-induced LM, respectively. Antagonists were

administered 15 minutes prior to quipazine injection.

Behavioural and kinematic analyses

We used two complementary methods to assess hindlimb movements. The first one is

based upon an on-line (live) assessment of LM described in detail elsewhere (Guertin,

2005). In brief, one LM was defined as a flexion followed by an extension or vice versa

(involving one or several articulations) occurring in both hindlimbs in alternation (Guertin,

2005; Lapointe et a l , 2006; Ung et a l , 2007). This approach allows an assessment of

bilaterally-alternated movements. LM frequency (number/minute) was assessed prior to

drug injection and 15 minutes after quipazine administration successively in two conditions

- air-stepping and treadmill (4 minutes of observation in each condition). In air-stepping-

condition, each animal was gently placed in a plastic cylinder with open extremities for the

hindlimbs to move freely. The cylinder was hold above ground-level to ensure that both

hindlimbs were completely suspended (see Fig.3Ai«-/v for illustration). The treadmill-

condition consisted of a custom-made, 10-track-adjustable-speed treadmill set to move at 8-

10 cm/sec. A harness placed around the hip and torso was used to maintain the animals in

front of the camera but not to provide body-weight support assistance (see Fig.3A.i-» for

illustration). Stimuli such as tail pinching were not provided in this study to avoid

unnecessary variations and unspecific (reflex-mediated) drug-induced effects. Kinematic

analyses were also performed as a complementary method to further characterize hindlimb

movement in representative cases. Animals were filmed using a digital video camera (Sony

105

DCR110, shutter speed: 1/4000; acquisition: 30 frames/sec) placed sideways. Digital

movies were stored on computer for subsequent off-line two-dimensional kinematic

analyses (i.e., angular excursions, stick diagrams, movement amplitude) using MaxTRAQ

and MaxMATE softwares (Innovision System, Columbiaville, MI).

Statistical analyses

Paired Student t-tests were used for comparing LM frequencies prior to versus after

quipazine as well as between air-stepping and treadmill conditions. One-way ANOVAs

followed by LSD post-hoc tests were used to compare drug-related groups. Same analysis

methods were used to compare the spatial and temporal 5-HTR2A expression changes post-

trauma. Values were expressed as mean ± SEM and considered statistically significant

when P < 0.05.

Results

5-HTR2A labelling

We decided to undertake in situ hybridization experiments in order to specifically

determine the pattern of mRNA expression in L1-L2 segments of the spinal cord. This was

done also to assess possible spatiotemporal changes of this pattern following a spinal cord

transection. 5-HTR2A transcripts were found essentially throughout the gray matter in non-

Tx mice (N = 7) (Fig.2A/)- In non-Tx animals, higher mRNA expression levels were found

in the ventral horn medially (44.3 ± 7.5 nCi/g) and laterally (56.4 ± 4.7 nCi/g) than in the

intermediate zone (medially and laterally) or dorsal horn (< 22.0 nCi/g,) (Fig2Z?/-v).

However, mRNA expression levels were found to largely augment after Tx specifically in

the lateral intermediate zone where a significant (P = 0.004) 2.5-fold increase was

measured at 3 hours compared with non-Tx mice (Fig.2fla). This up-regulated mRNA level

was highest at 7 days and remained elevated and significantly (P < 0.05) different from

non-Tx mice at all time points except at 28 days (P = 0.131). Representative pictograms

106

also show elevated mRNA levels in the lateral ventral horn that peaked at 7 days (Fig.2Av,

IBiv). No significant change post-Tx was found in the medial ventral horn, medial

intermediate zone and dorsal horn. In brief, mRNA expression data revealed detectable 5-

HTR2A transcripts throughout the gray matter with relatively high levels specifically in both

the ventral horn (both medially and laterally) and intermediate zone (laterally) areas.

However, following Tx, significantly increased mRNA expression levels were found only

in the lateral intermediate zone.

Effects of quipazine on hindlimb movement induction

Quipazine has already been shown to induce some LM (see Methods for a definition)

within 15 minutes of injection in low-thoracic Tx mice (Guertin, 2004a,b; Landry and

Guertin, 2004). Comparable results were found in this study following acute administration

of 1 mg/kg quipazine in 7-day-Tx mice examined in both treadmill and air-stepping

conditions. While no LM was found prior to injection (Fig.3/,/n), quipazine induced some

movements in treadmill and air-stepping conditions (Fig.3A//,/v). Video images from a

representative case clearly illustrate that quipazine-induced LM in treadmill-condition were

constituted of relatively small amplitude flexion-extension movements with no weight

bearing or plantar foot placement (Fig.3Aii). As shown below in Fig.3A/v, comparable

small amplitude rhythmic movements were found in air-stepping-condition (see also Table

1 for averaged LM amplitude data). A detailed analysis revealed several similarities and a

few differences between the induced-movements assessed in treadmill versus air-stepping

conditions. For instance, the phase-relationship was comparable in both conditions as

revealed by the angular excursion profile for a normalized flexion-extension cycle (Fig.3Z?//

versus Fig.3fi/v, middle and bottom traces for the knee and ankle, respectively). Note that

no movement was found at the hip level (Fig.3fi//,/v, upper traces). In turn, absolute angular

excursion values were typically different in both conditions, with greater angle excursions

in air-stepping than treadmill. At the knee, angular excursions ranged between 47 (most

extended position) and 40 degrees (most flexed position) in treadmill-condition and

between 112 (most extended position) and 85 degrees (most flexed position) in air-

stepping-condition (Fig.3fi//,/v, middle traces). Comparable differences were found at the

107

ankle level where values ranged between 112 and 63 degrees and between 121 and 62

degrees in treadmill and air-stepping conditions, respectively (Fig.3Bii,iv, bottom traces).

Differences between both conditions were found also with LM frequencies. Averaged data

from all animals of this group (N = 10) revealed that quipazine induced more (3-fold-

greater) LM in treadmill (7.5 ± 1.7 LM/min) than air-stepping (2.9 ± 1.6 LM/min) (P =

0.028, Fig.5A). This provides clear evidence that treadmill-related sensory inputs (e.g.,

hindlimb and foot skin rubbing against the moving treadmill belt) contribute to drug-

induced CPG-mediated effects. In contrast, averaged LM amplitude values (displacement

of the toes, in mm) for that group revealed no significant (P > 0.05) change between air-

stepping and treadmill conditions (12.6 ± 0.6 mm versus 11.9 ± 0.4 mm, respectively, see

Table 1).

Effects of quipazine in 5-HTR2A, 5-HTR2B and/or 5-//77?2c antagonist-pretreated animals

In clear contrast with the above results, quipazine completely failed to induce LM in

Tx mice pretreated 15 minutes earlier with MDL-100,907, a selective 5-HTR2A antagonist.

This is illustrated with video images and kinematic analyses from a typical animal

pretreated with MDL-100,907 (0.2 mg/kg, i.p.) that displayed absolutely no LM prior to

(Fig.4Ai,iii) or after quipazine (Fig.4A//,/v) administration both in treadmill and air-

stepping conditions. Note also the corresponding lack of angular excursion change post-

versus pre-quipazine administration (Fig. 4Bii,iv versus 4Bi,iU). Similar results were found

in all animals of that group (/V=10) that were pretreated with MDL-100,907 (Fig.55). In

turn, in a group of 10 animals pretreated instead with SB204741 (10 mg/kg, selective 5-

HTR2B antagonist), quipazine induced a few movements. Indeed, 1.9 ± 0.5 and 4.0 ± 0.9

LM/min were detected in air-stepping and treadmill conditions, respectively (Fig.5C). That

difference between air-stepping and treadmill was significant (P < 0.05). However, LM

amplitude was not significantly (P > 0.05) different between all groups (see Table 1). Some

LM were also induced in animals (Af = 10) pretreated with SB242084, a selective 5-HTR2C

antagonist. As in the quipazine-only treated group (Fig.5A), animals pretreated with

SB242084 (Fig.5D) displayed significantly more movements in treadmill (8.9 ± 3.0

LM/min) than air-stepping conditions (3.3 ±1.1 LM/min). However, LM amplitude did not

108

significantly (P > 0.05) differ between both conditions (Table 1). We also tested quipazine

in a last group (N = 10) pretreated with both SB204741 and SB242084 to examine whether

blocking all subtypes, but the 5-HTR2A, could produce similar effects (i.e., a complete

block) as those found in 5-HTR2A antagonist-pretreated animals. Quipazine remained

capable of inducing some movements in 5-HTR2B/2C antagonists-pretreated animals

(Fig.5A, air-stepping, 2.2 ± 0.7 LM/min; treadmill conditions, 4.2 ± 0.7 LM/min)

suggesting that among the 5-HTR2 subtypes, only the 5-HTR2A was critical to locomotor

network activity and LM generation induced by quipazine although some role may also

exist for 5-HTR2B. Indeed, significantly lower LM values were reported to be induced by

quipazine in SB204741-pretreated-mice than non-pretreated animals (see Figs. 5 and 6).

Comparisons between groups

In air-stepping-condition, a significant (P = 0.021) difference was found when

comparing all groups (Fig.6A). Compared to control (2.9 ± 1.6 LM/min with quipazine),

only the MDL-100,907-pretreated (no LM/min) was significantly different (P < 0.05).

Significant differences were found also between the MDL-100,907-pretreated and

SB204741-pretreated, SB242084-pretreated or SB204741+SB242084-pretreated groups

(Table 2, upper panel). In treadmill-condition statistical analyses also showed significant

difference (P = 0.006) between groups (Fig.6fi). Multiple paired-comparisons revealed that

both the 5-HTR2B antagonist-pretreated and the 5-HTR2B + 5-HTR2C antagonist-pretreated

groups displayed significantly (P < 0.05) lower scores (< 4.2 LM/min) compared to control

(7.5 ± 1.7 LM/min, see Fig.6 and Table 2, lower panel). Multiple paired post-hoc analyses

clearly show no difference specifically between the control (quipazine only) and 5-HTR2C-

pretreated groups, both in air-stepping and treadmill conditions. In turn, this, suggests that

5-HTR2C did not significantly contribute to the mediation of quipazine-induced effects.

Discussion

109

These results are the first to show spatiotemporal expression of 5-HTR2A mRNA in

the murine spinal cord. In the intermediate zone of L1-L2 segments, a significant increase

of that expression was found within a few days post-Tx (Th9/10 level). Quantitative

movement analyses combined with pharmacological tests using selective 5-HTR2A, 5-

HTR2B or 5-HTR2C antagonists in pre-treatment clearly established that 5-HTR2A is critical

to locomotor network activation since locomotor-like movements reported as LM failed to

be induced by quipazine only in 5-HTR2A antagonist-pretreated animals. This said, some

role was also played by 5-HTR2B as revealed by reduced (but not blocked) quipazine-

induced LM in 5-HT2B antagonist-pretreated mice.

Distribution of 5-HTR2 in motor areas of the spinal cord

All three subtypes of 5-HTR2 have been identified in rat, cat, monkey and human

spinal cords, suggesting a philogenetically well-conserved expression in mammals.

Specifically, spinal 5-HTR2A transcripts were identified mainly in the ventral horn area of

the gray matter (Pompeiano et a l , 1994; Fonseca et a l , 2001). Using

immunohistochemistry, clear 5-HTR2A expression was shown indeed near several pools of

motoneurons (Fuller et a l , 2005) with greater levels near hindlimb extensors than flexors in

rats (in lumbar cord, Vult von Steyern and Lomo, 2005). However, some expression was

found also elsewhere such as in the intermediate zone and dorsal horn areas (Maeshima et

al., 1998; Cornea-Hébert et a l , 1999; Doly et a l , 2004) as reported here also in mice (see

Fig.2).

In contrast, 5-HTR2B expression was identified near neuroepithelial cells in the

developing brain and spinal cord (embryos, Lauder et a l , 2000). In the CNS it is mainly

known for its role in migraine (Hamel, 1999). In turn, 5-HTR2C has been associated mainly

with chronic pain and is widely expressed throughout the spinal cord gray matter in most

mammalian species (Jeong et a l , 2004; Obata et a l , 2004, Molineaux et al., 1989; Mengod

et a l , 1990; Pompeiano et a l , 1994; Fonseca et a l , 2001), except in cats, suggesting

perhaps the existence of some species-specific differences (Helton et a l , 1994).

110

This study is also the first to report 5-HTR2A expression in L1-L2 which is of special

interest since these segments were shown using selective lesions to possess critical CPG

elements (mice, Nishimaru et a l , 2000). This specific expression of 5-HTR2A in motor and

pre-motor gray matter areas (see Fig.2) together with the lack of specific 5-HTR2B and 5-

HTR2c expression in the spinal cord (i.e., 5-HTR2B only near neuroepithelium cells; 5-

HTR2C widely expressed throughout the gray matter, Lauder et a l , 2000; Pompeiano et a l ,

1994) provide indirect evidence of a specific role for 5-HTR2A in the control of locomotion.

Supporting this line of evidence, the expression of 5-HTR2A in other segments of the cord

has been shown to be involved in the control of other motor functions including

micturition, erection and respiration (e.g., Xu et a l , 2007; Basura et a l , 2001). However,

although not directly assessed in this study, the present pharmacological results in 5-HT2B

antagonist-pretreated mice suggest the existence of some 5-HTR2B in the adult murine

spinal cord.

Spatiotemporal change of5-HTR2A expression sublesionally

The present results reveal a significant increase of 5-HTR2A mRNA levels in L1-L2

as early as 3 hours post-Tx (Th9/10). This upregulation (P < 0.05) was found to last for

several weeks specifically in the lateral intermediate zone of the gray matter. Although

apparently elevated also in the lateral ventral horn area, levels remained non-significantly

(P = 0.07) changed as in other laminar areas. This is in line with data from Fuller and

colleagues who showed by immunoblots a significant increase of 5-HTR2A expression in

sublesional segments (i.e., in C5 motor and pre-motor gray matter areas) within one week

following a C2 hemisection (Fuller et a l , 2005).

The reason why 5-HTR2A expression is upregulated post-Tx remains unclear.

Nonetheless, comparable upregulations of other transmembranal receptors have also been

found. For instance, in Tx cats, autoradiographic labelling using [3H] 8-hydroxy-

dipropylaminotetralin (5-HTRIA/7 agonist) found in laminae I-IV and X prior to trauma was

reported to largely increase only in laminae II, III and X after a low-thoracic transection

(Giroux et a l , 1999). Furthermore, these increases were transient, and returned to control

I l l

values by 30 days, similar to what was observed in murine intermediate zone laterally and

ventral horn laterally (although non-significant in the latter case). It remains unclear also

whether any of these changes in expression may contribute to the development of

spontaneous recovery (occasional small hindlimb movements, Guertin, 2005) or

pathological conditions (e.g., spasticity, clonus or restless leg syndrome, Benz et a l , 2005;

Lee e ta l , 1996).

Evidence of CPG-mediated effects

It has been clearly shown in several animal models that 5-HT and some

corresponding agonists can either activate or modulate spinal cord-mediated locomotor

activity. For instance, 5-HT used separately or in combination with other ligands was found

to induce locomotor rhythms in in vitro isolated spinal cord preparations (Cazalet et a l ,

1992, Cowley and Schmidt, 1994; Kiehn and Kjaerulff, 1996; Cina and Hochman, 2000;

Nishimaru et a l , 2000; Whelan et a l , 2000; Madriaga et a l , 2004). In in vivo models, its

precursor 5-HTP was reported to trigger locomotor-like activity in motor nerves of

anaesthetized and curarized Tx rabbits (Viala and Buser, 1971). In contrast, experiments in

Tx cats have shown that 5-HTP or 5-HT agonists (such as quipazine) can only modulate

hindlimb muscular activity (Barbeau and Rossignol, 1990; 1991). In turn, 5-HTR2 agonists

such as quipazine administered separately or in combination with noradrenergic

compounds were clearly shown to trigger locomotor-like movements in completely

suspended Tx animals in air-stepping-condition (rat pups, McEwen et a l , 1997; adult mice,

Guertin, 2004a) or treadmill-condition (Guertin, 2004b; Landry and Guertin, 2004) or full

weight-bearing stepping assisted by tail stimulation (Feraboli-Lohnherr et a l , 1999),

robotic devices (Fong et a l , 2006) or neural transplants (Kim et a l , 1999). In other

experiments where quipazine was chronically administered, the effects of regular treadmill

training on locomotor recovery were also enhanced (Feraboli-Lohnherr et a l , 1999; Antri

et a l , 2002; Fong et a l , 2005). Altogether, our data and those described above clearly

demonstrate that rhythmic locomotor-like activity triggered or promoted by 5-HT2 ligands

do not depend upon brain-mediated commands over CPG neurons and hindlimb

motoneurons (i.e. since found in completely Tx animals or isolated spinal cords).

112

Moreover, results from isolated spinal cords or obtained in air-stepping-condition (no

physical contact with the ground) suggest also that sensory inputs are not critical for 5-

HTR2 agonist-induced activation of the CPG.

This said, a contribution of sensory inputs to quipazine-induced effects on hindlimb

movement generation can not be excluded. For instance, ligands from other families such

as clonidine (alpha-2 agonist) were clearly shown to enhance locomotor activity and

recovery by modulating sensory afferent-evoked reflexes in spinal cord-injured patients

(Remy-Neris et a l , 1999; 2003). Afferent inputs from the hip were also shown to facilitate

flexion and to entrain CPG-mediated rhythmic movements in Tx cats (Andersson and

Grillner, 1983). Muscle proprioceptor (e.g., Ia- and Ib-afferents) activation was also found

in decerebrate cats to reset CPG rhythms and to largely enhance extensor activity (Guertin

et a l , 1995; Whelan et a l , 1995). Here, we found greater effects in treadmill than air-

stepping conditions (e.g., see Fig.5A) further suggesting some contribution of sensory

inputs (e.g., from the foot or the leg that are typically rubbing against the moving treadmill

belt) to LM generation.

Specific 5-HTR2A-tnediated effects

A straightforward demonstration of a role for 5-HTR2A in CPG activation would have

been facilitated if highly selective 5-HTR2A agonists had been developed. Since none are

currently available, we used quipazine, often referred to as a 5-HT2A/2C agonist (Barnes and

Sharp, 1999), as a tool to pharmacologically dissect the contribution of each receptor

subtypes (i.e., 2A, 2B, or 2C) in selective antagonist-pretreated animals. As mentioned

earlier, preliminary evidence from in in vitro isolated spinal cords has suggested a specific

contribution of 5-HTR2A to CPG activation since 5-HT-induced fictive locomotion is

blocked by 5-HTR2A antagonists (Madriaga et a l , 2004). This was supported also by Liu

and Jordan who showed that electrical or pharmacological stimulation of the parapyramidal

region can induce fictive locomotion blocked specifically by 5-HTR2A antagonists and not

by 5-HTR2C antagonists (Liu and Jordan, 2005).

113

The blocking effect induced by the 5HTR2A antagonist MDL100,907 could take place

at different levels. As mentioned above, several studies, including the present one, showed

that 5-HTR2 agonists can trigger LM (McEwen et a l 1997; Feraboli-Lohnherr et al., 1999;

Kim et a l , 1999; Antri et a l , 2002; Guertin, 2004a, b; Landry and Guertin, 2004) and that

some 5HTR2A immunohistochemistry were found in the intermediate zone and around the

central canal (Maeshima et al. 1998; Cornéa-Hébert et al. 1999; Doly et al., 2004). The full

blocking effect found in the 5-HTR2A antagonist-pretreated group may well include

hyperpolarizing actions at the motoneuronal level (i.e., entirely preventing movements from

being displayed such as in this case). This is supported to some extent by data showing 5-

HT-induced depolarization of spinal motoneurons (reviewed in Reckling et al., 2000). In

contrast, actions (hyperpolarization) solely at the motoneuronal level for 5-HT2B and 5-

HT2C antagonists (if any in the latter case) are unlikely since effects on movement

frequency (as shown in Figs.5-6) imply that the pattern or rhythm generator was affected or

modulated.

The present study provides clear in vivo data showing that only 5-HTR2A antagonists

(among the three subtypes) can block quipazine-induced LM in Tx mice. Other 5-HT

agonists such as TFMPP (5-HTR|B/2c) and m-CPP (5-HTR2B/2C) had been found a few

years ago not to induce LM in these animals (Landry and Guertin, 2004), which has led to

suggest that spinal 5-HTR2A are indeed critical for LM generation. Along this line of

evidence, Zhou and colleagues have also shown that a stimulation of the phrenic nerve and

a corresponding recovery of respiratory functions induced by DOI (5-HTR2A/2C agonist) in

C2 hemisected rats can be specifically blocked by 5-HTR2A antagonists and not by 5-

HTR2C antagonists (Zhou et a l , 2001). Even though the 5-HTR2B antagonist failed to

completely suppress quipazine-induced LM (see Fig.6), lower values were found in

SB204741- or SB204471+SB242084-pretreated animals (see also Table 2). The reasons

behind these apparently partial blocking effects (even with 10 mg/kg SB204741) remain

unclear although some in vitro data have reported SB204741 inhibitory effects over 5-HT

facilitation of NMDA-induced depolarization in frog motoneurons (Holohean and

Hackman, 2004).

114

In conclusion, quipazine was found to promote rhythmogenesis and LM generation

although these induced movements were not accompanied of weight-bearing or plantar foot

placement capabilities. This suggests that greater effects may be achieved by

simultaneously activating 5-HTR2A together with other families of receptors as suggested

by in vitro studies where glutamatergic, serotonergic and dopaminergic compounds are

often combined for potent and stable CPG activation (e.g. Jiang et a l , 1999, Whelan et a l ,

2000).

Acknowledgements

This work was supported by the Christopher Reeve Paralysis Foundation (CRPF) and

the Fond de Recherche en Santé du Quebec (FRSQ). We would like to thank Dr. Daniel

Levesque for his contribution to in situ hybridization experiments and to Sanofi-Aventis for

kindly providing MDL-100,907.

Abbreviations

5-HT, serotonin; 5-HTR, serotonin receptor; CPG, central pattern generator; Ctr,

control; ISH, In situ hybridization; i.p., intraperitoneally; LM, locomotor-like movements;

PFA, paraformaldehyde; PBS, phosphate-buffered saline; ROI, region of interest; SSC,

standard saline citrate; TEA, triethanolamine; Tx, spinal cord-transected.

115

References

ANDERSSON, O. & GRILLNER, S. (1983) Peripheral control of the cat's step cycle. H.

Entrainment of the central pattern generators for locomotion by sinusoidal hip

movements during fictive locomotion. Acta. Physiol Scand., 118, 229-239.

ANTRI, M., ORSAL, D. & BARTHE, J.Y. (2002) Locomotor recovery in the chronic spinal rat:

effects of long-term treatment with a 5-HT2 agonist. Eur. J. Neurosci., 16, 467-476.

BALLION, B., BRANCHEREAU, P., CHAPRON, J. & VIALA, D. (2002) Ontogeny of descending

serotonergic innervation and evidence for intraspinal 5-HT neurons in the mouse

spinal cord. Brain Res. Dev. Brain Res., 137, 81-88.

BARBEAU, H. & ROSSIGNOL, S. (1990) The effects of serotonergic drugs on the locomotor

pattern and on cutaneous reflexes of the adult chronic spinal cat. Brain Res., 514,

55-67.

BARBEAU, H. & ROSSIGNOL, S. (1991) Initiation and modulation of the locomotor pattern in

the adult chronic spinal cat by noradrenergic, serotonergic and dopaminergic drugs.

Brain Res., 546, 250-260.

BARNES, N.M. & SHARP, T. (1999) A review of central 5-HT receptors and their function.

Neuropharmacology, 38, 1083-1152.

BASURA, G.J., ZHOU, S.Y., WALKER, P.D. & GOSHGARIAN, H.G. (2001) Distribution of

serotonin 2A and 2C receptor mRNA expression in the cervical ventral horn and

phrenic motoneurons following spinal cord hemisection. Exp. Neurol., 169, 255-

263.

BEAUDRY, G., LANGLOIS, M.C, WEPPE, I., ROUILLARD, C. & LEVESQUE, D. (2000)

Contrasting patterns and cellular specificity of transcriptional regulation of the

nuclear receptor nerve growth factor-inducible B by haloperidol and clozapine in

the rat forebrain. J. Neurochem., 75, 1694-1702.

BENZ, E.N., HORNBY, T.G., BODE, R.K., SCHEIDT, R.A. & SCHMIT, B.D. (2005) A

physiologically based clinical measure for spastic reflexes in spinal cord injury.

Arch. Phys. Med. Rehabil., 86, 52-59.

116

CAZALETS, J.R., SQALLI-HOUSSAINI, Y. & CLARAC, F. (1992) Activation of the central

pattern generators for locomotion by serotonin and excitatory amino acids in

neonatal rat. J. Physiol, 455, 187-204.

CINA,C. & HOCHMAN, S. (2000) Diffuse distribution of sulforhodamine-labeled neurons

during serotonin-evoked locomotion in the neonatal rat thoracolumbar spinal cord.

J. Comp. Neurol., 423, 590-602.

CORNEA-HEBERT, V., RIAD, M., WU, C , SINGH, S.K., DESCARRIES, L., (1999) Cellular and

subcellular distribution of the serotonin 5-HT2A receptor in the central nervous

system of adult rat. J Comp Neurol 409, 187-209.

COWLEY, K.C & SCHMIDT, B.J. (1994) A comparison of motor patterns induced by N-

methyl-D-aspartate, acetylcholine and serotonin in the in vitro neonatal rat spinal

cord. Neurosci. Lett., 171, 147-150.

CYR, M., BOSSE R. & Di PAOLO T. (1998) Gonadal hormones modulate 5-

hydroxytryptamine2A receptors : emphasis on the rat frontal cortex. Neurosci., 83,

829-36.

DOLY, S., MADEIRA, A., FISCHER, J., BRISORGUEIL, M.J., DA VAL, G., BERNARD, R., VERGE,

D. & CONRATH, M. (2004) The 5-HT2A receptor is widely distributed in the rat

spinal cord and mainly localized at the plasma membrane of postsynaptic neurons.

J. Comp. Neurol, All, 496-511.

FERABOLI-LOHNHERR, D., BARTHE, J.Y. «fe ORSAL, D. (1999) Serotonin-induced activation

of the network for locomotion in adult spinal rats. J. Neurosci. Res., 55, 87-98.

FONG, A.J., CAI, L.L., OTOSHI, C.K., REINKENSMEYER, D.J., BURDICK, J.W., ROY, R.R. «fe

EDGERTON, V.R. (2005) Spinal cord-transected mice learn to step in response to

quipazine treatment and robotic training. J. Neurosci., 25, 11738-11747.

FONSECA, M.I., Ni, Y.G., DUNNING, D.D. «& MILEDI, R. (2001) Distribution of serotonin 2A,

2C and 3 receptor mRNA in spinal cord and medulla oblongata. Brain. Res. Mol

Brain Res., 89, 11-19.

FULLER, D.D., BAKER-HERMAN, T.L., GOLDER, F.J., DOPERALSKI, N.J., WAITERS, J.J. «fe

MITCHELL, G.S. (2005) Cervical spinal cord injury upregulates ventral spinal 5-

HT2A receptors. J. Neurotrauma, 22, 203-213.

117

GERASIMENKO, Y.P., ICHIYAMA, R.M., LAVROV, LA., COURTINE, G., CAI, L., ZHONG, H.,

ROY, R.R. <& EDGERTON, V.R. (2007) Epidural spinal cord stimulation plus

quipazine administration enable stepping in complete spinal adult rats. J.

Neurophysiol, 98, 2525-2536.

GIROUX, N., ROSSIGNOL, S. <& READER, T.A. (1999) Autoradiographic study of alphal- and

alpha2-noradrenergic and serotonin 1A receptors in the spinal cord of normal and

chronically transected cats. J. Comp. Neurol, 406, 402-414.

GRILLNER, S. <& ZANGGER, P. (1979) On the central generation of locomotion in the low

spinal cat. Exp. Brain Res., 34, 241-261.

GUERTIN, P., ANGEL, M.J., PERREAULT, M.C. «fe MCCREA, D.A. (1995) Ankle extensor

group I afférents excite extensors throughout the hindlimb during fictive locomotion

in the cat. J. Physiol, 487, 197-209.

GUERTIN, P.A. (2004a) Synergistic activation of the central pattern generator for

locomotion by l-beta-3,4-dihydroxyphenylalanine and quipazine in adult paraplegic

mice. Neurosci. Lett., 358, 71-74.

GUERTIN, P.A. (2004b) Role of NMDA receptor activation in serotonin agonist-induced air-

stepping in paraplegic mice. Spinal Cord, Al, 185-190.

GUERTIN, P.A. (2005) Semiquantitative assessment of hindlimb movement recovery

without intervention in adult paraplegic mice. Spinal Cord, A3, 162-166.

HAMEL, E. (1999) The biology of serotonin receptors: focus on migraine pathophysiology

and treatment. Can. J. Neurol. Sci., 3, S2-6.

HELTON, L.A., THOR, K.B. «fe BAEZ, M. (1994) 5-hydroxytryptamine2A, 5-

hydroxytryptamine2B, and 5-hydroxytryptamine2C receptor mRNA expression in

the spinal cord of rat, cat, monkey and human. Neuroreport, 5, 2617-2620.

HOLOHEAN, A.M. «fe HACKMAN, J.C. (2004) Mechanisms intrinsic to 5-HT2B receptor-

induced potentiation of NMDA receptor responses in frog motoneurones. Br. J.

Pharmacol, 143, 351-360.

JEONG, C.Y., CHOI, J.L «fe YOON, M.H. (2004) Roles of serotonin receptor subtypes for the

antinociception of 5-HT in the spinal cord of rats. Eur. J. Pharmacol, 502, 205-

211.

118

JIANG, Z., CARLIN, K.P. & BROWNSTONE, R.M. (1999) An in vitro functionally mature

mouse spinal cord preparation for the study of spinal motor networks. Brain Res.,

816, 493-9.

JORDAN, L.M., Liu, J., HEDLUND, P.B., AKAY, T. «fe PEARSON, K.G. (2008) Descending

command systems for the initiation of locomotion in mammals. Brain. Res. Rev.,

57, 183-191.

KIEHN, O. <& KJAERULFF, O. (1996) Spatiotemporal characteristics of 5-HT and dopamine-

induced rhythmic hindlimb activity in the in vitro neonatal rat. J. Neurophysiol, 75,

1472-1482.

KIM, D., ADIPUDI, V., SHIBAYAMA, M., GISZTER, S., TESSLER, A., MURRAY, M. <fe

SIMANSKY, K.J. (1999) Direct agonists for serotonin receptors enhance locomotor

function in rats that received neural transplants after neonatal spinal transection. J.

Neurosci, 19, 6213-6224.

KNIGHT, A.R., MISRA, A., QUIRK, K., BENWELL, K., REVELL, D., KENNETT, G. «fe

BICKERDIKE, M. (2004) Pharmacological characterisation of the agonist radioligand

binding site of 5-HT(2A), 5-HT(2B) and 5-HT(2C) receptors. Naunyn

Schmiedebergs Arch Pharmacol 370, 114-123.

LANDRY, E.S. «fe GUERTIN, P.A. (2004) Differential effects of 5-HT(l) and 5-HT(2) receptor

agonists on hindlimb movements in paraplegic mice. Prog. Neuropsychopharmacol.

Biol. Psychiatry, 28, 1053-1060.

LANGLOIS, M.C, BEAUDRY, G., ZEKKI, H., ROUILLARD, C «fe LEVESQUE, D. (2001) Impact

of antipsychotic drug administration on the expression of nuclear receptors in the

neocortex and striatum of the rat brain. Neuroscience, 106, 117-128.

LAPOINTE, N.P., UNG, R.V., BERGERON, M., COTE, M. <& GUERTIN, P.A. (2006) Strain-

dependent recovery of spontaneous hindlimb movement in spinal cord transected

mice (CD1, C57BL/6, BALB/c). Behav. Neurosci., 120, 826-834.

LAUDER, J.M., WILKIE, M.B., Wu, C. «fe SINGH, S. (2000) Expression of 5-HT(2A), 5-

HT(2B) and 5-HT(2C) receptors in the mouse embryo. Int. J. Dev. Neurosci., 18,

653-662.

LEE, M.S., CHOI, Y.C, LEE, S.H. «fe LEE, S.B. (1996) Sleep-related periodic leg movements

associated with spinal cord lesions. Mov. Disord., 11, 719-722.

119

Liu, J. <& JORDAN, L.M. (2005) Stimulation of the parapyramidal region of the neonatal rat

brain stem produces locomotor-like activity involving spinal 5-HT7 and 5-HT2A

receptors. J. Neurophysiol, 94, 1392-1404.

MADRIAGA, M.A., MCPHEE, L.C, CHERSA, T., CHRISTIE, KJ. «fe WHELAN, P.J. (2004)

Modulation of locomotor activity by multiple 5-HT and dopaminergic receptor

subtypes in the neonatal mouse spinal cord. J. Neurophysiol, 92, 1566-1576.

MAESHIMA, T., ITO, R., HAMADA, S., SENZAKI, K, HAMAGUCHI-HAMADA, K., SHUTOH, E,

OKADO, N. (1998) The cellular localization of 5-HT2A receptors in the spinal cord

and spinal ganglia of the adult rat. Brain Res 797, 118-124.

MCEWEN, M.L., VAN HARTESVELDT, C <& STEHOUWER, D.J. (1997) L-DOPA and

quipazine elicit air-stepping in neonatal rats with spinal cord transections. Behav.

Neurosci, 111, 825-833.

MENGOD, G., NGUYEN, H., LE, H., WAEBER, C , LUBBERT, H. «fe PALACIOS, J.M. (1990) The

distribution and cellular localization of the serotonin 1C receptor mRNA in the

rodent brain examined by in situ hybridization histochemistry. Comparison with

receptor binding distribution. Neuroscience, 35, 577-591.

MOLINEAUX, S.M., JESSELL, T.M., AXEL, R. «fe JULIUS, D. (1989) 5-HTlc receptor is a

prominent serotonin receptor subtype in the central nervous system. Proc. Natl.

Acad. Sci. USA., 86, 6793-6797.

NISHIMARU, H., TAKIZAWA, H. «fe KUDO, N. (2000) 5-Hydroxytryptamine-induced

locomotor rhythm in the neonatal mouse spinal cord in vitro. Neurosci. Lett., 280,

187-190.

OBATA, H., SAITO, S., SAKURAZAWA, S., SASAKI, M., USUI, T. «fe GOTO, F. (2004)

Antiallodynic effects of intrathecally administered 5-HT(2C) receptor agonists in

rats with nerve injury. Pain, 108, 163-169.

POMPEIANO, M., PALACIOS, J.M. «fe MENGOD, G. (1994) Distribution of the serotonin 5-HT2

receptor family mRNAs: comparison between 5-HT2A and 5-HT2C receptors.

Brain Res. Mol. Brain Res., 23, 163-178.

PRITCHETT, D.B., BACH, A.W.J., WOZNY, O., TALEB, R.D.T., SHIH, J.C. «fe SEEBURG, P.H.

(1988) Structure and functional expression of cloned rat serotonin 5HT-2 receptor.

Eur. Molec. Biol Org. J., 1, 4135-4140.

120

R E K L I N G , J . C , F U N K , G.D., B A Y L I S S , D.A., D O N G , X.W. <& F E L D M A N , J.L. (2000) Synaptic

control of motoneuronal excitability. Phys io l Rev., 80, 767-852.

R E M Y - N E R I S , O., B A R B E A U , H., D A N I E L , O., B O I T E A U , F. «fe B U S S E L , B . (1999) Effects of

intrathecal clonidine injection on spinal reflexes and human locomotion in

incomplete paraplegic subjects. Exp. Brain Res., 129 ,433-440.

R E M Y - N E R I S , O., D E N Y S , P., D A N I E L , O., B A R B E A U , H. «fe B U S S E L , B . (2003) Effect of

intrathecal clonidine on group I and group II oligosynaptic excitation in paraplegics.

Exp. Brain Res., 148, 509-514.

R U A T , M. , T R A I F F O R T , E., L E U R S , R., T A R D I V E L - L A C O M B E , J., D I A Z , J., A R R A N G , J .M. «fe

S C H W A R T Z , J.C. (1993) Molecular cloning, characterization, and localization of a

high-affinity serotonin receptor (5-HT7) activating c A M P formation. Proc. Natl.

Acad. Sci. USA., 90, 8547-8551.

U N G , R.V., L A N D R Y , E.S., L A P O I N T E , N. , R O U I L L A R D , C , L E V E S Q U E , D. «fe G U E R T I N , P.

(2005) Expression, activation and function of 5-HT2A receptor in the lumbar spinal

cord of adult paraplegic. Program No. 516.7. 2005 Abstract Viewer/Itinerary

Planner. Washington, DC: Society for Neuroscience.

U N G , R.V., L A P O I N T E , N.P. , T R E M B L A Y , C , L A R O U C H E , A. «fe G U E R T I N , P.A. (2007)

Spontaneous recovery of hindlimb movement in completely spinal cord transected

mice: a comparison of assessment methods and conditions. Spinal Cord, 45 , 367-

379.

V I A L A , D. «fe B U S E R , P. (1971) [Methods of obtaining locomotor rhythms in the spinal

rabbit by pharmacological treatments (DOPA, 5-HTP, D-amphetamine)]. Brain

Res . , 35 , 151-165.

VULT VON S T E Y E R N , F. «fe L O M O , T. (2005) Postnatal appearance of 5-HT2A receptors on

fast flexor and slow extensor rat motor neurons. Neuroscience, 136, 87-93.

W H E L A N , P., B O N N O T , A. «fe O ' D O N O V A N , M.J. (2000) Properties of rhythmic activity

generated by the isolated spinal cord of the neonatal mouse. J. Neurophysio l , 84,

2821-2833.

W H E L A N , P.J., H I E B E R T , G.W. «fe P E A R S O N , K.G. (1995) Stimulation of the group I extensor

afférents prolongs the stance phase in walking cats. Exp. Brain. Res., 103, 20-30.

121

Xu, C , GIULIANO, F., SUN, X.Q., BRISORGUEIL, M.J., LECLERC, P., VERGE, D. «fe CONRATH,

M. (2007) Serotonin 5-HT2A and 5-HT5A receptors are expressed by different

motoneuron populations in rat Onuf s nucleus. J. Comp. Neurol, 502,620-634.

ZHOU, S.Y., BASURA, G.J. «fe GOSHGARIAN, H.G. (2001) Serotonin(2) receptors mediate

respiratory

recovery after cervical spinal cord hemisection in adult rats. J. Appl. Physiol, 91,

2665-2673.

122

Figures and legends

Fig. 1 Acetic acid and DMSO control vehicle administered alone or in combination with

quipazine did not influence the induction of locomotor movements.

Fig. 2 Autoradiograms of 5-HTR2A mRNA (L1-L2) from control non-Tx mice (/') and Tx

animals (ii-vii). The left end side of each panel represents the autoradiogram levels

associated with 5-HTR2A labelling whereas the right end side shows a non-labelled cord to

ease identification of the corresponding white matter and gray matter laminae. Optical

autoradiograms densities were translated into nCi/g of tissue. In non-Tx animals, 5-HTR2A

transcripts revealed to be distributed mainly in the ventral horn (although non-exclusively).

In Tx animals, a strong upregulation was found specifically in the lateral intermediate zone

area (peaking at 7 days). Panel mi shows a representation of a spinal cord section that

clearly reveals where autoradiographic measurements were made (left-end side - from top

to bottom: dorsal horn area, lateral intermediate zone, medial intermediate zone, lateral

ventral horn, medial ventral horn) versus the laminar structure of the gray matter (right-end

side - from top to bottom: laminae I to DC with lamina X surrounding the central canal).

Fig. 3 Video images and kinematics analyses from a quipazine only-treated Tx animal. A.

Video images showing a typical cycle (or 4 sec-bout of recording when no movement)

prior to quipazine administration (i - treadmill, iii - air-stepping) as well as 15 min after 1

mg/kg quipazine (ii - treadmill, iv - air-stepping). B. Corresponding angular excursion

values (in degrees) at the hip, knee and ankle joints (top, middle and bottom panels,

respectively) calculated by averaging all cycles detected from each 4 min-bout of recorded

activity (no quipazine, i - treadmill, iii - air-stepping; with quipazine, ii - treadmill, iv -

air-stepping). The cycle period was normalized with respect to the average flexion-

extension cycle length. Time (sec) is shown at the bottom right of each video image.

Fig. 4 Video images and kinematic analyses from a representative 5-HTR2A antagonist-

pretreated Tx animal. A. Video images showing a typical cycle (or 4 sec-bout of recording

when no movement) prior to quipazine administration (î - treadmill, iii - air-stepping) as

123

well as 15 min after 1 mg/kg quipazine (//' - treadmill, iv - air-stepping) in a typical animal

that was pretreated with MDL-100,907 15 min prior to testing. B. Corresponding angular

excursion values (in degrees) at the hip, knee and ankle joints (top, middle and bottom

panels, respectively) calculated by averaging all cycles detected from each 4 min-bout of

recorded activity (no quipazine, î - treadmill, iii - air-stepping; with quipazine, ii -

treadmill, iv - air-stepping). The cycle period was normalized with respect to the average

flexion-extension cycle length. Time (sec) is shown at the bottom right of each video

image.

Fig. 5 Quipazine-induced effects in pretreated and non-pretreated animals: comparisons

between conditions (treadmill versus air-stepping). A. Quipazine was administered alone.

B-E. 15 minutes prior to quipazine administration, animals were pretreated either with 5-

HTR2A, 5-HTR2B, 5-HTR2C, 5-HTR2B + 5HTR2C, respectively. * P < 0.05

Fig. 6 Quipazine-induced effects in pretreated and non-pretreated animals: comparisons

between groups. A. Air-stepping. B. Treadmill condition. * P < 0.05; ** P < 0.01. For

further post-hoc analysis, see Table 2. All antagonists were administered 15 minutes prior

to quipazine administration.

Table 1 Summary of movement amplitude (in mm). Movement amplitude referred to as the

distance covered by the 2nd toe (longest), from the most extended to the most flexed

positions. Data obtained from several cycles were averaged for all tested animals.

Table 2 Summary of within-group comparisons. Statistical results from group-comparisons

between the different pharmacological treatments in air-stepping-condition (upper panel)

and treadmill-condition (lower panel). * P < 0.05; ** P < 0.01; NS non-significant.

124

Air-stepping 5 n

i4

>. u CT S 1

Acetic acid + DMSO Quipazine -

+ + + - +

B 12-

"ç 10-E 5 8H

>• fi-c B

Ë u. 2H

Treadmill

Acetic acid + DMSO Quipazine -

i

- + + + - +

Figure 1

125

B i

ô

Dorsal horn (area 1 )

fr-^e^g-D o o^Q i i i i i i i

Non-Tx 3hr 1d 3d 7d 14d 28d

//' Lateral intermediate zone (area 2)

100-1 *

80

60

40

I I I I I I Non-Tx 3hr 1d 3d 7d 14d 28d

iii Medial intermediate zone (area 3)

100-1

80

60

40 '

20-

0 I I I I I I 1 — Non-Tx 3hr 1d 3d 7d 14d 28d

O

IV

1001

80

60

40

20

Lateral ventral horn (area 4)

i i i i i i i Non-Tx 3hr 1d 3d 7d 14d 28d

100

80

60

40

20

Medial ventral horn (area 5)

*^4i-$\ i i i i i

Non-Tx 3hr 1d 3d 7d 14d 28d

Time

Figure 2

126

;' Non-treated (treadmill)

ii Quipazine (treadmill)

iii Non-treated (air-steppinq)

iv Quipazine (air-stepping)

a /' Non-treated

(treadmill) ii Quipazine

(treadmill) /'/;' Non-treated

(air-stepping) iv Quipazine

(air-stepping)

000 025 050 075 100 125 0.00 0 25 0 50 0 75 100 125 0 00 0 25 0 50 0 75 100 125 0.00 0.25

Normalized time (% flexion-extension cycle) 0.50 0 75 100 125

Figure 3

127

/' Non-treated (treadmill)

ii M100907 + Quipazine (treadmill)

/// Non-treated (air-stepping)

B

100 90

Q. 80 î

70 I eo Q 80

1 7°

E • S » 60 8* i 160

150 C 140. <

130

/' Non-treated (treadmill)

//' M100907 +Quipazine (treadmill)

120. 000

150 140 130 120

110

III

60 50 40 30

-I 20 100 90.

70. 60

Non-treated (air-stepping)

<

iv M100907 + Quipazine (air stepping)

140,

130

120.

110

-1 100

<

120. 120.

^ _ n n . 110.

100. 100.

110.

100. 100.

110.

100.

90. 90.

140

i 130

120

110

, 100 025 050 075 100 125 0 00 025 0 50 0 75 100 125 0 00 0 25 0 50 0 75 100 125 0 00 0 25 0 50 0 75 100 125

Normalized time (% flexion-extension cycle)

Figure 4

128

1 2 T

-ç- 10-E 5 8-

>. o c (U 3 4 -cr S u- 2H

Quipazine

Li i

. . < • * »

8 M100907 + Quipazine 12-

ID­

S'

6-

4-

2-

0

SB204741 + Quipazine D 12-

"ç 10-E 5 8-_J_

>< fi-o c d) ï

U. 2 -0- m

v* ->* *»

^ ^

12-

10-

8-

6-

4-

2

0

SB242084 + Quipazine *

Li Xs ^ $

12-

ç- 10-

1 i 8->. o c «> 3 4 -

u- 2-

SB204741 + SB242084 + Quipazine

m ^

Figure 5

129

Air-stepping

5-,

J 4-J

! 3H

2H

£ ^

X

^ ^ € ^ ^ ^ 0 » B Treadmill

i 2 n

10H

>> 6H

1 4 H

i t 2H

X

< ^ ^ 4> €> 4& Figure 6

Table 1. Movement amplitude (in mm)

_ . . M100907 SB204741 SB242084 ^ l ? ? ™ Quipazine _ . _ . _ . +SB242084 +Quipazine +Quipazine +Quipazine 0 .

Air-stepping 12.6 ±0.6 0.0 ±0.0 11.7 ±0.8 11.1 ±0.5 12.3 ±1.9

Treadmill 11.9±0.4 0.0±0.0 11.5±0.7 13.1 ±0.5 12.3±0.5

130

131

Table 2. Summary of with in-group compar isons

„ M100907 SB204741 SB242084 ^ S S Ï Ï L Groups . ^ . _. . ^_ . +SB242084

+Quipazine +Quipazine +Quipazine + Q .

Average raw data (LM/min)

Quipazine (2.9 ±1.6) * * NS NS NS

M100907 i n n n n ^ |> +Quipazine <

0 0 ± 0 0> * * * *

Q. % SB204741 u? +Quipazine (1.9 ±0.5) - - NS NS

SB242084 +Quipazine (3.3 ±1.1) - - NS

SB204741 +SB242084 (2.2 ± 0.7) - - - -

+Quipazine

Quipazine (7.5 ±1.7) * * * NS *

M100907 ♦Quipazine (0-0 ±0.0) - * * * * * *

1 SB204741 I +Quipazine (4.0 ±0.9) * NS ro

r

S r- SB242084

♦Quipazine (8.9 ±3.0) * SB204741

+SB242084 (4-

2 * °-7) _ - _ -

+Quipazine

132

C H A P I T R E V I I - E F F E C T S O F C O - A D M I N I S T R A T I O N O F C L E N B U T E R O L A N D

TESTOSTERONE PROPIONATE ON SKELETAL MUSCLE IN PARAPLEGIC MICE

La troisième étude consistait à évaluer les effets de l'administration de testosterone

et clenbuterol, au niveau de la composition corporelle et sur la récupération motrice

spontanée de souris paraplégiques. Les résultats nous ont permis de noter que ces deux

substances administrées seules ou en combinaison montraient de fortes propriétés

anaboliques au niveau musculaire, mais n'avaient aucune incidence sur le rétablissement

des fonctions locomotrices. Cette étude à été publiée dans Journal of Neurotrauma, 2010

27(6): 1129-42.

Abstract

Spinal cord injury (SCI) is generally associated with a rapid and significant decrease in

muscle mass and corresponding changes in skeletal muscle properties. Although 62-

adrenergic and androgen receptor agonists are anabolic substances clearly shown to prevent

or reverse muscle wasting in some pathological conditions, their effects in SCI patients

remain largely unknown. Here, we studied the effects of clenbuterol and testosterone

propionate administered separately or in combination on skeletal muscle properties and

adipose tissue amount in adult CD1 mice spinal cord-transected (Tx) at the low-thoracic

level (i.e., induced complete paraplegia). Administered shortly post-Tx, these substances

were found to differentially reduce loss in body weight, muscle mass and muscle fiber

cross-sectional area (CSA) values. Although all three treatments induced significant effects,

testosterone-treated animals were generally less protected against Tx-related changes.

However, none of the treatments prevented fat tissue loss or muscle fiber-type conversion

and functional loss generally found in Tx animals. These results provide evidence

suggesting that clenbuterol alone or combined with testosterone may constitute better

clinically-relevant treatments than testosterone to decrease muscle atrophy (mass and fiber

CSA) in SCI subjects.

Keywords: Spinal cord injury, therapeutic approaches for the treatment of CNS injury,

recovery, locomotor function.

133

Abbreviations: Spinal cord injury (SCI); spinal cord-transected (Tx); cross-sectional area

(CSA); central nervous system (CNS); locomotor movement (LM); non-locomotor

movement (NLM); extensor digitorum longus (EDL); myosin heavy chain fast and slow

(MHCf, MHCs); central pattern generator (CPG); Control (Ctr); Clenbuterol (Cb);

Testosterone propionate (Tp).

Introduction

Spinal cord injury (SCI) generally leads to an immediate and irreversible loss of sensory

and voluntary motor functions. The paralysis and lack of physical activity are also

associated with the development of significant health problems or so-called 'secondary

complications', including skeletal muscle atrophy among many medical problems (Bauman

et al., 1999; Bauman and Spungen, 2000; Cruse et al., 2000; Giangregorio and McCartney,

2006).

Some anabolic substances are known to prevent extended body weight loss and muscle

wasting in different disorders (Shahidi, 2001; Dudgeon et al., 2006; Koopman et al., 2009).

Among these substances, clenbuterol, a P2-adrenergic agonist, generally used to treat

asthma (Anderson and Wilkins, 1977), was also reported to stimulate muscle growth

(MacLennan and Edwards, 1989; Carter et al., 1991; Choo et al., 1992; Lynch et al., 1999)

or to prevent muscle atrophy in dystrophic, hindlimb unloaded or nerve-transected models

(Agbenyega et al., 1995; Zeman et al., 2000; Herrera et al., 2001; Teng et al., 2006).

Interestingly, we recently found that a significant decrease of testosterone level occurs

shortly after a spinal cord transection, suggesting a role for this hormone in rapid muscle

mass changes after SCI (Rouleau et al., 2007). Testesterone is well-known for its muscle-

building effects in athletes and normal individuals (Graham et al., 2008; Choong et al.,

2008; Bhasin et al., 1996) as well as for its role against muscle wasting in different

pathological conditions such as in burn individuals or HIV patients (Ferrando et al., 2001;

Dudgeon et al., 2006).

134

These two anabolic substances may thus be of special interests to SCI patients for their

muscle-building properties as well as for their potential neuroprotective effects, recently

suggested from data in CNS injured or SCI animal models (Kujawa et al., 1989; Brown et

al., 1999; Zhu et al., 2001; Junker et al., 2002; Zeman et al., 1999, 2004). Indeed,

neuroprotection was proposed for the underlying functional recovery induced by

clenbuterol or testosterone in incomplete SCI animals. However, it remains unknown if

these anabolic substances can also enable locomotor recovery in a complete Tx animal, by

facilitating locomotor network reorganisation at the spinal level. Moreover, to our

knowledge, their effects on muscle atrophy and adipose tissue (used separately or in

combination) have never been compared in animal model of SCI.

To investigate these questions, we assessed the effects induced by clenbuterol, testosterone

or both on body weight, skeletal muscle and fiber size, fiber phénotype distribution,

adipose tissue amount, and locomotor function recovery in low thoracic Tx mice, a reliable

experimental model of motor-complete paraplegia (e.g., Landry et al., 2004; Lapointe et al.,

2006; Ung et al., 2007).

Material and Methods

Animals and surgical procedures

All experimental procedures were conducted in accordance with the Canadian Council for

Animal Care guidelines and were accepted by the Laval University Animal Care and Use

Committee. Eight week-old male CD1 mice (n = 159, Charles River Canada, St-Constant,

Quebec, Canada) initially weighing 30-35 g were used for this study. Preoperative care

included subcutaneous injections of an analgesic (0.1 mg/kg, buprenorphine), an antibiotic

(5 mg/kg, enrofloxacin) and lactate-Ringer's solution (1 ml). All surgical procedures were

performed under aseptic conditions. Mice were anesthetised with 2.5% isoflurane. A small

incision was made on their back in order to expose some thoracic segments. The spinal

cord was then completely transected inter-vertebrally with microscissors inserted between

the 9th and 10th thoracic vertebrae (Lapointe et al., 2006; Ung et al., 2007). To ensure that

135

complete Tx was achieved, the inner vertebral walls were explored and entirely scraped

several times with small scissor tips. The opened skin area was sutured and animals were

placed on heating pads for a few hours. Mice were left in their cage with food and water ad

libitum. Post-operative cares provided for 4 days included subcutaneous injections of

buprenorphine (0.2 mg/kg/day), enrofloxacin (5 mg/kg/day) and lactate-Ringers's solution

( 2 x 1 ml/day). Bladders were manually expressed twice daily for the first week post-Tx

and once daily thereafter. Complete Tx was confirmed by 1) initial full paralysis of the

hindlimbs, 2) post-mortem visual and microscopic examination of the spinal cord lesion,

and 3) histological assessment of coronal or midsagittal spinal cord sections stained with

luxol fast blue/cresyl violet for myelinated axons and Nissl substance, respectively.

Groups and treatments

Tx mice were randomly divided into four main groups (treatments): 1) vehicle-treated

(control Tx), 2) clenbuterol hydrochloride-treated, 3) testosterone propionate-treated, 4)

clenbuterol hydrochloride+testosterone propionate-treated. In each main group, animals

were further divided into four subgroups (treatment durations) treated daily (first treatment

given 1 hour post-surgery) during 1, 2, 4 or 8 weeks prior to sacrifice. Clenbuterol

hydrochloride and testosterone propionate was subcutaneously injected at 1 mg/kg/day

(e.g., Zeman et al., 2000) and 5 mg/kg/day (e.g., Pansarasa et al., 2002), respectively.

Clenbuterol hydrochloride and testosterone propionate, purchased from Sigma Chemical

Co. (St-Louis, MO, USA), were dissolved in saline, and 0.5 % benzyl alcohol + vegetable

oil, respectively. Corresponding vehicle-treated groups were injected either with saline or

0.5% benzyl alcohol + vegetable oil. Since data from saline- or oil-treated animals were not

significantly different, they were subsequently pooled for statistical purposes. Furthermore

10 uninjured non-treated mice were added for body weight comparison at 8 weeks.

Body weight and hindlimb movement recovery

During the first week, animals were weighed daily and weekly thereafter. One day prior to

sacrifice, mice were placed on a motor-driven treadmill (8-10 cm/sec) and locomotor

136

function recovery was assessed using two complementary methods. Each assessment

method was conducted for two minutes. For the first assay, we used a semi-quantitative

method called Average Combined Score or ACOS that was developed for assessment of

hindlimb movements in Tx rodents (Guertin, 2005; Lapointe et al., 2006; Ung et al., 2007).

ACOS is composed of non-locomotor movement (NLM, number/min), locomotor-like

movement (LM, number/min) and amplitude. Values are arithmetically combined as

follows: [NLM + (2 x LM)] x amplitude. One LM was defined as a flexion-extension

movement occurring bilaterally in alternation. We decided to exclude LM occurring during

bowel movements to avoid taking into account sacral reflex-induced LM (Strauss and Lev-

Tov, 2003). One NLM was defined as a non-bilaterally alternating movement including

movements such as jerks, fast-paw shaking and twitches. Amplitude was characterized by

assigning one of three values; 0 - no movement; 1 - movements that were less than half the

range of motion of normal steps; 2 - movements were greater than half the range of motion

of normal steps. For the second assay, we used a method referred to as Antri, Orsal &

Barthe or AOB that was specifically designed for complete Tx rodents (Antri et al., 2002).

In brief, the AOB scale consists of 22 scores: 0 - no movement; 1 - weak limb jerks; 2 -

weak rhythmic movements with no bilateral alternation; 3 - large rhythmic movements

with no bilateral alternation; 4 - weak rhythmic movements with occasional bilateral

alternation; 5 - large rhythmic movements with occasional bilateral alternation. For

additional levels, see Antri and colleagues (2002). The ACOS and AOB scales were

preferred over other methods such as the Basso mouse scale (Basso et al., 2006) based on

results from this laboratory demonstrating greater sensitivity and specificity in completely

spinal cord-transected mice (Ung et al., 2007).

Tissue collection and measures

Upon sacrifice, extensor digitorum longus (EDL) and soleus muscles were dissected and

weighed. Area-specific adipose tissues were collected from the abdominal subcutaneous,

inguinal, and visceral regions. Seminal vesicles were removed, weighed and used as

positive control tissue for testosterone efficacy.

137

Immunohistochemistry

Upon dissection, soleus and EDL were frozen in melting isopentane and stored at -80°C

until further use. 10 pm-thick serial sections from the mid-belly were cut with a cryostat

maintained at -20°C (2800E Frigocut, Leica Instruments, Germany) and mounted on

Superfrost® plus glass slides (VWR Canlab, Mississauga, ON, Canada). Sections were

incubated for 2 hours in either Myosin Heavy Chain slow or fast (MHCs or MHCf) primary

antibodies, specific for MHC isoform type I and isoform type II, respectively (Vector

Laboratories, Burlingame, CA, USA) (dilution 1/50 in 0.1M PBS containing 1% rabbit

serum and 1% triton X-100). For control, some sections were not incubated with the

corresponding primary antibodies. Sections were then rinsed in PBS 0.1 M before

incubation with a goat anti-mouse IgG (H+L) Alexa Fluor 488 secondary antibody

(Molecular Probes, Eugene, OR, USA) (dilution 1/500 in 0.1M PBS containing 1% rabbit

serum and 1% triton X-100). Slides were then rinsed in PBS 0.1 M and mounted with PBS-

Glycerol (50-50). Immunofluorescence labeling was visualized with an Olympus BX61WI

confocal microscope using a lOx water-immersion objective. Images were captured using

FluoView 300 (Olympus Canada Inc., Markham, ON, Canada) and analysed with ImageJ

(ImageJ 1.40, Research Services Branch, NIH, Bethesda, MD, USA). Analyses consisted of

determining muscle fiber phénotype, measuring corresponding CSA (averages were from

50 fibers per fiber type/muscle, when available), and calculating relative fiber type

distribution per muscle. Type I fibers displayed labeling only with MHCs antibodies, type

II fibers displayed labeling only with MHCf antibodies (no distinction between type Ha, IIx

or lib isoforms was made), whereas hybrid fibers displayed labeling with both antibodies

(adjacent sections).

Data analyses

Differences between groups for locomotor recovery, adipose tissue, muscle mass, fiber

CSA and fiber type distribution were analyzed using a One-Way ANOVA followed by a

Bonferroni post-hoc. Body weight changes were analyzed using a two-way repeated

138

measures ANOVA followed by a Bonferroni post-hoc. Results are reported as Mean ±

SEM.

Results

Prior to assessing the effects of anabolic treatments on muscles, testosterone bioavailability

and efficacy had been confirmed in standard tissue control (testosterone-induced seminal

vesicle hypertrophy). The results showed that testosterone with or without clenbuterol

significantly (P < 0.05) increased by 35% on average seminal vesicle size and weight

compared with control Tx or clenbuterol-treated animals (data not shown).

Body weight

As previously shown elsewhere, a significant loss in body weight is normally found within

the first week post-Tx in paraplegic animals (e.g., Landry et al., 2004; Ung et al., 2007).

This finding was again confirmed in the present study. By combining data from all

subgroups, Fig.lA reveals that only vehicle-treated Tx mice (control Tx) displayed a

significantly greater decrease in body weight that reached its peak (84.1 ± 0.9% of initial

body weight) at 5 days post-Tx (Fig.lA, solid black line includes control Tx mice from all

four subgroups). Tx animals treated with clenbuterol (solid grey line), testosterone (dashed

grey line), or both (dashed black line) also displayed significant body weight loss that

reached its peak (88-89% of initial body weight) at day 3 post-Tx (Fig.lA). Thus, the

extent of body weight loss in control Tx animals was significantly (P < 0.05) greater at 3, 5

and 7 days post-Tx compared with treated animals. For figure 1B-E, only mice that were

sacrificed on that specific week were taken for analyses. Data from each of the four main

groups (treatments and control Tx) examined individually showed decreased body weight

loss and increased body weight values at 1, 2, 4 and 8 weeks post-Tx (Fig. 1B-E).

Differences that reached significant levels (P < 0.05) were found specifically at 4 and 8

weeks post-Tx in clenbuterol-treated (e.g., 105% of initial body weight) and

clenbuterol+testosterone-treated mice (e.g., 103% of initial body weight) compared with

vehicle-treated mice and testosterone-treated animals (Fig 1D-E). At 8 weeks, uninjured

139

mice showed average body weight of 122.4 ± 2.2 %, which represented a 2.8 % increase

per week (Fig. IF). Conversely, after Tx, control Tx (1.4 %), clenbuterol (1.7 %),

testosterone (1.5 %) and clenbuterol+testosterone (1.8 %) groups showed lower average

body weight gain per week. These results show that comparable short-term effects (i.e.,

within 1 week post-Tx) on body weight were induced by all three treatments, although only

clenbuterol or clenbuterol+testosterone induced significantly greater body weight re-

increase at 4 and 8 weeks post-Tx.

Muscle mass

Soleus and EDL were dissected out and weighed. When comparing subgroups (treatment

durations), soleus values were significantly (P < 0.05) greater in clenbuterol-treated at 4

and 8 weeks (0.0244 ± 0.0024% and 0.0207 ± 0.0017% of body weight, respectively) than

at 1 week (0.0158 ± 0.009% of body weight) (Fig.2AII). Comparable results were found

with soleus in testosterone-treated or clenbuterol+testosterone-treated animals with

significant (P < 0.01) increases at 4 and 8 weeks post-Tx than at 1 week (Fig.2AIII and

Fig.2AIV). The greatest significant (P < 0.05) increase of soleus mass, which was revealed

by comparing treatments, was identified in clenbuterol+testosterone-treated animals with

values reaching 0.0265 ± 0.0020% of body weight (table 1).

EDL also increased in mass with time in all subgroups, as shown in Fig.2BI-IV. In contrast

with soleus, EDL masses were significantly higher in both clenbuterol+testosterone-treated

(0.0416 ± 0.0009% of body weight) and clenbuterol-treated (0.0424 ± 0.0020% of body

weight) animals compared with control Tx at 2 weeks for instance (0.0347 ± 0.0016% of

body weight) (table 1 ). This indicates that comparable effects on EDL mass were induced

by clenbuterol alone or clenbuterol+testosterone.

EDL and soleus fiber distribution

140

We also studied potential changes in fiber type conversion since a shift from slow-twitch to

fast-twitch fibers (e.g., type I — hybrid fibers —> type II fibers) is generally known to occur

after a spinal cord transection (Lieber et al., 1986a,b).

The results showed that significantly (P < 0.001) less hybrid fibers were found in EDL with

time (Fig.3B). Consequently, this was associated with a progressive increase in type II

fibers (Fig.3A). No type I fibers were found in EDL. When comparing treatments, we

found significantly (P < 0.05) greater percentages of hybrid fibers in

clenbuterol+testosterone-treated mice than in control Tx animals at 1 week and 2 weeks

(table 1). At 2 weeks, greater percentages of hybrid fibers were also found in clenbuterol-

treated animals compared with testosterone-treated or vehicle-treated mice. However, at 4

and 8 weeks, only low percentages of hybrid fibers (< 3% of total fiber number) were found

with no significant difference between groups. In contrast, percentages of type II fibers kept

increasing over time with relatively high values at 8 weeks that ranged between 97.5 ± 0.9

and 98.6 ± 0.5 % of total fiber number (table 1). Results in this section showed that no

long-term difference between treatments was found.

In soleus, where all three main types of fibers (type I, type II, and hybrid) were found, the

results showed also significant changes in fiber type conversion although different than

those in EDL. Across all treatments, percentages of type I fibers ranging between 26.6 ±

4.8% and 44.5 ± 3.1% of total fiber number at 1 week were found to decrease to values

ranging between 0 and 5.3 ± 1.1% b 8 weeks (Fig.4AI-AIV, table 1). Type II fiber

percentages were found to remain relatively unchanged between 40% and 50% when

comparing week 1 and 8. However, at 4 weeks a transient decrease in percentage was found

in all groups except in clenbuterol+testosterone-treated mice (Fig.4BI-BrV). In clear

contrast, percentages of hybrid fibers were found to increase with time in all groups

(generally from 5-20% up to 40-60% of total fiber number, Fig.4CI-CIV). When comparing

groups (treatments), differences were only found at week 8 for all fiber type. Testosterone-

treated group had greater percentage of type I fibers than clenbuterol and

clenbuterol+testosterone-treated mice. Greater percentages of type I and type II fibers were

also found in vehicle-treated groups compared to clenbuterol and clenbuterol+testosterone-

141

treated mice. In contrast, more hybrid fibers were labeled in clenbuterol and

clenbuterol+testoterone-treated mice (table 1). Data in this section show that none of the

treatments could prevent the drastic loss of type I fibers or the significant increase in type

I+II (hybrid) fibers. This said, clenbuterol and clenbuterol+testosterone further exacerbated

the loss of type I fibers which completely disappeared (none remained labeled at 8 weeks).

EDL and soleus fiber CSA

The results showed, in all subgroups, a progressive increase of EDL fiber CSA values

(Figs.5). In vehicle-treated animals, type II fiber CSA values went from 579.7 ± 7.6 pm at 1

week up to 938.9 ± 17.3 pm at 8 weeks (Fig. 5AI). Type II fiber CSA values progressively

increased also in clenbuterol-treated, testosterone-treated and clenbuterol+testosterone-

treated animals with values that reached 1032.7 ± 16.9 pm, 926.6 ± 13.8 pm and 1207.9 ±

24.7 pm at 8 weeks, respectively (Fig.5AII-AIV). Comparable changes were found in

hybrid fiber CSAs with values lower than 400 pm in vehicle-treated up to 650 pm in

clenbuterol or clenbuterol+testosterone-treated mice (Fig. 5BI-BIV). When comparing

treatments at 8 weeks, the results revealed that greater (P < 0.001) augmentations of type II

fiber CSA values were found in clenbuterol+testosterone-treated mice compared with the

other groups (table 1). Clenbuterol-treated mice also showed greater type II fiber CSA

values than testosterone and control Tx animals. For hybrid fibers, CSA values were not

significantly different between clenbuterol-treated and clenbuterol+testosterone-treated

groups. However, both groups showed significantly greater CSA values (P < 0.001)

compared with vehicle-treated or testosterone-treated mice (table 1). These data show that

clenbuterol and clenbuterol+testosterone induce greater effects on EDL CSA values.

For soleus, CSA values were also found to generally increase with time excepted for type I

fibers in clenbuterol and clenbuterol+testosterone-treated animals where at 8 weeks, no

more type I fibers were found as mentioned previously (Fig.6AII, IV, table 1). In vehicle-

treated animals, type I fiber CSA values ranged around 450 pm (at 1 week) and 620 pm (at

8 weeks) whereas in clenbuterol or clenbuterol+testosterone-treated mice, values ranged

between 485 pm and 950 pm (e.g. at 4 weeks in clenbuterol+testosterone-treated animals,

142

Fig.6ArV). When comparing groups, we found that greater augmentations were found at 4

weeks in clenbuterol+testosterone-treated animals compared with the other groups (P <

0.001, table 1) or at 8 weeks in testosterone-treated compared with vehicle-treated mice (P

< 0.001, table 1). Regarding type II and hybrid fiber types, CSA values generally increased

over time (Fig.6). Comparisons between treatments showed significantly greater

augmentations at 8 weeks in clenbuterol+testosterone-treated mice (approximately 1100

pm) compared with the other groups (table 1). Data in this section show that greater effects

on soleus fiber size are induced by clenbuterol+testosterone.

Adipose tissue

Adipose tissues were collected from specific regions of the body to study the extent to

which these treatments could selectively alter fat storage in Tx mice (table 2). In brief, fat

tissue amounts were generally reduced in clenbuterol and clenbuterol+testosterone groups

compared to vehicle-treated mice. This was found at all time points although significant

levels were reached only at 2 and 4 weeks (e.g., reduced from 2.81% to less than 2.0% of

body weight at 4 weeks). At 4 weeks, compared to control Tx mice, testosterone-treated

mice also had lower adipose tissue. No significant difference was found between treatments

at 8 weeks. In brief, no trend or clear region-specific loss was identified.

Hindlimb motor recovery

We found in all groups a significant increase with time of motor/locomotor scores (i.e.,

involuntary hindlimb movements) assessed with the ACOS and AOB scales (P < 0.05),

both designed for assessing hindlimb movements in completely spinal cord-transected

animals. ACOS scores ranged between 0 and 0.1 ± 0.1 at 1 week and increased up to scores

ranging between 8.3 ± 3.6 and 14.9 ± 3.8 at 8 weeks (table 3). Comparable results were

found with the AOB scale with scores ranging between 0 and 0.2 ±0.1 at 1 week that

increased up to scores ranging between 2.0 ± 0.6 and 3.1 ± 0.6 at 8 weeks (table 3).

However, no significant (P > 0.05) difference was found between treatments.

143

Discussion

This study is the first to directly assess in comparable conditions, the effects of

testosterone, clenbuterol or both in the same animal model of SCI. The results clearly show

that 1) clenbuterol and clenbuterol+testosterone (but not testosterone alone) induced

significantly greater body weight increase at 4 and 8 weeks post-Tx; 2) greater effects were

induced by clenbuterol+testosterone on soleus mass compared with testosterone or control

Tx; 3) greater effects were elicited by clenbuterol alone and clenbuterol+testosterone on

EDL mass compared with testosterone or control Tx; 4) no difference between treatments

was found for EDL fiber type conversion after 8 weeks; 5) no treatments prevent or reverse

the drastically reduced proportion of soleus type I fibers and the increased proportion of

soleus hybrid fibers (although clenbuterol and clenbuterol+testosterone exacerbated the loss

of type I fibers and induced increased hybrid fiber phénotype); 6) clenbuterol and

clenbuterol+testosterone induced greater effects on EDL CSA values; 7) greater effects on

soleus CSA values were induced by clenbuterol+testosterone, 8) the loss of fat tissue

amounts was generally greater in all treated-groups than control Tx; 9) the progressive

occurrence of involuntary motor/locomotor hindlimb movements was not altered by these

treatments.

In other words, testosterone did not generally alter most changes normally found in SCI

mice although it significantly increased fat tissue loss, as the other treatments. In contrast,

clenbuterol reduced and reversed body weight loss (although it decreased adipose tissue

storage), EDL muscle mass and fiber atrophy. In turn, greater effects, although not often

significantly greater than clenbuterol alone, were generally induced by

clenbuterol+testosterone which decreased and reversed body weight loss, EDL and soleus

muscle mass as well as fiber atrophy.

Effects of anabolic treatments on overall body composition

Drastic changes in overall body composition are generally known to occur after SCI. For

instance, a rapid decrease in body weight was reported in both SCI patients (Cox et al.,

144

1985) and Tx mice (Landry et al., 2004). Here, although all three treatments reduced body

weight loss within the first week post-Tx, only clenbuterol and clenbuterol+testosterone

increased body weight gain after 1 or 2 months of treatment (e.g., 105% of initial body

weight in clenbuterol-treated animals). This effect on body weight was most likely

attributed to corresponding changes in muscles rather than fat tissues since clenbuterol and

clenbuterol+testosterone both largely increased skeletal muscle mass and muscle fiber CSA

values whereas fat tissue amount decreased with all treatments. It was suggested that

testosterone and clenbuterol induced proliferation, differentiation, recruitment of satellite

cells into muscle fibers to promote muscle hypertrophy or inhibit muscle atrophy (Sinha-

Hikim et al., 2002; Spurlock et al., 2006). However, the detailed mechanisms underlying

this effect on muscle atrophy/hypertrophy is not completely understood, but regulation of

protein synthesis and breakdown may include inhibition of ubiquitin-proteasome pathways,

inhibition of 3-methylhistidine and upregulated expression of MAFbx, as shown in other

models of unloading (Benson et al., 1991; Yimlamai et al., 2005; Zhao et al., 2008 a,b, for

reviews see Herbst and Bashin 2004, Lynch and Ryall, 2008).

This said, these drugs could have decreased weight loss through actions on other systems

such as on bones (not assessed in this study) which are known to undergo extensive tissue

loss after SCI, in both humans and rodents (Modlesky et al, 2004; Zehnder et al., 2004). In

fact, in Tx mice, Picard and colleagues (2008) reported a rapid (i.e., within 1 month post-

Tx) 22% and 14% reduction of femoral bone volume and femoral bone mineral content,

respectively. In line with this, clenbuterol was reported to prevent bone tissue loss under

conditions of disuse such as in the denervated hindlimb or tail-suspended model (Zeman et

al., 1991; Apseloff et al., 1993; Bloomfield et al., 1997). There is also evidence suggesting

that testosterone can prevent bone tissue loss in rats (Stuermer et al., 2009) and increase

bone density in hypogonadal men (Katznelson et al., 1996; Snyder et al., 2000). Therefore,

clenbuterol and clenbuterol+testosterone induced effects on body weight may have been

partially caused also by protecting bone tissue loss in Tx animals.

Clenbuterol and testosterone are also well-known to increase lean body mass through

lipolytic actions on fat tissue (Kearns et al., 2001; Katznelson et al., 1996). Results in this

145

study could suggest lipolytic effects induced in Tx animals since amounts of specific fat

tissues were reduced by all treatments, especially on week 4. This is of particular clinical

interest specifically in late chronic SCI patients (> 1 year post-SCI) since increased fat

tissue and obesity are often found several months to several years post-trauma (Cox et al.,

1985). This said, we can not conclude, from our data, that lipolytic effects were found in

these animals since caloric intake or recompartmentalization of adipose tissues has not been

examined by Dual-X-Ray-Absorptiometry.

Effects on muscle fiber type distribution and size

To our knowledge, no study, prior to this one, has tested clenbuterol-induced effects on

muscle fiber type distribution and size after SCI. Results from various animal models

showed that clenbuterol can prevent muscle mass loss, although it remains unclear what its

actions are at the single cell level in individual fibers (Zeman et al., 1997; 1999). In turn,

some tests were conducted with testosterone in SCI patients but mainly to study

spermatogenesis, muscle strength and locomotor recovery (Huang et al., 1999; Gregory et

al., 2003; Clark et al., 2008).

In contrast with data from Gregory and colleagues (2003), we found that testosterone did

not to prevent type I fiber conversion in hindlimb muscles (e.g, soleus). Moreover,

clenbuterol exacerbated this phase shift since no more type I fibers were found in the soleus

of clenbuterol-treated and clenbuterol+testosterone-treated mice after 2 months of

treatment. In contrast, hybrid fibers were increased in proportion suggesting a phase shift

from type I towards type I+II fibers. This is supported by previous studies showing that

clenbuterol can promote a slow to fast fiber transition in the soleus of normal or hindlimb

unloaded rats (Criswell et al. 1996; Picquet et al. 2004, Oishi et al. 2002). On the other

hand, in EDL, where type I fibers do not exist in rodents, a different but comparable

conversion was found - less hybrid fibers and more type II fibers in treated and control Tx

animals.

146

Regarding fiber CSA changes, it is well known that muscle activity level is positively

correlated with muscle mass and strength, especially in SCI patients where functional

electrical stimulation, passive movement and locomotor training were shown to induce

beneficial effects on muscle size and strength recovery (Gerrits et al., 2001; Krause et al.,

2008; Forrest et al., 2008). However, the amount of spontaneously and progressively

occurring hindlimb movements normally found in Tx mice (Lapointe et al., 2006; Ung et

al., 2007) could not explain levels of muscle mass and CSA values found in this study

since, clenbuterol-treated and clenbuterol+testosterone-treated animals displayed greater

levels of muscle growth than vehicle-treated mice, whereas locomotor recovery in all

groups was similar. In other words, fiber CSA increases were most probably directly

induced by clenbuterol and clenbuterol+testosterone.

Some reports have provided data suggesting that slow-twitch muscles may be more

sensitive to testosterone treatment than other types of muscles (Axell et al., 2006) whereas,

clenbuterol would primarily affect fast-twitch muscle groups (Ryall et al., 2002). Our

results provided additional evidence supporting the idea of differential effects between

testosterone and clenbuterol on slow- vs. fast-twitch fibers since we found that EDL fibers

were larger (CSA) in clenbuterol or clenbuterol+testosterone treated mice. However, the

combination of clenbuterol+testosterone induced greater hypertrophic effects (e.g., Soleus

mass and fiber CSA values) than testosterone alone suggesting that synergistic actions may

exist between these compounds.

Locomotor recovery

As mentioned earlier, a phenomenon called spontaneous hindlimb movement recovery has

been found to occur progressively (after one week post-Tx) in motor-complete paraplegic

mice (Lapointe et al., 2006; Ung et al., 2007). However, only very low levels of activity

were found including non-locomotor-like and locomotor-like movements of small

amplitude without weight bearing capabilities. In this study, clenbuterol, testosterone or

both did not enhance that level of spontaneous recovery since all groups (treated and

control Tx) displayed non-significantly different scores at 8 weeks post-Tx. This said, it is

147

likely that further muscle activity induced by specific locomotor training or

pharmacological approaches (CPG-activating drugs) would have affected locomotor

recovery levels (Ichiyama et al., 2008; Landry et al., 2006b; Fong et al., 2005). This is in

contrast with data found with clenbuterol on functional recovery in incompletely SCI

rodents (Zeman et al., 1999) probably because clenbuterol-induced effects in their study

was mainly produced by spinal cord repair-mediated functional recovery rather than

sublesional network-mediated property changes (e.g., increased CPG excitability).

Perspectives and future directions

Although testosterone propionate did not induce large effects on the SCI-related secondary

complications as shown in this study, this does not exclude the possibility that testosterone-

related compounds such as specific anabolic steroids may possibly induce significant

effects in SCI patients. In fact, oxandrolone (Anavar), an anabolic steroid with a high

therapeutic index (low androgenic side effects and high myogenic desired effects) was

found in SCI patients to improve skin wound healing (pressure sores) and blood

coagulation and homeostasis (Spungen et al., 2001; Kahn et al., 2006). Given the results

obtained in our study, this may even suggest that greater effects may perhaps be elicited

with combination treatments such as clenbuterol and anabolic steroids (e.g, Anavar) or

selective androgen receptor modulators (SARMs which are currently being developed by

pharmaceutical companies such as GlaxoSmithKline and Schering-Plough).

Acknowledgments

This study was supported by the National Science and Engineering Research Council of

Canada (NSERC). We wish to thank Dr. Nicolas Lapointe for his contribution to some

experiments and tissue collection.

Author disclosure statement

No competing financial interests exist

148

References

Agbenyega, E. T., Morton, R. H., Hatton, P. A., and Wareham, A. C. (1995). Effect of the

beta 2-adrenergic agonist clenbuterol on the growth of fast- and slow-twitch skeletal

muscle of the dystrophic (C57BL6J dy2J/dy2J) mouse. Comp Biochem Physiol C

Pharmacol Toxicol Endocrinol 111, 397-403.

Anderson, G., and Wilkins, E. (1977). A trial of clenbuterol in bronchial asthma. Thorax

32,717-719.

Antri, M., Orsal, D., and Barthe, J. Y. (2002). Locomotor recovery in the chronic spinal rat:

effects of long-term treatment with a 5-HT2 agonist. Eur J Neurosci 16, 467-476.

Apseloff, G., Girten, B., Walker, M., Shepard, D. R., Krecic, M. E., Stern, L. S., and

Gerber, N. (1993). Aminohydroxybutane bisphosphonate and clenbuterol prevent bone

changes and retard muscle atrophy respectively in tail-suspended rats. J Pharmacol Exp

Ther264, 1071-1078.

Axell, A. M., MacLean, H. E., Plant, D. R., Harcourt, L. J., Davis, J. A., Jimenez, M.,

Handelsman, D. J., Lynch, G. S., and Zajac, J. D. (2006). Continuous testosterone

administration prevents skeletal muscle atrophy and enhances resistance to fatigue in

orchidectomized male mice. Am J Physiol Endocrinol Metab 291, E506-516.

Bauman, W.A., Kahn, N.N., Grimm, D.R., and Spungen, A.M. (1999). Risk factors for

atherogenesis and cardiovascular autonomic function in persons with spinal cord injury.

Spinal Cord 37, 601-616.

Bauman, W.A., and Spungen, A.M. (2000). Metabolic changes in persons after spinal cord

injury. Phys Med Rehabil Clin N Am 11, 109-140.

149

Basso, D.M., Fisher, L.C, Anderson, A.J., Jakeman, L.B., McTigue, D.M., and Popovich,

P.G. (2006). Basso Mouse Scale for locomotion detects differences in recovery after spinal

cord injury in five common mouse strains. J Neurotrauma 23, 635-659.

Benson, D. W., Foley-Nelson, T., Chance, W. T., Zhang, F. S., James, J. H., and Fischer, J.

E. (1991). Decreased myofibrillar protein breakdown following treatment with clenbuterol.

J Surg Res 50, 1-5.

Bhasin, S., Storer, T. W., Berman, N., Callegari, C , Clevenger, B., Phillips, J., Bunnell, T.

J., Tricker, R., Shirazi, A., and Casaburi, R. (1996). The effects of supraphysiologic doses

of testosterone on muscle size and strength in normal men. N Engl J Med 335, 1-7.

Bloomfield, S. A., Girten, B. E., and Weisbrode, S. E. (1997). Effects of vigorous exercise

training and beta-agonist administration on bone response to hindlimb suspension. J Appl

Physiol 83, 172-178.

Brown, T. J., Khan, T., and Jones, K. J. (1999). Androgen induced acceleration of

functional recovery after rat sciatic nerve injury. Restor Neurol Neurosci 15, 289-295.

Carter, W. J., Dang, A. Q., Faas, F. H., and Lynch, M. E. (1991). Effects of clenbuterol on

skeletal muscle mass, body composition, and recovery from surgical stress in senescent

rats. Metabolism 40, 855-860.

Choo, J. J., Horan, M. A., Little, R. A., and Rothwell, N. J. (1992). Anabolic effects of

clenbuterol on skeletal muscle are mediated by beta 2-adrenoceptor activation. Am J

Physiol 263, E50-56.

Choong, K., Lakshman, K.M., and Bhasin, S. (2008). The physiological and

pharmacological basis for the ergogenic effects of androgens in elite sports. Asian J Androl

10,351-63.

150

Clark, M. J., Petroski, G. F., Mazurek, M. O., Hagglund, K. J., Sherman, A. K., Lammy, A.

B., Childers, M. K., and Acuff, M. E. (2008). Testosterone replacement therapy and motor

function in men with spinal cord injury: a retrospective analysis. Am J Phys Med Rehabil

87, 281-284.

Cox, S. A., Weiss, S. M., Posuniak, E. A., Worthington, P., Prioleau, M., and Heffley, G.

(1985). Energy expenditure after spinal cord injury: an evaluation of stable rehabilitating

patients. J Trauma 25, 419-423.

Criswell, D. S., Powers, S. K., and Herb, R. A. (1996). Clenbuterol-induced fiber type

transition in the soleus of adult rats. Eur J Appl Physiol Occup Physiol 74, 391-396.

Cruse, J.M., Lewis, R.E., Dilioglou, S., Roe, D.L., Wallace, W.F., and Chen, R.S. (2000).

Review of immune function, healing of pressure ulcers, and nutritional status in patients

with spinal cord injury. J Spinal Cord Med 23, 129-135.

Dudgeon, W. D., Phillips, K. D., Carson, J. A., Brewer, R. B., Durstine, J. L., and Hand, G.

A. (2006). Counteracting muscle wasting in HIV-infected individuals. HIV Med 7, 299-

310.

Ferrando, A. A., Sheffield-Moore, M., Wolf, S. E., Herndon, D. N., and Wolfe, R. R.

(2001).

Testosterone administration in severe burns ameliorates muscle catabolism. Crit Care Med

29, 1936-1942.

Fong, A. J., Cai, L. L., Otoshi, C. K., Reinkensmeyer, D. J., Burdick, J. W., Roy, R. R., and

Edgerton, V. R. (2005). Spinal cord-transected mice learn to step in response to quipazine

treatment and robotic training. J Neurosci 25, 11738-11747.

Forrest, G. F., Sisto, S. A., Barbeau, H., Kirshblum, S. C , Wilen, J., Bond, Q., Bentson, S.,

Asselin, P., Cirnigliaro, C. M., and Harkema, S. (2008). Neuromotor and musculoskeletal

151

responses to locomotor training for an individual with chronic motor complete AIS-B

spinal cord injury. J Spinal Cord Med 31, 509-521.

Gerrits, H. L., de Haan, A., Sargeant, A. J., van Langen, H., and Hopman, M. T. (2001).

Peripheral vascular changes after electrically stimulated cycle training in people with spinal

cord injury. Arch Phys Med Rehabil 82, 832-839.

Giangregorio, L., and McCartney, N. (2006). Bone loss and muscle atrophy in spinal cord

injury: Epidemiology, fracture prediction, and rehabilitation strategies. J Spinal Cord Med

29,489-500.

Graham, M.R., Davies, B., Grace, F.M., Kicman, A., and Baker, J.S. (2008). Anabolic

steroid use: patterns of use and detection of doping. Sports Med 38, 505-25.

Gregory, C. M., Vandenbome, K., Huang, H. F., Ottenweller, J. E., and Dudley, G. A.

(2003). Effects of testosterone replacement therapy on skeletal muscle after spinal cord

injury. Spinal Cord 41, 23-28.

Guertin, P. A. (2005). Semiquantitative assessment of hindlimb movement recovery

without intervention in adult paraplegic mice. Spinal Cord 43, 162-166.

Herbst, K.L., and Bhasin, S. (2004). Testosterone action on skeletal muscle. Curr Opin Clin

Nutr Metab Care 7, 271-277.

Herrera, N. M., Jr., Zimmerman, A. N., Dykstra, D. D., and Thompson, L. V. (2001).

Clenbuterol in the prevention of muscle atrophy: a study of hindlimb-unweighted rats. Arch

Phys Med Rehabil 82, 930-934.

Huang, H. F., Li, M. T., Giglio, W., Anesetti, R., Ottenweller, J. E., and Pogach, L. M.

(1999). The detrimental effects of spinal cord injury on spermatogenesis in the rat is

152

partially reversed by testosterone, but enhanced by follicle-stimulating hormone.

Endocrinology 140, 1349-1355.

Ichiyama, R. M., Courtine, G., Gerasimenko, Y. P., Yang, G. J., van den Brand, R., Lavrov,

I. A., Zhong, H., Roy, R. R., & Edgerton, V. R. (2008). Step training reinforces specific

spinal locomotor circuitry in adult spinal rats. J Neurosci, 28, 7370-7375.

Junker, V., Becker, A., Huhne, R., Zembatov, M., Ravati, A., Culmsee, C , and Krieglstein,

J. (2002). Stimulation of beta-adrenoceptors activates astrocytes and provides

neuroprotection. Eur J Pharmacol 446, 25-36.

Kahn, N.N., Sinha, A.K., Spungen, A.M., and Bauman, W.A. (2006). Effects of

oxandrolone, an anabolic steroid, on hemostasis. Am J Hematol 81, 95-100.

Katznelson, L., Finkelstein, J. S., Schoenfeld, D. A., Rosenthal, D. I., Anderson, E. J., and

Klibanski, A. (1996). Increase in bone density and lean body mass during testosterone

administration in men with acquired hypogonadism. J Clin Endocrinol Metab 81, 4358-

4365.

Kearns, C. F., McKeever, K. H., Malinowski, K., Struck, M. B., and Abe, T. (2001).

Chronic administration of therapeutic levels of clenbuterol acts as a repartitioning agent. J

Appl Physiol 91, 2064-2070.

Koopman, R., Ryall, J.G., Church, J.E., and Lynch, G.S. (2009). The role of beta-

adrenoceptor signalling in skeletal muscle: therapeutic implications for muscle wasting

disorders. Curr Opin Clin Nutr Metab Care 12, 601-6.

Krause, P., Szecsi, J., and Straube, A. (2008). Changes in spastic muscle tone increase in

patients with spinal cord injury using functional electrical stimulation and passive leg

movements. Clin Rehabil 22, 627-634.

153

Kujawa, K. A., Kinderman, N. B., and Jones, K. J. (1989). Testosterone-induced

acceleration of recovery from facial paralysis following crush axotomy of the facial nerve

in male hamsters. Exp Neurol 105, 80-85.

Landry, E., Frenette, J., and Guertin, P. A. (2004). Body weight, limb size, and muscular

properties of early paraplegic mice. J Neurotrauma 21, 1008-1016.

Landry, E. S., Lapointe, N. P., Rouillard, C , Levesque, D., Hedlund, P. B., and Guertin, P.

A. (2006). Contribution of spinal 5-HT 1A and 5-HT7 receptors to locomotor-like

movement induced by 8-OH-DPAT in spinal cord-transected mice. Eur J Neurosci 24, 535-

546.

Lapointe, N. P., Ung, R. V., Bergeron, M., Cote, M., and Guertin, P. A. (2006). Strain-

dependent recovery of spontaneous hindlimb movement in spinal cord transected mice

(CD1, C57BL/6, BALB/c).

Behav Neurosci 120, 826-834.

Lieber, R.L, Johansson, C.B., Vahlsing, H.L., Hargens, A.R., and Feringa, E.R. (1986a).

Long-term effects of spinal cord transection on fast and slow rat skeletal muscle. 1.

Contractile properties. Exp Neurol 91,423-34.

Lieber, R.L., Friden, J.O., Hargens, A.R., and Feringa, E.R. (1986b). Long-term effects of

spinal cord transection on fast and slow rat skeletal muscle. II. Morphometric properties.

Exp Neurol 91, 435-48.

Lynch, G. S., Hinkle, R. T., and Faulkner, J. A. (1999). Year-long clenbuterol treatment of

mice increases mass, but not specific force or normalized power, of skeletal muscles. Clin

Exp Pharmacol Physiol 26, 117-120.

Lynch, G.S., and Ryall, J.G. (2008). Role of beta-adrenoceptor signaling in skeletal muscle:

implications for muscle wasting and disease. Physiol Rev 88, 729-767.

154

MacLennan, P. A., and Edwards, R. H. (1989). Effects of clenbuterol and propranolol on

muscle mass. Evidence that clenbuterol stimulates muscle beta-adrenoceptors to induce

hypertrophy. Biochem J 264, 573-579.

Modlesky, C. M., Majumdar, S., Narasimhan, A., and Dudley, G. A. (2004). Trabecular

bone microarchitecture is deteriorated in men with spinal cord injury. J Bone Miner Res 19,

48-55.

Oishi, Y., Imoto, K, Ogata, T., Taniguchi, K, Matsumoto, H., and Roy, R. R. (2002).

Clenbuterol induces expression of multiple myosin heavy chain isoforms in rat soleus

fibres. Acta Physiol Scand 176, 311-318.

Pansarasa, O., D'Antona, G., Gualea, M. R., Marzani, B., Pellegrino, M. A., and Marzatico,

F. (2002). "Oxidative stress": effects of mild endurance training and testosterone treatment

on rat gastrocnemius muscle. Eur J Appl Physiol 87, 550-555.

Picard, S., Lapointe, N. P., Brown, J. P., and Guertin, P. A. (2008). Histomorphometric and

densitometric changes in the femora of spinal cord transected mice. Anat Rec (Hoboken)

291,303-307.

Picquet, F., De-Doncker, L., and Falempin, M. (2004). Enhancement of hybrid-fiber types

in rat soleus muscle after clenbuterol administration during hindlimb unloading. Can J

Physiol Pharmacol 82, 311-318.

Rouleau, P., Ung, R. V., Lapointe, N. P., and Guertin, P. A. (2007). Hormonal and

immunological changes in mice after spinal cord injury. J Neurotrauma 24, 367-378.

Ryall, J. G., Gregorevic, P., Plant, D. R., Sillence, M. N., and Lynch, G. S. (2002). Beta 2-

agonist fenoterol has greater effects on contractile function of rat skeletal muscles than

clenbuterol. Am J Physiol Regul Integr Comp Physiol 283, R1386-1394.

155

Sinha-Hikim, I., Artaza, J., Woodhouse, L., Gonzalez-Cadavid, N., Singh, A.B., Lee, M.I.,

Storer, T.W., Casaburi, R., Shen, R., and Bhasin, S. (2002). Testosterone-induced increase

in muscle size in healthy young men is associated with muscle fiber hypertrophy. Am J

Physiol Endocrinol Metab 283, El54-164.

Shahidi, N.T. (2001). A review of the chemistry, biological action, and clinical applications

of anabolic-androgen steroids. Clin Ther 23, 1355-90.

Snyder, P. J., Peachey, H, Berlin, J. A., Hannoush, P., Haddad, G., Dlewati, A., Santanna,

J., Loh, L., Lenrow, D. A., Holmes, J. H., Kapoor, S.C., Atkinson, L.E., and Strom, B.L.

(2000). Effects of testosterone replacement in hypogonadal men. J Clin Endocrinol Metab

85, 2670-2677.

Spungen, A.M., Koehler, K.M., Modeste-Duncan, R., Rasul, M., Cytryn, A.S., and

Bauman, W.A. (2001). 9 clinical cases of nonhealing pressure ulcers in patients with spinal

cord injury treated with an anabolic agent: a therapeutic trial. Adv Skin Wound Care 14,

139-44.

Spurlock, D.M., McDaneld, T.G., and Mclntyre, L.M. (2006). Changes in skeletal muscle

gene expression following clenbuterol administration. BMC Genomics 7, 320.

Strauss, I., and Lev-Tov, A. (2003). Neural pathways between sacrocaudal afférents and

lumbar pattern generators in neonatal rats. J Neurophysiol 89, 773-784.

Stuermer, E. K., Sehmisch, S., Tezval, M., Tezval, H., Rack, T., Boekhoff, J., Wuttke, W.,

Herrmann, T. R., Seidlova-Wuttke, D., and Stuermer, K. M. (2009). Effect of testosterone,

raloxifene and estrogen replacement on the microstructure and biomechanics of

metaphyseal osteoporotic bones in orchiectomized male rats. World J Urol.

156

Teng, Y. D., Choi, H., Huang, W., Onario, R. C , Frontera, W. R., Snyder, E. Y., and

Sabharwal, S. (2006). Therapeutic effects of clenbuterol in a murine model of amyotrophic

lateral sclerosis. Neurosci Lett 397, 155-158.

Ung, R. V., Lapointe, N. P., Tremblay, C , Larouche, A., and Guertin, P. A. (2007).

Spontaneous recovery of hindlimb movement in completely spinal cord transected mice: a

comparison of assessment methods and conditions. Spinal Cord 45, 367-379.

Yimlamai, T., Dodd, S. L., Borst, S. E., and Park, S. (2005). Clenbuterol induces muscle-

specific attenuation of atrophy through effects on the ubiquitin-proteasome pathway. J Appl

Physiol 99, 71-80.

Zehnder, Y., Luthi, M., Michel, D., Knecht, H., Perrelet, R., Neto, I., Kraenzlin, M., Zach,

G., and Lippuner, K. (2004). Long-term changes in bone metabolism, bone mineral density,

quantitative ultrasound parameters, and fracture incidence after spinal cord injury: a cross-

sectional observational study in 100 paraplegic men. Osteoporos Int 15, 180-189.

Zeman, R. J., Feng, Y., Peng, H., and Etlinger, J. D. (1999). Clenbuterol, a beta(2)-

adrenoceptor agonist, improves locomotor and histological outcomes after spinal cord

contusion in rats. Exp Neurol 159, 267-273.

Zeman, R. J., Hirschman, A., Hirschman, M. L., Guo, G., and Etlinger, J. D. (1991).

Clenbuterol, a beta 2-receptor agonist, reduces net bone loss in denervated hindlimbs. Am J

Physiol 261, E285-289.

Zeman, R. J., Peng, H., Danon, M. J., and Etlinger, J. D. (2000). Clenbuterol reduces

degeneration of exercised or aged dystrophic (mdx) muscle. Muscle Nerve 23, 521-528.

Zeman, R. J., Peng, H., and Etlinger, J. D. (2004). Clenbuterol retards loss of motor

function in motor neuron degeneration mice. Exp Neurol 187, 460-467.

157

Zeman, R. J., Zhang, Y., and Etlinger, J. D. (1997). Clenbuterol, a beta2-adrenoceptor

agonist, reduces scoliosis due to partial transection of rat spinal cord. Am J Physiol 272,

E712-715.

Zhao, W., Pan, J., Wang, X., Wu, Y., Bauman, W. A., and Cardozo, C. P. (2008a).

Expression of the

muscle atrophy factor muscle atrophy F-box is suppressed by testosterone. Endocrinology

149, 5449-5460.

Zhao, W., Pan, J., Zhao, Z., Wu, Y., Bauman, W. A., and Cardozo, C. P. (2008b).

Testosterone protects against dexamethasone-induced muscle atrophy, protein degradation

and MAFbx upregulation. J Steroid Biochem Mol Biol 110, 125-129.

Zhu, Y., Culmsee, C , Roth-Eichhorn, S., and Krieglstein, J. (2001). Beta(2)-adrenoceptor

stimulation enhances latent transforming growth factor-beta-binding protein-1 and

transforming growth factor-beta 1 expression in rat hippocampus after transient forebrain

ischemia. Neuroscience 107, 593-602.

158

Figures and Legends

Fig. 1 For the first 7 days, body weights were averaged from all mice in all subgroups

combined (A). For (B-E), only weights from mice that were sacrificed on that specific

week were averaged. During the first 7 days, mice that received Cb, Tp and Cb + Tp

showed lower body weight loss than control animals. After the 4th week, Cb and Cb + Tp

had bigger weight gain. * P < 0.05, ** P < 0.01. (F) Average body weight increase per

week after spinal cord transection, in Ctr Tx, Cb, Tp, Cb + Tp and in uninjured mice.

Fig. 2 Soleus (A) and EDL (B) muscle mass. Comparisons between subgroups (treatment

durations). * P < 0.05, ** P < 0.01, *** P < 0.001.

Fig. 3 EDL fiber phénotype distribution. Comparisons between subgroups (treatment

durations). (A) type II and (B) hybrid fibers is shown throughout the 8 week period. * P <

0.05, ** P < 0.01, *** P < 0.001.

Fig. 4 Soleus fiber phénotype distribution. Comparisons between subgroups (treatment

durations). (A) type I, (B) type II and (C) hybrid fibers throughout the 8 week period. (D)

Typical example showing immunohistolabeling in soleus muscle, from a control Tx mouse

after 1 (I-III) and 8 (IV-VI) weeks. * P < 0.05, ** P < 0.01, *** P < 0.001.

Fig. 5 EDL fiber CSA values. Comparisons between subgroups (treatment durations). (A)

type II and (B) hybrid fibers throughout the 8 week period. * P < 0.05, **P<0.01,***P<

0.001.

Fig. 6 Soleus fiber CSA values. Comparisons between subgroups (treatment durations). (A)

type I, (B) type II and (C) hybrid fibers. * P < 0.05, ** P < 0.01, *** P < 0.001.

Table 1 Whole muscle mass (% of body weight), fiber number (% of total fiber) and fiber

CSA (pm ) from EDL and soleus are shown at 1,2 4 and 8 weeks. Comparison between

treatments revealed significant differences (P < 0.05, bold). For post-hoc analyses,

159

compared to control Tx mice * P < 0.05, ** P < 0.01, *** P < 0.001; compared to

clenbuterol treated mice # P < 0.05, ## P < 0.01, ### P < 0.001; and compared to

testosterone treated mice t P < 0.05, t t P < 0.01, t t t P < 0.001.

Table 2 Adipose tissue results expressed in % of body weight. Significant differences

(bold) were only noted for total adipose tissue as all specific area showed similar

differences. For post-hoc analyses, compared to control Tx mice * P < 0.05, ** P < 0.01,

*** P < 0.001, compared to clenbuterol treated mice t t t P < 0.001.

Table 3 All groups showed significant recovery from the 1st to the 8th week (P < 0.05). No

difference between treatments was found after 8 week.

160

B f 100 5 95-

E >• f i 90 > ni

f 85 £ 80

- © - Control Tx - © - Clenbuterol - © - Testosterome - © - Clenbuterol + Testosterone

g 105,

| 100.

Jf ô m xi 5 râ

95-

s 90 85

1 3 5 7 Time (days)

1 s t week 2 n d week 105-i

100-

5 I n 9 5 ' 5 S ° 90. l * ë ë

o c

CtrTx Cb Tp Cb + Tp CtrTx Cb Tp Cb + Tp

4^- week 110-i

8 th week * ** ** x r5 I

. - J 105-

' B 1 0 0 ' 5 95-

** x r5 1 *

CtrTx Cb Tp Cb + Tp CtrTx Cb Tp Cb + Tp

Figi re. 1

3 g tn CD 0)

ë 1

Average body weight gain per week

CJ* & x > * ^

sC

o II / a /

* 1 * sîi

>°

c / * / * \ * \ * a

LU

* Ifl J *

2

o

T -

o d n a.

* * * X

o _ o

3

(lH&a/w Apoq (0 %) W6K»M

CO (in&8« Apoqjo %)

162

A i 100

? r 95

| 90

| 85-

c 80

II 100

g 9b

a) -Q b - J

90 C

1 85

80

III 100

? r 95 E f 90 c I 85 iZ

80 IV

100-, ? r 95 È I 90 c j j 85 L.

80

Type II control Tx

T *

clenbuterol

— n **

i I f

I * * i

n **

i I

**

B

*** I

I s

*** I *** I I s

*** I .

I s

1

clenbuterol + testosterone

I ***

2 4 Weeks

15

10

5

0

I 15n

10

5

0

III 15

10

5

0

Type I + II control Tx

** 1 I —

** " "1 1 * - i - i 1

clenbuterol

testosterone

IV clenbuterol + testosterone 15n

Figure. 3

163

Ë

ê

ill

60

40

20

0

60

40

20-

Type I control Tx B,

clenbuterol

"f i

60 testosterone

5 40 a

1 20

5 5

60

40

20

0

I 60

40

20

0

60

40

20

Type II control Tx

| * * * T

*** x *

1 T *** \ ?

T . 1

clenbuterol " *

ï -\ ï ** 1

1 T

\ ***

1

80-

60-

40

20

0

I 80-1

60-

40-

20- i

Type I + Il control Tx

clenbuterol

testosterone

IV

g I E 3

60

40

20

clenbuterol + testosterone |V clenbuterol + testosterone |V 60

| *** ;

40

20 H

0

80-,

60-

40-

20-

0.

80 T

60

40-|

20

0

testosterone

*** z *** _ I

* 1 I I

* 1

i ***

clenbuterol + testosterone

***

r 1 2 4

Weeks

Control MHCs

2 4 Weeks

MHCf

2 4 Weeks

- Type I fiber -Type II fiber - * Type I + II fiber

Figure. 4

164

_1300

w 900-O | 700-u.

500-

Type II

control Tx

II clenbuterol _1300 CM

I 1100-

O 900

i— tu

-O 700 LL

500

testosterone

B

*** .

1

*** ***,'

1

*** *** | *** H

1 1

*** 1 1

I 700!

600

500

400

300

II ***

*** i

1 I~J

. ^*_ r

700 n

600

500

400

300

^ 1 3 0 0 -

| 1100- *** w 900-O

*** m w 900-O II

•

1 * * * l l

]> 700-iZ

son.

II • *** 1 ]> 700-

iZ son.

II •

7001

600

5001

400

300

IV

Type I + II control Tx

clenbuterol

*** ***

1 *

***

i

*** 1

** 1

*

*** ***

1

*

***

testosterone

*** i i

r ***l i i

r ***|

CN

E

1300-1

1100-

w 900 O jjj 700 u.

500

clenbuterol + testosterone IV *** *** I

***l *** * * * 1

l_ * * * 1

***

1 2 4 Weeks

700 600 500 400 300

— i i i i

clenbuterol + testosterone * * * T

*** ***

1 2 4 8 Weeks

Figure. 5

165

Type I control Tx

1000-,

E 3 800 < o I

6 0 0

clenbuterol

400

1000-,

E 3 800 < w o . 6 0 0

testosterone

400

B 1UUU-

E 3 800-

< *** Ï 6 0 0 -

*** _ 1 Ï 6 0 0 - 1 ***l

I

Li.

I

A n n .

I

1100-1

900-

700

500

Type II control Tx

*** _ «**

: •

1100

900

700

500-

clenbuterol

*** I *** 1

'' ***

*** ' *** *** ***

1100

900

700

500

testosterone

*** r i _ —

*** ***s * * * i

i _ —

***

IV 1000

c\T E 3 800 < w o | 800. i l

400'

clenbuterol + testosterone IV clenbuterol + testosterone 'V 1UU-

:

***

900-

:

*** *** 700-

500-: ***

* * *

: : * * *

: : 1 2 4

Weeks 2 4 Weeks

I 1200

1000

800

600

400

II 1200

1000

800

600

400

1200

1000

800

600

400

1200-

1000-

800-

600-

400-

III

Type I + II control Tx

*** 1

* * * _ * * * l l ***

clenbuterol

*** I I *** H i »

testosterone

*** ***

1*2*.. • „ ***

clenbuterol + testosterone

*** B - *** _ ***1 -

1 2 4 Weeks

Figure. 6

a 5 S

c o o

- C c/î c o ISI

a. S o

C3

E

ET o E 3 O

- ) Û LU

| E E 3

JZl

H

i -

- S * 5 i

s i

â = II

2

s i

tfi *

? ! d °

S § 2

.O v i OO d

w^ » i - : d

» ». 1^ o

» O

(N * 00 o

ï«-

-H

3 _•

H

i ^ d

-H V I » .

• -H • -H

3 » -H S O - ' § O »

( N ° 9

I" I: jo-qiunu J.x|i |

» r- £ -H r- •» * o =>

2-H : sM

\© rô (N —

2-H

M

: I-

VI " î ob *

: -H

«1 3 oc 53

O0 'Ç

<•". f . ^C o -f d r^ d r i O OC V " ,

-H +--H

O OC +- v-. s^

i -rn * * â r-'

c* *r.

H 44 - M s C r»« T r i r i c 3-'

-H + o « 00 - H ~ M ffi »

î -H

m 0O

S 2

t -H r -H ♦• « * s * » » î r i r * fc «

-H fc ».

+1

£'

-H

•*: — o i t r i

= — -H .

vs3 J«IU

o i d s

* s rs O

s O d

-H r-o ( N O

r-

© O d

-H

< * . £

o 3 O

o d

i r-cs o

r i

o d

-H CJ T f ( N O

<y,

© S

o d

4 j

CS ©

n

o d

+ r-

CS O

3 o

o d

-H r-O

O s o O d

-H G-

C4

o o d

-H ( N

o O o o d

41

oc oo 8 o O

o d

-H

5 o o d d

-H

§ O ©

d d

-H oo

O 3

d d

-H oo o 5 © d

tn

« 3 S J* C y:

-H « 4 4 o X — X

44

O <N

î '

m r*

m fc

4 ) W P

-H so r i

-H _ 00

oo =

H

— -» C

m. 00 ^ fi

1- "t d *

d «

H « »

» *.

d m

fc f " ! ^ <N

« H "O » fc •• C ^ < ^ O —

-H

d N

-H -H -H -,0 oo w i - ^ ^ 0ï ^j ^ O " » ( N «*1

» ». 00 *

£" S-S:

joqiunu J.x)i j

1-5

-H

ae M

3 =

-H

oo fc

* * "** . *" _ -H

S ^

oo o

: -H oe * S 2

* V ) fc

4H

3 2

• 0 00

oo a»

f i = .

4(

: -H

ffifc

» o v. oé

ï r j — * « w>

« - ! *

es oo z fc

i&!

B K X c

-T c" * f S

• ~ -H

. -H

« r- r *

* ri rK # m * 3C

+ 3 C c-K r i o =c

+ — - T

H r^! sC r^

> -H

« = * * ri g • m * s^

+ h» r i r^ r i *zt r~~

-H +

— v. r-ï 3* r~~ « + 2C \ r ,

» d i r , -C

^ -H

• r s n ! -

îd & -O

4 - tfî

•:- -H

':~ i r , sC

* sC 30

# •e

* * -H

* i r . r-

s oc

« + r i "T 3 oc I V ,

« -H

c r s " r i <N

T.

41 3 C C:

£ !> 9 - f

i-4- -H +- t* r-

* ^ d * *

44 <3 r-, r n sC r~~ -T

D ^ 5,

C' +

U E x

H E o

to ~o o

fc^J

o j y Q

y cq

'43 U 09 o a

'5 ca

E E

CZ]

r i j g -c

u H

c

U

Q

u

s - t

ê *

J £ u tu is

X)

U

o.

u

x;

U

D.

O

O 3 ' M 2 o .-oS «

5 o

OO o

1 =

3

OO o

6 °

o

o d

oo o d °

- î :

44

3 3

4*

is m o

2°

O o 2; o 6 =

—> O

-h -fl 41 -r o o S C* q

t r , o d d d

e d o

-H i * 44 fc - r i

fO r i

o r i o d

o o d d

r i o d

d d d

+ -H H m oc n

^r o o o O r i .c d d c d d d

-H oo -H

0 0 44 fc

oc o co m o c . d 9) d r j d

+ +1 r o " fc O o

r i d q d d ~ 4 o 4-1 *

fc r i r i

o

52

ON —

0 0 Csl

3 C

Û d

2"> -* o Z; d

2 o

ê l 1 1

2* oo — ^ d

ON

~2 ^r d o q — o oc --w

r i sO 1/1

d d d d d o d

-H oc +1 oc -H fc +

oc o ( f l >/-) o m -r ro d q d r i c r i d d d d r i

2s -» o

m 2

3 d rs

41 41 -H H ♦- 41 N O ro ro ro .;- i r . o o o fc t q fr- M r j "V " , r i sD

fr-»TA

d d d d d d d d d * *— 4)

o + r-0

41 i n 41 «* -H

oo r i o o oc o r i q O Cl d en d r i d oc d d d C r i

4H 44 -H +1 : - 41 ^ t o t n r i : -

O q <J~J o eo q r i q : -

N O r i o

: -

N C

d d d d o d d d d * "" -H

00 4i t r .

41 -r i 41 *

-o o TT o r i O N o r i r i

fc d n d q d —, d c*3 d d d —

o r i

4-

00 r i fc

n r i d

+ S

fc t r , d

fcl t n

5«

c r-

X

y

CS

y

00 C

CQ O <

C/3 O U < M)

# C "ça

-o u M BJ OJ uo

1> > O U U

o o E c y c

i l c

s E G

J c G

lO

oc

= .

H

a DC

-S

o c D

_ij

|É

if l

-S 00

5

r i

S » "

4i O

41 w-i q fci o

-H O

41 NO O . (N NO

4H

41

41

-H r i

y. C <

- t d

c-

d

d

t . d

41 CN

41

NO

~J2 d

41 O

a»

ca c <

169

C H A P I T R E V I I I - N O N - A S S I S T E D T R E A D M I L L T R A I N I N G D O E S N O T I M P R O V E

LOCOMOTOR RECOVERY AND BODY COMPOSITION IN SPINAL CORD-TRANSECTED MICE.

Ce chapitre avait pour but d'évaluer un protocole d'entraînement sur tapis roulant

qui n'impliquait aucune autre forme d'assistance ou de stimulation autre que celles

provenant du tapis roulant. Nous avons montré que sans assistance, l'entraînement sur tapis

roulant ne permet pas une amélioration des performances motrices. De plus les souris

entraînées ont montré une plus grande atrophie au niveau de l'aire des fibres muscaulaires

et ont perdu d'avantage de masse adipeuse. Ceci suggère que la récupération des fonctions

locomotrices et de l'amélioration de la constitution corporelle seraient dépendante d'une

assistance et/ou de stimulations adéquates. Cet article a été publié dans Spinal Cord, 2010,

Epub sorti avant impression.

Abstract

Study design: Experiments in a mouse model of complete paraplegia

Objectives: To evaluate the effect of non-assisted treadmill training on motor recovery and

body composition in completely spinal cord-transected mice.

Settings: Laval University Medical Center, Neuroscience Unit, Quebec city, Quebec,

Canada.

Methods: Following a complete low-thoracic (Th9/10) spinal transection (Tx), mice were

divided into two groups that were either untrained or trained with no assistance. Training

consisted of placing the mice during fifteen minutes with no further intervention (i.e., no

tail pinching or body weight support, etc.) on a motorized treadmill (8-10 cm/sec) five

times/week for five weeks. Locomotor performances were assessed weekly in both groups

using two complementary locomotor rating scales. After five weeks, all mice were

sacrificed and adipose tissue, soleus and extensor digitorum longus muscles were dissected

for analyses.

Results: No significant difference in locomotor performances or in muscle fiber type

conversion was found between trained and untrained mice. In contrast, body weight,

170

adipose tissue, whole muscle and individual fiber cross-sectional area (CSA) values were

significantly lower in trained compared to untrained animals.

Conclusion: Non-assisted treadmill training in these conditions did not improve motor

performances and contributed to further accentuate body composition changes post-Tx

suggesting that assistance provided manually, robotically or pharmacologically may be key

to spinal learning and recovery of locomotor function and body composition.

Keywords: Spinal cord injury, hindlimb movements, training, plasticity, learning muscle

atrophy.

Introduction

It has been well-established that adult paraplegic animals can recover some locomotor

functions using regular treadmill training (TT) typically assisted with sensory stimulation

and/or pharmacological aids (Guertin 2008; Rossignol et al., 2001). In fact, even without

pharmacological aids, regular TT assisted with a weight-supporting harness and/or tail or

sexual organ pinching can enable the expression of involuntary 'reflex' stepping in low-

thoracically spinal cord-transected (Tx) animals (Barbeau and Rossignol, 1987; Lovely et

al., 1986; Antri et al., 2002). However, without assisted training, some spontaneously

occurring small amplitude hindlimb movements were also shown to progressively occur a

few weeks post-Tx in untrained Tx mice (Guertin, 2005; Lapointe et al., 2006; Ung et al.,

2008), suggesting that basic spontaneous sublesional plasticity and spinal learning events

may occur even in absence of assisted training. Untrained Tx mice were also found to

undergo significant changes in body composition - e.g., body weight (-24%) and soleus

fiber properties (-32% in mass and maximal force) within a few days post-Tx (Landry et

al., 2004) whereas assisted TT was found to partially prevent muscular property changes in

Tx cats and rats (Dupont-Versteegden et al., 1998; Roy et al., 1998). All and all, it remains

unclear the extent to which TT, in its most basic form (i.e. without assistance), may

improve locomotor performances and prevent or restore body composition changes post-

Tx. In this study, we assessed and compared locomotor performances and body

composition (body weight, region-specific adipose tissue mass, muscle/fiber cross sectional

area and phénotype) between untrained Tx mice and non-assisted trained Tx mice.

171

Materials and Methods

Animal model and surgical procedures

All experimental procedures were conducted in accordance with the Canadian Council

for Animal Care guidelines and accepted by the Laval University Animal Care and Use

Committee. A total of 22 male CD1 mice (Charles River Canada, St-Constant, QC) initially

weighing 30-35 g were used for this study. Pre-operative cares included administration of

lactate-Ringer's solution (1 ml, s.c), an analgesic (0.1 mg/kg buprenorphine, s.c.) and an

antibiotic (5 mg/kg enrofloxacin, s.c). All surgical procedures were performed under

aseptic conditions. Mice were anesthetised with 2.5% isoflurane. A small incision was

made on their back in order to expose the mid-to-low thoracic vertebrae. The spinal cord

was then completely transected intervertebrally with microscissors inserted between the 9th

and 10th thoracic vertebrae (Guertin, 2005; Lapointe et al., 2006; Ung et al., 2008). To

ensure that complete Tx was achieved, the inner vertebral walls were explored and entirely

scraped with small scissors tips. The opened skin area was sutured and animals were placed

for a few hours on heating pads. Mice were left in their cage with food and water ad libitum

during 3 days post-surgery. Post-operative cares included administration of lactate-Ringer's

solution ( 2 x 1 ml/day, s.c), buprenorphine (2 x 0.1 mg/kg/day, s.c.) and enrofloxacin (5

mg/kg/day, s.c). Bladders were manually emptied twice daily for the first week and once a

day throughout the study. Complete Tx was confirmed by 1) initial full paralysis of the

hindlimbs, 2) post-mortem visual and microscopic examination of the spinal cord lesion,

and 3) histological examination of coronal or midsagittal spinal cord sections stained with

luxol fast blue/cresyl violet for myelinated axons and Nissl substance, respectively.

Training procedures

Animals were randomly divided into untrained Tx mice (control, N = 12), and non-

assisted TT Tx mice (N = 10). TT began on the 3rd day post-surgery (to allow a few days of

recovery post-surgery) and ended after 5 weeks. Trained mice were simply placed for 15

min on a motorized treadmill belt moving at 8-10 cm/sec without any other forms of

172

assistance or stimulation. A harness placed around the torso and attached ahead was used to

maintain the animals on the treadmill belt without providing weight support.

Assessment of locomotor function recovery

On the 3rd day, 1st, 2nd, 3rd 4th and 5th week post-Tx, hindlimb motor and locomotor

movements were assessed using two complementary methods. Each assessment session

was performed on the treadmill and lasted no more than two minutes to minimize potential

training-induced effects of the assessment per se in the untrained group. Hindlimb

movements were assessed 'live' using a qualitative motor scale, referred to as AOB, which

has been specifically developed for complete Tx rodents. In brief, the scale consists of 22

scores: 0 - no movement; 1 - weak limb jerks; 2 - weak rhythmic movements with no

bilateral alternation; 3 - large rhythmic movements with no bilateral alternation; 4 - weak

rhythmic movements with occasional bilateral alternation; 5 - large rhythmic movements

with occasional bilateral alternation. Higher levels of recovery are described in detail in the

original article (Antri et al., 2002). We also used a semi-quantitative assay called Average

Combined Score (ACOS) that has been developed in our laboratory for 'live' semi­

quantitative assessment of hindlimb movements in Tx rodents (Guertin, 2005; Lapointe et

al., 2006; Ung et al., 2008). ACOS is designed to assess non-locomotor movements (NLM,

number/min), locomotor-like movements (LM, number/min), and amplitude arithmetically

combined as follows: [NLM + (2 x LM)] x amplitude. One NLM was defined as a non-

bilaterally alternating movement including jerks, fast-paw shaking and twitches. One LM

was defined as a flexion followed by an extension (or vice versa) occurring bilaterally in

alternation, not necessarily including weight bearing capabilities. We did not consider LMs

or NLMs induced during bowel movements to avoid non-related afferent (sacral)

stimulation-induced movements (Strauss and Lev-Tov, 2003). Amplitude was characterized

by assigning one of three values; 0 - no movement; 1 -movements considered to be less

than half the range of motion of a normal step; 2 - movements considered to be greater than

half the range of motion of a normal step.

Muscle immunohistochemistry

173

Upon sacrifice, left soleus and extensor digitorum longus (EDL) muscles were

dissected, frozen in melting isopentane and stored at -80°C until further use. Serial cross

sections of 12[im-thick from the muscle mid-portion were cut with a cryostat maintained at

-20°C (2800E Frigocut, Leica Instruments, Germany) and mounted on Superfrost® plus

glass slides (VWR Canlab, Mississauga, ON, Canada). For individual fiber labelling

Myosin Heavy Chain slow or fast primary antibodies were used (MHCs or MHCf, specific

for MHC isoform type I and type II, respectively) (Vector Laboratories, Burlingame, CA).

First, cross sections were incubated 1 hour in a blocking solution containing 10% rabbit

serum, 1% triton X-100 and 0.1M phosphate-buffered saline (PBS). Cross sections were

then washed in PBS 0.1 M and incubated for 2 hours in a solution containing MHCs or

MHCf primary antibodies (dilution 1/50 in 0.1 M PBS containing 1% rabbit serum and 1%

triton X-100). Sections were rinsed in PBS 0.1 M before incubation with a goat anti-mouse

IgG(H+L) Alexa Fluor 488 secondary antibodies (Molecular Probes, Eugene, OR) (dilution

1/500 in 0.1M PBS containing 1% rabbit serum and 1% triton X-100). Slides were then

rinsed in PBS 0.1 M and mounted with PBS-Glycerol (50-50). Some sections were treated

as above, excepted that the primary antibodies were omitted from the incubation solution as

control. Immunofluorescence labelling was visualized on an Olympus BX61WI confocal

microscope with a lOx water-immersion objective. Images were captured using FluoView

300 (Olympus Canada Inc., Markham, ON, Canada) and analysed with ImageJ (ImageJ

1.40, Research Services Branch, NIH, Bethesda, MD). Analyses consisted of determining

soleus and EDL muscle fiber type composition, whole muscle cross sectional area (CSA)

and fiber CSA for fiber type I, II and I + II (type I + II is a hybrid fiber isoform, labelled

with both MHCs and MHCf antibodies). For the latter analysis, 50 fibers/isoform/muscle

were averaged. In cases where the number of fibers was less than 50 fibers/isoform/muscle,

CSA was calculated using all available fibers.

Data analyses

Comparisons between untrained and trained Tx mice for locomotor recovery levels

and body weight were assessed using a Two-Way ANOVA followed by a Bonferroni post-

hoc. Differences in adipose tissues, muscle and fiber CSA as well as fiber type relative

174

distribution were examined using an unpaired Student T-Test. Results are reported as Mean

±SEM.

Results

Hindlimb movement recovery assessed with the AOB and ACOS rating scale

Occasional hindlimb movements of small amplitude (NLMs and LMs with no weight

bearing capabilities) have already been shown to progressively occur spontaneously in

untrained Tx mice (Guertin, 2005; Lapointe et al., 2006; Ung et al., 2008). Here, Tx mice

from the untrained group displayed hindlimb movements corresponding to AOB scores no

greater than T (Fig.lA). Similar results were found in the trained mice. For both groups,

Two-Way ANOVA revealed a significant increase in score (P < 0.001). However, no

difference between groups were found throughout the 5 weeks (P = 0.723). Similarly, from

the 3rd day through the 5th week, scores from ACOS locomotor rating scale showed a

moderate but significant increase (Fig.IB, from 0 at 3 days post-Tx to less than 4 at 5

weeks post-Tx, Fig.IB, P = 0.002). Again, no difference between the untrained and trained

Tx mice was observed (P = 0.321). Note that no LM was found throughout this study (i.e.,

only NLMs mainly constituted of small amplitude unilateral flexions were observed).

Body weight and site-specific adipose tissue

Body weight was measured prior to surgery and before each assessment session

(Fig.2). During the 1st week, comparable data were found in untrained and trained Tx mice

which displayed significant body weight losses (P < 0.001). Subsequently, between the 2nd

and the 5th week, untrained Tx mice progressively recovered some weight (approx. 1

gram/week) whereas at 2 weeks post-Tx, trained Tx mice continued loosing weight (75.1 ±

1.2 %), prior to a progressive regain. Region-specific adipose tissue masses were examined

post-mortem in all animals. Tissues from the interscapular, subcutaneous abdominal and

intra-peritoneal regions were collected (Fig.3). When combined, total adipose tissue values

were found to be 29 % lower (P < 0.05) in the trained than in the untrained group (Fig.3A).

175

This difference was largely due to changes in intra-peritoneal fat tissue since no difference

was observed in subcutaneous abdominal and interscapular fat tissues.

Hindlimb muscle size and composition

Morphometric analyses of soleus (extensor) and EDL (flexor) fibers were performed to

assess the effects of non-assisted TT on hindlimb muscle atrophy, composition and fiber

type conversion levels. Both muscles displayed lower CSA values in the trained group

compared to the untrained group, although significant differences (P < 0.05) were found

only with EDL (Fig.4). In order to assess more specifically individual fiber type distribution

(conversion) and CSA, we also studied immunohistochemically-identified fibers from the

soleus and EDL muscles. It is well-known that muscle unloading typically induces fiber

phénotype conversion. For both of these muscles, the relative fiber type distribution was

comparable (P > 0.05) between groups (Fig. 5) suggesting that preservation or reversal of

fiber phénotype conversion did not occur in non-assisted TT Tx mice. In clear contrast

though, large differences between groups were found with respect to muscle fiber CSA. For

soleus and EDL, fibers of all phénotypes (type I, II and hybrid) displayed CSA values in the

trained Tx mice that were significantly lower than those in the untrained animals (Fig.6).

Discussion

This study revealed that Tx mice placed on a motorized treadmill with no other form

of assistance (non-assisted treadmill training) did not display greater locomotor

performances than the untrained Tx mice. In fact, this study clearly showed that this basic

form of training, under these conditions (15 minutes, 5 times per week), negatively affected

body composition. Indeed, body weight, adipose tissue, whole muscle and individual fiber

CSA values were lower in trained than untrained Tx mice, indicating greater muscular

atrophy and less fat tissue in non-assisted TT Tx mice.

Effects of treadmill training on locomotor recovery

It has been demonstrated that TT, combined with various forms of stimulation

(manual, robotic and/or pharmacological assistance) leads to some recovery of locomotor

176

functions. In this study, we determined to examine the effect of TT alone (without

assistance) on locomotor recovery. This experimental protocol was driven by the idea that

progressively occurring motor and locomotor-like movements displayed typically in

untrained Tx animals (Guertin, 2005; Lapointe et al., 2006; Ung et al., 2008), if combined

with the stimulation provided by the moving treadmill belt, may suffice to further promote

locomotor function recovery. However, as shown here, non-assisted TT did not enable the

expression of large amplitude 'reflex' stepping movements in Tx mice as AOB and ACOS

scores remained lower than 1 and 4, respectively (see Fig.l). In fact, involuntary LMs were

not even expressed after 5 weeks of training. The lack of differences in hindlimb movement

recovery between TT and untrained mice may be supported by the idea that this form of

basic training could have been inadequate. This is supported by data from Grau and

colleagues who suggested, using the electric shock model in Tx rats, that appropriate

stimulation is key to significant instrumental learning in the spinal cord (Grau et al., 2004).

Previous reports showed also, using different modes of training, that spinal learning is task-

dependent (Edgerton et al., 1997; Bigbee et al., 2007). Furthermore, some data suggest that

the spinal cord itself needs to have reached a significant level of excitability for weight-

bearing spinal stepping to be expressed (Rossignol et al., 2008; Edgerton et al., 2008). It

was also shown that the amount of hindlimb loading (using robotic-controlled weight

support assistance) as well as the amount of training (number of steps) significantly alter

stepping quality promoted by training (Timoszyk et al., 2005; Cha et al., 2007). Other

reasons why non-assisted TT mice did not displayed higher locomotor recovery values than

untrained Tx mice may be associated with mild skin irritations that can potentially induce

some pain. In Tx rats, there is evidence suggesting that central sensitization triggered by

nociceptive stimuli can interfere with instrumental spinal learning of motor task (Ferguson

et al., 2006). In the present study, none of the mice did show signs of hindlimb irritations.

However, it can not be ruled out that pain-related pathways may have been activated by

non-visually detectable signs of damage or inflammation. Furthermore, it is intuitive to

think that body degradation (decreased weight, muscle CSA, fat tissue) as seen in trained

Tx mice is unlikely to promote locomotor function recovery. It may be also that the early

timing of implementation for training was inappropriate. Indeed, in traumatic brain injury

models, it was found that motor recovery was enhanced when training begins only 2 to 3

177

weeks post-injury (Griesbach et al., 2004). Since we began training these animals at 3 days

post-Tx, this may have contributed to further body degradation as well as to the apparent

lack of spinal learning and locomotor function recovery found here. This said, other SCI

models reported that training starting immediately after partial SCI contributed to enhanced

locomotor function recovery (Nome et al., 2005).

Effects of treadmill training on body composition

Previous work from our laboratory has shown important body weight losses after Tx in

mice.9 One week after Tx, mice generally lose 25% of their body weight followed by a

gradual re-increase in weight subsequently. In the present study, untrained Tx mice showed

comparable weight losses whereas, surprisingly, TT Tx mice underwent greater losses and

subsequent lower weight re-increases. Given the results obtained with adipose tissue

changes in the trained animals, it is possible to conclude that some of their body weight

losses were due to a significant decrease of adipose tissue from the intraperitoneal area.

This could have been induced by training which is known generally to decrease body fat

levels (Irving et al., 2008). This greater weight loss in trained mice was also associated with

muscular atrophy, since TT Tx animals displayed further muscle atrophy than the untrained

animals (as seen with soleus and EDL lower muscle CSA values). This may perhaps be

taken as preliminary signs of overtraining or other stress-related factors. Another stress

factor that could contribute to weight loss is the harness used to keep the trained mice in a

perpendicular plan on the treadmill. Untrained mice were not subjected to this device. The

purpose of the harness is somewhat similar to a restrainer device, which can induce stress

in animals (Jarillo-Luna et al., 2007), at least in the initial days of training.

Conclusion

This study supports the idea that training-dependent spinal learning and related sublesional

plasticity in the lumbar spinal cord of completely low-thoracic spinal cord-transected

mammals does not simply depend upon training in its most basic form (non-assisted). In

fact, together with results from other studies, our findings strongly suggest that spinal

learning and recovery critically depend upon various training modalities including type,

quality, quantity and, perhaps, onset (post-injury).

178

Acknowledgements

This study was supported by the Natural Sciences and Engineering Research Council of

Canada (NSERC).

Conflict of interest

The authors declare no conflict of interest

179

References

Antri M, Orsal D, Barthe JY. Locomotor recovery in the chronic spinal rat: effects of long-

term treatment with a 5-HT2 agonist. Eur J Neurosci 2002; 16: 467-476.

Barbeau H, Rossignol S. Recovery of locomotion after chronic spinalization in the adult

cat. Brain Res 1987; 412: 84-95.

Bigbee AJ, Crown ED, Ferguson AR, Roy RR, Tillakaratne NJ, Grau JW, et al. Two

chronic motor training paradigms differentially influence acute instrumental learning in

spinally transected rats. Behav Brain Res 2007; 180, 95-101.

Cha J, Heng C, Reinkensmeyer DJ, Roy RR, Edgerton VR, De Leon RD. Locomotor

ability in spinal rats is dependent on the amount of activity imposed on the hindlimbs

during treadmill training. J Neurotrauma 2007; 24: 1000-1012.

Dupont-Versteegden EE, Houle JD, Gurley CM, Peterson CA. Early changes in muscle

fiber size and gene expression in response to spinal cord transection and exercise. Am J

Physiol 1998; 275: 1124-33.

Edgerton VR, Courtine G, Gerasimenko YP, Lavrov I, Ichiyama RM, Fong AJ, et al.

Training locomotor networks. Brain Res Rev 2008; 57: 241-254.

Edgerton VR, de Leon RD, Tillakaratne N, Recktenwald MR, Hodgson J A, Roy RR.. Use-

dependent plasticity in spinal stepping and standing. Adv Neurol 1997; 72: 233-247.

Ferguson AR, Crown ED, Grau JW. Nociceptive plasticity inhibits adaptive learning in the

spinal cord. Neuroscience 2006; 141: 421-431.

180

Grau JW, Washburn SN, Hook MA, Ferguson AR, Crown ED, Garcia G, et al.

Uncontrollable stimulation undermines recovery after spinal cord injury. J Neurotrauma

2004; 21: 1795-1817.

Griesbach GS, Hovda DA, Molteni R, Wu A, Gomez-Pinilla F. Voluntary exercise

following traumatic brain injury: brain-derived neurotrophic factor upregulation and

recovery of function. Neuroscience 2004, 125, 129-139

Guertin PA. A technological platform to optimize combinatorial treatment design and

discovery for chronic spinal cord injury. J Neurosci Res 2008; 86: 3039-51.

Guertin PA. Semi-quantitative assessment of hindlimb movement recovery without

intervention in adult paraplegic mice. Spinal Cord 2005; 43: 162-6.

Irving BA, Davis CK, Brock DW, Weltman JY, Swift D., Barrett EJ. Effect of exercise

training intensity on abdominal visceral fat and body composition. Med Sci Sports Exerc

2008; 40: 1863-1872.

Jarillo-Luna A, Rivera-Aguilar V, Garfias HR, Lara-Padilla E, Kormanovsky A, Campos-

Rodriguez R. Effect of repeated restraint stress on the levels of intestinal IgA in mice.

Psychoneuroendocrinology 2007; 32: 681-692.

Landry E, Frenette J, Guertin PA. Body weight, limb size, and muscular properties of early

paraplegic mice. J Neurotrauma 2004; 21: 1008-1016.

Lapointe NP, Ung RV, Bergeron M, Côté M, Guertin PA. Strain-dependent recovery of

spontaneous hindlimb movement in spinal cord transected mice (CD1, C57BL/6, BALB/c).

Behav Neurosci 2006; 120: 826-834.

Lovely RG, Gregor RJ, Roy RR, Edgerton VR. Effects of training on the recovery of full-

weight-bearing stepping in the adult spinal cat. Expe Neurol 1986; 92: 421-435.

181

Norrie BA, Nevett-Duchcherer JM, Gorassini MA. Reduced functional recovery by

delaying motor training after spinal cord injury. J Neurophysiol 2005; 94: 255-264.

Rossignol S, Barrière G, Frigon A, Barthélémy D, Bouyer L, Provencher J, et al. Plasticity

of locomotor sensorimotor interactions after peripheral and/or spinal lesions. Brain Res Rev

2008; 57: 228-240.

Rossignol S, Giroux N, Chau C. Marcoux J, Brustein E, Reader TA. Pharmacological aids

to locomotor training after spinal cord injury. J Physiol 2001 ; 533: 65-74.

Roy RR, Talmadge RJ, Hodgson JA, Zhong H, Baldwin KM, Edgerton VR. Training

effects on soleus of cats spinal cord transected (T12-13) as adults. Muscle & Nerve 1998;

21: 63-71.

Strauss I, Lev-Tov A. Neural pathways between sacrocaudal afférents and lumbar pattern

generators in neonatal rats. J Neurophysiol 2003; 89: 773-84.

Timoszyk WK, Nessler JA, Acosta C, Roy RR, Edgerton VR, Reinkensmeyer DJ, et al.

Hindlimb loading determines stepping quantity and quality following spinal cord

transection. Brain Res 2005; 1050: 180-189.

Ung RV, Lapointe NP, Tremblay C, Larouche A, Guertin PA. Spontaneous recovery of

hindlimb movement in completely spinal cord transected mice: a comparison of assessment

methods and conditions. Spinal Cord 2007; 45: 367-379.

182

Figures and legends

Figure 1. Hindlimb locomotor movements assessed with the Antri-Orsal-Barthe (AOB)

scale and Average combined Scores (ACOS). On the 3rd day, none of the groups showed

hindlimb movements. Both AOB and ACOS score showed moderate, but significant

increase in score over time. However, no difference between groups was found.

Figure 2. Body weight values in untrained and trained spinal cord Tx mice. Values are

reported as percentages of pre-surgery values. On the 1st week, values ranged between 76.8

± 1.2 % and 78.0 ± 1.4 %. On the 5th week weight from untrained and trained mice reached

85.2 ± 0.8 % and 80.4 ± 1.9 %, respectively. ** P < 0.01, *** P < 0.001

Figure 3. Adipose tissues from the intra-peritoneal (B), subcutaneous abdominal (C) and

inter-scapular (D) regions were collected and weighed immediately after sacrifice.

Differences between groups for the total adipose tissue changes (A) were attributed mainly

to changes in intra-peritoneal fat. * P < 0.05, ** P < 0.01

Figure 4. Whole muscle cross-sectional area (CSA) measurements of (A) soleus and (B)

EDL. Specifically, soleus muscles CSA values from the untrained and trained Tx mice

reached, 0.56 ± 0.04 mm2 and 0.45 ± 0.02 mm2, respectively. In EDL, muscle CSA values tj -y

were of 0.95 ± 0.07 mm and 0.73 ± 0.07 mm (untrained and trained mice, respectively)

which was significantly different. * P < 0.05.

Figure 5. Soleus and EDL individual fiber relative distribution. MHC type I (A), type II

(B, D) and type I + II or hybrid (C, D) showed no difference between groups. Untrained

and trained groups displayed over 2-fold higher increases in percentages of Type I + II

fibers (ranged between 60.3 ± 6.5 % and 69.5 ±3.1 %) than Type I (ranged between 4.7 ±

1.6 % and 7.0 ± 3.0 %) or Type II fibers (ranged between 25.8 ± 3.1 % and 33.5 ± 5.3 %).

In contrast, for EDL where no Type I fiber was found, there were more Type II (~ 98.5 %)

than Type I + II fibers (~ 1.5 %).

183

Figure 6. Soleus and EDL individual fiber cross-sectional area (CSA) values.

Immunolabelling for MHC type I (A), type II (B, D) and type I + D or hybrid (C, D). All

non-assisted TT mice showed significantly lower fiber CSA than untrained mice. * P <

0.05, ** P < 0.01, *** P < 0.001.

184

AOB

2 3 Weeks

untrained Tx mice trained Tx mice

Figure 1.

1 0 0 ,

185

* * * * *

Untrained Tx mice Trained Tx mice

Figure 2.

186

Total adipose tissue

O) 1.00

«5 0.75

m 050

0.25

0.00

Subcutaneous abdominal fat

B Intra-peritoneal fat

a 0.4

<0 n 1 eo 0.1

Inter-scapular fat

Untrained Tx mice Trained Tx mice

Figure 3.

187

A Soleus 0.81

B EDL

eo 0.2 H

Untrained Tx mice Trained Tx mice

Figure 4.

188

Soleus EDL

Type I

& 10

B Type II D Type II ~ 80T _ 100

Type I + Il Type I + Il

g 8 0 " | 6 0 '

Ë AO-

iï 20-11

Untrained Tx mice Trained Tx mice

Figure 5.

189

Soleus Type I

EDL

800

•S. 700

< 600 O v 500

B Type II

800

E 700

w 600 O i— S S 500

Type

1000

E 900

co 800

a> S 700

Type I + Type I + Il

600

500

o aS 400

Untrained Tx mice Trained Tx mice

Figure 6.

190

C H A P I T R E I X - F U N C T I O N A L A N D P H Y S I O L O G I C A L E F F E C T S I N D U C E D B Y

SPINAL LOCOMOTOR NETWORK-ACTIVATING TRITHERAPY AND CLENBUTEROL IN PARAPLEGIC MICE

La cinquième étude consistait à évaluer les effets de l'entraînement locomoteur

induit par l'activation du CPG en administrant de la buspirone + L-DOPA + Carbidopa et

par la supplementation de clenbuterol. Les effets étaient observés au niveau de la

récupération locomotrice, de la composition corporelle et des propriétés musculaires de

souris paraplégiques. Les résultats nous ont permis de noter qu'il y avait un rétablissement

locomoteur graduel et que l'ajout de clenbuterol permettait d'augmenter la masse

musculaire au niveau des pattes arrière. Cependant l'ajout de clenbuterol semblait favoriser

la dégradation osseuse au niveau du fémur. Cet article est en révision dans Journal of

Neurotrauma (2010).

Abstract

A complete transection (Tx) of the low-thoracic spinal cord entirely eliminates

voluntary motor and locomotor functions below injury level. This state of paralysis is

generally accompanied of secondary health problems and life-threatening complications.

We have recently demonstrated that the central pattern generator (CPG) for locomotion

located at the lumbar cord level can be potently reactivated temporarily following

administration of a novel tritherapy in low-thoracic Tx animals. Here, we investigated the

effects of an eight week-treatment (3 times/week) of this tritherapy composed of buspirone,

levodopa and carbidopa in clenbuterol-treated Tx mice. We found in non-assisted and

previously untrained Tx mice that weight-bearing stepping induced by the tritherapy

significantly improved over time although no significant difference was reported between

clenbuterol-treated (daily injected) and non-clenbuterol-treated groups. Soleus muscle

mass, soleus and extensor digitorum longus fiber cross-sectional only increase in

tritherapy-treated Tx mice, whereas fiber type conversion, bone mineral density remained

unchanged between Tx groups. These findings suggest that locomotor recovery and

specific muscle properties can be largely enhanced within a few weeks by combining both

191

tritherapy-elicited locomotor training and anabolic aids such as clenbuterol in completely

paraplegic animals.

Keywords

Spinal cord injury, CPG, locomotor training, locomotor recovery, 5-HT, L-DOPA, muscle

atrophy

Abbreviations

Average Combined Score locomotor rating scale (ACOS); Antri-Orsal-Barthe locomotor

rating scale (AOB); Bone mineral content (BMC); Bone mineral density (BMD); Central

pattern generator (CPG); Coefficient of variation (CV); Cross-sectional area (CSA); Dual-

energy X-ray absortiometry (DEXA); Extensor digitorum longus (EDL); Locomotor

movement (LM); Non-locomotor movement (NLM); Phosphate-buffered saline (PBS);

Spinal cord injury (SCI); Spinal cord transaction (TX).

192

Introduction

Spinal cord injury (SCI) generally leads to a partial or complete loss of motor and

sensory functions below the level of injury. For over two decades, sensori-stimulation

(e.g., tail pinching, skin rubbing and/or sexual organs squeezing) and body weight-support

approaches have been shown with or without serotonin (5-HT) or alpha-2 adrenergic

receptor ligands to promote the recovery of involuntary hindlimb stepping activity in

completely spinal cord transected (Tx) cats (Lovely et al., 1986; Barbeau and Rossignol,

1987). Some of these findings have supported the development of related rehabilitation

strategies in incompletely SCI patients (Wernig et al., 1992; Barbeau and Fung 2001;

Dobkin et al., 2006). Additional studies have been conducted recently to further explore the

potential of more clinically relevant approaches pharmacologically or robotically-driven in

locomotor function recovery after SCI (Edgerton et al., 2008; Guertin 2009).

The idea that synergistic and full CPG-activating effects may be induced by drug

cocktails was supported by results with various combinations of drugs (Antri et al., 2005;

Lapointe and Guertin 2008; Courtine et al., 2009). The detailed mechanisms underlying this

presumed CPG-activating effect remains unclear (Guertin, 2009), but several specific

subsets of monoamine receptors have been shown to be associated with some CPG-

activating effects (Liu and Jordan, 2005; Madriaga et al., 2004; Landry and Guertin, 2004;

Lapointe et al., 2009; Noga et al., 2009). For instance, 5-HTIA/7 receptor agonists were

shown to produce some locomotor-like movements in Tx rodents (Landry et al., 2006;

Courtine et al., 2009). Levodopa (L-DOPA), a noradrenergic and dopaminergic precursor

was also shown to elicit fictive locomotor-like activity recorded from motor nerves in

anesthetised Tx animals (Viala and Buser, 1969). Administered in vivo L-DOPA was also

able to induce some locomotor-like movements (McEwen et al., 1997; Guertin, 2004). We

have recently shown that systemic administration of a tritherapy, delivered orally,

composed of a 5-HT]A receptor agonist (buspirone), a noradrenergic/dopaminergic

precursor (levodopa), and a decarboxylase inhibitor (carbidopa) can temporarily elicit

weight-bearing stepping activity in previously untrained and non-sensori-stimulated Tx

animals (Guertin et al., 2010). However, it remains unclear the extent to which regular

193

training episodes elicited pharmacology with this tritherapy during several weeks can

further improve the motor and locomotor systems (bones, muscles and performances).

Indeed, secondary complications associated with lean body mass, bone and muscle

losses, as well as immune system dysregulations have been fully characterized in Tx mice

(Rouleau et al. 2007; Picard et al., 2008; Ung et al 2008). Given that clenbuterol, a 62-

adrenergic agonist, has been shown to partially prevent muscle wasting in several

experimental model of disuse and muscular atrophy (Agbenyega et al., 1995; Zeman et al.,

2000; Herrera et al., 2001; Teng et al., 2006), we decided to explore the potential benefits

on the motor and locomotor systems of combining clenbuterol and the tritherapy in Tx

mice. We recently showed, indeed, that clenbuterol effects on muscle properties were

superior to those induced by other anabolic aids such as testosterone propionate in non-

tritherapy-treated Tx animals (Ung et al., 2010).

Material and methods

Animal model

All experimental procedures were conducted in accordance with the Canadian Council

for Animal Care guidelines and were accepted by the Laval University Animal Care and

Use Committee. Male CD1 mice (Charles River Canada, St-Constant, Quebec) initially

weighing 30-35 g (approximately eight week-old) were used for this study. Pre-operative

cares included subcutaneous injections of an analgesic (0.1 mg/kg, buprenorphine), an

antibiotic (5 mg/kg, enrofloxacin) and lactate-Ringer's solution (1 ml). All surgical

procedures were performed under aseptic conditions. Mice were anesthetised with 2.5%

isoflurane. A small incision was made on their back in order to expose some thoracic

segments. The spinal cord was then completely transected intervertebrally with

microscissors inserted between the 9th and 10th thoracic vertebrae (Th9/10). To ensure that

complete Tx was achieved, the inner vertebral walls were explored and entirely scraped

several times with small scissor tips. The opened skin area was sutured and animals were

placed for a few hours on heating pads. Mice were left in their cage with food and water ad

194

libitum. Post-operative cares provided for 4 days included subcutaneous injections of

buprenorphine (0.2 mg/kg/day), enrofloxacin (5 mg/kg/day) and lactate-Ringers's solution

(2 x 1 ml/day). Bladders were manually expressed twice daily for the first week post-Tx

and once daily thereafter. Complete low-thoracic Tx was confirmed by 1) initial full

paralysis of the hindlimbs, 2) post-mortem visual and microscopic examination of the

spinal cord lesion, and 3) histological assessment of coronal or midsagittal spinal cord

sections stained with luxol fast blue/cresyl violet for myelinated axons and Nissl substance,

respectively.

Drug treatment and experimental design

Buspirone hydrochloride, L-DOPA and carbidopa (co-administered with L-DOPA to

increase bio-availability of L-DOPA centrally, Lotti and Porter, 1970) were purchased from

Sigma Chemical Co. (St-Louis, Mo., USA). All drugs were dissolved in sterile saline.

Animals were randomly divided in 4 experimental groups: 1) untrained non-Tx mice (N =

11), 2) untrained Tx mice (N = 11), 3) tritherapy-trained Tx mice (N = 11) and 4)

tritherapy-trained + clenbuterol-treated Tx mice (N = 10). The tritherapy composed of

buspirone (3 mg/kg), L-DOPA (50 mg/kg) and carbidopa (12.5 mg/kg) were administered

intraperitoneally 15 minutes prior to training sessions whereas clenbuterol (1 mg/kg) was

administered subcutaneoulsy on a daily basis. Untrained Tx mice received vehicle (saline)

injection instead of the tritherapy. Tritherapy treatment and training began 7 days after Tx

to allow sufficient recovery from surgery. Mice were trained 3 times/week on a motor-

driven treadmill set to a speed of 8-10 cm/sec Each training session lasted 15 minutes. The

overall study lasted 8 weeks. Note that a harness placed around the torso and attached to the

treadmill was used to maintain the animals in front of the camera (perpendicularly

positioned). No weight support was provided with this system.

Hormonal profile

On the 8th week, mice were sacrificed by overdose of ketamine-xylazine. Immediately

upon sacrifice, a cardiac puncture was performed and approximately 850 uL of blood was

collected. For hormone profile measurements, sera were isolated from blood samples by

centrifugation at 500g for 5 minutes at room temperature. Insulin levels were measured

195

from serum using a rat/mouse insulin ELISA kit (EZRMI-13K, Millipore Corporation,

Billerica, MA). This Sandwich ELISA is the standard quantification method for non­

radioactive quantification of rodent insulin levels, which has an effective range between 0.2

and 10 ng/ml and accuracy levels of 1.2-8.4% (intra-assay) and 6.0-17.9% (inter-assay).

IGF-1 levels were measured using a mouse/rat Insulin-like Growth Factor-1 ELISA kit (22-

IG1MS-E01, ALPCO Diagnostics, Salem, NH) which has a reference range between 0.5-

18 ng/ml). DHEA levels were quantified with a competitive immunoassay DHEA Enzyme

Immunoassay Kit (900-093, Assay Designs, Inc., Ann Arbor, MI), which has a reference

range between 12.21-50 000 pg/ml and a precision levels varying between 4.8% and 6.4%

(intra-assay) or between 6.5% and 8.4% (inter-assay). PTH levels were quantified using an

ELISA kit (31-IPTMS-E01, ALPCO Diagnostics, Salem, NH), which has a reference range

of 36-3300 pg/ml and a precision levels varying between 2.5% and 3.9% (intra-assay) or

between 7.8% and 8.9% (inter-assay). All other reagents were of ACS grade and were

obtained from Sigma (St-Louis, MO).

Anatomical modification

Body weight was monitored daily during the first week and once a week subsequently.

Upon sacrifice, the hindlimb muscles extensor digitorum longus (EDL), soleus, biceps

femoris and quadriceps were dissected out and weighed. Forelimb muscles including the

biceps brachii and triceps brachii were also collected. Region-specific adipose tissues were

collected from the abdominal subcutaneous, inguinal, visceral, retroperitoneal and inter­

scapular areas.

Densitometry

Femoral bones were dissected and cleaned of soft tissue. The femoral bones were

wrapped in saline-soaked gauze and frozen at -20 °C in sealed vials until testing. On the

day of testing, the femoral bones were slowly (4 hours) thawed at 4 °C. Dual-energy X-ray

absortiometry (DEXA) measurements (PIXImus 2, Lunar Corp., Madison, WI) were

performed using the femora of animal from all groups. Calibration of the apparatus was

conducted according to the manufacturer's protocol. Bone mineral density (BMD) and bone

mineral content (BMC) were measured within a predetermined common region of interest

196

near the metaphyseal area for all specimens. These experiments were performed by the

McGill's Centre for Bone and Periodontal Research.

Assessment of locomotor movement

Each week, mice were placed on a motor-driven treadmill in order to assess locomotor

performances induced pharmacologically using several complementary methods,.

Movements were assessed during 2 minutes immediately prior to and during 2 minutes

after tritherapy administration. First, we used a semi-quantitative method called Average

Combined Score (ACOS) that has been developed for on-line assessment of hindlimb

movements in Tx rodents (Guertin, 2005; Lapointe et al., 2006; Ung et al., 2007). ACOS

comprises the assessment of hindlimb non-locomotor movements (NLMs, number/min),

locomotor-like movements (LMs, number/min) and movement amplitude expressed as

follows: [NLM + (2 x LM)] x amplitude (Guertin, 2005). One NLM was defined as a non-

bilaterally alternating movement including movements such as jerks, fast-paw shaking and

twitches. One LM is defined as a flexion- extension movement occurring bilaterally in

alternation. We decided to exclude NLMs and LMs occurring during bowel movements to

avoid taking into account sacral reflex-induced movements (Strauss and Lev-Tov, 2003).

Amplitude was assessed by assigning one of following three values; 0 -no movement; 1 -

movements that were less than half the range of motion of normal steps; 2 -movements that

were greater than half the range of motion of normal steps. Hindlimb movements were also

assessed on-line using a locomotor rating scale referred to as AOB (Antri, Orsal and

Barthe) that has been specifically developed for complete Tx rodents (Antri et al., 2002). In

brief, the scale consists of 22 scores. Scores from 0 to 9 assess the frequency of right-left

hindlimb alternating movements and their amplitude. Scores from 10 to 22 assess the

occurrence and quality of weight-bearing steps and of plantar foot placement. Kinematic

analyses were also performed as a complementary method to characterize, in greater

details, some of the induced movements. Animals were filmed using a digital video camera

(JVC GZ-MG330, shutter speed: 1/4000 and acquisition: 60 frame/sec) placed sideways.

Data were stored on computer for subsequent off-line two-dimensional kinematic analyses.

Movement amplitude and angular excursion of the hip, knee and ankle were specifically

197

quantified using MaxTRAQ and MaxMATE softwares (Innovision System, Columbiaville,

MI).

Muscle immunohistochemistry

Upon sacrifice, left soleus and EDL muscles were dissected, frozen in melting isopentane

and stored at -80°C until further use. Serial cross sections (12(xm-thick) prepared from the

muscle mid-portion were cut with a cryostat maintained at -20°C (2800E Frigocut, Leica

Instruments, Germany) and mounted on Superfrost® plus glass slides (VWR Canlab,

Mississauga, ON, Canada). For individual fiber labelling Myosin Heavy Chain slow or fast

primary antibodies were used (MHCs or MHCf, specific for MHC isoform type I and type

II, respectively) (Vector Laboratories, Burlingame, CA). First, cross sections were

incubated 1 hour in a blocking solution containing 10% rabbit serum in 0.1 M phosphate-

buffered saline (PBS). Cross sections were then washed in PBS 0.1M and incubated for 2

hours in a solution containing MHCs or MHCf primary antibodies (dilution 1/50 in 0.1 M

PBS containing 1% rabbit serum). Sections were rinsed in PBS 0.1 M before incubation

with goat anti-mouse IgG(H+L) Alexa Fluor 488 secondary antibodies (dilution 1/500 in

0.1M PBS containing 1% rabbit serum) (Molecular Probes, Eugene, OR). Slides were

mounted with PBS-Glycerol (50-50). For control, some sections were treated as above,

except that the primary antibodies were omitted from the incubation solution.

Immunofluorescence labelling was visualized with a lOx water-immersion objective placed

on an Olympus BX61WI confocal microscope. Images were captured using FluoView 300

(Olympus Canada Inc., Markham, ON, Canada) and analysed with ImageJ (ImageJ 1.40,

Research Services Branch, NH, Bethesda, MD). For each muscle, we measured muscle

CSA, determined muscle fiber phénotype relative distribution and measured corresponding

single fiber CSA (50 fibers per fiber type/muscle were averaged, when available). Type I

fiber was labelled with MHCs, type II fiber with MHCf (no distinction between type Ha,

IIx or lib isoforms was made) whereas hybrid fiber isoforms were labelled with both

antibodies.

Data analyses

198

Locomotor movement and body weight values were analysed using a Two-Way

ANOVA followed by LSD post-hoc tests. Differences between groups for muscle mass,

soleus and EDL muscle CSA, fiber type distribution, single fiber type CSA, region-specific

adipose tissue amounts and femoral BMD and BMC values were analysed using a One-

Way ANOVA followed by LSD post-hoc tests. Statistical significance was determined at P

< 0.05. Results were reported as Mean ± SEM.

Results

Locomotor movement induction

AOB, ACOS and LM values were significantly different between groups (P < 0.001)

over time (Fig. 1 AJ-AJIT). At all time points, differences (P < 0.001) between untrained,

and tritherapy-trained or tritherapy-trained + clenbuterol Tx mice were found, but not

specifically between both trained groups. NLM values between groups (P < 0.001) and

times (P = 0.019) significantly differed between untrained Tx and tritherapy-trained Tx or

tritherapy-trained + clenbuterol Tx animals only on week 1 and 2 (Fig. 1 AJV).

In detail, in untrained Tx mice, a significant (P < 0.001) improvement of performances

over time was found. On the 1st week, no movement was observed whereas AOB and

ACOS scores began to progressively increase until week 3, where a plateau was reached

(AOB and ACOS scores ~ 3 and 25, respectively). LMs also spontaneously occurred over

time in untrained Tx mice, although no significant level was reached. In contrast, NLM

frequency significantly increased from 0 to ~ 12 movements/min. As mentioned above,

tritherapy-trained and tritherapy-trained + clenbuterol Tx mice displayed comparable

performances during the whole training protocol. For AOB rating scales, on the 1st week

averaged scores were near 7 movements/min. Scores further increased on week 2 (~ 10 and

12, Tx trained and trained + clenbuterol, respectively), on week 3 (~ 14 for both trained

groups) and reached a plateau in performances subsequently until the end (~ 16 for both

groups of trained mice). Note that weight-bearing steps corresponding with AOB scores of

10 or more were clearly elicited by the 2nd week. Similarly, in both groups of tritherapy-

trained mice, ACOS scores progressively increased over time to reach values « 350 by

week 8. For LMs, in Tx tritherapy-trained mice, average frequencies progressively

199

increased from 26.3 ± 3.8 to 74.1 ± 7.2 movements/min on week 8. Similarly, tritherapy-

trained + clenbuterol Tx mice displayed LM scores ranged from 17.8 ± 3.5 on week 1 to ~

60 movements/min by the week 4 and 77.0 ± 6.8 by week 8 (Fig. IA///). No significant

change in NLM frequency was found throughout the 8 week training period between the

two tritherapy-trained groups with average frequency values in the range of 15

movements/min (Fig 1.A/V).

On the 8th week, differences between the three Tx groups and non-Tx mice were

analyzed. Tx tritherapy-trained alone, Tx tritherapy-trained + clenbuterol and non-Tx mice

showed similar ACOS scores and LM frequency (Fig. 1 BH-IH). However, AOB scores in

both groups of tritherapy-trained animals were lower than non-Tx animals (P < 0.001, Fig.

IB/). It is noteworthy to mention that in tritherapy-trained animals during the last 3 weeks,

13/21 mice displayed AOB scores greater than 19, which correspond literally to 'large

amplitude locomotor movements with frequent-to-consistent right-left hindlimb alternation,

consistent weight-bearing capabilities and occasional-to-consistent plantar foot placement.

Except for NLMs, untrained Tx mice displayed lower AOB, ACOS and LM scores than all

other groups.

To further characterize movements induced by the tritherapy-training, angular

excursion at the hip, knee and ankle, as well as movement amplitude values were analysed.

Typical examples of hindlimb kinematics are shown in figure 2. Hip, knee and ankle

angular displacement showed similar patterns in non-Tx, tritherapy-trained alone and

tritherapy-trained + clenbuterol Tx animals (Fig. 2A, 2C and 2D, top 3 panels). However,

non-Tx animals displayed greater joint angle displacement values (hip: 29-117°, knee: 47-

123°, ankle: 41-132°) than both tritherapy-trained groups, which showed similar values

(hip: 43-79°, knee: 47-74°, ankle: 20-80°). On the other hand, untrained Tx animals

displayed a consistent lack of angular excursion at the hip level although some

displacements were found at the knee and ankle levels (hip: 85°, knee: 30-47°, ankle: 28-

125°). Hindlimb movement amplitude values measured by calculating toe displacement in

X and Y axis (step "length" and "height") (Fig.2A-2D, bottom 2 panels and Table 1)

revealed that non-Tx mice had greater (P < 0.001) step length values than both tritherapy-

200

trained Tx groups. On the other hand, both groups of tritherapy-trained Tx animals showed

similar step length values which were significantly greater than those in untrained Tx mice.

The coefficient of variation (CV) was higher in untrained Tx mice. Non-Tx, tritherapy-

trained and tritherapy-trained + clenbuterol Tx mice showed similar step height values.

However, differences were found in the variability of the step height, as shown by CV,

where non-Tx animals displayed less variability than the other groups of tritherapy-trained

animals. No Y axis movement amplitude was observed in Tx untrained since no weight-

bearing movement are normally expressed spontaneously as shown elsewhere (Guertin,

2005; Lapointe et al., 2006; Ung et al., 2007).

Overall, a significant increase in performances over time was observed in tritherapy-

trained groups as LM and ACOS values were comparable with those from non-Tx mice on

week 8.

Body composition

Body weights were monitored weekly to assess overall body composition changes

(Fig.3A). Tx mice underwent approximately a 15% decrease in weight within the first few

days, with a progressive re-increase subsequently. By the week 4 and 5, tritherapy-trained +

clenbuterol Tx mice showed higher weight gains than Tx untrained mice (represented by *)

and tritherapy-trained Tx mice (represented by t), respectively. Throughout the 8 week-

period, non-Tx mice displayed weight increase corresponding to 122 ± 2.2 % of initial body

weight. At all time points after Tx, non-Tx weight was greater than all Tx groups (not

shown for clarity reasons).

In Tx mice, lower body weight values were partly attributed to lower adipose tissue

amounts (Fig. 3B). Non-Tx mice showed > 3.5-fold higher total adipose tissue amounts

than all Tx groups. When comparing only Tx mice, these data also revealed that tritherapy-

trained + clenbuterol Tx mice had lower adipose tissue amounts than untrained Tx mice

(0.599 ± 0.051 g vs 0.946 ± 0.116 g; P < 0.05), especially in visceral and retroperitoneal

areas. Differences in body weight were also partly attributable to muscle mass changes (Fig

3H-3M). Quadriceps from non-Tx mice were heavier than those from untrained and

201

tritherapy-trained Tx mice. Biceps femoris in all groups of Tx mice showed lower (P <

0.001) muscle mass than non-Tx mice (0.263 ± 0.015 g). In addition, untrained Tx animals

(0.179 ± 0.008 g) and tritherapy-trained Tx mice (0.185 ± 0.007 g) had lower biceps

femoris mass than tritherapy-trained + clenbuterol Tx mice (0.221 ± 0.007 g) mice. At the

ankle level, no difference was observed in EDL mass. However, soleus muscles were found

to display marked (P < 0.009) tritherapy training-induced effect, as untrained Tx mice

(0.0070 ± 0.0004 g) had lower mass than tritherapy-trained Tx animals (0.0084 ± 0.0003

g), tritherapy-trained + clenbuterol Tx mice (0.0103 ± 0.0007 g) and non-Tx mice (0.0095

± 0.0003 g). No difference was found between non-Tx and tritherapy-trained + clenbuterol

Tx mice. In forelimb muscles, non-Tx (0.217 ± 0.005 g) and tritherapy-trained +

clenbuterol Tx mice (0.225 ± 0.005 g) displayed similar triceps mass values, which were

bigger than untrained Tx (0.189 ± 0.004 g) and tritherapy-trained alone Tx mice (0.196 ±

0.003 g). Muscle mass values from biceps brachii were similar between all groups. Thus,

comparison between all groups of Tx mice revealed that tritherapy-trained + clenbuterol

treatment increased in body weight could partly be explained by a rather large increase in

muscle mass specifically at the biceps femoris, soleus and triceps levels.

Rapid bone loss is a well-known secondary complication occurring after SCI. BMD and

BMC were measured using femurs in order to address whether tritherapy-training alone or

combine with clenbuterol can prevent or at least reduce bone loss normally found in

untrained Tx mice (Picard et al. 20008). However, in all groups of Tx animals, important

losses were found (Fig. 4). Untrained Tx mice (BMD: 0.0767 ± 0.0010 g/cm2, BMC:

0.0381 ± 0.0008 g) and tritherapy-trained Tx animals (BMD: 0.0766 ± 0.0011 g/cm2, BMC:

0.0378 ± 0.0009 g) showed comparable values whereas in Tx trained + clenbuterol groups,

femoral BMD (0.0731 ± 0.0012) and BMC (0.0349 ± 0.0009) further decreased. However,

significance was only reached on femur BMC, by comparing it to untrained Tx mice (P <

0.036).

Muscle and fiber type morphometry

Morphometric analyses of soleus and EDL were performed in order to further

characterize specific muscular property changes in all groups. Muscle CSA, fiber type-

202

specific CSA and relative distribution values were analysed. For soleus CSA, untrained and

tritherapy-trained Tx mice had significantly lower muscle CSA values than non-Tx (P <

0.01) and tritherapy-trained + clenbuterol (P < 0.001) Tx groups. However soleus CSA in

untrained Tx mice was not significantly lower than tritherapy-trained Tx animals. For EDL,

in contrast with muscle mass changes, CSA values showed statistical differences between

groups (P < 0.001). Tritherapy-trained + clenbuterol Tx mice showed higher EDL CSA

values than all the other groups. Untrained and tritherapy-trained Tx mice had lower CSA

values than non-Tx animals (Table 2).

More differences were found when analysing individually fiber type-specific CSA

values. Typical illustrations of muscle immunohistochemistry labelling performed on

soleus and EDL sections are shown in Figure 5. For soleus and EDL, when comparing all

groups, significant differences were observed (P < 0.001) in all fiber types. Specifically, for

soleus fiber types (Fig.6A-6C), all three fiber types from tritherapy-trained + clenbuterol

Tx animals displayed larger CSA values than all other groups (type I: 1656.7 ± 80.8 [im2,

type II: 987.2 ± 16.7 [im2, hybrid: 1145.5 ± 18.0 [im2). Conversely, untrained Tx mice

displayed the lowest soleus fiber type CSA of all groups (type I: 783.1 ± 15.1 [im2, type II:

753.2 ± 9.1 [im2, hybrid: 750.0 ± 8.1 [im2). Fiber type CSA from tritherapy-trained Tx mice

(type I: 827.4 ± 28.9 nm2, type II: 804.8 ± 12.2 [im2, hybrid: 847.7 ± 11.9 [im2) were

significantly lower than non-Tx group (type I: 949.8 ±11.5 [im2, type II: 948.7 ± 12.3 yon2,

hybrid: none). Comparison between Tx untrained and tritherapy-trained mice showed

differences only for type II and hybrid fibers. In EDL (Fig. 6D-E), type II fiber CSA

differences between groups were similar to soleus type II (non-Tx: 1063.5 ± 15.9 [im2, Tx

untrained: 908.1 ±11.4 [im , Tx trained: 963.4 ± 10.9 [im2, Tx trained + clenbuterol:

11.65.2 ± 17.9 [im2). Few hybrid fibers were found in EDL (Table 3). For those fibers, non-

Tx mice had larger CSA values than other groups (non-Tx, 542.2 ± 15.2; Tx untrained,

406.3 ± 7.9; Tx trained, 426.3 ± 13.7; Tx trained + clenbuterol, 487.4 ± 13.7).

It is well-known that there is an important shift in fiber phénotype distribution after SCI

even more so in soleus (Lieber et al., 1986; Talmadge et al., 2002). Generally, slow fibers

tend to change for a faster phénotype. None of the non-Tx mice had hybrid fibers. After Tx,

203

50-55% of the slow type fibers showed important fiber type conversion, shifting to a hybrid

isoform. Fiber type conversion was not observed in EDL muscle (Table 3).

Overall, untrained Tx mice had lower muscle CSA and fiber-type specific CSA values

than other groups, whereas tritherapy-trained Tx mice combined with daily administration

of clenbuterol showed generally higher CSA values. However after 2 months, none of the

treatment significantly influenced fiber type conversion in Tx mice.

Hormonal profile

Immediately upon sacrifice, a cardiac puncture was performed. Sera were isolated from

whole blood sample and a hormonal profile was assessed (Table 4). Differences were only

observed with insulin and IGF-1 hormones. Indeed, non-Tx animals had more than 2.5-fold

higher insulin levels than all Tx mice. No difference was observed between Tx groups.

Conversely, average IGF-1 levels in non-Tx and Tx untrained groups were respectively the

lowest and highest. Tritherapy-trained and tritherapy-trained + clenbuterol Tx groups IGF-1

values were significantly different compared with non-Tx and untrained Tx mice.

Discussion

We found that treadmill training elicited 3 times per week during 8 weeks by a novel

tritherapy leads to improved locomotor performances over time with frequent weight-

bearing steps, bilateral alternation, and plantar foot placement. Moreover, comparable

hindlimb kinematic patterns were observed between both tritherapy-trained groups and

non-Tx mice, although differences in the overall quality were found, as shown by AOB

scores, stepping amplitude and movement variability. Combining tritherapy-training with

daily administration of clenbuterol clearly reversed muscular atrophy and even induced

hypertrophy compared with intact animals. Furthermore, this combinatorial approach led to

decreased adipose tissue amounts but also to decreased femoral BMD and BMC. Insulin

levels remained low in all Tx animals compared to non-Tx animals, but IGF-1 was higher

in Tx untrained mice.

204

Training combined with pharmacological stimulation induces significant locomotor

recovery

In adult rodents, spontaneous hindlimb movement (spontaneously occurring over time

without treatment) essentially composed of non-locomotor movements (NLMs) have been

reported after a complete Tx, although significant LMs or weight-bearing stepping has

never been found (Guertin, 2005; Lapointe et al 2006; Ung et al., 2007). Potent CPG-

activating effects using combinatory drug treatments have been proposed using a few

specific compounds (McEwen et al., 1997, Antri et al., 2005, Courtine et al., 2009,

Lapointe and Guertin, 2008; Guertin et al., 2010). Although, the use of cocktails was

clearly shown to elicit superior locomotor movement in Tx animals, the detailed

mechanism underlying their synergistic actions remains unclear. It is possible that several

receptor sub-types have to be activated simultaneously in order to fully activate the CPG

and corresponding hindlimb stepping. In this study, in order to stimulate locomotor

network activation, we used as part of a tritherapy, buspirone, L-DOPA and carbidopa

(Guertin et al., 2010). Buspirone binds to a 5-HTIA receptor, which was shown to be

involved in locomotor-like movement induction (Antri et al. 2003; Landry et al., 2006;

Lapointe and Guertin 2008; Noga et al., 2009). L-DOPA, a noradrenalin and dopamine

precursor, has also been shown to induce some rhythmicity in various animal models

(McEwen et al. 1997; Viala and Buser 1969; Guertin et al., 2010). Carbidopa is not a priori

a CPG-activating compound but instead a decarboxylase inhibitor typically used clinically

in combination with L-DOPA to reduce its peripheral decarboxylation, thus increasing its

bioavailability centrally (Lotti and Porter, 1970).

Another recent study has shown remarkable locomotor recovery by combining epidural

stimulation, quipazine, 8-OH-DPAT and robotically assisted treadmill training, (Courtine et

al., 2009). That study showed that complete Tx rats displayed locomotor performances

comparable with those found in non-Tx animals. Several differences are found between our

study and theirs, one of them being the training approach. They used a form of bipedal

treadmill training whereas we used quadrupedal locomotion, a more natural position for

mice. They also used robotic devices and epidural stimulation in addition to

pharmacological tools. We did not have to use those devices to elicit full CPG activation

205

and corresponding locomotor performances probably because of the synergistic effects

induced with our tritherapy (Guertin et al., 2010).

Musculoskeletal adaptations after SCI and tritherapy-training.

Muscular deterioration is among several of the secondary consequences of SCI. The

lost of neural activation induce atrophy and changes in muscle properties. Hindlimb

muscular alteration is believed to negatively affect stepping abilities, as muscles are more

weak and fatigable. Preserving or reversing muscle changes after SCI could be beneficial

for improving locomotor recovery. We showed that tritherapy-trained Tx mice had higher

soleus muscle mass, type I and hybrid fiber CSA and EDL type II CSA than untrained

animals. This corroborates previous works showing that training increased soleus muscle

mass and fiber CSA (Dupont-Versteegden et al., 1988; Roy et al., 1998). However, in this

study, training alone only partially reverse atrophy. To further maximize muscular mass

increase with decided to administer clenbuterol. We have previously shown that in Tx

mice, administration of clenbuterol alone had hypertrophic properties (Ung et al., 2010). In

this study, combined with training, clenbuterol helped increase muscle mass, whole muscle

CSA and fiber CSA. In fact, those values were similar or higher than non-Tx mice. This

increase also partly explained body weight differences found across all Tx mice. However,

no correlation was found between muscle morphometric data and locomotor recovery (data

not shown). In other words, tritherapy-trained Tx mice with higher locomotor score did not

necessarily have higher soleus or EDL mass, muscle or fiber CSA.

The use of clenbuterol after SCI has remained poorly investigated. Although it was

found beneficial on muscle anabolism in some studies with other models of disuse (Zeman

et al., 1999, 2000; Herrera et al., 2001; Teng et al., 2006; Picquet et al., 2004), it is also

known to display less desirable effects. For example, it can induce a slow to fast muscle

phénotype transition especially in soleus (Oishi et al., 2002; Ryall et al., 2002). We showed

that tritherapy trained + clenbuterol Tx mice had the lowest proportion of type I fiber in

soleus, which also corroborates our previous report (Ung et al., 2010). Although fiber type

proportion was not significantly different between all Tx groups two months after SCI,

differences could have been more pronounced early after SCI. Clenbuterol may also induce

206

or accelerate bone tissue loss as confirmed also in this study (Fig. 4) and elsewhere

(Kitaura et al., 2002; Bonnet et al., 2005). It remains unclear if such deleterious effect on

bone property may be prevented with co-administration of bisphosphonates typically used

for age-related osteoporosis (Russell et al. 2008; Recker et al., 2009).

Although exercise itself is known to possess osteogenic potential (Guadalupe-Grau et

al., 2009), we did not find any benefits on femoral bones during 2 months of treadmill

training. Another case-report with a SCI person revealed negligible training-induced effect

on femoral bone property (Coupaud et al., 2009). Because of the severe bone loss rapidly

induced after SCI, increased fracture incidence is generally found in SCI patients (Lazo et

al., 2001; Sabo et al., 2001). Clearly, other strategies and approaches need to be

investigated to prevent or reverse bone loss after SCI.

Overall serum profile

We have previously reported strong decrease of insulin levels soon after SCI (Rouleau

et al., 2007). Likewise in this study, all Tx mice had lower insulin levels compared to non-

Tx animals. In all Tx groups, neither of the treatment significantly altered insulin levels.

Only a few studies have examined levels of IGF-1 after SCI. Previous reports have

shown lower IGF-1 in SCI patients (Bauman et al., 1994; Shetty et al., 1993). However, a

recent study performed on monozygotic twins discordant for SCI, reported no difference in

IGF-1 levels between SCI and able bodied person (Bauman et al., 2007). In contrast, Tx

mice displayed higher serum IGF-1 levels compared to non-Tx animal at least shortly after

SCI (within the first 2 months post-Tx). Reasons for this discrepancy remains unclear but

may be associated with time post-injury, (i.e., a progressively return to normal levels may

perhaps occur in late chronic subjects). On the other hand, compared to untrained Tx mice,

lower plasma IGF-1 levels are found in Tx trained and Tx trained + clenbuterol mice. In

those trained groups, IGF-1 upregulation could have been found more locally (i.e. at the

muscular level). Indeed, it was reported that clenbuterol or training increased mRNA

encoding IGF-1 in EDL and soleus (Mounier et al., 2007; Awede et al., 2002).

207

Clinical implications

This study has provided strong evidences that such combination have great potential for

drug development that helped SCI persons train and recover some locomotor functions.

Indeed, all molecules used for locomotor network activation are already approved by FDA

as treatment for other neurological problems (Parkinson's Disease and anxiety). However,

other adjunct approaches may need to be identified to prevent not only muscular atrophy

but also bone loss and immune system changes.

Acknowledgements

This study was supported by the Canadian Institutes of Health Research (CIHR), the Fonds

de Recherche en Santé du Quebec (FRSQ) and Nordic Life Science Pipeline.

Disclosure Statement

Pierre A Guertin is the president and CEO of Nordic Life Science Pipeline. Nordic Life

Science Pipeline has an in-licencing agreement with Laval University to develop and

commercialise this technology

208

References

Agbenyega ET, Morton RH, Hatton PA, Wareham AC. 1995. Effect of the beta 2-

adrenergic agonist clenbuterol on the growth of fast- and slow-twitch skeletal

muscle of the dystrophic (C57BL6J dy2J/dy2J) mouse. Comp Biochem Physiol C

Pharmacol Toxicol Endocrinol 111(3):397-403.

Antri M, Barthe JY, Mouffle C, Orsal D. 2005. Long-lasting recovery of locomotor

function in chronic spinal rat following chronic combined pharmacological

stimulation of serotonergic receptors with 8-OHDPAT and quipazine. Neurosci Lett

384(1-2): 162-167.

Antri M, Mouffle C, Orsal D, Barthe JY. 2003. 5-HT 1A receptors are involved in short-

and long-term processes responsible for 5-HT-induced locomotor function recovery

in chronic spinal rat. Eur J Neurosci 18(7): 1963-1972.

Antri M, Orsal D, Barthe JY. 2002. Locomotor recovery in the chronic spinal rat: effects of

long-term treatment with a 5-HT2 agonist. Eur J Neurosci 16(3):467-476.

Awede BL, Thissen JP, Lebacq J. 2002. Role of IGF-I and IGFBPs in the changes of mass

and phénotype induced in rat soleus muscle by clenbuterol. Am J Physiol

Endocrinol Metab 282(l):E31-37.

Barbeau H, Fung J. 2001. The role of rehabilitation in the recovery of walking in the

neurological population. Curr Opin Neurol 14(6):735-740.

Barbeau H, Rossignol S. 1987. Recovery of locomotion after chronic spinalization in the

adult cat. Brain Res 412(1):84-95.

Bauman WA, Spungen AM, Zhong YG, Tsitouras PD. 1994. Chronic baclofen therapy

improves the blunted growth hormone response to intravenous arginine in subjects

with spinal cord injury. J Clin Endocrinol Metab 78(5): 1135-1138.

Bauman WA, Zhang RL, Spungen AM. 2007. Provocative stimulation of growth hormone:

a monozygotic twin study discordant for spinal cord injury. J Spinal Cord Med

30(5):467-472.

Bonnet N, Benhamou CL, Brunet-Imbault B, Arlettaz A, Horcajada MN, Richard O, Vico

L, Collomp K, Courteix D. 2005. Severe bone alterations under beta2 agonist

209

treatments: bone mass, microarchitecture and strength analyses in female rats. Bone

37(5):622-633.

Coupaud S, Jack LP, Hunt KJ, Allan DB. 2009. Muscle and bone adaptations after

treadmill training in incomplete Spinal Cord Injury: a case study using peripheral

Quantitative Computed Tomography. J Musculoskelet Neuronal Interact 9(4):288-

297.

Courtine G, Gerasimenko Y, van den Brand R, Yew A, Musienko P, Zhong H, Song B, Ao

Y, Ichiyama RM, Lavrov I, Roy RR, Sofroniew MV, Edgerton VR. 2009.

Transformation of nonfunctional spinal circuits into functional states after the loss

of brain input. Nat Neurosci 12(10): 1333-1342.

Dobkin B, Apple D, Barbeau H, Basso M, Behrman A, Deforge D, Ditunno J, Dudley G,

Elashoff R, Fugate L, Harkema S, Saulino M, Scott M. 2006. Weight-supported

treadmill vs over-ground training for walking after acute incomplete SCI.

Neurology 66(4):484-493.

Dupont-Versteegden EE, Houle JD, Gurley CM, Peterson CA. 1998. Early changes in

muscle fiber size and gene expression in response to spinal cord transection and

exercise. Am J Physiol 275(4 Pt 1):C1124-1133.

Edgerton VR, Courtine G, Gerasimenko YP, Lavrov I, Ichiyama RM, Fong AJ, Cai LL,

Otoshi CK, Tillakaratne NJ, Burdick JW, Roy RR. 2008. Training locomotor

networks. Brain Res Rev 57(1 ):241-254.

Guadalupe-Grau A, Fuentes T, Guerra B, Calbet JA. 2009. Exercise and bone mass in

adults. Sports Med 39(6):439-468.

Guertin PA. 2004. Synergistic activation of the central pattern generator for locomotion by

l-beta-3,4-dihydroxyphenylalanine and quipazine in adult paraplegic mice. Neurosci

Lett 358(2):71-74.

Guertin PA. 2005. Semiquantitative assessment of hindlimb movement recovery without

intervention in adult paraplegic mice. Spinal Cord 43(3): 162-166.

Guertin PA. 2009. Recovery of locomotor function with combinatory drug treatments

designed to synergistically activate specific neuronal networks. Curr Med Chem

16(11):1366-1371.

210

Guertin PA, Ung RV, Rouleau P. 2010. Oral administration of a tri-therapy for central

pattern generator activation in paraplegic mice: proof-of-concept of efficacy.

BiotechnolJ5(4):421-426.

Herrera NM, Jr., Zimmerman AN, Dykstra DD, Thompson LV. 2001. Clenbuterol in the

prevention of muscle atrophy: a study of hindlimb-unweighted rats. Arch Phys Med

Rehabil 82(7):930-934.

Kitaura T, Tsunekawa N, Kraemer WJ. 2002. Inhibited longitudinal growth of bones in

young male rats by clenbuterol. Med Sci Sports Exerc 34(2):267-273.

Landry ES, Guertin PA. 2004. Differential effects of 5-HTQ) and 5-HT(2) receptor

agonists on hindlimb movements in paraplegic mice. Prog Neuropsychopharmacol

Biol Psychiatry 28(6): 1053-1060.

Landry ES, Lapointe NP, Rouillard C, Levesque D, Hedlund PB, Guertin PA. 2006.

Contribution of spinal 5-HT 1A and 5-HT7 receptors to locomotor-like movement

induced by 8-OH-DPAT in spinal cord-transected mice. Eur J Neurosci 24(2):535-

546.

Lapointe NP, Guertin PA. 2008. Synergistic effects of Dl/5 and 5-HT 1 A/7 receptor

agonists on locomotor movement induction in complete spinal cord-transected

mice. J Neurophysiol 100(1): 160-168.

Lapointe NP, Rouleau P, Ung RV, Guertin PA. 2009. Specific role of dopamine DI

receptors in spinal network activation and rhythmic movement induction in

vertebrates. J Physiol 587(Pt 7): 1499-1511.

Lazo MG, Shirazi P, Sam M, Giobbie-Hurder A, Blacconiere MJ, Muppidi M. 2001.

Osteoporosis and risk of fracture in men with spinal cord injury. Spinal Cord

39(4):208-214.

Lieber RL, Friden JO, Hargens AR, Feringa ER. 1986. Long-term effects of spinal cord

transection on fast and slow rat skeletal muscle. II. Morphometric properties. Exp

Neurol 91(3):435-448.

Liu J, Jordan LM. 2005. Stimulation of the parapyramidal region of the neonatal rat brain

stem produces locomotor-like activity involving spinal 5-HT7 and 5-HT2A

receptors. J Neurophysiol 94(2): 1392-1404.

211

Lotti VJ, Porter CC. 1970. Potentiation and inhbition of some central actions of L(-)-dopa

by decarboxylase inhibitors. J Pharmacol Exp Ther 172(2):406-415.

Lovely RG, Gregor RJ, Roy RR, Edgerton VR. 1986. Effects of training on the recovery of

full-weight-bearing stepping in the adult spinal cat. Exp Neurol 92(2):421-435.

Madriaga MA, McPhee LC, Chersa T, Christie KJ, Whelan PJ. 2004. Modulation of

locomotor activity by multiple 5-HT and dopaminergic receptor subtypes in the

neonatal mouse spinal cord. J Neurophysiol 92(3): 1566-1576.

McEwen ML, Van Hartesveldt C, Stehouwer DJ. 1997. L-DOPA and quipazine elicit air-

stepping in neonatal rats with spinal cord transections. Behav Neurosci 111(4):825-

833.

Mounier R, Cavalie H, Lac G, Clottes E. 2007. Molecular impact of clenbuterol and

isometric strength training on rat EDL muscles. Pflugers Arch 453(4):497-507.

Noga BR, Johnson DM, Riesgo MI, Pinzon A. 2009. Locomotor-activated neurons of the

cat. I. Serotonergic innervation and co-localization of 5-HT7, 5-HT2A, and 5-HT 1A

receptors in the thoraco-lumbar spinal cord. J Neurophysiol 102(3): 1560-1576.

Oishi Y, Imoto K, Ogata T, Taniguchi K, Matsumoto H, Roy RR. 2002. Clenbuterol

induces expression of multiple myosin heavy chain isoforms in rat soleus fibres.

Acta Physiol Scand 176(4):311-318.

Picard S, Lapointe NP, Brown JP, Guertin PA. 2008. Histomorphometric and densitometric

changes in the femora of spinal cord transected mice. Anat Rec (Hoboken)

291(3):303-307.

Picquet F, De-Doncker L, Falempin M. 2004. Enhancement of hybrid-fiber types in rat

soleus muscle after clenbuterol administration during hindlimb unloading. Can J

Physiol Pharmacol 82(5):311-318.

Recker RR, Lewiecki EM, Miller PD, Reiffel J. 2009. Safety of bisphosphonates in the

treatment of osteoporosis. Am J Med 122(2 Suppl):S22-32.

Rouleau P, Ung RV, Lapointe NP, Guertin PA. 2007. Hormonal and immunological

changes in mice after spinal cord injury. J Neurotrauma 24(2):367-378.

Roy RR, Talmadge RJ, Hodgson JA, Zhong H, Baldwin KM, Edgerton VR. 1998. Training

effects on soleus of cats spinal cord transected (T12-13) as adults. Muscle Nerve

21(1):63-71.

212

Russell RG, Watts NB, Ebetino FH, Rogers MJ. 2008. Mechanisms of action of

bisphosphonates: similarities and differences and their potential influence on

clinical efficacy. Osteoporos Int 19(6):733-759.

Ryall JG, Gregorevic P, Plant DR, Sillence MN, Lynch GS. 2002. Beta 2-agonist fenoterol

has greater effects on contractile function of rat skeletal muscles than clenbuterol.

Am J Physiol Regul Integr Comp Physiol 283(6):R 1386-1394.

Sabo D, Blaich S, Wenz W, Hohmann M, Loew M, Gemer H J. 2001. Osteoporosis in

patients with paralysis after spinal cord injury. A cross sectional study in 46 male

patients with dual-energy X-ray absorptiometry. Arch Orthop Trauma Surg 121(1-

2):75-78.

Shetty KR, Sutton CH, Mattson DE, Rudman D. 1993. Hyposomatomedinemia in

quadriplegic men. Am J Med Sci 305(2):95-100.

Strauss I, Lev-Tov A. 2003. Neural pathways between sacrocaudal afférents and lumbar

pattern generators in neonatal rats. J Neurophysiol 89(2):773-784.

Talmadge RJ, Roy RR, Caiozzo VJ, Edgerton VR. 2002. Mechanical properties of rat

soleus after long-term spinal cord transection. J Appl Physiol 93(4): 1487-1497.

Teng YD, Choi H, Huang W, Onario RC, Frontera WR, Snyder EY, Sabharwal S. 2006.

Therapeutic effects of clenbuterol in a murine model of amyotrophic lateral

sclerosis. Neurosci Lett 397(1-2): 155-158.

Ung RV, Lapointe NP, Tremblay C, Larouche A, Guertin PA. 2007. Spontaneous recovery

of hindlimb movement in completely spinal cord transected mice: a comparison of

assessment methods and conditions. Spinal Cord 45(5):367-379.

Ung RV, Rouleau P, Guertin PA. 2010. Effects of co-administration of clenbuterol and

testosterone propionate on skeletal muscle in paraplegic mice. J Neurotrauma

27(6): 1129-1142.

Viala D, Buser P. 1969. The effects of DOPA and 5-HTP on rhythmic efferent discharges

in hind limb nerves in the rabbit. Brain Res 12(2):437-443.

Wernig A, Muller S. 1992. Laufband locomotion with body weight support improved

walking in persons with severe spinal cord injuries. Paraplegia 30(4):229-238.

213

Zeman RJ, Feng Y, Peng H, Etlinger JD. 1999. Clenbuterol, a beta(2)-adrenoceptor agonist,

improves locomotor and histological outcomes after spinal cord contusion in rats.

Exp Neurol 159(l):267-273.

Zeman RJ, Peng H, Danon MJ, Etlinger JD. 2000. Clenbuterol reduces degeneration of

exercised or aged dystrophic (mdx) muscle. Muscle Nerve 23(4):521-528.

214

Figure legends

Fig. 1 Locomotor recovery assessed using 2 complementary locomotor rating scales, AOB

and ACOS. The ACOS is a combination of non-locomotor movements (NLM), locomotor

movements (LM) and amplitude. In post-drugs results, a two-way ANOVA revealed

significant differences between both tritherapy-trained groups compared to untrained Tx

mice. On the 8lh week, comparison between non-Tx mice and both tritherapy-trained mice

showed no difference in scores excepted for AOB. * P < 0.05, *** P < 0.001.

Fig. 2 Representative hindlimb kinematics. For non-Tx (A), untrained Tx (B), tritherapy-

trained Tx (C) and tritherapy-trained + clenbuterol Tx (D) joint angle displacement were

averaged in 10 consecutive steps (or extension-flexion). Hip, knee and ankle joint angular

displacement are shown in top 3 panels. Movement amplitude in X an Y axis (step length

and height at the toe level) are shown in bottom 2 panels.

Fig. 3 Mice body weights, adipose tissue accumulations and muscle masses. Weights were

averaged on a weekly basis in order to give a general insight for body composition (A).

Two-way ANOVA for body weight was performed only between Tx groups. Non-tx mice

showed greater body weight, partly due to higher adipose tissue accumulation (B-G), and

generally higher muscle mass (H-M). * P < 0.05, ** P < 0.01, P < 0.001 compared to non-

tx; # P < 0.05 compared to untrained Tx; f P < 0.05, f t P < 0.01, t t t P < 0.001 compared

to tritherapy-trained + clenbuterol.

Fig. 4 Bone mineral density (BMD) and bone mineral content (BMC) quantification.

Measurements were performed by dual X-ray absorptiometry (PIXImus). All Tx mice

showed lower BMD and BMC values than non-Tx mice. *** P < 0.001 compared to non-

Tx; t P < 0.05 compared to Tx untrained mice.

Fig. 5 Muscle immunolabelling. Typical example of myosin heavy chain slow and fast

(MHCs, MHCf) labelling on (A) soleus and (B) EDL muscles. In soleus, specific cell type

is shown by arrows. EDL is mostly constituted of type II fibers (MHCf).

215

Fig. 6 Muscle fiber type cross-sectional area (CSA). Comparisons between groups for

soleus (A-B-C) and EDL (D-E) type I, II and hybrid fiber CSA. No EDL type I fiber were

found. * P < 0.05, *** P < 0.001 compared to non-Tx; # P < 0.05 compared to tritherapy-

trained Tx; t t t P < 0.001 compared to tritherapy-trained + clenbuterol.

Table 1 Movement amplitude in X and Y axis (step length and height at the toe level) and

coefficient of variation (CV).

Table 2 Soleus and EDL muscle cross-sectional area were evaluated at mid-portion of

muscles. Data are expressed in mm2. * P < 0.01, ** P < 0.(

t t P < 0.001, compared to tritherapy-trained + clenbuterol.

muscles. Data are expressed in mm2. * P < 0.01, ** P < 0.001, compared to non-Tx group;

Table 3 Soleus and EDL type I, II and hybrid fiber phénotype were counted and expressed

as percentage of total muscle fiber to quantify the relative distribution of each fiber

phénotype within a muscle. Difference between groups were only found in soleus type I

and hybrid fiber phénotype. * P < 0.001

Table 4 ** P < 0.01, *** P < 0.001 compared to non-Tx; # P < 0.05, ## P < 0.01, compared

to untrained Tx

216

Post-drugs

AOB

400n

300-

200-

100-

Locomotor movements 100'

80'

60

40

Locomotor movements 100n

20-|

IV Non-locomotor movements Non-locomotor movements 30n

20-

10-

4 6 8 Weeks

- O - Tx untrained - Ô - Tx tritherapy-trained - O - Tx tritherapy trained + clenbuterol

&>

Figure 1

217

E E (D -a

E ra c eu E > O E

I > Il

Non-tx Tx untrained C Tx tritherapy-trained Tx tritherapy-trained + Clenbuterol

75 100 125 100 125 0 25

Step cycle (%) 75 100 125 0 75 100 125

A / W N / N / W V V N AAAAAAAAAA AAAAAAAAM

Figure 2

218

A „ Weight

B Total adipose tissue H 4

3

2

1

0

1 0 0.8

3 0.4 E 0.2

p < 0.001

p < 0.001

D 2.0

~ 1.5

8 1 ° m E 0.5

0 8

™ ° 6

Ï 0 . 4 ra E 0.2

05 0.4

S 0.3 S 0.2 E

0 . 1 0

Visceral p < 0.001

Retro-peritoneal p < 0.001

T

***

Inter-scapular

p < 0.001

0.4

0 3

0 2

0.1 •

0

Quadriceps p = 0.004

1D1IQ Sub-cutaneous Biceps femoris

0.3

02

0.1

0

. p < 0.001

+ f r"i

Inguinal J EDL 0.0151 p = 0.078

p < 0.001 *** 0 010-1

0 005-

K m

Soleus

0 0 1 5 , p< 0.009 »

0010 .

0.005-

* * * t t t t t "

Triceps p < 0.001

*** ** t t t t t t

0 5 . . 0.4 5 0.3 0

3 02 : 0.1

0 Non-Tx Tx tritherapy-trained Tx untrained Tx tntherapy-trained + Cb

M Biceps brachii

0 04-j p = 0.387

0.03

0 0 2

0.01 EUE Figure 3

E o

0.10

0.09

219

£ 0.08 0) Q

0.07

B BMC 0.0501

BMD p < 0.001

S 0.045 -I J ,

<û 0.040 A c O 0.035

0.030

p< 0.001

l l là D Non-Tx Tx tritherapy-trained Tx untrained QTx tritherapy-trained + Cb

Figure 4

220

Control Soleus

MHCs MHCf Control MHCf

S 1 MX

— Type I fiber -* Type II fiber - » Hybrid fiber

Figure 5

221

CM

E

< V J O i_ <D

B CM E 3 < C/3 O 0)

CN E

< CO

o I— <D

-O

1800

15001

1200

900

600

1200

1000

800

600

1200-1

1000

800

600

Soleus Type I

www

n *** i

Type II

#

I- 1 n

1 * * * ; * * * t t t

II Hybrid

Hi t t t

EDL

D 1200

1000

800

600

E 600

500

400

300

Type II

i * * * * * * t t t

ill Hybrid

* * * J L * * * t t t

lilQ D Non-Tx Tx tritherapy-trained Tx untrained QTx tritherapy-trained + Cb

Figure 6

Table 1. Step or movement amplitude

Non-Tx Tx untrained T x | r i t h e^Py- Tx tritherapy-trained trained + clen

X axis (mm) 57.4 ±1.0 15.1 ± 0 . 7 * * t

27.7 ±0.5 29.0 ±0.4** CV (%) 12.8 25.7 10.9 11.0

Y axis (mm) 9.1 ± 0.2 f 0.0 ±0.0 8 .9±0.2 t 8.4 ± 0.3T

CV (%) 11.5 NA 16.1 18.3

222

223

Table 2. Soleus and EDL muscle cross-sectional area Non-Tx Tx untrained ^ ^ ^ Tx tritherapy-

trained trained + clen ** *

Soleus 0.85 ±0.05 0.56±0.05tt 0.64 ± 0.04n 0.93 ± 0.06 ** ** **

EDL 1.10 ±0.04 0.95 ±0.04™ 0.96 ± 0.04n 1.25 ±0.05

224

Table 3. Fiber type relative distribution

Non-Tx Tx untrained Tx tritherapy-trained

Tx tritherapy-trained + clen

Soleus Fiber type %

Type 1 54 6 ±2 .6* 2.9 ±1.5 5.4 ±1.9 0.3 ±0.2

Type II 45.4 ± 2.6 46.5 ± 2.5 44.5 ± 3.6 42.8 ± 3.4

Hybrid EDL

0.0 ± 0.0 * 50.7 ± 3.3 50.1 ±4.1 56.9 ± 3.3

Fiber type %

Type II 98.7 ± 0.3 98.7 ± 0.6 98.3 i 0.5 98.2 10.5

Hybrid 1.3 ±0.3 1.3 ±0.6 1.7 ±0.5 1.8 ±0.5

Table 4. Hormonal profile .. - - Txtntherapy- Tx tntherapy-Non-Tx Tx untrained . " . , trained trained + clen

Insulin (ng/ml)

IGF-1 (ng/ml)

1.57 ±0.23 784 ±28

0.58 ± 0.05 ***

1057 ±35 0.55 ± 0.04

918 ±35**

0.48 ± 0.02 925 ± 38**

PTH (ng/ml) 137 ±10 133112 121 ±6 125 ±7

DHEA (pg/ml) 210± 15 209 ± 25 216 ±27 188 ±40

225

226

CHAPITRE x - DISCUSSION ET CONCLUSION

Une des conséquences les plus notables des personnes blessées à la moelle épinière

est la perte des fonctions locomotrices causée par une rupture de communication avec les

centres supraspinaux. Pour l'instant, il est impossible de remplacer les connections perdues.

De ce fait, plusieurs laboratoires étudient les raisons de cette incapacité et tentent de trouver

une façon de rétablir ces connections. Les études effectuées dans le laboratoire du Dr Pierre

Guertin n'ont pas cette visée, mais tentent plutôt de trouver des traitements palliatifs dans

l'attente d'une percée majeure dans le domaine de la régénérescence de la moelle épinière.

Les travaux présentés pour cette thèse s'inscrivent dans cette optique. Nous avons tout

d'abord montré que chez la souris paraplégique, une récupération motrice spontanée mais

limitée existait. Dans le but d'améliorer le rétablissement locomoteur, nous avons évalué

plusieurs approches individuellement puis, nous avons combiné ces traitements pour une

approche multidisciplinaires.

10.1 Résumé des études de la thèse

L'article de revue proposé dans l'introduction (Ung et al., 2008) caractérise les

différents changements observés après une lésion complète de la moelle épinière chez la

souris. Les similitudes et différences entre notre modèle animal et les recherches chez

l'humain y sont notées.

La première étude (Ung et al., 2007) consistait à évaluer le rétablissement moteur

spontané en utilisant une variété de méthodes d'évaluation qualitative et semi-quantitative.

Nous y avons montré que chez la souris spinale, il existe une récupération motrice

spontanée mais limitée et que l'ACOS était la méthode d'évaluation la plus discriminative

pour évaluer ce faible niveau de recouvrement moteur.

Pour la deuxième étude (Ung et al., 2008), nous avons pharmacologiquement

disséqué la contribution des différents sous-types de récepteurs 5-HT2 durant l'induction de

LMs induits par la quipazine. Nous avons montré que les récepteurs 5-HT2A étaient

227

essentiels à l'induction de mouvements de type locomoteurs, alors que les récepteurs 5-

HT2B et 5-HT2C ne le sont pas. Nous avons également montré le décours spatio-temporel de

l'expression de l'ARNm des récepteurs 5-HT2A dans les heures et jours qui suivent la

transsection complète de la moelle épinière.

La troisième étude (Ung et al., 2010a) consistait à évaluer les effets de

l'administration de testosterone et de clenbuterol sur la composition corporelle et la

récupération motrice spontanée de souris paraplégiques. Les résultats nous ont permis de

noter que ces deux substances administrées seules ou en combinaison montraient de fortes

propriétés anaboliques au niveau musculaire, mais n'avaient aucune incidence sur le

rétablissement des fonctions locomotrices.

Par la suite, nous avons évalué l'effet de l'entraînement locomoteur sur tapis roulant,

sans aucune autre forme d'assistance, au niveau de la récupération motrice. Cependant,

cette forme d'entraînement n'a pas permis d'améliorer les performances motrices, ni la

composition corporelle des souris paraplégiques (Ung et al., 2010b).

Finalement, nous avons évalué les effets d'un entraînement régulier sur tapis roulant,

assisté par l'administration de L-Dopa + carbidopa + buspirone, et combiné à

l'administration de clenbuterol, sur le rétablissement des fonctions locomotrices et la

constitution corporelle générale. Nous avons montré que les performances locomotrices se

sont améliorées avec l'entraînement. L'ajout de clenbuterol permettait l'augmentation de la

masse musculaire, mais induisait une détérioration plus prononcée des os (article 5 de cette

thèse).

10.2 Limitations des études

10.2.1 Différence de rétablissement moteur entre souris mâles et femelles.

L'un des premiers sous-objectifs pour les travaux de la thèse consistait à évaluer les

différences de rétablissement moteur spontanné entre les souris mâles et femelles (Ung et

228

al., 2007, chapitre 5). Le rationnel derrière ce sous-objectif était que certaines conditions

pathologiques pouvaient être fortement influencées par le sexe (apnée du sommeil,

ostéoporose, sclérose en plaque, arthrite rhumatoïde). Au niveau des traumatismes du SNC

ou SNP, il n'y a pas de consensus en soi. Les hormones sexuelles mâles (testosterone) et

femelles (œstrogène/progestérone) montrent des effets neuroprotecteurs et réparateurs sur

les différents systèmes étudiés (lésion de nerf, croissance de neurites, plasticité,

synaptogénèse, augmentation de la synthèse de facteurs neurotrophiques, etc.). Mais en

générale, ces études sont très spécifiques et ne ciblent qu'un élément de la multitude

d'événements suivant un traumatisme du SNC ou du SNP. Ceci ne permet pas de donner

une image d'ensembl d'une blessure à la moelle épinière et du rétablissement moteur qui

s'en suit.

Une étude plus spécifique s'est consacrée aux différences de rétablissement

locomoteur entre les mâle/femelles chez le rat et la souris (Hauben et al., 2002). Les auteurs

ont montré que les femelles avaient un meilleur score locomoteur que les mâles suite à des

lésions partielles de moelle épinière.

En résumé, suite à la lésion :

• Les mâles récupéraient moins bien que les femelles

• Les femelles traitées à la dyhydrotestostérone récupéraient moins bien que des

femelles contrôles

• Les mâles non-castrés récupéraient moins bien que les mâles castrés.

Ils ont montré que les différences étaient reliées à la réponse immunitaire car chez

les souris nues, qui possèdent un système immunitaire affaibli ou absent, le rétablissement

moteur suite à la blessure à la moelle épinière était très faible, comparativement aux souris

normales contrôles. De plus, aucune différence de rétablissement moteur n'a été observée

entre les souris mâles et femelles chez les souris nues. Les auteurs ont alors suggéré qu'une

réponse immunitaire plus adéquate serait en place suite à une blessure à la moelle épinière

chez les femelles, ce qui favoriserait la récupération motrice. Nos données, ne permettent

pas d'observer une différence mâle/femelle pour le rétablissement de la motricité au niveau

des pattes arrière pour des lésions complètes. Ce type de lésion est extrêmement sévère, ce

229

qui limite en quelque sorte le rétablissement moteur. Nous n'avons cependant pas étudié les

processus moléculaire dans la moelle épinière suite à la lésion. Peut-être que les femelles

montre une meilleure réponse immunitaire et une « neuroprotection » de la moelle épinière

suite à la lésion. Par exemple, en évaluant les conséquences d'une lésion complète de la

moelle épinière peut-être aurait-il été possible d'observer moins de tissus cicatriciel, plus de

préservation des tissus de moelle épinière « sain » dans les régions avoisinant la lésion et

moins de dégénérescence dendritique ou axonal chez les souris femelles. Une étude récente

semble montrer qu'effectivement, la moelle épinière de souris femelles seraient mieux

conservées que celle de mâles après une blessure partielle de la moelle épinière (Farooque

et al., 2006).

10.2.2 Évaluation de la récupération locomotrice

Afin de quantifier et qualifier la récupération motrice, nous avons tout d'abord

utilisé des grilles d'évaluation locomotrice. Suite à l'étude proposée par Lapointe et

collaborateurs (2006) et l'article 1 de cette thèse, nous avons déterminé que l'utilisation de

deux grilles d'évaluation, ACOS et AOB, offraient une évaluation plus détaillée, rapide et

efficace que toutes autres grilles d'évaluations présentes dans la littérature, à cette date,

pour évaluer le rétablissement locomoteur chez la souris ayant une lésion complète de la

moelle épinière. Tout d'abord, ces deux méthodes d'évaluation ont spécifiquement été

développées pour les animaux ayant une transsection complète de la moelle épinière. Ainsi,

ces deux grilles offrent une meilleure résolution lorsque les niveaux de rétablissement

locomoteur sont faibles. D'autres méthodes d'évaluation telles que BMS (Basso et al., 2006)

ou la grille d'évaluation HiJK (Hillyer-Joynes Kinematics scale, Hillyer et Joynes, 2009)

pourraient s'avérer utiles à des niveaux de rétablissement locomoteur plus élevé, mais leurs

caractéristiques sont essentiellement les mêmes que l'AOB. De plus, aucune des méthodes

mentionnées ci-haut n'évaluent pas directement la fréquence de mouvements (i.e. le nombre

absolus de mouvements, LM ou NLM) générés, ce que propose l'ACOS. Nous croyons que

ce paramètre est une caractéristique importante du rétablissement locomoteur, puisque nous

avons montré que le nombre absolus de mouvements augmentait avec le temps (Guertin

230

2005; Lapointe et al., 2006; article 1 de cette thèse). L'ACOS et l'AOB sont donc utilisées

pour leur complémentarité. D'autres auteurs suggèrent également l'utilisation de plusieurs

méthodes d'évaluation du recouvrement locomoteur (Pajoohesh-Ganji et al., 2010).

Les analyses de cinématiques et des EMGs, souvent utilisées chez le chat

paraplégique, sont de plus en plus utilisées chez le rat et la souris (Leblond et al., 2003;

Fong et al., 2005; Liu et al., 2009; article 5 de cette thèse pour la cinématique). Pour nos

études, nous n'avons pas effectué d'analyse d'EMGs parce que nous avons préalablement

déterminé que les grilles d'évaluations (ACOS et AOB) combinées à des analyses de

cinématique étaient suffisantes pour nous informer sur les paramètres importants du

rétablissement locomoteur chez la souris paraplégique, soit l'augmentation de la fréquence

et de la qualité du mouvement. Nous avions aussi déterminé que l'utilisation d'EMGs à des

fins d'analyses statistiques occasionnerait d'importantes complications techniques. Pour une

acquisition rapide des données, il nous faudrait utiliser des électrodes de surface ou des

électrodes implantées de façon transdermique dans les muscles à étudier. Ces deux

méthodes sont inadéquates pour notre modèle animal puisque les mouvements de la souris,

combinés au mouvement du tapis pourraient facilement déplacer les électrodes. D nous

faudrait alors effectuer une implantation chronique par une méthode similaire à celle

proposée par Pearson et collaborateurs (2005). Mais l'implantation chronique d'électrodes

pose également problème car, à long terme, celles-ci pourraient être déplacées durant les

sessions d'entraînement ou à tout autre moment dans leur cage. Néanmoins, il demeure que

ces types d'analyses (EMG et cinématique) sont utiles pour une évaluation plus approfondie

du rétablissement moteur. Les informations recueillies nous informent davantage sur les

détails subtils, mais tout aussi importants, du rétablissement locomoteur, par exemple :

l'augmentation de l'amplitude de mouvement, l'augmentation de la coordination et de

l'amplitude de l'activité musculaire, les changements dans la variabilité du cycle de marche,

etc. Chez l'humain, ces analyses sont couramment utilisées (Leroux et al., 1999; Ladouceur

et Barbeau, 2000; Liinenburger et al., 2006; Forrest et al., 2008) et peuvent assurément

complementer d'autres formes d'évaluation qualitative et semi-quantitative du mouvement

(Dobkin et al., 2006, 2007; Ditunno et al., 2007).

231

Plus récemment, le laboratoire du Dr Edgerton a proposé une autre forme d'analyse

en cumulant 135 variables différentes provenant de la cinématique et des EMGs

enregistrées dans les pattes arrière gauches et droites de rats spinaux. Les auteurs ont

effectué une analyse des composantes principales (principal component analyses ou PCA).

Ce type d'analyse permet de déterminer les quelques variables qui expliquent les

différences entre les conditions expérimentales ou qui sont fortement impliquées dans le

rétablissement des fonctions locomotrices (Courtine et al., 2009). Ils ont démontré de

manière statistique que la réduction de la variabilité et l'amélioration de la stabilité du cycle

de marche, l'augmentation de l'amplitude des EMGs des pattes arrière et le retour de la

capacité à soutenir le poids lors de la locomotion étaient les critères qui expliquaient le

rétablissement locomoteur induit par leur combinaison thérapeutique. Cette information

plus détaillée complémenterait nos analyses concernant le retour des fonctions

locomotrices.

10.2.3 Pharmaco- et hormono-théraptes.

Des études in vivo chez le rongeur (Antri et al., 2002; Fong et al., 2003; de Leon et

Acosta, 2006; Courtine et al., 2009) ont montré que la quipazine pouvait aider le retour des

fonctions locomotrices. Suite à notre étude (Ung et al., 2008) constituant l'article 2, nous

avons montré que la quipazine induit son effet pro-locomoteur via l'activation des

récepteurs 5-HT2A- Des récentes études de marquage ont révélé une forte augmentation de

la densité de l'ARNm (Ung et al., 2008) et des récepteurs 5-HT2A (Kong et al., 2010) dans

la moelle épinière suite à une spinalisation (expliquant en partie le rétablissement moteur

spontané ou la supersensibilité causée par la dénervation) et que la modulation

sérotoninergique de la locomotion impliquerait les récepteurs 5-HT2A, 5-HTIA et 5-HT7

(Noga et al., 2009). Cependant la quipazine n'a pas été utilisée dans le cocktail

pharmacologique pour les études subséquentes (Article 5, Guertin et al., 2010, Guertin et

al., 2010 Epub). La principale raison pour cette omission est que cet agoniste 5-HT2 à large

spectre, administré seul de façon aiguë, n'induit pas un effet pro-locomoteur puissant. La

figure 10.1 montre 2 courbes dose-réponse chez la souris spinale. En air stepping et sur

tapis roulant, la quipazine adminstré à 1 mg/kg induit une fréquence maximale de 4

232

LMs/min et 10 LMs/min, respectivement. Ces mouvements sont généralement de faible

amplitude. D'autres agonistes sérotononinergiques utilisés dans notre laboratoire induisent

une plus grande incidence de LMs de grande amplitude. Par exemple le 8-OH-DPAT et la

buspirone induisent jusqu'à 30 LMs/min (Landry et al., 2006b; Guertin et al., 2010). De

plus, puisque les effets pro-locomoteurs sont principalement induits par l'activation des

récepteurs 5-HT2A et que la quipazine se lie aussi aux autres sous-types de récepteurs 5-

HT2, il se pourrait que l'activation des récepteurs 5-HT2B et 5-HT2C interfère avec

l'induction de LMs. En effet, aucun LM n'était observé suite à l'activation des récepteurs 5-

HT2B et 5-HT20 seuls des NLMs étaient induits (Landry et Guertin, 2004). Pour l'instant, il

n'existe pas d'agoniste qui se lie spécifiquement et avec une grande affinité au récepteur 5-

HT2A seul.

Il reste qu'une combinaison de drogues, telle qu'utilisée par Dr Edgerton et

collaborateurs (2009) pourrait s'avérer avantageux. Toutefois d'autres résultats provenant de

notre laboratoire (Guertin et al., 2010; article 5 de cette thèse) montrent que la combinaison

de L-Dopa + Carbidopa + Buspirone est très efficace. Il serait intéressant de voir si l'ajout

de quipazine ou d'un éventuel agoniste spécifique pour les récepteurs 5-HT2A en

combinaison avec la L-Dopa + Carbidopa + Buspirone permettrait d'obtenir de meilleures

performances locomotrices ou un retour plus rapide des fonctions locomotrices.

Certains auteurs suggèrent que le maintient ou l'augmentation de la masse

musculaire après BME serait bénéfique pour le rétablissement locomoteur (Stewart et al.,

2004). Nos études ne semblent pas corroborer ces résultats. Nos souris ayant reçu des

injections de clenbuterol et/ou testosterone propionate n'ont pas présenté une meilleure

récupération motrice spontanée (Ung et al., 2010a) et, nos souris entraînées avec ou sans

clenbuterol ont montré un niveau similaire de récupération (article 5 de cette thèse).

233

g E 2 c ai E ai > o E

8 -I

i l 2

O O ­OO

E c OJ E ® 10 o g

| 5

9

o o ­oo

Air stepping O Locomotor movements

L 1,0 2,0 3.0

Concentration (mg/kg)

Treadmill

2,0 3,0

Concentration (mg/kg)

R2= 0,182

40

A R"= 0,355

Figure 10.1 Courbe dose-réponse de mouvements locomoteurs pour la quipazine, en air stepping (graphique

du haut) et sur tapis roulant (graphique du bas). Il est à noter que ces mouvements locomoteurs ne comporte

ni placement plantaire, ni support de poids. La faible fréquence de movements induits nous dans l'élaboration

d'un cocktail pharmacologique.

Les liens entre les systèmes musculaire et osseux sont reconnus (pour une revue,

voir Gross et al., 2010). Plusieurs hormones et facteurs de croissances, tel que l'IGF-l, les

androgènes et l'estrogène modulent à la fois le développement musculaire et osseux. De

plus, l'activité musculaire est nécessaire au bon développement et au maintient des

propriétés osseuses. Ainsi, les augmentations du volume et de l'activité musculaire sont

234

correlés à des augmentations en masse osseuse (Zanchetta et al., 1995). D'autre part, les

diminutions de masse musculaire associées à des pathologies ou au vieillissement sont

généralement accompagnées de détérioration osseuse. Lors d'une blessure à la moelle

épinière, la paralysie des membres inférieurs diminue ou d'éliminé l'activité musculaire et

la prise en charge du poids par les jambes, ayant pour conséquence l'atrophie musculaire et

la détérioration osseuse. Ceci qui se traduit par un risque plus élevé de fractures chez la

population de blessés médullaires (Giangregorio et McCartney, 2006). Nous pensions que

l'entraînement assisté par l'administration de L-Dopa + Carbidopa + Buspirone avec ou sans

ajout de clenbuterol aurait renversé les pertes osseuses, ce qui n'a pas été le cas. Les souris

entraînées sans clenbuterol ont présenté une densité minérale osseuse (DMO) et un contenu

minéral osseux (CMO) similaires aux souris non-entrainées, corroborant certaines données

chez l'humain (Giangregorio et al., 2006). Davantage de détérioration osseuse a été

observée chez les souris entraînées qui ont reçu du clenbuterol (article 5 de cette thèse). À

l'exception de quelques études (Zeman et al., 1991; 1997; Apseloff et al., 1993) qui

montrent que l'ajout de clenbuterol est bénéfique pour l'os (ou n'induit pas de perte

osseuse), il est reconnu que le clenbuterol accentue la détérioration osseuse (Kitaura et al.,

2002; Bonnet et al., 2005). De ce point de vue, les résultats qui impliquent l'administration

de clenbuterol ne sont pas surprenants. Par contre, les raisons pour lesquelles l'entraînement

sur tapis roulant combiné à l'administration de L-DOPA + Carbidopa + Buspirone n'a pas

permis de contrer la détérioration osseuse demeurent incomprises. Certaines études

suggèrent que les améliorations en CMO et DMO seraient dépendantes du type d'exercice,

de l'intensité et de la durée de l'entraînement (pour une revue, voir Guadaloupe-Grau et al.,

2009). Il se pourrait que notre méthode d'entraînement n'ait pas été optimale en ce sens.

Peut-être aurait-il fallu entraîner nos animaux sur une plus longue période de temps ou

augmenter la durée et la fréquence des entraînements afin d'observer des effets bénéfiques

sur les propriétés osseuses. Une autre explication, plus indirecte, pour expliquer que

l'entraînement combiné à la pharmacologie n'a pas permis de contrer la détérioration

osseuse pourrait impliquer l'administration d'agonistes sérotoninergiques. De plus en plus

d'évidences montrent que la 5-HT aurait un rôle important sur la régulation de l'os. Des

récepteurs 5-HT ont été retrouvés sur certaines populations de cellules osseuses, telles que

les osteoblasts, ostéocytes et osteoclasts (Westbroek et al., 2001). La stimulation de ces

235

récepteurs pourrait avoir un effet inhibiteur sur la formation de l'os (Yadav et al., 2008). En

effet, une étude a montré une correlation négative entre le taux de 5-HT sanguin et le DMO

au niveau du fémur (Môdder et al., 2010). Par ailleurs, des personnes qui utilisent

régulièrement des inhibiteurs de la recapture de la sérotonine, ce qui augmente les

concentrations de sérotonine dans le sang, présentent un risque plus élevé de fractures

(Richards et al., 2007). Il serait possible que l'administration exogène d'agonistes

sérotoninergiques puisse être un autre facteur qui contribue à la détérioration osseuse. Cette

hypothèse reste toutefois à être validée sur notre modèle animal.

Il importe donc de développer d'autres stratégies afin de pallier à ces problèmes

secondaires. À ce sujet, la nandrolone reconnue pour ses propriétés anaboliques

musculaires (Joumaa et al., 2002; Zhao et al., 2008a) préserverait également la densité

osseuse suite à une dénervation de patte (Cardozo et al., 2010). La surexpression ou la

supplementation d'IGF-1 aurait des propriétés bénéfiques similaires (Clemmons, 2009; Elis

et al., 2010). D'autres stratégies n'impliquant pas l'utilisation de substances

pharmacologiques pourrait être utilisées, telle que la stimulation électrique fonctionnelle.

Celle-ci consiste à stimuler électriquement les nerfs ou les muscles paralysés de façon

séquentielle, de manière à reproduire une activation musculaire appropriée pour l'exécution

d'une tâche motrice. Cette thérapie a montré des bénéfices pour contrer l'atrophie

musculaire, la détérioration osseuse et faciliter le retour de la locomotion (Baldi et al.,

1998; Barbeau et al., 2002; Johnston et al., 2008; Ashe et al., 2010).

10.3 Perspectives futures pour notre approche multidisciplinaires

Afin de rétablir la locomotion ou de contrer les effets secondaires d'une BME, de

plus en plus d'études utilisent des combinaisons d'approches thérapeutiques en passant par

l'entraînement locomoteur, la stimulation électrique fonctionnelle, l'utilisation de drogues

ciblant le CPG, l'ajout de facteurs de croissance ou de cellules souches et le retrait de

molécules inhibitrices à la croissance axonale (Petersen et al., 2000; Barbeau et al., 2002;

236

Nothias et al., 2005; Vavrek et al., 2007; Kubasak et al., 2008; Haastert et al., 2008;

Courtine et al., 2009; Tom et al., 2009; Maier et al., 2009; Sandrow-Feinberg, 2010).

Dans les protocoles utilisant l'entraînement locomoteur, notre approche se distingue

des autres à plusieurs égards. Tout d'abord, nous entraînons les animaux en utilisant une

pharmacologie qui active le réseau locomoteur. L'entraînement locomoteur se fait de façon

quadrupède, une démarche plus naturelle pour l'animal. La combinaison de l'entraînement

et de la stimulation pharmacologique, à elle seule, est suffisante pour permettre le retour

des fonctions locomotrices avec support de poids. Le maintient de l'équilibre est toutefois

nécessaire (Guertin et al., 2010, Oct 15 Epub ahead of print; article 5 de cette thèse). La

pharmacologie proposée par le laboratoire du Dr Edgerton n'active pas le CPG, mais en

diminue son seuil d'activation (Fong et al., 2005; Courtine et al., 2009). Avec leur

approche, le poids des animaux doit être supporté car ceux-ci sont entraînés de façon

bipède. Les mouvements locomoteurs sont induits par un robot (rat stepper) et par

stimulations électriques épidurales. Ainsi, notre approche nécessite beaucoup moins

d'assistance pour obtenir un retour des fonctions locomotrices.

Il demeure que les paramètres d'entraînement proposés dans l'article 5 de cette

thèse pourraient être davantage optimisés. Plusieurs études montrent qu'il y aurait une

fenêtre temporelle à respecter pour bénéficier davantage d'un entraînement moteur. Suite à

un traumatisme crânien qui induisait une paralysie partielle d'une patte avant, les animaux

entraînés à utiliser cette patte une journée après la lésion montraient une plus faible

récupération motrice que ceux qui étaient entraînés 7 jours après la lésion. De plus, les

dommages corticaux reliés à l'entraînement hâtif était exacerbés (Risedal et al., 1999). Des

résultats similaires ont été notés chez l'animal partiellement lésé à la moelle épinière

(Girgis et al 2007; Krajacic et al. 2009). Les animaux qui étaient entraînés 4 ou 12 jours

post-chirurgie montraient des améliorations motrices similaires à la tâche pour laquelle ils

étaient entraînés (reaching task). Cependant les animaux dont l'entraînement avait débuté à

4 jours présentaient davantage de déficits à l'exécution d'une tâche locomotrice,

comparativement au groupe d'animaux qui a commencé l'entraînement 12 jours post­

chirurgie. Les dommages à la matière blanche étaient également moins importants pour ce

237

dernier groupe. Pour notre stratégie thérapeutique, il faudrait alors déterminer le meilleur

moment pour débuter l'entraînement. À titre d'exemple, des comparaisons sur les

performances locomotrices, la constitution corporelle et les dommages et la préservation de

la moelle épinière pourraient être effectuées pour des entraînement qui débuteraient à 3

jours, 1, 2 et 4 semaines suivant une lésion de la moelle épinière.

Dans un même ordre d'idée, il faudrait déterminer une fréquence d'entraînement

optimale qui permettrait d'obtenir les meilleures améliorations au niveau du rétablissement

locomoteur, des systèmes musculaire et osseux. Pour l'instant, la majorité des études chez

l'animal utilise des entraînements 5 fois semaines. Mais nous n'avons pas nécessairement

les preuves que ce nombre de session d'entraînement est nécessairement plus bénéfique

pour notre thérapie. Afin de ne pas sur- ou sous-entraînés les animaux, il serait intéressant

de poursuivre nos recherches en proposant différentes fréquences et durée d'entraînement

locomoteur.

Il faudrait peut-être optimiser la pharmacologie. La L-DOPA qui est un précurseur

de la dopamine (DA) et de la noradrenaline (NA) montre son effet pro-locomoteur

lorsqu'elle est administrée en combinaison avec la buspirone. Toutefois les récepteurs qui

sont ciblés par la L-DOPA ne sont pas connus de façon précise. Premièrement, il faudrait

déterminer si l'action de la L-DOPA passe par l'activation des récepteurs dopaminergiques,

adrénergiques ou par les 2 systèmes de neurotransmission. Pour répondre à cette question il

faudrait administrer la L-DOPA et 1) bloquer la conversion de la DA en NA et en évaluer

les effets sur le rétablissement locomoteur. On pourrait ainsi discriminer les effets de

l'activation des récepteurs NA. 2) utiliser un ou des antagonistes spécifiques pour tous les

sous-types de récepteurs DA, ce qui discriminerait les effets de l'activation des récepteurs

DA. 3) Utiliser des antagonistes ou une combinaison d'antagonistes qui sont spécifiques à

chaque sous-type de récepteurs NA et DA présents dans la moelle épinière pour évaluer

l'effet de l'activation (et du blocage) de chacun des récepteurs NA et DA. Les résultats

obtenus pourraient servir à cibler plus spécifiquement les sous-types de récepteurs

impliqués dans l'activation et la modulation des patrons locomoteurs induits. Il faudrait

également vérifier si l'ajout d'agonistes spécifiques pour les récepteurs 5-HT2A et 5-HT7

238

pourrait améliorer le traitement pharmacologique puisque ces deux récepteurs ainsi que le

récepteur 5-HTiA sont impliqués dans l'activation et la modulation des patrons locomoteurs

(Liu et Jordan, 2005; Landry et al., 2007; Noga et al., 2009).

Éventuellement, il faudrait évaluer l'effet de cette thérapie sur un modèle animal de

lésion partielle parce que la prévalence de ce type de blessure est plus élevée chez la

population de blessés médullaires. Même si cette pharmacologie est très efficace pour

activer le CPG locomoteur sur une moelle épinière complètement isolée, nous ne

connaissons pas les effets de cette thérapie en présence de l'influence partielle des centres

supraspinaux. Par exemple, sur ce modèle, il faudrait savoir si l'entraînement combiné à la

trithérapie accélère ou améliore le rétablissement locomoteur, comparativement à un

entraînement locomoteur induit manuellement sans pharmacologie activatrice du CPG.

Une approche qui vise à combiner plusieurs stratégies et/ou agents

pharmacologiques est à préconiser. Dans la majorité des cas, les combinaisons

thérapeutiques utilisées améliorent considérablement la locomotion, mais certaines

interactions s'avèrent nuisibles (Maier et al., 2009). En effet, alors que l'ajout d'anticorps

anti-NOGO-A et l'entraînement locomoteur amélioraient indépendamment différentes

caractéristiques de la locomotion, l'interaction des deux stratégies diminuait le

recouvrement fonctionnel. D importe donc de continuer à investiguer les interactions

possibles entre les différentes avenues thérapeutiques proposées, non seulement pour le

retour des fonctions locomotrices, mais aussi sur la santé générale des individus.

Que ce soit chez l'animal ou l'humain peu d'études ont évalué les effets de

l'entraînement locomoteur sur la santé générale, autres que la détérioration osseuse et

l'atrophie musculaire. Nous avons montré et répertorié les nombreux changements

biochimiques suite à une paralysie (chapitre 3; Rouleau et al., 2007; Rouleau et Guertin,

2010b). Chez la souris, mis à par une diminution de l'IGF-l, l'entraînement locomoteur

assisté par l'administration de L-DOPA + Carbidopa + Buspirone n'a pas eu d'effet

significatif sur les variations d'hormones (DHEA, insuline, PTH) et de cellules

immunitaires du sang (leucocytes, lymphocytes, monocytes, neutrophiles, éosinophiles et

239

basophiles) (observations non-publiées et Ung et al., 2010c). Un différent protocole

d'entraînement plus optimisé pourrait cependant influencer davantage ces variations.

Concernant la dysréflexie autonomique, les résultats sont contradictoires. Une étude a

montré que l'entraînement sur tapis roulant diminuait la distension colorectale (mécanisme

déclenchant les épisodes de dysréflexie), le rythme cardiaque et la pression artérielle

(Collins et Dicarlo, 2002) des rats spinaux. Des résultats contraires ont toutefois été notés

en suivant un protocole d'entraînement similaire (Laird et al., 2009). D'autre part,

l'entraînement locomoteur diminuerait également la douleur neuropathique (Hutchinson et

al., 2004).

Chez l'humain, le BWSTT a montré beaucoup d'effets bénéfiques sur la santé

générale des patients. Les personnes entraînées présentaient des améliorations au niveau du

rythme cardiaque, de la pression artérielle, de la circulation sanguine et du profil lipidique

(Jacobs et al., 2001; Ditor et al., 2005a,b; Stewart et al., 2004). Par contre, aucune évidence

ne permet pour l'instant de conclure que cette forme d'entraînement affecterait les tissus

adipeux. Par ailleurs, la dysréflexie autonomique n'a jamais été systématiquement évaluée

suite à des séances d'entraînement. Pour une étude (Forrest et al., 2008), des épisodes de

dysréflexie autonomique ont été observés chez un patient, alors qu'aucun des autres sujets

n'en n'ont eu. Outre la santé physique, la santé psychologique des patients peut aussi être

améliorée par l'entraînement. Une méta-analyse nous a rapporté que l'activité physique, peu

importe sa forme, améliorait significativement la qualité de vie générale des patients

(Martin Ginis et al., 2010). Le BWSTT a aussi montré des effets positifs sur le bien-être

psychologique (Hicks et al., 2005).

10.4 Vers une transition de ces approches chez le patient

Chez l'humain, les évidences d'effets bénéfiques de l'entraînement locomoteur

s'accumulent, mais malgré les nombreuses études chez l'animal, l'utilisation de substances

pharmacologiques pour aider le retour des fonctions locomotrices chez l'humain n'a que

240

très peu été étudiée (Fung et al., 1990, Wainberg et al., 1990; Stewart et al., 1991; Marie et

al., 2008).

Un avantage considérable de notre approche thérapeutique consiste en la

pharmacologie utilisée pour activer le CPG. Toutes les molécules présentes dans ce

cocktail, soit la buspirone, la L-Dopa et la carbidopa sont approuvées par les instances

Nord-Américaines en santé, ce qui pourrait s'avérer avantageux dans l'optique du

développement de ce médicament pour l'humain. Tous les tests toxicologiques ont déjà été

effectués pour ces molécules administrées seules, il ne resterait qu'à évaluer la toxicologie

du cocktail en soit. Des résultats préliminaires chez un patient monoplégique semblent

montrer que cette combinaison pharmacologique serait sécuritaire (Guertin et Brochu,

2009). De plus, une récente étude a rapporté l'efficacité de ce traitement administré de

façon orale, ce qui faciliterait son utilisation chez l'humain (Guertin et al., 2010).

Nous avons montré que l'administration du cocktail pharmacologique, à lui seul,

pouvait induire efficacement la locomotion ce qui permettait l'entraînement locomoteur. De

ce fait, cette approche se veut moins invasive que d'autres utilisant une combinaison de

drogues et de stimulations électriques épidurales (Courtine et al., 2009). De la même façon,

étant donné que cette pharmacologie est efficace pour induire une locomotion avec support

de poids, du moins chez la souris, la transition de cette thérapie chez l'humain réduirait

significativement les coûts associés à l'entraînement locomoteur qui nécessite l'intervention

de plusieurs techniciens et physiothérapeutes ou d'assistance robotisée (Morrison et Backus,

2007). Avec l'espérance de vie qui s'accroît chez la population de blessés médullaires et une

incidence plus élevée d'hospitalisations due à des complications secondaires, l'entraînement

locomoteur par l'assistance pharmacologique pourrait être un outil efficace pour le

maintient de cette population en santé et augmenter sa qualité de vie. Ainsi, les dépenses

reliées aux hospitalisations et aux médicaments utilisés (plus de 300 répertoriés, Rouleau et

Guertin, 2010a) pourraient être diminuées.

Un autre avantage relié à l'utilisation de cette pharmacologie est qu'elle induit une

locomotion chez la souris ayant une lésion complète de la moelle épinière. En transposant

241

chez l'humain, il serait fort possible que cette même combinaison puissent aider les blessés

médullaires classifies ASIA A et ASIA B et pour lesquelles l'entraînement locomoteur seul

n'est que rarement utilisé puisqu'aucun contrôle moteur n'est exprimé chez ces classes de

patients. L'administration de ce médicament pourrait d'autant plus faciliter l'entraînement et

le retour des fonctions locomotrices de patients étant classifies ASIA B, C et D. Des tests

sur des modèles animaux de lésion incomplète à la moelle épinière devraient être

préalablement effectués pour vérifier son efficacité sur ce type de blessure. Plusieurs

paramètres restent toutefois à être optimisés. Chez l'humain, les concentrations des

différentes molécules utilisées dans le cocktail, le développement d'une dépendance ou

d'une tolérance sont encore inconnues. Comme précédemment discuté, les paramètres reliés

à l'entraînement offrant le meilleur ratio coût énergétique/bénéfices reste à être déterminé.

Tout ceci doit également être évalué en fonction du type de blessure, de l'âge des patients,

et du temps depuis la blessure (Barbeau et al., 2006).

Conclusion

Peu de traitements existent pour contrer les effets dévastateurs d'une blessure à la

moelle épinière. Cette thèse avait pour but d'approfondir nos connaissances en ce domaine.

En utilisant un modèle de souris ayant une lésion complète de la moelle épinière, nous

avons montré que malgré le rétablissement moteur spontané, la récupération de la

locomotion est très limitée. Celle-ci est facilitée lorsque des stimulations adéquates, par

entraînement sur tapis roulant combiné à une pharmacologie capable d'activer le réseau

locomoteur, sont appliquées. Les différentes modalités d'entraînement (moment

d'implantation, fréquence, durée, intensité) restent à investiguer afin de déterminer leurs

influences sur le retour des fonctions locomotrices ainsi que sur l'état de santé générale.

Les influences corticales sur le recouvrement locomoteur ne sont pas non plus à négliger.

Puisque la majorité des blessés médullaires ont des lésions partielles de la moelle épinière,

il reste donc des connections avec les centres supraspinaux. D faudra travailler également

sur une réorganisation corticale fonctionnelle, mettre à profit ces connections restantes puis

évaluer leurs influences sur les réseaux spinaux, le rétablissement moteur et locomoteur.

242

D'autres troubles moteurs importants doivent être également ciblés pour favoriser le

rétablissement locomoteur. Entre autre la spasticité qui, non-traitée adéquatement, peut

nuire au rétablissement locomoteur en interférant avec la génération de mouvements. Il faut

noter que les traitements pharmacologiques présentement utilisés pour contrôler la

spasticité sont généralement proscrits lors d'entraînement locomoteur puisque leur

utilisation interfère avec l'intégration des inputs sensoriels dans la moelle épinière.

D'autres traitements doivent donc être recherchés afin de contrer les nombreux effets

secondaires d'une blessure à la moelle épinière. Nous espérons que ces recherches

contribueront à développer une thérapie palliative qui permettra aux blessés médullaires

d'améliorer leurs fonctions motrices, locomotrices et leur bilan de santé générale.

243

BIBLIOGRAPHIE

Anderson G, Wilkins E (1977) A trial of clenbuterol in bronchial asthma. Thorax 32:717-

719.

Antri M, Orsal D, Barthe JY (2002) Locomotor recovery in the chronic spinal rat: effects of

long-term treatment with a 5-HT2 agonist. Eur J Neurosci 16:467-476.

Antri M, Mouffle C, Orsal D, Barthe JY (2003) 5-HT 1A receptors are involved in short-

and long-term processes responsible for 5-HT-induced locomotor function recovery

in chronic spinal rat. Eur J Neurosci 18:1963-1972.

Antri M, Barthe JY, Mouffle C, Orsal D (2005) Long-lasting recovery of locomotor

function in chronic spinal rat following chronic combined pharmacological

stimulation of serotonergic receptors with 8-OHDPAT and quipazine. Neurosci Lett

384:162-167.

Apseloff G, Girten B, Walker M, Shepard DR, Krecic ME, Stern LS, Gerber N (1993)

Aminohydroxybutane bisphosphonate and clenbuterol prevent bone changes and

retard muscle atrophy respectively in tail-suspended rats. J Pharmacol Exp Ther

264:1071-1078.

Ashby P, Verrier M, Lightfoot E (1974) Segmental reflex pathways in spinal shock and

spinal spasticity in man. J Neurol Neurosurg Psychiatry 37:1352-1360.

Ashe MC, Eng JJ, Krassioukov AV, Warburton DE, Hung C, Tawashy A (2010) Response

to functional electrical stimulation cycling in women with spinal cord injuries using

dual-energy X-ray absorptiometry and peripheral quantitative computed

tomography: a case series. J Spinal Cord Med 33:68-72.

Axell AM, MacLean HE, Plant DR, Harcourt LJ, Davis JA, Jimenez M, Handelsman DJ,

Lynch GS, Zajac JD (2006) Continuous testosterone administration prevents

skeletal muscle atrophy and enhances resistance to fatigue in orchidectomized male

mice. Am J Physiol Endocrinol Metab 29LE506-516.

Baldi JC, Jackson RD, Moraille R, Mysiw WJ (1998) Muscle atrophy is prevented in

patients with acute spinal cord injury using functional electrical stimulation. Spinal

Cord 36:463-469.

244

Barbeau H, Rossignol S (1987) Recovery of locomotion after chronic spinalization in the

adult cat. Brain Res 412:84-95.

Barbeau H, Rossignol S (1990) The effects of serotonergic drugs on the locomotor pattern

and on cutaneous reflexes of the adult chronic spinal cat. Brain Res 514:55-67.

Barbeau H, Rossignol S (1991) Initiation and modulation of the locomotor pattern in the

adult chronic spinal cat by noradrenergic, serotonergic and dopaminergic drugs.

Brain Res 546:250-260.

Barbeau H, Rossignol S (1994). Enhancement of locomotor recovery following spinal cord

injury. Curr Opin Neurol 7(6):517-524.

Barbeau H, Wainberg M, Finch L (1987a) Description and application of a system for

locomotor rehabilitation. Med Biol Eng Comput 25:341-344.

Barbeau H, Julien C, Rossignol S (1987b) The effects of clonidine and yohimbine on

locomotion and cutaneous reflexes in the adult chronic spinal cat. Brain Res 437:83-

96.

Barbeau H, Chau C, Rossignol S (1993) Noradrenergic agonists and locomotor training

affect locomotor recovery after cord transection in adult cats. Brain Res Bull

30:387-393.

Barbeau H, Nadeau S, Garneau C (2006) Physical determinants, emerging concepts, and

training approaches in gait of individuals with spinal cord injury. J Neurotrauma

23:571-585.

Barbeau H, Ladouceur M, Mirbagheri MM, Kearney RE (2002) The effect of locomotor

training combined with functional electrical stimulation in chronic spinal cord

injured subjects: walking and reflex studies. Brain Res Brain Res Rev 40:274-291.

Barnes NM, Sharp T (1999) A review of central 5-HT receptors and their function.

Neuropharmacology 38:1083-1152.

Basso DM, Fisher LC, Anderson AJ, Jakeman LB, McTigue DM, Popovich PG (2006)

Basso Mouse Scale for locomotion detects differences in recovery after spinal cord

injury in five common mouse strains. J Neurotrauma 23:635-659.

Basura GJ, Zhou SY, Walker PD, Goshgarian HG (2001) Distribution of serotonin 2A and

2C receptor mRNA expression in the cervical ventral horn and phrenic motoneurons

following spinal cord hemisection. Exp Neurol 169:255-263.

245

Baxter G, Kennett G, Blaney F, Blackburn T (1995) 5-HT2 receptor subtypes: a family re­

united? Trends Pharmacol Sci 16:105-110.

Behrman AL, Harkema SJ (2000) Locomotor training after human spinal cord injury: a

series of case studies. Phys Ther 80:688-700.

Behrman AL, Lawless-Dixon AR, Davis SB, Bowden MG, Nair P, Phadke C, Hannold

EM, Plummer P, Harkema SJ (2005) Locomotor training progression and outcomes

after incomplete spinal cord injury. Phys Ther 85:1356-1371.

Bélanger M, Drew T, Provencher J, Rossignol S (1996) A comparison of treadmill

locomotion in adult cats before and after spinal transection. J Neurophysiol 76:471-

491.

Bennett DJ, De Serres S J, Stein RB (1996) Regulation of soleus muscle spindle sensitivity

in decerebrate and spinal cats during postural and locomotor activities. J Physiol

495 ( Pt 3):835-850.

Bhasin S, Storer TW, Berman N, Callegari C, Clevenger B, Phillips J, Bunnell TJ, Tricker

R, Shirazi A, Casaburi R (1996) The effects of supraphysiologic doses of

testosterone on muscle size and strength in normal men. N Engl J Med 335:1-7.

Bhasin S, Woodhouse L, Casaburi R, Singh AB, Bhasin D, Berman N, Chen X, Yarasheski

KE, Magliano L, Dzekov C, Dzekov J, Bross R, Phillips J, Sinha-Hikim I, Shen R,

Storer TW (2001) Testosterone dose-response relationships in healthy young men.

Am J Physiol Endocrinol Metab 281 :E1172-1181.

Birder L, de Groat W, Mills I, Morrison J, Thor K, Drake M (2010) Neural control of the

lower urinary tract: peripheral and spinal mechanisms. Neurourol Urodyn 29:128-

139.

Bonnet N, Brunet-Imbault B, Arlettaz A, Horcajada MN, Collomp K, Benhamou CL,

Courteix D (2005) Alteration of trabecular bone under chronic beta2 agonists

treatment. Med Sci Sports Exerc 37:1493-1501.

Bouyer LJ, Rossignol S (2003a) Contribution of cutaneous inputs from the hindpaw to the

control of locomotion. I. Intact cats. J Neurophysiol 90:3625-3639.

Bouyer LJ, Rossignol S (2003b) Contribution of cutaneous inputs from the hindpaw to the

control of locomotion. II. Spinal cats. J Neurophysiol 90:3640-3653.

246

Brocard F, Ryczko D, Fenelon K, Hatem R, Gonzales D, Auclair F, Dubuc R (2010) The

transformation of a unilateral locomotor command into a symmetrical bilateral

activation in the brainstem. J Neurosci 30:523-533.

Brown TJ, Khan T, Jones KJ (1999) Androgen induced acceleration of functional recovery

after rat sciatic nerve injury. Restor Neurol Neurosci 15:289-295.

Brustein E, Rossignol S (1999) Recovery of locomotion after ventral and ventrolateral

spinal lesions in the cat. II. Effects of noradrenergic and serotoninergic drugs. J

Neurophysiol 81:1513-1530.

Burke RE, Degtyarenko AM, Simon ES (2001) Patterns of locomotor drive to motoneurons

and last-order interneurons: clues to the structure of the CPG. J Neurophysiol

86:447-462.

Bussel B, Roby-Brami A, Azouvi P, Biraben A, Yakovleff A, Held JP (1988) Myoclonus in

a patient with spinal cord transection. Possible involvement of the spinal stepping

generator. Brain 111 ( Pt 5): 1235-1245.

Cai LL, Fong AJ, Otoshi CK, Liang Y, Burdick JW, Roy RR, Edgerton VR (2006)

Implications of assist-as-needed robotic step training after a complete spinal cord

injury on intrinsic strategies of motor learning. J Neurosci 26:10564-10568.

Calancie B, Needham-Shropshire B, Jacobs P, Wilier K, Zych G, Green BA (1994)

Involuntary stepping after chronic spinal cord injury. Evidence for a central rhythm

generator for locomotion in man. Brain 117 ( Pt 5): 1143-1159.

Capaday C (2002) The special nature of human walking and its neural control. Trends

Neurosci 25:370-376.

Capaday C, Stein RB (1986) Amplitude modulation of the soleus H-reflex in the human

during walking and standing. J Neurosci 6:1308-1313.

Capaday C, Stein RB (1987) Difference in the amplitude of the human soleus H reflex

during walking and running. J Physiol 392:513-522.

Cardozo CP, Qin W, Peng Y, Liu X, Wu Y, Pan J, Bauman WA, Zaidi M, Sun L (2010)

Nandrolone slows hindlimb bone loss in a rat model of bone loss due to

denervation. Ann N Y Acad Sci 1192:303-306.

247

Carter WJ, Dang AQ, Faas FH, Lynch ME (1991) Effects of clenbuterol on skeletal muscle

mass, body composition, and recovery from surgical stress in senescent rats.

Metabolism 40:855-860.

Cavalie H, Lac G, Lebecque P, Chanteranne B, Davicco MJ, Barlet JP (2002) Influence of

clenbuterol on bone metabolism in exercised or sedentary rats. J Appl Physiol

93:2034-2037.

Cazalets JR, Bertrand S (2000) Ubiquity of motor networks in the spinal cord of

vertebrates. Brain Res Bull 53:627-634.

Cazalets JR, Borde M, Clarac F (1995) Localization and organization of the central pattern

generator for hindlimb locomotion in newborn rat. J Neurosci 15:4943-4951.

Chau C, Barbeau H, Rossignol S (1998) Effects of intrathecal alpha 1- and alpha2-

noradrenergic agonists and norepinephrine on locomotion in chronic spinal cats. J

Neurophysiol 79:2941-2963.

Choi H, Liao WL, Newton KM, Onario RC, King AM, Désilets FC, Woodard EJ, Eichler

ME, Frontera WR, Sabharwal S, Teng YD (2005) Respiratory abnormalities

resulting from midcervical spinal cord injury and their reversal by serotonin IA

agonists in conscious rats. J Neurosci 25:4550-4559.

Choo JJ, Horan MA, Little RA, Rothwell NJ (1992) Anabolic effects of clenbuterol on

skeletal muscle are mediated by beta 2-adrenoceptor activation. Am J Physiol

263:E50-56.

Choong K, Lakshman KM, Bhasin S (2008) The physiological and pharmacological basis

for the ergogenic effects of androgens in elite sports. Asian J Androl 10:351-363.

Cina C, Hochman S (2000) Diffuse distribution of sulforhodamine-labeled neurons during

serotonin-evoked locomotion in the neonatal rat thoracolumbar spinal cord. J Comp

Neurol 423:590-602.

Clarac F (2008) Some historical reflections on the neural control of locomotion. Brain Res

Rev 57:13-21.

Clarac F, Pearlstein E (2007) Invertebrate preparations and their contribution to

neurobiology in the second half of the 20th century. Brain Res Rev 54:113-161.

Clemmons DR (2009) Role of IGF-I in skeletal muscle mass maintenance. Trends

Endocrinol Metab 20:349-356.

248

Collins HL, Dicarlo SE (2002) Acute exercise reduces the response to colon distension in

T(5) spinal rats. Am J Physiol Heart Circ Physiol 282:H1566-1570.

Conway BA, Hultborn H, Kiehn O (1987) Proprioceptive input resets central locomotor

rhythm in the spinal cat. Exp Brain Res 68:643-656.

Coonan JR, Greferath U, Messenger J, Hartley L, Murphy M, Boyd AW, Dottori M, Galea

MP, Bartlett PF (2001) Development and reorganization of corticospinal projections

in EphA4 deficient mice. J Comp Neurol 436:248-262.

Cornea-Hebert V, Riad M, Wu C, Singh SK, Descarries L (1999) Cellular and subcellular

distribution of the serotonin 5-HT2A receptor in the central nervous system of adult

rat. J Comp Neurol 409:187-209.

Courtine G, Gerasimenko Y, van den Brand R, Yew A, Musienko P, Zhong H, Song B, Ao

Y, Ichiyama RM, Lavrov I, Roy RR, Sofroniew MV, Edgerton VR (2009)

Transformation of nonfunctional spinal circuits into functional states after the loss

of brain input. Nat Neurosci 12:1333-1342.

Cowley KC, Schmidt BJ (1994) A comparison of motor patterns induced by N-methyl-D-

aspartate, acetylcholine and serotonin in the in vitro neonatal rat spinal cord.

Neurosci Lett 171:147-150.

Crone SA, Quinlan KA, Zagoraiou L, Droho S, Restrepo CE, Lundfald L, Endo T, Setlak J,

Jessell TM, Kiehn O, Sharma K (2008) Genetic ablation of V2a ipsilateral

interneurons disrupts left-right locomotor coordination in mammalian spinal cord.

Neuron 60:70-83.

Dahlstrom A, Fuxe K (1964) Localization of monoamines in the lower brain stem.

Experientia 20:398-399.

Dai X, Noga BR, Douglas JR, Jordan LM (2005) Localization of spinal neurons activated

during locomotion using the c-fos immunohistochemical method. J Neurophysiol

93:3442-3452.

de Leon RD, Acosta CN (2006) Effect of robotic-assisted treadmill training and chronic

quipazine treatment on hindlimb stepping in spinally transected rats. J Neurotrauma

23:1147-1163.

De Leon RD, Hodgson JA, Roy RR, Edgerton VR (1998) Full weight-bearing hindlimb

standing following stand training in the adult spinal cat. J Neurophysiol 80:83-91.

249

de Leon RD, Hodgson JA, Roy RR, Edgerton VR (1998) Locomotor capacity attributable

to step training versus spontaneous recovery after spinalization in adult cats. J

Neurophysiol 79:1329-1340.

de Leon RD, Reinkensmeyer DJ, Timoszyk WK, London NJ, Roy RR, Edgerton VR

(2002a) Use of robotics in assessing the adaptive capacity of the rat lumbar spinal

cord. Prog Brain Res 137:141-149.

de Leon RD, Kubasak MD, Phelps PE, Timoszyk WK, Reinkensmeyer DJ, Roy RR,

Edgerton VR (2002b) Using robotics to teach the spinal cord to walk. Brain Res

Brain Res Rev 40:267-273.

De Mello MT, Esteves AM, Tufik S (2004) Comparison between dopaminergic agents and

physical exercise as treatment for periodic limb movements in patients with spinal

cord injury. Spinal Cord 42:218-221.

Deliagina TG, Beloozerova IN, Zelenin PV, Orlovsky GN (2008) Spinal and supraspinal

postural networks. Brain Res Rev 57:212-221.

Delvolve I, Gabbay H, Lev-Tov A (2001) The motor output and behavior produced by

rhythmogenic sacrocaudal networks in spinal cords of neonatal rats. J Neurophysiol

85:2100-2110.

Dietz V, Colombo G, Jensen L, Baumgartner L (1995) Locomotor capacity of spinal cord

in paraplegic patients. Ann Neurol 37:574-582.

Dietz V, Wirz M, Curt A, Colombo G (1998) Locomotor pattern in paraplegic patients:

training effects and recovery of spinal cord function. Spinal Cord 36:380-390.

Dimitrijevic MR, Gerasimenko Y, Pinter MM (1998) Evidence for a spinal central pattern

generator in humans. Ann N Y Acad Sci 860:360-376.

Ditor DS, Kamath MV, MacDonald MJ, Bugaresti J, McCartney N, Hicks AL (2005a)

Effects of body weight-supported treadmill training on heart rate variability and

blood pressure variability in individuals with spinal cord injury. J Appl Physiol

98:1519-1525.

Ditor DS, Macdonald MJ, Kamath MV, Bugaresti J, Adams M, McCartney N, Hicks AL

(2005b) The effects of body-weight supported treadmill training on cardiovascular

regulation in individuals with motor-complete SCI. Spinal Cord 43:664-673.

250

Ditunno JF, Little JW, Tessler A, Burns AS (2004) Spinal shock revisited: a four-phase

model. Spinal Cord 42:383-395.

Ditunno JF, Jr., Barbeau H, Dobkin BH, Elashoff R, Harkema S, Marino RJ, Hauck WW,

Apple D, Basso DM, Behrman A, Deforge D, Fugate L, Saulino M, Scott M, Chung

J (2007) Validity of the walking scale for spinal cord injury and other domains of

function in a multicenter clinical trial. Neurorehabil Neural Repair 21:539-550.

Dobkin B, Barbeau H, Deforge D, Ditunno J, Elashoff R, Apple D, Basso M, Behrman A,

Harkema S, Saulino M, Scott M (2007) The evolution of walking-related outcomes

over the first 12 weeks of rehabilitation for incomplete traumatic spinal cord injury:

the multicenter randomized Spinal Cord Injury Locomotor Trial. Neurorehabil

Neural Repair 21:25-35.

Dobkin B, Apple D, Barbeau H, Basso M, Behrman A, Deforge D, Ditunno J, Dudley G,

Elashoff R, Fugate L, Harkema S, Saulino M, Scott M (2006) Weight-supported

treadmill vs over-ground training for walking after acute incomplete SCI.

Neurology 66:484-493.

Dobkin BH, Havton LA (2004) Basic advances and new avenues in therapy of spinal cord

injury. Annu Rev Med 55:255-282.

Dodd SL, Koesterer TJ (2002) Clenbuterol attenuates muscle atrophy and dysfunction in

hindlimb-suspended rats. Aviat Space Environ Med 73:635-639.

Doly S, Fischer J, Brisorgueil MJ, Verge D, Conrath M (2005) Pre- and postsynaptic

localization of the 5-HT7 receptor in rat dorsal spinal cord: immunocytochemical

evidence. J Comp Neurol 490:256-269.

Doly S, Madeira A, Fischer J, Brisorgueil MJ, Daval G, Bernard R, Verge D, Conrath M

(2004) The 5-HT2A receptor is widely distributed in the rat spinal cord and mainly

localized at the plasma membrane of postsynaptic neurons. J Comp Neurol

472:496-511.

Drew T (1988) Motor cortical cell discharge during voluntary gait modification. Brain Res

457:181-187.

Duysens J, Pearson KG (1980) Inhibition of flexor burst generation by loading ankle

extensor muscles in walking cats. Brain Res 187:321-332.

251

Edgerton VR, Roy RR (2009a) Activity-dependent plasticity of spinal locomotion:

implications for sensory processing. Exerc Sport Sci Rev 37:171-178.

Edgerton VR, Roy RR (2009b) Robotic training and spinal cord plasticity. Brain Res Bull

78:4-12.

Edgerton VR, Leon RD, Harkema SJ, Hodgson JA, London N, Reinkensmeyer DJ, Roy

RR, Talmadge RJ, Tillakaratne NJ, Timoszyk W, Tobin A (2001) Retraining the

injured spinal cord. J Physiol 533:15-22.

Edgerton VR, Courtine G, Gerasimenko YP, Lavrov I, Ichiyama RM, Fong AJ, Cai LL,

Otoshi CK, Tillakaratne NJ, Burdick JW, Roy RR (2008) Training locomotor

networks. Brain Res Rev 57:241-254.

Elis S, Courtland HW, Wu Y, Rosen CJ, Sun H, Jepsen KJ, Majeska RJ, Yakar S (2010)

Elevated serum levels of IGF-1 are sufficient to establish normal body size and

skeletal properties even in the absence of tissue IGF-1. J Bone Miner Res 25:1257-

1266.

Engesser-Cesar C, Ichiyama RM, Nefas AL, Hill MA, Edgerton VR, Cotman CW,

Anderson AJ (2007) Wheel running following spinal cord injury improves

locomotor recovery and stimulates serotonergic fiber growth. Eur J Neurosci

25:1931-1939.

Eriksson J, Olausson B, Jankowska E (1996) Antispastic effects of L-dopa. Exp Brain Res

111:296-304.

Farooque M, Suo Z, Arnold PM, Wulser MJ, Chou CT, Vancura RW, Fowler S, Festoff

BW. (2006) Gender-related differences in recovery of locomotor function after

spinal cord injury in mice. Spinal Cord. 44:182-7.

Fedirchuk B, Nielsen J, Petersen N, Hultborn H (1998) Pharmacologically evoked fictive

motor patterns in the acutely spinalized marmoset monkey (Callithrix jacchus). Exp

Brain Res 122:351-361.

Ferrando AA, Tipton KD, Doyle D, Phillips SM, Cortiella J, Wolfe RR (1998)

Testosterone injection stimulates net protein synthesis but not tissue amino acid

transport. Am J Physiol 275:E864-871.

Filip M, Bader M (2009) Overview on 5-HT receptors and their role in physiology and

pathology of the central nervous system. Pharmacol Rep 61:761-777.

252

Fong AJ, Cai LL, Otoshi CK, Reinkensmeyer DJ, Burdick JW, Roy RR, Edgerton VR

(2005) Spinal cord-transected mice learn to step in response to quipazine treatment

and robotic training. J Neurosci 25:11738-11747.

Fong AJ, Roy RR, Ichiyama RM, Lavrov I, Courtine G, Gerasimenko Y, Tai YC, Burdick

J, Edgerton VR (2009) Recovery of control of posture and locomotion after a spinal

cord injury: solutions staring us in the face. Prog Brain Res 175:393-418.

Fonseca MI, Ni YG, Dunning DD, Miledi R (2001) Distribution of serotonin 2A, 2C and 3

receptor mRNA in spinal cord and medulla oblongata. Brain Res Mol Brain Res

89:11-19.

Forrest GF, Sisto SA, Barbeau H, Kirshblum SC, Wilen J, Bond Q, Bentson S, Asselin P,

Cirnigliaro CM, Harkema S (2008) Neuromotor and musculoskeletal responses to

locomotor training for an individual with chronic motor complete AIS-B spinal cord

injury. J Spinal Cord Med 31:509-521.

Forssberg H (1979) Stumbling corrective reaction: a phase-dependent compensatory

reaction during locomotion. J Neurophysiol 42:936-953.

Forssberg H, Grillner S, Rossignol S (1975) Phase dependent reflex reversal during

walking in chronic spinal cats. Brain Res 85:103-107.

Forssberg H, Grillner S, Rossignol S (1977) Phasic gain control of reflexes from the

dorsum of the paw during spinal locomotion. Brain Res 132:121-139.

Forssberg H, Grillner S, Halbertsma J (1980a) The locomotion of the low spinal cat. I.

Coordination within a hindlimb. Acta Physiol Scand 108:269-281.

Forssberg H, Grillner S, Halbertsma J, Rossignol S (1980b) The locomotion of the low

spinal cat. II. Interlimb coordination. Acta Physiol Scand 108:283-295.

Frigon A, Rossignol S (2008) Adaptive changes of the locomotor pattern and cutaneous

reflexes during locomotion studied in the same cats before and after spinalization. J

Physiol 586:2927-2945.

Frigon A, Sirois J, Gossard JP (2010) Effects of ankle and hip muscle afferent inputs on

rhythm generation during fictive locomotion. J Neurophysiol 103:1591-1605.

Fung J, Stewart JE, Barbeau H (1990) The combined effects of clonidine and

cyproheptadine with interactive training on the modulation of locomotion in spinal

cord injured subjects. J Neurol Sci 100:85-93.

253

Giangregorio L, McCartney N (2006) Bone loss and muscle atrophy in spinal cord injury:

epidemiology, fracture prediction, and rehabilitation strategies. J Spinal Cord Med

29:489-500.

Giangregorio LM, Webber CE, Phillips SM, Hicks AL, Craven BC, Bugaresti JM,

McCartney N (2006) Can body weight supported treadmill training increase bone

mass and reverse muscle atrophy in individuals with chronic incomplete spinal cord

injury? Appl Physiol Nutr Metab 31:283-291.

Girgis J, Merrett D, Kirkland S, Metz GA, Verge V, Fouad K (2007) Reaching training in

rats with spinal cord injury promotes plasticity and task specific recovery. Brain

130:2993-3003.

Giroux N, Rossignol S, Reader TA (1999) Autoradiographic study of alphal- and alpha2-

noradrenergic and serotonin 1A receptors in the spinal cord of normal and

chronically transected cats. J Comp Neurol 406:402-414.

Giroux N, Reader TA, Rossignol S (2001) Comparison of the effect of intrathecal

administration of clonidine and yohimbine on the locomotion of intact and spinal

cats. J Neurophysiol 85:2516-2536.

Giuliano F, Rampin O (2004) Neural control of erection. Physiol Behav 83:189-201.

Giuliano F, Clement P (2005) Physiology of ejaculation: emphasis on serotonergic control.

Eur Urol 48:408-417.

Gomez-Pinilla F, Ying Z, Roy RR, Hodgson J, Edgerton VR (2004) Afferent input

modulates neurotrophins and synaptic plasticity in the spinal cord. J Neurophysiol

92:3423-3432.

Gooren LJ, Bunck MC (2004) Androgen replacement therapy: present and future. Drugs

64:1861-1891.

Gorassini MA, Prochazka A, Hiebert GW, Gauthier MJ (1994) Corrective responses to loss

of ground support during walking. I. Intact cats. J Neurophysiol 71:603-610.

Gosgnach S, Quevedo J, Fedirchuk B, McCrea DA (2000) Depression of group Ia

monosynaptic EPSPs in cat hindlimb motoneurones during fictive locomotion. J

Physiol 526 Pt 3:639-652.

254

Gosgnach S, Lanuza GM, Butt SJ, Saueressig H, Zhang Y, Velasquez T, Riethmacher D,

Callaway EM, Kiehn O, Goulding M (2006) VI spinal neurons regulate the speed

of vertebrate locomotor outputs. Nature 440:215-219.

Goulding M, Lanuza G, Sapir T, Narayan S (2002) The formation of sensorimotor circuits.

Curr Opin Neurobiol 12:508-515.

Graham MR, Davies B, Grace FM, Kicman A, Baker JS (2008) Anabolic steroid use:

patterns of use and detection of doping. Sports Med 38:505-525.

Graham Brown T, (1911) The intrinsic factors in the act of progression in mammal. Proc.

Roy. Soc. B 84, 308-319.

Graham Brown T (1914) On the nature of the fundamental activity of the nervous centres;

together with an analysis of the conditioning of rhythmic activity in progression,

and a theory of the evolution of function in the nervous system. J. Physiol. (London)

48, 18^6.

Griesbach GS, Gomez-Pinilla F, Hovda DA (2007) Time window for voluntary exercise-

induced increases in hippocampal neuroplasticity molecules after traumatic brain

injury is severity dependent. J Neurotrauma 24:1161-1171.

Griffin L, Decker MJ, Hwang JY, Wang B, Kitchen K, Ding Z, Ivy JL (2009) Functional

electrical stimulation cycling improves body composition, metabolic and neural

factors in persons with spinal cord injury. J Electromyogr Kinesiol 19:614-622.

Grillner S (1981) Control of locomotion in bipeds, tetrapods, and fish in : Handbook of

Physiology, J.M. Brookhart, V.B. Mountcastle (Eds). Handbook of Physiology, The

nervous system II 2:1179-1236.

Grillner S, Rossignol S (1978) On the initiation of the swing phase of locomotion in

chronic spinal cats. Brain Res 146:269-277.

Grillner S, Zangger P (1979) On the central generation of locomotion in the low spinal cat.

Exp Brain Res 34:241-261.

Gross TS, Poliachik SL, Prasad J, Bain SD (2010) The effect of muscle dysfunction on

bone mass and morphology. J Musculoskelet Neuronal Interact 10:25-34.

Grossman SD, Rosenberg LJ, Wrathall JR (2001) Relationship of altered glutamate

receptor subunit mRNA expression to acute cell loss after spinal cord contusion.

Exp Neurol 168:283-289.

255

Grossman SD, Wolfe BB, Yasuda RP, Wrathall JR (2000) Changes in NMDA receptor

subunit expression in response to contusive spinal cord injury. J Neurochem

75:174-184.

Grotewiel MS, Sanders-Bush E (1999) Differences in agonist-independent activity of 5-

Ht2A and 5-HT2c receptors revealed by heterologous expression. Naunyn

Schmiedebergs Arch Pharmacol 359:21-27.

Guadalupe-Grau A, Fuentes T, Guerra B, Calbet JA (2009) Exercise and bone mass in

adults. Sports Med 39:439-468.

Guertin P, Angel MJ, Perreault MC, McCrea DA (1995) Ankle extensor group I afférents

excite extensors throughout the hindlimb during fictive locomotion in the cat. J

Physiol 487 ( Pt 1): 197-209.

Guertin PA (2004a) Synergistic activation of the central pattern generator for locomotion

by l-beta-3,4-dihydroxyphenylalanine and quipazine in adult paraplegic mice.

Neurosci Lett 358:71-74.

Guertin PA (2004b) Role of NMDA receptor activation in serotonin agonist-induced air-

stepping in paraplegic mice. Spinal Cord 42:185-190.

Guertin PA (2005) Semiquantitative assessment of hindlimb movement recovery without

intervention in adult paraplegic mice. Spinal Cord 43:162-166.

Guertin PA (2009) The mammalian central pattern generator for locomotion. Brain Res

Rev 62:45-56.

Guertin PA, Brochu C (2009) Preliminary evidence of safety following administration of L-

DOPA and buspirone in an incomplete monoplegic patient. Spinal Cord 47:91-92.

Guertin PA, Steuer I (2009) Key central pattern generators of the spinal cord. J Neurosci

Res 87:2399-2405.

Guertin PA, Ung RV, Rouleau P (2010) Oral administration of a tri-therapy for central

pattern generator activation in paraplegic mice: proof-of-concept of efficacy.

Biotechnol J 5:421-426.

Haastert K, Ying Z, Grothe C, Gomez-Pinilla F (2008) The effects of FGF-2 gene therapy

combined with voluntary exercise on axonal regeneration across peripheral nerve

gaps. Neurosci Lett 443:179-183.

256

Halbertsma JM (1983) The stride cycle of the cat: the modelling of locomotion by

computerized analysis of automatic recordings. Acta Physiol Scand Suppl 521:1-75.

Hartgens F, Kuipers H, Wijnen JA, Keizer HA (1996) Body composition, cardiovascular

risk factors and liver function in long-term androgenic-anabolic steroids using

bodybuilders three months after drug withdrawal. Int J Sports Med 17:429-433.

Hashim MA, Bieger D (1987) Excitatory action of 5-HT on deglutitive substrates in the rat

solitary complex. Brain Res Bull 18:355-363.

Hauben E, Mizrahi T, Agranov E, Schwartz M (2002) Sexual dimorphism in the

spontaneous recovery from spinal cord injury: a gender gap in beneficial

autoimmunity? Eur J Neurosci. 2002 16(9): 1731-40.

Hayes A, Williams DA (1994) Long-term clenbuterol administration alters the isometric

contractile properties of skeletal muscle from normal and dystrophin-deficient mdx

mice. Clin Exp Pharmacol Physiol 21:757-765.

Hedlund PB, Kelly L, Mazur C, Lovenberg T, Sutcliffe JG, Bonaventure P (2004) 8-OH-

DPAT acts on both 5-HT 1A and 5-HT7 receptors to induce hypothermia in rodents.

Eur J Pharmacol 487:125-132.

Helton LA, Thor KB, Baez M (1994) 5-hydroxytryptamine2A, 5-hydroxytryptamine2B,

and 5-hydroxytryptamine2C receptor mRNA expression in the spinal cord of rat,

cat, monkey and human. Neuroreport 5:2617-2620.

Herbst KL, Bhasin S (2004) Testosterone action on skeletal muscle. Curr Opin Clin Nutr

Metab Care 7:271-277.

Hicke L, Dunn R (2003) Regulation of membrane protein transport by ubiquitin and

ubiquitin-binding proteins. Annu Rev Cell Dev Biol 19:141-172.

Hicks AL, Adams MM, Martin Ginis K, Giangregorio L, Latimer A, Phillips SM,

McCartney N (2005) Long-term body-weight-supported treadmill training and

subsequent follow-up in persons with chronic SCI: effects on functional walking

ability and measures of subjective well-being. Spinal Cord 43:291-298.

Hillyer JE, Joynes RL (2009) A new measure of hindlimb stepping ability in neonatally

spinalized rats. Behav Brain Res 202:291-302.

257

Hinckley CA, Hartley R, Wu L, Todd A, Ziskind-Conhaim L (2005) Locomotor-like

rhythms in a genetically distinct cluster of interneurons in the mammalian spinal

cord. J Neurophysiol 93:1439-1449.

Hochman S, Garraway SM, Machacek DW, Shay BL (2001) 5-HT receptors and the

neuromodulatory control of spinal cord function, in: Motor Neurobiology of the

spinal cord/ ed. by Timothy C. Cope. CRC Press LLC47-87.

Holohean AM, Hackman JC (2004) Mechanisms intrinsic to 5-HT2B receptor-induced

potentiation of NMDA receptor responses in frog motoneurones. Br J Pharmacol

143:351-360.

Holohean AM, Hackman JC, Shope SB, Davidoff RA (1992) Serotonin 1A facilitation of

frog motoneuron responses to afferent stimuli and to N-methyl-D-aspartate.

Neuroscience 48:469-477.

Huang HJ, Ferris DP (2004) Neural coupling between upper and lower limbs during

recumbent stepping. J Appl Physiol 97:1299-1308.

Huang J, Forsberg NE (1998) Role of calpain in skeletal-muscle protein degradation. Proc

Natl Acad Sci U S A 95:12100-12105.

Hughes GM, Wiersma CAG (1960) The co-ordination of swimmeret movements in the

crayfish, Procambarus clarkia (Girard). J. Exp. Biol. 37, 657-670.

Hultbom H, Nielsen JB (2007) Spinal control of locomotion—from cat to man. Acta Physiol

(Oxf) 189:111-121.

Hutchinson KJ, Gomez-Pinilla F, Crowe MJ, Ying Z, Basso DM (2004) Three exercise

paradigms differentially improve sensory recovery after spinal cord contusion in

rats. Brain 127:1403-1414.

Jackman RW, Kandarian SC (2004) The molecular basis of skeletal muscle atrophy. Am J

Physiol Cell Physiol 287:C834-843.

Jackson DA, White SR (1990) Receptor subtypes mediating facilitation by serotonin of

excitability of spinal motoneurons. Neuropharmacology 29:787-797.

Jacobs PL, Nash MS, Rusinowski JW (2001) Circuit training provides cardiorespiratory

and strength benefits in persons with paraplegia. Med Sci Sports Exerc 33:711-717.

Jagoe RT, Goldberg AL (2001) What do we really know about the ubiquitin-proteasome

pathway in muscle atrophy? Curr Opin Clin Nutr Metab Care 4:183-190.

258

Jankowska E, Jukes MG, Lund S, Lundberg A (1967a) The effect of DOPA on the spinal

cord. 6. Half-centre organization of interneurones transmitting effects from the

flexor reflex afférents. Acta Physiol Scand 70:389-402.

Jankowska E, Jukes MG, Lund S, Lundberg A (1967b) The effect of DOPA on the spinal

cord. 5. Reciprocal organization of pathways transmitting excitatory action to alpha

motoneurones of flexors and extensors. Acta Physiol Scand 70:369-388.

Jessell TM (2000) Neuronal specification in the spinal cord: inductive signals and

transcriptional codes. Nat Rev Genet 1:20-29.

Johansen JA, Yu Z, Mo K, Monks DA, Lieberman AP, Breedlove SM, Jordan CL (2009)

Recovery of function in a myogenic mouse model of spinal bulbar muscular

atrophy. Neurobiol Dis 34:113-120.

Johnston TE, Smith BT, Oladeji O, Betz RR, Lauer RT (2008) Outcomes of a home cycling

program using functional electrical stimulation or passive motion for children with

spinal cord injury: a case series. J Spinal Cord Med 31:215-221.

Johnston TE, Smith BT, Mulcahey MJ, Betz RR, Lauer RT (2009) A randomized

controlled trial on the effects of cycling with and without electrical stimulation on

cardiorespiratory and vascular health in children with spinal cord injury. Arch Phys

Med Rehabil 90:1379-1388.

Joumaa WH, Bouhlel A, Leoty C (2002) Post-effects of nandrolone decanoate treatment on

contractile responses of rat skeletal muscles. Jpn J Physiol 52:479-487.

Kadi F (2008) Cellular and molecular mechanisms responsible for the action of testosterone

on human skeletal muscle. A basis for illegal performance enhancement. Br J

Pharmacol 154:522-528.

Kandarian SC, Jackman RW (2006) Intracellular signaling during skeletal muscle atrophy.

Muscle Nerve 33:155-165.

Kao T, Shumsky JS, Jacob-Vadakot S, Himes BT, Murray M, Moxon KA (2006) Role of

the 5-HT2C receptor in improving weight-supported stepping in adult rats

spinalized as neonates. Brain Res 1112:159-168.

Kiehn O, Quinlan KA, Restrepo CE, Lundfald L, Borgius L, Talpalar AE, Endo T (2008)

Excitatory components of the mammalian locomotor CPG. Brain Res Rev 57:56-63.

259

Kim CM, Eng JJ, Whittaker MW (2004) Level walking and ambulatory capacity in persons

with incomplete spinal cord injury: relationship with muscle strength. Spinal Cord

42:156-162.

Kim D, Murray M, Simansky KJ (2001) The serotonergic 5-HT(2C) agonist m-

chlorophenylpiperazine increases weight-supported locomotion without

development of tolerance in rats with spinal transections. Exp Neurol 169:496-500.

Kim D, Adipudi V, Shibayama M, Giszter S, Tessler A, Murray M, Simansky KJ (1999)

Direct agonists for serotonin receptors enhance locomotor function in rats that

received neural transplants after neonatal spinal transection. J Neurosci 19:6213-

6224.

Kinkead R, Belzile O, Gulemetova R (2002) Serotonergic modulation of respiratory motor

output during tadpole development. J Appl Physiol 93:936-946.

Kinkead R. (2009). Phylogenetic trends in respiratory rhythmogenesis: insights from

ectothermic vertebrates. Respir Physiol Neurobiol 168(l-2):39-48.

Kitaura T, Tsunekawa N, Kraemer WJ (2002) Inhibited longitudinal growth of bones in

young male rats by clenbuterol. Med Sci Sports Exerc 34:267-273.

Kjaerulff O, Kiehn O (1996) Distribution of networks generating and coordinating

locomotor activity in the neonatal rat spinal cord in vitro: a lesion study. J Neurosci

16:5777-5794.

Kjaerulff O, Barajon I, Kiehn O (1994) Sulphorhodamine-labelled cells in the neonatal rat

spinal cord following chemically induced locomotor activity in vitro. J Physiol 478

( Pt 2):265-273.

Kong XY, Wienecke J, Hultborn H, Zhang M (2010) Robust upregulation of serotonin 2A

receptors after chronic spinal transection of rats: an immunohistochemical study.

Brain Res 1320:60-68.

Kovacheva EL, Hikim AP, Shen R, Sinha I, Sinha-Hikim I (2010) Testosterone

supplementation reverses sarcopenia in aging through regulation of myostatin, c-Jun

NH2-terminal kinase, Notch, and Akt signaling pathways. Endocrinology 151:628-

638.

260

Krajacic A, Ghosh M, Puentes R, Pearse DD, Fouad K (2009) Advantages of delaying the

onset of rehabilitative reaching training in rats with incomplete spinal cord injury.

Eur J Neurosci 29:641-651.

Kubasak MD, Jindrich DL, Zhong H, Takeoka A, McFarland KC, Munoz-Quiles C, Roy

RR, Edgerton VR, Ramon-Cueto A, Phelps PE (2008) OEG implantation and step

training enhance hindlimb-stepping ability in adult spinal transected rats. Brain

131:264-276.

Kuhn RA (1950) Functional capacity of the isolated human spinal cord. Brain 73:1-51.

Kujawa KA, Kinderman NB, Jones KJ (1989) Testosterone-induced acceleration of

recovery from facial paralysis following crush axotomy of the facial nerve in male

hamsters. Exp Neurol 105:80-85.

Kullander K, Butt SJ, Lebret JM, Lundfald L, Restrepo CE, Rydstrom A, Klein R, Kiehn O

(2003) Role of EphA4 and EphrinB3 in local neuronal circuits that control walking.

Science 299:1889-1892.

Ladouceur M, Barbeau H (2000) Functional electrical stimulation-assisted walking for

persons with incomplete spinal injuries: changes in the kinematics and

physiological cost of overground walking. Scand J Rehabil Med 32:72-79.

Laird AS, Carrive P, Waite PM (2009) Effect of treadmill training on autonomic

dysreflexia in spinal cord—injured rats. Neurorehabil Neural Repair 23:910-920.

Lajoie K, Andujar JE, Pearson K, Drew T (2010) Neurons in area 5 of the posterior parietal

cortex in the cat contribute to interlimb coordination during visually guided

locomotion: a role in working memory. J Neurophysiol 103:2234-2254.

Landry E, Frenette J, Guertin PA (2004) Body weight, limb size, and muscular properties

of early paraplegic mice. J Neurotrauma 21:1008-1016.

Landry ES, Guertin PA (2004) Differential effects of 5-HT(l) and 5-HT(2) receptor

agonists on hindlimb movements in paraplegic mice. Prog Neuropsychopharmacol

Biol Psychiatry 28:1053-1060.

Landry ES, Rouillard C, Levesque D, Guertin PA (2006a) Profile of immediate early gene

expression in the lumbar spinal cord of low-thoracic paraplegic mice. Behav

Neurosci 120:1384-1388.

261

Landry ES, Lapointe NP, Rouillard C, Levesque D, Hedlund PB, Guertin PA (2006b)

Contribution of spinal 5-HT 1A and 5-HT7 receptors to locomotor-like movement

induced by 8-OH-DPAT in spinal cord-transected mice. Eur J Neurosci 24:535-546.

Langlet C, Leblond H, Rossignol S (2005) Mid-lumbar segments are needed for the

expression of locomotion in chronic spinal cats. J Neurophysiol 93:2474-2488.

Lanuza GM, Gosgnach S, Pierani A, Jessell TM, Goulding M (2004) Genetic identification

of spinal interneurons that coordinate left-right locomotor activity necessary for

walking movements. Neuron 42:375-386.

Lapointe NP, Guertin PA (2008) Synergistic effects of Dl/5 and 5-HT1A/7 receptor

agonists on locomotor movement induction in complete spinal cord-transected

mice. J Neurophysiol 100:160-168.

Lapointe NP, Ung RV, Rouleau P, Guertin PA (2008a) Tail pinching-induced hindlimb

movements are suppressed by clonidine in spinal cord injured mice. Behav

Neurosci 122:576-588.

Lapointe NP, Ung RV, Rouleau P, Guertin PA (2008b) Effects of spinal alpha(2)-

adrenoceptor and I(l)-imidazoline receptor activation on hindlimb movement

induction in spinal cord-injured mice. J Pharmacol Exp Ther 325:994-1006.

Lapointe NP, Rouleau P, Ung RV, Guertin PA (2009) Specific role of dopamine DI

receptors in spinal network activation and rhythmic movement induction in

vertebrates. J Physiol 587:1499-1511.

Lapointe NP, Ung RV, Bergeron M, Cote M, Guertin PA (2006) Strain-dependent recovery

of spontaneous hindlimb movement in spinal cord transected mice (CD1, C57BL/6,

BALB/c). Behav Neurosci 120:826-834.

Lauder JM, Wilkie MB, Wu C, Singh S (2000) Expression of 5-HT(2A), 5-HT(2B) and 5-

HT(2C) receptors in the mouse embryo. Int J Dev Neurosci 18:653-662.

Leblond H, L'Espérance M, Orsal D, Rossignol S (2003) Treadmill locomotion in the intact

and spinal mouse. J Neurosci 23:11411-11419.

Lee MS, Choi YC, Lee SH, Lee SB (1996) Sleep-related periodic leg movements

associated with spinal cord lesions. Mov Disord 11:719-722.

Leroux A, Fung J, Barbeau H (1999) Adaptation of the walking pattern to uphill walking in

normal and spinal-cord injured subjects. Exp Brain Res 126:359-368.

262

Li XL, Zhang W, Zhou X, Wang XY, Zhang HT, Qin DX, Zhang H, Li Q, Li M, Wang TH

(2007) Temporal changes in the expression of some neurotrophins in spinal cord

transected adult rats. Neuropeptides 41:135-143.

Lieber RL, Friden JO, Hargens AR, Feringa ER (1986a) Long-term effects of spinal cord

transection on fast and slow rat skeletal muscle. II. Morphometric properties. Exp

Neurol 91:435-448.

Lieber RL, Johansson CB, Vahlsing HL, Hargens AR, Feringa ER (1986b) Long-term

effects of spinal cord transection on fast and slow rat skeletal muscle. I. Contractile

properties. Exp Neurol 91:423-434.

Liu J, Jordan LM (2005) Stimulation of the parapyramidal region of the neonatal rat brain

stem produces locomotor-like activity involving spinal 5-HT7 and 5-HT2A

receptors. J Neurophysiol 94:1392-1404.

Liu J, Akay T, Hedlund PB, Pearson KG, Jordan LM (2009) Spinal 5-HT7 receptors are

critical for alternating activity during locomotion: in vitro neonatal and in vivo adult

studies using 5-HT7 receptor knockout mice. J Neurophysiol 102:337-348.

Llewellyn-Smith IJ, Weaver LC (2001) Changes in synaptic inputs to sympathetic

preganglionic neurons after spinal cord injury. J Comp Neurol 435:226-240.

Lovely RG, Gregor RJ, Roy RR, Edgerton VR (1986) Effects of training on the recovery of

full-weight-bearing stepping in the adult spinal cat. Exp Neurol 92:421-435.

Lundfald L, Restrepo CE, Butt SJ, Peng CY, Droho S, Endo T, Zeilhofer HU, Sharma K,

Kiehn O (2007) Phénotype of V2-derived interneurons and their relationship to the

axon guidance molecule EphA4 in the developing mouse spinal cord. Eur J

Neurosci 26:2989-3002.

Lunenburger L, Bolliger M, Czell D, Muller R, Dietz V (2006) Modulation of locomotor

activity in complete spinal cord injury. Exp Brain Res 174:638-646.

Lynch GS, Ryall JG (2008) Role of beta-adrenoceptor signaling in skeletal muscle:

implications for muscle wasting and disease. Physiol Rev 88:729-767.

Lynch GS, Hinkle RT, Faulkner JA (1999) Year-long clenbuterol treatment of mice

increases mass, but not specific force or normalized power, of skeletal muscles. Clin

Exp Pharmacol Physiol 26:117-120.

263

MacLennan PA, Edwards RH (1989) Effects of clenbuterol and propranolol on muscle

mass. Evidence that clenbuterol stimulates muscle beta-adrenoceptors to induce

hypertrophy. Biochem J 264:573-579.

Madriaga MA, McPhee LC, Chersa T, Christie KJ, Whelan PJ (2004) Modulation of

locomotor activity by multiple 5-HT and dopaminergic receptor subtypes in the

neonatal mouse spinal cord. J Neurophysiol 92:1566-1576.

Maegele M, Muller S, Wemig A, Edgerton VR, Harkema SJ (2002) Recruitment of spinal

motor pools during voluntary movements versus stepping after human spinal cord

injury. J Neurotrauma 19:1217-1229.

Maeshima T, Ito R, Hamada S, Senzaki K, Hamaguchi-Hamada K, Shutoh F, Okado N

(1998) The cellular localization of 5-HT2A receptors in the spinal cord and spinal

ganglia of the adult rat. Brain Res 797:118-124.

Magnuson DS, Smith RR, Brown EH, Enzmann G, Angeli C, Quesada PM, Burke D

(2009) Swimming as a model of task-specific locomotor retraining after spinal cord

injury in the rat. Neurorehabil Neural Repair 23:535-545.

Maier IC, Ichiyama RM, Courtine G, Schnell L, Lavrov I, Edgerton VR, Schwab ME

(2009) Differential effects of anti-Nogo-A antibody treatment and treadmill training

in rats with incomplete spinal cord injury. Brain 132:1426-1440.

Maltin CA, Reeds PJ, Delday MI, Hay SM, Smith FG, Lobley GE (1986) Inhibition and

reversal of denervation-induced atrophy by the beta-agonist growth promoter,

clenbuterol. Biosci Rep 6:811-818.

Marcoux J, Rossignol S (2000) Initiating or blocking locomotion in spinal cats by applying

noradrenergic drugs to restricted lumbar spinal segments. J Neurosci 20:8577-8585.

Marie O, Zorner B, Dietz V (2008) Levodopa therapy in incomplete spinal cord injury. J

Neurotrauma 25:1303-1307.

Martin Ginis KA, Jetha A, Mack DE, Hetz S (2010) Physical activity and subjective well-

being among people with spinal cord injury: a meta-analysis. Spinal Cord 48:65-72.

Maynard FM, Reynolds GG, Fountain S, Wilmot C, Hamilton R (1979) Neurological

prognosis after traumatic quadriplegia. Three-year experience of California

Regional Spinal Cord Injury Care System. J Neurosurg 50:611-616.

264

McCrea DA, Rybak IA (2008) Organization of mammalian locomotor rhythm and pattern

generation. Brain Res Rev 57:134-146.

McEwen ML, Van Hartesveldt C, Stehouwer DJ (1997) L-DOPA and quipazine elicit air-

stepping in neonatal rats with spinal cord transections. Behav Neurosci 111:825-

833.

McVea DA, Donelan JM, Tachibana A, Pearson KG (2005) A role for hip position in

initiating the swing-to-stance transition in walking cats. J Neurophysiol 94:3497-

3508.

Mehrholz J, Kugler J, Pohl M (2008) Locomotor training for walking after spinal cord

injury. Spine (Phila Pa 1976) 33:E768-777.

Mengod G, Nguyen H, Le H, Waeber C, Lubbert H, Palacios JM (1990) The distribution

and cellular localization of the serotonin IC receptor mRNA in the rodent brain

examined by in situ hybridization histochemistry. Comparison with receptor

binding distribution. Neuroscience 35:577-591.

Meuser T, Pietruck C, Gabriel A, Xie GX, Lim KJ, Pierce Palmer P (2002) 5-HT7

receptors are involved in mediating 5-HT-induced activation of rat primary afferent

neurons. Life Sci 71:2279-2289.

Mitsui T, Fischer I, Shumsky JS, Murray M (2005) Transplants of fibroblasts expressing

BDNF and NT-3 promote recovery of bladder and hindlimb function following

spinal contusion injury in rats. Exp Neurol 194:410-431.

Môdder UI, Achenbach SJ, Amin S, Riggs BL, Melton LJ 3rd, Khosla S. (2010) Relation

of serum serotonin levels to bone density and structural parameters in women. J

Bone Miner Res. 25:415-22.

Molineaux SM, Jessell TM, Axel R, Julius D (1989) 5-HTlc receptor is a prominent

serotonin receptor subtype in the central nervous system. Proc Natl Acad Sci U S A

86:6793-6797.

Mori S, Sakamoto T, Ohta Y, Takakusaki K, Matsuyama K (1989) Site-specific postural

and locomotor changes evoked in awake, freely moving intact cats by stimulating

the brainstem. Brain Res 505:66-74.

Morrison SA, Backus D (2007) Locomotor training: is translating evidence into practice

financially feasible? J Neurol Phys Ther 31:50-54.

265

Morton SM, Bastian AJ (2004) Cerebellar control of balance and locomotion.

Neuroscientist 10:247-259.

Multon S, Franzen R, Poirrier AL, Scholtes F, Schoenen J (2003) The effect of treadmill

training on motor recovery after a partial spinal cord compression-injury in the adult

rat. J Neurotrauma 20:699-706.

Nadeau S, Jacquemin G, Fournier C, Lamarre Y, Rossignol S (2010) Spontaneous motor

rhythms of the back and legs in a patient with a complete spinal cord transection.

Neurorehabil Neural Repair 24:377-383.

Nakamura Y, Katakura N, Nakajima M. (1999) Generation of rhythmical ingestive

activities of the trigeminal, facial, and hypoglossal motoneurons in in vitro CNS

preparations isolated from rats and mice. J Med Dent Sci 46(2):63-73.

Nessler JA, Timoszyk W, Merlo M, Emken JL, Minakata K, Roy RR, de Leon RD,

Edgerton VR, Reinkensmeyer DJ (2005) A robotic device for studying rodent

locomotion after spinal cord injury. IEEE Trans Neural Syst Rehabil Eng 13:497-

506.

Nicoll RA, Malenka RC, Kauer JA (1990) Functional comparison of neurotransmitter

receptor subtypes in mammalian central nervous system. Physiol Rev 70:513-565.

Nielsen JB, Crone C, Hultborn H (2007) The spinal pathophysiology of spasticity—from a

basic science point of view. Acta Physiol (Oxf) 189:171-180.

Nishimaru H, Takizawa H, Kudo N (2000) 5-Hydroxytryptamine-induced locomotor

rhythm in the neonatal mouse spinal cord in vitro. Neurosci Lett 280:187-190.

Noga BR, Kettler J, Jordan LM (1988) Locomotion produced in mesencephalic cats by

injections of putative transmitter substances and antagonists into the medial

reticular formation and the pontomedullary locomotor strip. J Neurosci 8:2074-

2086.

Noga BR, Johnson DM, Riesgo MI, Pinzon A (2009) Locomotor-activated neurons of the

cat. I. Serotonergic innervation and co-localization of 5-HT7, 5-HT2A, and 5-HT 1A

receptors in the thoraco-lumbar spinal cord. J Neurophysiol 102:1560-1576.

Norman KE, Pepin A, Ladouceur M, Barbeau H (1995) A treadmill apparatus and harness

support for evaluation and rehabilitation of gait. Arch Phys Med Rehabil 76:772-

778.

266

Norrie BA, Nevett-Duchcherer JM, Gorassini MA (2005) Reduced functional recovery by

delaying motor training after spinal cord injury. J Neurophysiol 94:255-264.

Nothias JM, Mitsui T, Shumsky JS, Fischer I, Antonacci MD, Murray M (2005) Combined

effects of neurotrophin secreting transplants, exercise, and serotonergic drug

challenge improve function in spinal rats. Neurorehabil Neural Repair 19:296-312.

Orlovsky GN (1991) Gravistatic postural control in simpler systems. Curr Opin Neurobiol

1:621-627.

Pajoohesh-Ganji A, Byrnes KR, Fatemi G, Faden AI (2010) A combined scoring method to

assess behavioral recovery after mouse spinal cord injury. Neurosci Res 67:117-

125.

Pearlstein E, Mabrouk FB, Pflieger JF, Vinay L (2005) Serotonin refines the locomotor-

related alternations in the in vitro neonatal rat spinal cord. Eur J Neurosci 21:1338-

1346.

Pearson KG, Acharya H, Fouad K (2005) A new electrode configuration for recording

electromyographic activity in behaving mice. J Neurosci Methods 148:36-42.

Perret C, Cabelguen JM (1980) Main characteristics of the hindlimb locomotor cycle in the

decorticate cat with special reference to bifunctional muscles. Brain Res 187:333-

352.

Perrier JF, Alaburda A, Hounsgaard J (2003) 5-HT 1A receptors increase excitability of

spinal motoneurons by inhibiting a TASK-1-like K+ current in the adult turtle. J

Physiol 548:485-492.

Peterson CA, Murphy RJ, Dupont-Versteegden EE, Houle JD (2000) Cycling exercise and

fetal spinal cord transplantation act synergistically on atrophied muscle following

chronic spinal cord injury in rats. Neurorehabil Neural Repair 14:85-91.

Pompeiano M, Palacios JM, Mengod G (1994) Distribution of the serotonin 5-HT2 receptor

family mRNAs: comparison between 5-HT2A and 5-HT2C receptors. Brain Res

Mol Brain Res 23:163-178.

Quevedo J, Stecina K, Gosgnach S, McCrea DA (2005) Stumbling corrective reaction

during fictive locomotion in the cat. J Neurophysiol 94:2045-2052.

267

Remy-Neris O, Barbeau H, Daniel O, Boiteau F, Bussel B (1999) Effects of intrathecal

clonidine injection on spinal reflexes and human locomotion in incomplete

paraplegic subjects. Exp Brain Res 129:433-440.

Rexed B (1952) The cytoarchitectonic organization of the spinal cord in the cat. J Comp

Neurol 96:414-495.

Richards JB, Papaioannou A, Adachi JD, Joseph L, Whitson HE, Prior JC, Goltzman D

(2007) Effect of selective serotonin reuptake inhibitors on the risk of fracture. Arch

Intern Med 167:188-194.

Risedal A, Zeng J, Johansson BB (1999) Early training may exacerbate brain damage after

focal brain ischemia in the rat. J Cereb Blood Flow Metab 19:997-1003.

Rossignol S (2000) Locomotion and its recovery after spinal injury. Curr Opin Neurobiol

10:708-716.

Rossignol S (2006) Plasticity of connections underlying locomotor recovery after central

and/or peripheral lesions in the adult mammals. Philos Trans R Soc Lond B Biol Sci

361:1647-1671.

Rossignol S, Barrière G, Alluin O, Frigon A (2009) Re-expression of locomotor function

after partial spinal cord injury. Physiology 24:127-139.

Rossignol S, Bouyer L, Barthélémy D, Langlet C, Leblond H (2002) Recovery of

locomotion in the cat following spinal cord lesions. Brain Res Brain Res Rev

40:257-266.

Rossignol S, Dubuc R, Gossard JP (2006) Dynamic sensorimotor interactions in

locomotion. Physiol Rev 86(1):89-154

Roudet C, Savasta M, Feuerstein C (1993) Normal distribution of alpha-1-adrenoceptors in

the rat spinal cord and its modification after noradrenergic denervation: a

quantitative autoradiographic study. J Neurosci Res 34:44-53.

Roudet C, Mouchet P, Feuerstein C, Savasta M (1994) Normal distribution of alpha 2-

adrenoceptors in the rat spinal cord and its modification after noradrenergic

denervation: a quantitative autoradiographic study. J Neurosci Res 39:319-329.

Rouleau P, Guertin PA (2010a) Traumatic and nontraumatic spinal-cord-injured patients in

Quebec, Canada. Part 3: pharmacological characteristics. Spinal Cord.

268

Rouleau P, Guertin PA (2010b) Traumatic and non-traumatic spinal cord-injured patients in

Quebec, Canada. Part 2: biochemical profile. Spinal Cord.

Rouleau P, Ung RV, Lapointe NP, Guertin PA (2007) Hormonal and immunological

changes in mice after spinal cord injury. J Neurotrauma 24:367-378.

Rybak IA, Shevtsova NA, Lafreniere-Roula M, McCrea DA (2006a) Modelling spinal

circuitry involved in locomotor pattern generation: insights from deletions during

fictive locomotion. J Physiol 577:617-639.

Rybak IA, Stecina K, Shevtsova NA, McCrea DA (2006b) Modelling spinal circuitry

involved in locomotor pattern generation: insights from the effects of afferent

stimulation. J Physiol 577:641-658.

Sandrow-Feinberg HR, Izzi J, Shumsky JS, Zhukareva V, Houle JD (2009) Forced exercise

as a rehabilitation strategy after unilateral cervical spinal cord contusion injury. J

Neurotrauma 26:721-731.

Sandrow-Feinberg HR, Zhukareva V, Santi L, Miller K, Shumsky JS, Baker DP, Houle JD

(2010) PEGylated interferon-beta modulates the acute inflammatory response and

recovery when combined with forced exercise following cervical spinal contusion

injury. Exp Neurol 223:439-451.

Schadt JC, Barnes CD (1980) Motoneuron membrane changes associated with spinal shock

and the Schiff-Sherrington phenomenon. Brain Res 201:373-383.

Schmidt BJ, Jordan LM (2000) The role of serotonin in reflex modulation and locomotor

rhythm production in the mammalian spinal cord. Brain Res Bull 53:689-710.

Sherrington CS (1910) Flexion-reflex of the limb, crossed extension-reflex, and reflex

stepping and standing. J Physiol 40:28-121.

Shik ML, Severin FV, Orlovskii GN (1966) [Control of walking and running by means of

electric stimulation of the midbrain]. Biofizika 11:659-666.

Sinha-Hikim I, Artaza J, Woodhouse L, Gonzalez-Cadavid N, Singh AB, Lee MI, Storer

TW, Casaburi R, Shen R, Bhasin S (2002) Testosterone-induced increase in muscle

size in healthy young men is associated with muscle fiber hypertrophy. Am J

Physiol Endocrinol Metab 283:E154-164.

269

Smith RR, Burke DA, Baldini AD, Shum-Siu A, Baltzley R, Bunger M, Magnuson DS

(2006) The Louisville Swim Scale: a novel assessment of hindlimb function

following spinal cord injury in adult rats. J Neurotrauma 23:1654-1670.

Sribnick EA, Matzelle DD, Ray SK, Banik NL. (2006) Estrogen treatment of spinal cord

injury attenuates calpain activation and apoptosis. J Neurosci Res. 84:1064-75.

Sribnick EA, Wingrave JM, Matzelle DD, Wilford GG, Ray SK, Banik NL. (2005)

Estrogen attenuated markers of inflammation and decreased lesion volume in acute

spinal cord injury in rats. J Neurosci Res. 82:283-93.

Stelzner DJ, Ershler WB, Weber ED (1975) Effects of spinal transection in neonatal and

weanling rats: survival of function. Exp Neurol 46:156-177.

Stewart BG, Tarnopolsky MA, Hicks AL, McCartney N, Mahoney DJ, Staron RS, Phillips

SM (2004) Treadmill training-induced adaptations in muscle phénotype in persons

with incomplete spinal cord injury. Muscle Nerve 30:61-68.

Stewart JE, Barbeau H, Gauthier S (1991) Modulation of locomotor patterns and spasticity

with clonidine in spinal cord injured patients. Can J Neurol Sci 18:321-332.

Strauss I, Lev-Tov A (2003) Neural pathways between sacrocaudal afférents and lumbar

pattern generators in neonatal rats. J Neurophysiol 89:773-784.

Talmadge RJ, Roy RR, Edgerton VR (1999) Persistence of hybrid fibers in rat soleus after

spinal cord transection. Anat Rec 255:188-201.

Tanzer L, Jones KJ (1997) Gonadal steroid regulation of hamster facial nerve regeneration:

effects of dihydrotestostérone and estradiol. Exp Neurol. 1997 146(l):258-64.

Teng YD, Bingaman M, Taveira-DaSilva AM, Pace PP, Gillis RA, Wrathall JR (2003)

Serotonin IA receptor agonists reverse respiratory abnormalities in spinal cord-

injured rats. J Neurosci 23:4182-4189.

Thor KB, Blitz-Siebert A, Helke CJ (1992) Autoradiographic localization of 5HT1 binding

sites in the medulla oblongata of the rat. Synapse 10:185-205.

Tillakaratne NJ, de Leon RD, Hoang TX, Roy RR, Edgerton VR, Tobin AJ (2002) Use-

dependent modulation of inhibitory capacity in the feline lumbar spinal cord. J

Neurosci 22:3130-3143.

270

Timoszyk WK, De Leon RD, London N, Roy RR, Edgerton VR, Reinkensmeyer DJ (2002)

The rat lumbosacral spinal cord adapts to robotic loading applied during stance. J

Neurophysiol 88:3108-3117.

Tom VJ, Sandrow-Feinberg HR, Miller K, Santi L, Connors T, Lemay MA, Houle JD

(2009) Combining peripheral nerve grafts and chondroitinase promotes functional

axonal regeneration in the chronically injured spinal cord. J Neurosci 29:14881-

14890.

Tresch MC, Kiehn O (1999) Coding of locomotor phase in populations of neurons in rostral

and caudal segments of the neonatal rat lumbar spinal cord. J Neurophysiol

82:3563-3574.

Tuladhar BR, Womack MD, Naylor RJ (2000) Pharmacological characterization of the 5-

HT receptor-mediated contraction in the mouse isolated ileum. Br J Pharmacol

131:1716-1722.

Ung RV, Lapointe NP, Guertin PA (2008a) Early adaptive changes in chronic paraplegic

mice: a model to study rapid health degradation after spinal cord injury. Spinal Cord

46:176-180.

Ung RV, Rouleau P, Guertin PA (2010a) Effects of co-administration of clenbuterol and

testosterone propionate on skeletal muscle in paraplegic mice. J Neurotrauma

27:1129-1142.

Ung RV, Lapointe NP, Rouleau P, Guertin PA (2010b) Non-assisted treadmill training does

not improve motor recovery and body composition in spinal cord-transected mice.

Spinal Cord.

Ung R\^,Lapointe NP, Tremblay C, Larouche A, Guertin PA (2007) Spontaneous recovery

of hindlimb movement in completely spinal cord transected mice: a comparison of

assessment methods and conditions. Spinal Cord 45:367-379.

Ung RV, Landry ES, Rouleau P, Lapointe NP, Rouillard C, Guertin PA (2008b) Role of

spinal 5-HT2 receptor subtypes in quipazine-induced hindlimb movements after a

low-thoracic spinal cord transection. Eur J Neurosci 28:2231-2242.

Van de Crommert HW, Mulder T, Duysens J (1998) Neural control of locomotion: sensory

control of the central pattern generator and its relation to treadmill training. Gait

Posture 7:251-263.

271

Vavrek R, Pearse DD, Fouad K (2007) Neuronal populations capable of regeneration

following a combined treatment in rats with spinal cord transection. J Neurotrauma

24:1667-1673.

Viala D, Buser P (1969) The effects of DOPA and 5-HTP on rhythmic efferent discharges

in hind limb nerves in the rabbit. Brain Res 12:437-443.

Vita G, Dattola R, Girlanda P, Oteri G, Lo Presti F, Messina C (1983) Effects of steroid

hormones on muscle reinnervation after nerve crush in rabbit. Exp Neurol. 80:279-

87.

Wainberg M, Barbeau H, Gauthier S (1990) The effects of cyproheptadine on locomotion

and on spasticity in patients with spinal cord injuries. J Neurol Neurosurg

Psychiatry 53:754-763.

Waters RL, Adkins R, Yakura J, Vigil D (1994) Prediction of ambulatory performance

based on motor scores derived from standards of the American Spinal Injury

Association. Arch Phys Med Rehabil 75:756-760.

Weaver RA, Landau WM, Higgins JF (1963) Fusimotor Function. Ii. Evidence of

Fusimotor Depression in Human Spinal Shock. Arch Neurol 9:127-132.

Wernig A, Muller S (1992) Laufband locomotion with body weight support improved

walking in persons with severe spinal cord injuries. Paraplegia 30:229-238.

Wernig A, Muller S, Nanassy A, Cagol E (1995) Laufband therapy based on 'rules of spinal

locomotion' is effective in spinal cord injured persons. Eur J Neurosci 7:823-829.

Westbroek I, van der Plas A, de Rooij KE, Klein-Nulend J, Nijweide PJ. (2001) Expression

of serotonin receptors in bone. J Biol Chem. 276:28961-8

Whelan P, Bonnot A, O'Donovan MJ (2000) Properties of rhythmic activity generated by

the isolated spinal cord of the neonatal mouse. J Neurophysiol 84:2821-2833.

Wilson DM (1961) The central nervous control of flight in a locust. J. Exp. Biol. 38,471-

490.

Wilson JM, Hartley R, Maxwell DJ, Todd AJ, Lieberam I, Kaltschmidt JA, Yoshida Y,

Jessell TM, Brownstone RM (2005) Conditional rhythmicity of ventral spinal

interneurons defined by expression of the Hb9 homeodomain protein. J Neurosci

25:5710-5719.

272

Wirz M, Colombo G, Dietz V (2001) Long term effects of locomotor training in spinal

humans. J Neurol Neurosurg Psychiatry 71:93-96.

Yang JF, Stein RB, James KB (1991) Contribution of peripheral afférents to the activation

of the soleus muscle during walking in humans. Exp Brain Res 87:679-687.

Yoshihara H, Shumsky JS, Neuhuber B, Otsuka T, Fischer I, Murray M (2006) Combining

motor training with transplantation of rat bone marrow stromal cells does not

improve repair or recovery in rats with thoracic contusion injuries. Brain Res

1119:65-75.

Zehr EP, Komiyama T, Stein RB (1997) Cutaneous reflexes during human gait:

electromyographic and kinematic responses to electrical stimulation. J Neurophysiol

77:3311-3325.

Zeman RJ, Ludemann R, Etlinger JD (1987) Clenbuterol, a beta 2-agonist, retards atrophy

in denervated muscles. Am J Physiol 252:E152-155.

Zeman RJ, Zhang Y, Etlinger JD (1997) Clenbuterol, a beta2-adrenoceptor agonist, reduces

scoliosis due to partial transection of rat spinal cord. Am J Physiol 272:E712-715.

Zeman RJ, Feng Y, Peng H, Etlinger JD (1999) Clenbuterol, a beta(2)-adrenoceptor

agonist, improves locomotor and histological outcomes after spinal cord contusion

in rats. Exp Neurol 159:267-273.

Zeman RJ, Hirschman A, Hirschman ML, Guo G, Etlinger JD (1991) Clenbuterol, a beta 2-

receptor agonist, reduces net bone loss in denervated hindlimbs. Am J Physiol

261:E285-289.

Zhang P, Chen X, Fan M (2007) Signaling mechanisms involved in disuse muscle atrophy.

Med Hypotheses 69:310-321.

Zhang Y, Narayan S, Geiman E, Lanuza GM, Velasquez T, Shanks B, Akay T, Dyck J,

Pearson K, Gosgnach S, Fan CM, Goulding M (2008) V3 spinal neurons establish a

robust and balanced locomotor rhythm during walking. Neuron 60:84-96.

Zhao J, Zhang Y, Zhao W, Wu Y, Pan J, Bauman WA, Cardozo C (2008a) Effects of

nandrolone on denervation atrophy depend upon time after nerve transection.

Muscle Nerve 37:42-49.

273

Zhao W, Pan J, Zhao Z, Wu Y, Bauman WA, Cardozo CP (2008b) Testosterone protects

against dexamethasone-induced muscle atrophy, protein degradation and MAFbx

upregulation. J Steroid Biochem Mol Biol 110:125-129.

Zhu H, Clemens S, Sawchuk M, Hochman S (2007) Expression and distribution of all

dopamine receptor subtypes (D(l)-D(5)) in the mouse lumbar spinal cord: a real­

time polymerase chain reaction and non-autoradiographic in situ hybridization

study. Neuroscience 149:885-897.

Zitzmann M, Nieschlag E (2000) Hormone substitution in male hypogonadism. Mol Cell

Endocrinol 161:73-88.