Title: Inducing hindlimb locomotor recovery in adult rat after a complete thoracic spinal 1

cord section using repeated treadmill training with perineal stimulation only. 2

Abbreviated title: Treadmill training in adult complete spinal rat. 3

Authors: Olivier Alluin1,*,2, Hugo Delivet-Mongrain1,* and Serge Rossignol1,* 4

Authors’ affiliation: 5

1 Department of Neuroscience and Groupe de Recherche sur le Système Nerveux Central 6

(GRSNC), Faculty of Medicine, Université de Montréal, P.O. Box 6128, Montreal, 7

Quebec, Canada. 8

*SensoriMotor Rehabilitation Research Team of the Canadian Institute of Health 9

Research, Montreal, Quebec, Canada. 10

2Current affiliation: Institut des Sciences du Mouvement (UMR 7287), Aix-Marseille 11

Université, CNRS, Marseille, France. 12

Corresponding author: 13

Serge Rossignol: Université de Montréal - Département de Neurosciences - C.P. 6128, 14

Succ. Centre-Ville - Montréal (Québec), H3C 3J7 - Canada 15

E-mail: serge.rossignol@umontreal.ca - Phone: (514) 343-6371 - Fax : (514) 343-7972 16

Number of pages: 50 17

Number of figures: 11 18

Number of multimedia: 1 19

Number of words for abstract: 246 20

Number of words for Introduction: 505 21

Number of words for discussion: 1734 22

Articles in PresS. J Neurophysiol (July 22, 2015). doi:10.1152/jn.00416.2015

Copyright © 2015 by the American Physiological Society.

Acknowledgements: We thank Philippe Drapeau and Claude Gagner for their technical 23

assistance in software design and electronics. This work was supported by the operating 24

grants from the Canadian Institutes of Health Research (CIHR) to SR (Sensorimotor 25

Rehabilitation Research Team - SMRRT – team grant and Canada Research Chair). OA 26

was supported by a postdoctoral fellowship from the Fonds de Recherche du Québec - 27

Santé (FRQS) and by SMRRT. The authors declare no competing financial interests. 28

1

Abstract 29

Although a complete thoracic spinal cord section in various mammals induces 30

paralysis of voluntary movements, the spinal lumbosacral circuitry below the lesion 31

retains its ability to generate hindlimb locomotion. This important capacity may 32

contribute to the overall locomotor recovery after partial spinal cord injury (SCI). In rats 33

it is usually triggered by pharmacological and/or electrical stimulation of the cord while a 34

robot sustains the animals in an upright posture. In the current study we daily trained a 35

group of adult spinal rats (T7) to walk with the hindlimbs for 10 weeks (10 min/day for 5 36

days/week) using only perineal stimulation. Kinematic analysis and terminal 37

electromyographic recordings revealed a strong effect of training on the re-expression of 38

hindlimb locomotion. Indeed, trained animals gradually improved their locomotion while 39

untrained ones worsened throughout the post-SCI period. Kinematic parameters such as 40

averaged and instant swing phase velocity, step cycle variability, foot drag duration, off 41

period duration or the relationship between the swing features returned to normal values 42

only in trained animals. The present results clearly demonstrate that treadmill training 43

alone, in a normal horizontal posture, elicited by non-invasive perineal stimulation is 44

sufficient to induce a persistent hindlimb locomotor recovery without the need for more 45

complex strategies. This provides a baseline level that should be clearly surpassed if 46

additional locomotor enabling procedures are added. Moreover, it has a clinical value 47

since intrinsic spinal reorganization induced by training should contribute to improve 48

afferent feedback to higher centres in patients with incomplete SCI. 49

50

2

Introduction 51

In mammals, complete spinal cord injury (SCI) induces permanent deficits of 52

voluntary movements below the lesion. When such SCI is located at the thoracic level, 53

the lumbosacral neuronal network dedicated to hindlimbs locomotion, known as central 54

pattern generator (CPG), becomes isolated from supraspinal structures leading to 55

permanent paralysis of the lower body parts (Rossignol et al., 2009). Despite this motor 56

impairment, the CPG retains its intrinsic property of generating alternate rhythmic 57

activity as shown by electrophysiological and kinematic recordings in neonatal rats and 58

adult cats (for review, see Rossignol and Frigon, 2011; Rossignol et al., 2014). 59

The challenge of functionally triggering the spinal circuitry to re-express 60

hindlimbs locomotion after complete SCI in adult animals has been extensively 61

confronted using sensorimotor rehabilitation (Rossignol, 1996; Timoszyk et al., 2002; de 62

Leon et al., 2002; Tillakaratne et al., 2010; Ichiyama et al., 2011; Martinez et al., 2012), 63

administration of serotonin agonists as Quipazine or 8-OH-DPAT (Feraboli-Lohnherr et 64

al., 1999; Orsal et al., 2002; Antri et al., 2005; Fong et al., 2005), graft of embryonic 65

raphe cells or fetal brainstem 5-HT neurons (Gimenez y Ribotta et al., 2000; Slawinska et 66

al., 2013) or combined strategies including pharmacology and spinal electrostimulation 67

(Rossignol et al., 2001; Ichiyama et al., 2008; Courtine et al., 2009; Musienko et al., 68

2011). Two to three weeks after a complete thoracic spinalization, cats trained on a 69

treadmill with the help of perineal stimulation can recover the locomotor capability of the 70

hindlimbs and can readily adapt to the belt velocity (Rossignol et al., 2006). Some 71

previous studies tried to replicate these results in complete spinal rats with no success 72

(Orsal et al., 2002; Courtine et al., 2009; Slawinska et al., 2012; Slawinska et al., 2013). 73

3

All these studies show that serotonergic strategies, spinal electrostimulation or postural 74

changes are needed to revive the lumbosacral locomotor network in complete spinal rats. 75

On the other hand, we have previously shown using a clip compression model, 76

resulting in very large compressive lesions that damage the greatest part of the thoracic 77

spinal cord making them almost completely spinal, rats can develop an impressive 78

capability to walk adequately with the hindlimbs in quadrupedal posture without any 79

additional treatment (Alluin et al., 2011). These results suggest that the re-emergence of 80

hindlimb locomotion probably results mainly from the re-expression of the lumbosacral 81

spinal CPG (Rossignol et al., 2009). Although a positive effect of treadmill training after 82

SCI in humans and rodents have been demonstrated in combination with other strategies 83

(Courtine et al., 2009; Harkema et al., 2011; Van den Brand et al., 2012), the intrinsic 84

potential of the spinal circuitry causing the recovery of locomotion in adult complete 85

spinal rats remains poorly known. 86

In the present study, we induced daily treadmill training in natural horizontal 87

posture using sustained perineal stimulation to trigger the spinal locomotor circuitry and 88

re-express locomotion in complete spinalized adult rats without any additional treatment. 89

This work was deemed important since it provides a behavioral baseline that should be 90

surpassed to claim the need for additional enabling procedures to restore spinal 91

locomotion. 92

93

4

Materials and methods 94

Animal care 95

Twenty one adult female Wistar rats (250-275 g) from Charles River Laboratory 96

(Quebec, Canada) were involved in the present study. Animals were housed in standard 97

plastic cages at 22°C before SCI and 26°C after SCI in a 12:12h light/dark photoperiod. 98

Food (Agribrands Purina®, Ontario, Canada) and drinking water were available ad 99

libitum. Hardwood sawdust bedding (PWI brand, Quebec, Canada) was used before SCI 100

and was then replaced by soft paper bedding (Diamond Soft Bedding #7089, Harlan™ 101

Teklad) after SCI to prevent skin lesions. Animals were examined twice a day and 102

evaluated by a veterinarian when necessary. After SCI, the bladder was expressed twice a 103

day until the recovery of spontaneous bladder function, generally occurring between 7 104

and 14 days post-injury. All animal procedures were approved by the University of 105

Montreal Research Ethics Committee and were conducted according to the Canadian 106

Guide to the Care and Use of Experimental Animals (Canadian Council on Animal Care). 107

Experimental framework 108

After a one-week period of quarantine in the animal facility, rats were accustomed 109

to walk consistently with a quadrupedal gait on the treadmill at different speeds for 30 110

minutes daily, 5 days a week and during 3 weeks (see Alluin et al., 2011 for details). All 111

rats were then video recorded for an additional three week period to obtain baseline 112

kinematic values. A complete transection of the thoracic spinal cord was next performed 113

and rats were assigned to 2 groups: 1) rats daily trained to walk on treadmill during the 114

whole recovery period (Trained group; n = 13) and 2) no training (Untrained group; n = 115

8). Treadmill locomotion elicited by perineal stimulation in trained and untrained rats 116

5

was video recorded once a week during the following ten-week period for kinematic 117

analysis. It is thus important to remember that all rats were documented with videos 118

exactly in the same condition (perineal pinching) so that the difference between the 119

behavior of trained and untrained rats must be due to a training effect. At the end of the 120

experiment, electromyography (EMG) of selected hindlimbs muscles was recorded. 121

Finally, animals were then perfused with a fixative and the spinal cord was harvested for 122

histological evaluation of the lesions. Tibialis anterior and gastrocnemius muscles were 123

removed and weighted. 124

It is important to note that double-blind assessments are not possible in such 125

chronic studies because, with time and daily handling, the experimenters learn to 126

recognize the animals according to their individual specificities (shaving for kinematic 127

recordings, hair color tone, personality, weight, size, body marks…etc.). 128

Surgical procedures 129

All surgeries were performed in aseptic conditions and under general gas 130

anaesthesia consisting of O2/isoflurane (1-4 %) mixture given through a mask integrated 131

in a surgical stereotaxic frame associated with a heating pad set to 37 °C. Immediately 132

after surgery, rats were placed under a heating lamp until they recovered consciousness. 133

Animals were given systematic postoperative analgesia (40 µg/kg Temgesic®, Schering-134

Plough, Hertfordshire, UK) and saline (5 ml) subcutaneously to prevent pain and 135

dehydration. They also received antibiotic in drinking water (Clavamox® drops, Pfizer 136

Animal Health) from 3 days before to 1 week after SCI. 137

Prior to the SCI, the skin of the middle back was shaved and disinfected with a 138

1:1 mixture of 70 % alcohol and povidone-iodine (Betadine®, Purdue Pharma L.P., 139

6

Stamford, USA). A skin incision was made above the spinal processes between T4-T10 140

(≈ 3 cm). The superficial muscles from both sides were cut along the spine and retracted 141

from each side of the surgical area. A laminectomy was then performed at T7 and the 142

spinal cord was completely transected using micro scissors (Vannas Spring Scissors; 2.5 143

mm Blades; Fine Science Tools Inc.). After ensuring that the rostral and caudal parts of 144

the spinal cord were separated, a piece of sterile absorbable gelatin sponge (Gelfoam®, 145

Pfizer Inc.) was placed between the 2 spinal cord stumps and on top of the lesion between 146

T6-T8. Finally, muscles were sutured (Prolene 3-0; Ethicon) and skin was closed using 147

stainless steel wound clips (9 mm AutoClips; MikRon Precision, Inc.; USA) which were 148

removed 1 week later. 149

Treadmill training protocol after SCI 150

Spinal rats were trained to walk with the hindlimbs on the motorized treadmill 151

belt from day 1 to week 10 following SCI, while the forelimbs were actively maintained 152

on a platform 5 mm above the belt to maintain a horizontal posture. This training 153

procedure was performed 5 days a week and consisted in one 10-min walking session 154

daily on the treadmill at increasing speeds ranging from 14 to 26 m.min–1 depending on 155

the locomotor capability of each rat. Perineal stimulation, consisting in manually 156

pinching the perineum area just below the base of the tail, was used to facilitate the 157

hindlimb locomotion throughout the training sessions. During training, the belt speed was 158

initially set at 14 m.min–1 and the trunk of the animals was manually supported to reduce 159

the lateral body movements. Early after SCI, all rats were incapable of producing 160

hindlimb stepping even with perineal stimulation. This evoked, in most cases, only erratic 161

bilateral flexion/extension without paw placement on the treadmill belt. As soon as rats 162

7

were capable of producing a more regular hindlimb locomotor pattern, the belt speed was 163

incremented in steps of 2 m.min–1 every 1.5 minutes until the maximum speed that could 164

be followed by the rats was reached (i.e. in the range defined above). Then, depending of 165

the fatigue of the animals, the treadmill speed was progressively decreased up to the end 166

of the training session if necessary. 167

Kinematic recordings 168

Kinematic baseline data were recorded at 14, 20 and 26 m.min–1 for each rat to 169

obtain control values before the spinal lesion. The locomotor performance on treadmill 170

was then recorded weekly for 10 weeks following SCI. Kinematic recording protocol was 171

described previously in detail (Alluin et al., 2011). Briefly, before each recording session, 172

the left hindquarter of the rats was shaved and 3 black dots were set on the skin at the 173

bony landmarks of the ilium, great trochanter and lateral malleolus with a felt pen while 2 174

light-reflecting markers were glued on the skin of metatarso-phalanx [MTP] and the tip 175

of the third toe. A left side view of the rat walking on the treadmill with the hindlimbs 176

was captured using a high frequency video camera (120 Hz). The kinematic data were 177

generated from the [x, y] coordinates of each marker and from the paw contact/lift events 178

while the knee joint position was extrapolated using triangulation. 179

Perineal stimulation 180

The manual stimulation consisted in a tonic pinching of the skin on the perineum 181

area (right under the base of the tail) between thumb and index while performing rubbing 182

movements at a frequency between 1 and 2 Hz (for illustration, see finger movements of 183

the experimenter in Movie 1). Such stimulation was used in all rats from both groups for 184

every locomotor sequence recorded for kinematic evaluation throughout the post-SCI 185

8

period since, whether trained or untrained, rats performed only occasional hindlimb 186

movements on treadmill without stimulation. Although, the stimulation intensity was not 187

quantified in the present study, the strength applied on the skin by the experimenters was 188

just enough to trigger locomotion depending on the rat responsiveness. Specifically, the 189

experimenters started the recording session by applying a minimal strength on the 190

perineal skin and progressively increased the strength of pinching until it was sufficient 191

to induce locomotion in the rat. Once hindlimb locomotor movements started, the 192

experimenter kept the same intensity until the end of recording. For instance, at the end 193

of the experimental series, some rats in the trained group started to walk on the treadmill 194

with minimal intensity while some untrained rats were unresponsive even at a much 195

higher pinch strength which was kept, of course, at a level below which skin bruises 196

could be produced, a counterproductive situation in daily training rats. 197

Individual variability assessment 198

The method of individual variability assessment is detailed in Alluin et al., 2011. 199

Briefly, the coefficient of variation (CV) of every Cartesian kinematic parameter and the 200

circular dispersion of the circular data were used to assess the intrinsic variability of each 201

rat during locomotion. Since the analysis of locomotion of a given rat was based on 202

consecutive sequences of several step cycles, the kinematic data from these several steps 203

were averages and their respective standard deviations (SD) were calculated. Thereby, in 204

the present study, the CV of a given parameter is calculated from the individual SD 205

expressed in percent of the individual mean of this parameter and then averaged by 206

group. The group SEM associated to the CV group mean was then calculated. 207

208

9

Homologous hindlimb coupling 209

The averaged homologous hindlimb coupling is calculated (Alluin et al., 2011) 210

and illustrated on circular charts on which the temporal relationship of the left lift, left 211

contact and right lift are expressed relative to the right contact (i.e. the right step cycle as 212

reference) on the periphery of circles. The circular distance between contact and lift from 213

the same side represent the proportion of the stance phase of this side in the reference 214

step cycle, while the distance between lift and contact represent the proportion of the 215

swing phases. For instance, in intact rats walking on treadmill, the sequence of events 216

(i.e. foot contact and lift) from one hindlimb in a given step cycle is out of phase (i.e. 217

opposite position on the circle) with the sequence from the contralateral hindlimb. 218

Consequently, 4 symmetric locomotor periods lasting about one quarter of the whole step 219

cycle duration each can be highlighted, 1) a first bilateral stance period from the foot 220

contact of the reference side to the foot lift of the contralateral side, 2) the swing phase of 221

the contralateral side from the lift to the contact of this side while the other side keep the 222

stance phase going, 3) the second bilateral stance period from the contact of the 223

contralateral side to the lift of the reference side and 4) the swing phase of the reference 224

side from the lift to the contact of this side while the other side keep the stance phase 225

going (Figure 9A1). 226

After SCI the inter-limb coupling can change and the new temporal relationship 227

between limbs is reflected in the new proportions and positions of the stance and swing 228

phases around the circles. 229

230

231

10

Electromyographic recording 232

Prior to the sacrifice, the ends of isolated monofilament electrodes (Magnet wire; 233

0.22 mm; Cooner Wire Company, CA, USA) were stripped and implanted (2 per muscle) 234

in gastrocnemius lateralis (GL) and tibialis anterior (TA) muscles through the skin of 235

both hindlimbs to record EMG signals during treadmill locomotion in 7 trained and 6 236

untrained rats. Signals from bipolar electrodes were differentially amplified, filtered (100 237

– 3 KHz bandpass) and recorded on a computer through custom-made software for 238

further analysis. 239

Spinal cord processing and analysis 240

At the end of experiments, rats were deeply anesthetized with sodium 241

pentobarbital (80 mg/kg, intraperitoneally) and then were perfused transcardially with 60 242

ml of cold phosphate buffered saline (0.1 M PBS at 4 °C) followed by 180 ml of 4% 243

paraformaldehyde solution (PFA in 0.1 M PBS, pH 7.4). After perfusion, a 2 cm length 244

of the spinal cord centered at the epicentre of the injury was dissected and post-fixed in 245

the perfusing solution with 10% sucrose overnight at 4°C. The cords were then 246

cryoprotected in PBS with 20% sucrose for 48 h at 4°C. The cords were embedded in 247

mounting media (HistoPrep, Fisher Scientific) on dry ice. Cryostat sections (25 µm) were 248

cut transversely and stained with the Cresyl violet to confirm the completeness of the 249

spinal section. 250

Muscle mass assessment 251

Following perfusion, the left and right tibialis anterior and gastrocnemius were 252

also dissected, weighted and then normalized to the total body weight of the animal in 253

order to correct for the inter-individual variability. 254

11

Statistical analysis 255

Individual kinematic data were averaged from the number of locomotor cycles 256

that the rats were capable to execute consecutively in a range from 6 to 20. Two-way 257

repeated measure ANOVA followed by pairwise multiple comparison (Holm-Sidak post 258

hoc test) when ANOVA was significant were used to compare the standard Cartesian 259

data between the different periods and between groups (SigmaPlot, Systat software Inc., 260

San Jose, USA). To compare circular data (i.e. coordination measurement) the 261

multisample Watson-Williams F-test was used (Oriana software, Kovach Computing 262

Services, Wales, UK). Averaged Cartesian data are reported as arithmetic mean ± SEM 263

and the circular data as circular mean ± circular dispersion (equivalent of standard 264

deviation in the circular model). The significance threshold was set to p ≤ 0.05. *, ** and 265

*** represent p ≤ 0.05, 0.01 and 0.001 respectively compared to the group average 266

baseline while +, ++ and +++ represent p ≤ 0.05, 0.01 and 0.001 respectively for 267

comparison between experimental groups at the same time point after SCI. 268

269

12

Results 270

Overview of the locomotor recovery 271

In the 1st week following the complete SCI, although strong perineal stimulation 272

elicited few erratic flexion and extension cycles of small amplitude, flaccid paralysis of 273

the hindlimbs was observed in all animals in their cage. Perineal stimulation was always 274

needed, throughout the 10-week post-SCI period, to elicit locomotor movements during 275

training and recording sessions. Without such stimulation the rats performed only 276

sporadic hindlimb flexion movements. As soon as week 2 after SCI, about 70 % of rats in 277

both groups were capable of performing alternating locomotor movements on the 278

treadmill at 14 m.min-1 (Figure 1A). At the same period, about 45 % and 35 % of trained 279

rats could follow the treadmill belt with plantar placement of the paw at 20 and 26 m.min-280

1 respectively while none of the untrained ones were capable of walking on the belt at 281

these velocities, not even with dorsal foot placement (Figure 1B and C). With time, the 282

proportion of trained animals capable to walk on treadmill at the 3 velocities 283

progressively increased to reach, at week 10 after SCI, 100 % at 14 m.min-1 (Figure 1A), 284

80 % at 20 m.min-1 and 26 m.min-1 (Figure 1B and C). On the other hand, the number of 285

untrained rats capable of following the treadmill belt decreased drastically with time to 286

reach about 10 % at week 8 after SCI at 14 m.min-1 and even 0 % at week 10 at all speeds 287

(Figure 1A, B and C). In fact, training exerted a strong positive effect on locomotion 288

from week 3 and stabilized the locomotor performance as of week 4 while the locomotion 289

of untrained rats was deteriorated from week 3 until the end of experiment. From week 7 290

after SCI, the great majority of untrained animals performed only bilateral alternated 291

flexion and extension with no placement of the paw on the belt (Movie 1). 292

13

Although the intensity of perineal stimulation was not measured in the present 293

study, it was very clear to the experimenters that the strength of pinching needed to elicit 294

locomotion was drastically reduced with time in trained rats, from maximal intensity 295

immediately after SCI to minimal in some trained rats at the end of experiment. However, 296

the trend was reversed in untrained animals. The strength of pinching increased with 297

time, so that in several animals of this group only, the maximum strength applied did not 298

elicit any organized locomotor movement at the end of the experimental series (see 299

Methods). In addition, in trained rats no limitation of the consecutive number of step 300

cycles was noted during training session from week 3 to the end of experiment (i.e. they 301

walked throughout the training session). In contrast, the average number of consecutive 302

steps that untrained rats were capable to produce at the end of experiment during 303

recording sessions was clearly decreased (14.1 ± 2.2 step cycles at week 10) compared to 304

trained animals (p < 0.05) even with strong perineal stimulation (see the progressive 305

exhaustion of locomotor movements despite the perineal stimulation in typical untrained 306

rat at 14 m.min-1 week 7 in Movie 1). Finally, no improvement of spontaneous treadmill 307

locomotion (i.e. without perineal stimulation) was observed in animals from either group 308

throughout the experiment. 309

Treadmill training improves the spatial characteristics of the step cycle 310

Spatial characteristics of locomotion such as the step cycle length, the foot contact 311

and lift position relative to the hip and the maximum foot height during the swing phase 312

were assessed and are illustrated in figure 2. 313

After SCI the step cycle length decreased drastically throughout the recovery 314

period in untrained rats while treadmill training clearly restricted this decrease (Figure 315

14

2A). In addition, the recovery of the step cycle length remained greatly improved in 316

trained compared to untrained rats until the end of experiment (Figure 2A). The step to 317

step variability of the step length (represented by the coefficient of variation or CV) is an 318

indicator of the rats’ capability to maintain a constant step cycle length throughout the 319

locomotor session. In normal rat, the CV of the step cycle length is about 7 % and can be 320

attributed to the intrinsic biological and experimental noises. After complete SCI, trained 321

rats could keep the same degree of step length variability than on the normal state all 322

along the recovery period while the CV of untrained rats increased drastically early after 323

SCI (Figure 2B week 3), then, after some fluctuations, remained high until the end of 324

experiment compared to trained animals (Figure 2B week 10). 325

The foot contact and lift position relative to the hip during walking was assessed 326

weekly in all rats throughout the experiment. The position of the foot relative to the rest 327

of the body during the stance phase give information about the efficiency of the forward 328

propulsion. As indicated before, in spinalized rats the hindlimbs’ locomotor movements 329

shifted backward of the body as demonstrated by the displacement of the relative contact 330

and lift position compared to the baseline in both groups throughout the recovery period 331

(Figure 2C). In trained animals this shift remained constant until the end of experiment 332

and its impact on locomotion was limited by the preservation of the foot contact in front 333

of the hip vertical projection (Figure 2C). In untrained rats however, the position of the 334

foot contact was highly degraded from about – 15 mm compared to trained rats week 2 335

after SCI to about – 50 mm week 10 (Figure 2C). Although the foot lift position was 336

similar between trained and untrained animals week 2 after SCI, it worsened in untrained 337

rats from week 3 to be significantly shifted backward at week 5 and finally reach – 15 338

15

mm compared to trained animals week 10 after SCI (Figure 2C). Consequently, the 339

position of the ground contact period of untrained rats was shifted backward, shorter and 340

the degradation increased with time after SCI compared to the trained ones. 341

We also measured the vertical amplitude of the foot trajectory during locomotion 342

(Figure 2D). This parameter decreased in both groups after SCI. Although the decrease 343

was much greater and stable compared to the baseline in untrained animals throughout 344

the post-injury period, it remained present in trained ones with a trend for improvement 345

by the end of the recovery period (Figure 2D). Finally, the recovery of the vertical 346

amplitude performance was clearly greater in trained animals all along the experiment 347

(average of 58.5 % of the baseline value) compared to the untrained ones (average of 348

22.5 % of the baseline value; Figure 2D). 349

Positive effect of treadmill training on the recovery of angular excursion and 350

amplitude of hindlimb joints. 351

In order to assess the dynamic angular properties of rats’ locomotion we measured 352

the angular excursion, minimum angle, maximum angle and amplitude of hip, knee, ankle 353

and MTP throughout the post-SCI period as depicted in figure 3 and 4. 354

At the end of the experiment (week 10) the global posture of the hindquarters 355

during locomotion had shifted in all animals. The locomotor movements shifted 356

backward relative to the hip due to pelvic anteversion (i.e. posterior rotation; compare the 357

first stick at the top of the stick illustrations representing the orientation of the hip in 358

Figure 3A). These changes in the biomechanical characteristics of the locomotor 359

apparatus changed the hindlimb angular properties during the step cycle in trained 360

animals toward foot contact occurring more caudally in front of the hip and foot lift 361

16

occurring further back the hip (compare Figure 3A left and middle stick illustration). 362

Consequently, the limb extension part of the swing phase (defined as E1 in Philippson, 363

1905) was clearly reduced (compare Figure 3B left and middle panel) and its extension 364

counterpart at the beginning of the stance phase (defined as E2, corresponding to weight 365

acceptance) was so reduced that it was not possible to identify. Aside from these changes, 366

the angular coordination (i.e. coordination of the flexion and extension phases between 367

joints) was recovered in trained animals (compare the synchronicity of flexions and 368

extensions between joints in E3, F and E1 subphases in Figure 3B left and middle panels) 369

as well as the plantar placement of the foot during the stance phase. The whole step cycle 370

amplitude also returned close to the normal values in trained animals (compare Figure 3A 371

left and middle panels). 372

In untrained animals, the locomotor movements occurred far behind the hip at the 373

end of the recovery period and their amplitude was drastically reduced compared to 374

trained rats (Figure 3A right panel vs middle panel). In addition, the plantar placement 375

was lost (Figure 3A right panel), an “off period” with no active movement emerged at the 376

stance-swing transition (Figure 3B right panel) and the E1 and E2 subphases was 377

completely gone (Figure 3B right panel). This difference of the whole step cycle 378

amplitude between trained and untrained rats is clearly demonstrated by the angle 379

amplitudes of hip, ankle and MTP joints. The amplitudes of these joints were much 380

higher in trained group week 10 after SCI (Figure 4A, C and D). These greater 381

amplitudes are explain by the obvious higher flexion capacity of hip and ankle joints 382

(Figure 4A and C) together with the higher extension capacity of the MTP joint (Figure 383

4D). Interestingly, the amplitude of the knee joint remained reduced in similar 384

17

proportions in trained and untrained rats week 10 after SCI (Figure 4B) but the increased 385

ankle flexion capacity in trained group at the same time point and consequently the 386

increased ankle amplitude (Figure 4C) demonstrate that trained animals compensated the 387

knee deficit by increasing ankle amplitude. 388

Taken together these results show that our training protocol greatly improved the 389

overall kinematic of the angular joint excursions toward a return of coordination and 390

amplitude of the angles. 391

Treadmill training improves the temporal characteristics of locomotion 392

Although the swing phase length remained longer in trained group throughout the 393

recovery period (swing length is the exact half of the step cycle length depicted in Figure 394

2A) the swing phase duration was similar in both groups except week 5 and 6 (Figure 395

5A) and reached normal values (about 120 ms) at the end of experiment. The proportion 396

of the 2 constitutive subphases, F (about 37 % of the normal swing phase duration) and 397

E1 (about 63 %), were increased and decreased respectively compared to the baseline 398

throughout the recovery period in both groups (Figure 5B and C). However, the treadmill 399

training improved the recovery of F and E1 subphases’ proportion all along the post-SCI 400

period to reach about 70 % and 30 % respectively week 10 after SCI in trained group 401

while the untrained rats, stagnating around 95 % and 5 % respectively and failed to show 402

any change (Figure 5B and C). 403

A foot drag period, consisting of dragging forward the dorsal surface of the foot 404

on the treadmill belt, appeared in the first part of the swing phase after SCI in all animals. 405

In untrained rats, the duration of this deficit remained at around 120 ms throughout the 406

post-SCI period (Figure 5D) which was 100 % of the swing phase duration at week 10 407

18

after SCI (Figure 5A). In contrast, the duration of the foot drag period decreased week 408

after week with training so that it was not significantly higher than baseline in trained rats 409

as soon as week 3 after SCI to reach about 17 % of the swing phase duration at week 10 410

(Figure 5D). In addition, the treadmill training decreased the foot drag duration as early 411

as week 2 until it was greatly reduced week 10 after SCI so that it completely disappeared 412

in 33 % of the trained group while at the same time point all animals in untrained group 413

were highly affected by this deficit (Figure 5D). This is a clear case of a locomotor 414

parameter (foot drag) improving with training week after week. 415

Our results also show, in untrained animals, the appearance of a period of 416

inactivity at the transition between the end of the stance phase and the onset of the swing 417

phase that we have named “off period” during which no active movement of the 418

hindlimbs were present despite the continuous perineal stimulation. The duration of the 419

off period increased in untrained rats throughout the recovery to reach 352.6 ± 96.1 ms 420

(294 % of the swing phase duration) at the end of experiment (Figure 5E). The 421

consequence of this increase in the period of inactivity in untrained rats was a growing 422

incapacity to follow the treadmill speed. In contrast, treadmill training prevented such 423

period of inactivity at the end of stance (Figure 5E). 424

The variability of the step cycle duration was also a function of treadmill training 425

after SCI. The step to step variability of the cycle duration greatly increased throughout 426

the recovery period in untrained rats to reach 32.7 ± 7.4 % at week 10 (Figure 5F) while 427

the trained animals remained at the baseline level during the whole post-SCI period with 428

a CV of 8.2 ± 0.9 % at the end of experiment (Figure 5F). These data suggest that 429

19

treadmill training prevented the increase of the step duration instability and kept the 430

spinal trained animal at a level of variability similar to the intact (baseline) condition. 431

The foot trajectory during the swing phase improved in trained rats 432

We also assessed the averaged trajectory in the sagittal plan of the limb endpoint 433

in trained and untrained rats during the swing phase of treadmill locomotion at the normal 434

state and week 2, 6 and 8 after SCI as depicted in Figure 6. In normal rats, the foot leaves 435

the ground at the onset of the swing phase, reaches the greatest height in the second part 436

of the forward movement (E1 subphase) and finishes the swing phase by a vertical 437

trajectory corresponding to the final extension of the ankle until the foot hits the ground 438

(Figure 6A). As soon as week 2 after the spinal cord transection the trajectory of the foot 439

in trained animals showed a greatest vertical amplitude in the second part of the 440

movement and more regularity in the group than in untrained one (Figure 6B and E). 441

Then, the trajectory of the foot continued to improve throughout the recovery period in 442

trained animals including greatest vertical amplitude and regularity in the trajectories 443

themselves (Figure 6 C-D) while the foot trajectories of untrained rats worsened, became 444

short, disorganized and almost flat near the end of experiment (Figure 6 F-G). 445

Treadmill training improves the velocity of the swing phase of locomotion. 446

The instant and averaged velocity of the limb endpoint were assessed during the 447

swing phase and illustrated in figure 7. Briefly, in intact rats, the instant velocity of the 448

swing phase can be divided in 3 main parts, acceleration, deceleration and rebound (noted 449

, and respectively in Figure 7A1). As early as week 2 after SCI, trained rats 450

showed a return of the 3-part organisation of the instant velocity of the swing in similar 451

proportions to the baseline while the untrained rats remained deficient with a low and 452

20

long acceleration followed by a short deceleration and the disappearance of the third part 453

(compare subdivisions , and in Figure 7A2 and A5). This disorganization 454

remained present without improvement all along the recovery period (Figure 7A6 and 455

A7). On the other hand, in trained rats, the instant velocity improved throughout the post-456

SCI period so that week 8 after the lesion, not only the normalized duration of each part 457

were recovered compared to the baseline, but also the magnitude of the acceleration peak 458

increased (compare Figure 7A4 and A1). 459

Similarly to the instant velocity, whether the subparts or the whole swing phase, 460

the averaged velocity of the limb endpoint in untrained rats was strongly deficient and no 461

improvement was present throughout the 10-week recovery period compared to the 462

baseline (Figure 7B-D). In trained rats however, the averaged velocity of the F subphase 463

was improved as early as week 2 compared to untrained group and became similar to the 464

baseline from week 5 to the end of experiment (Figure 7B). Although the E1 subphase 465

velocity remained deficient in both groups at the end of the experiment, this parameter 466

was partially recovered with training from week 5 to the end of experiment (Figure 7C). 467

Finally, the averaged velocity of the overall swing phase in trained group became similar 468

to the baseline from week 3 after SCI and was fully recovered at the end of experiment 469

(Figure 7D). 470

These results show that treadmill training strongly improves the instant and 471

averaged velocity of the foot during the swing phase early after SCI and throughout the 472

10-week recovery period. 473

474

475

21

Treadmill training improves the recovery of swing phase components relationship 476

The swing phase duration depends on the length of the foot trajectory and the 477

hindlimb’s velocity in the air that are driven by the spinal cord. Consequently, the swing 478

phase features are very relevant indicators of the modulation exerted by the spinal 479

locomotor network because they are essentially dependent of the sensorimotor control 480

capacity of the animals. 481

Although the swing phase duration remained similar between groups at the end of 482

experiment (Figure 5A) we have shown that its principle constituents, namely swing 483

phase length, height and velocity had been strongly improved by our training protocol 484

(Figure 4A, D and Figure 7D). This apparent inconsistency between a parameter 485

(dependent variable) and its constituents (independent variables) can be only explained 486

by compensatory changes of the constituents’ relationship keeping constant the parameter 487

value. In order to explore the change of this relationship, we have computed polynomial 488

multiple regressions in normal, trained and untrained rats week 10 after SCI as illustrated 489

by the regression hyperplanes in figure 8. 490

In normal rats, the 4 variables (i.e. swing duration, length, velocity and height) are 491

strongly interrelated (R2 = 0.93) meaning that the swing duration can be reliably 492

predicted by the values of the 3 other variables. Ten weeks after SCI, in trained animals 493

the 4 variables remained correlated (R2 = 0.78) and the fitting hyperplane remained in the 494

same orientation and position than in normal animals (compare Figure 8A and hyperplane 495

corresponding to trained rats in Figure 8B). Even if the coefficient of determination was 496

decreased compared to the normal animals, (which can be explained by the increase of 497

the measurement variability after SCI), it remained high enough and significant to attest 498

22

of a strong relationship between the different variables similarly to the baseline. In 499

untrained rats, only 3 variables were interrelated week-10 after SCI (R2 = 0.78), the 500

swing height remained unrelated and the orientation and position in the 3D graph is 501

clearly different to normal animals (compare Figure 8A and hyperplane corresponding to 502

untrained rats in Figure 8B). These data show that our training model allowed to 503

maintain, in complete spinalized rats, similar interrelations than in normal rats between 504

different swing phase components while these interrelations were changed without 505

training. 506

Effect of training on the recovery of hindlimb coupling 507

The inter-limb coordination and variability was measured at the time points of 508

interest and illustrated in figure 9 (see method part for details about coordination 509

calculation and illustration). 510

Two weeks after SCI the out of phase coordination of the hindlimbs locomotor 511

movements was recovered in both groups (Figure 9A2 and A5). However, the foot lifts 512

occurred earlier after the contralateral contact compared to the baseline (Figure 9A2 and 513

A5). Consequently, the proportion of bilateral stance periods were decreased in both 514

groups although this effect was more pronounced in trained animals. The proportion of 515

bilateral stance periods then increased progressively in trained group to reach values 516

similar to untrained group week 6 after SCI (Figure 9A3 and A6). By the end of the 517

recovery period, the hindlimb coordination remained similar between groups while the 518

left and right lift still occurred earlier in both groups compared to the baseline keeping 519

shortened the bilateral stance periods (Figure 9A4 and A7). In trained animals, each 520

locomotor period and event tends to return progressively to the baseline value after SCI 521

23

while the dispersion around the mean of each event decreased (see dotted lines in Figure 522

9A4) demonstrating the improvement of the rats’ homogeneity in this group which is not 523

the case in untrained animals (see dotted lines in Figure 9A7). 524

The step to step variability of the left lift, left contact and right lift coordination in 525

each rat during locomotion was averaged by group and plotted against each time point in 526

figure 9B, C and D respectively. In trained rats, the regularity of left lift and left contact 527

coordination remained similar to the baseline all along the post-SCI period while a non-528

significant upward trend in right lift variability was present (Figure 9B-D). However, 529

these parameters were greatly increased in untrained rats throughout the recovery period 530

(Figure 9B, C and D). In addition, the inconstancy of coordination was clearly higher in 531

untrained group compared to the trained one all along the recovery period for the left lift 532

(Figure 9B) and the left contact (week 3-4 and 6-10 in Figure 9C). Although the right lift 533

variability was higher in untrained group during almost the whole post-SCI period (week 534

3-4 and 8) it became similar week 10 after SCI (Figure 9D). 535

These results show that the averaged coordination of both hindlimbs was similar 536

between trained and untrained rats after SCI, independently from the other kinematic 537

parameters highly degraded in untrained compared to trained rats. In addition, the step to 538

step coordination in each rat and the coordination regularity between rats in the group 539

were more consistent in trained than untrained animals. 540

Recovery of EMG pattern 541

Week 11 after SCI we recorded, through transcutaneous electrodes, the EMG 542

activity of tibialis anterior and gastrocnemius muscles on both hindlimbs during 543

locomotion on treadmill to assess the recovery of central pattern output. Comparison of 544

24

raw EMG activity in normal, trained and untrained rats at the end of the recovery period 545

is depicted in figure 10. 546

At the end of the 10-week recovery period, concomitantly with the kinematic 547

parameters, electromyographic recording showed differential muscle activity patterns 548

between groups during locomotion. In trained rats, the hindlimb locomotor EMGs were 549

well organized, coordinated and the regularity of bursting was so similar to those 550

recorded in normal rats that it was impossible to visually differentiate (compare Figure 551

10A and B). In addition, the recovered EMG pattern in trained animals was not restricted 552

in short bouts of locomotion but was very stable over the whole locomotor session 553

similarly to their kinematic locomotor pattern. The extensor activity (i.e. related to the 554

weight bearing and forward propulsion during locomotion) was also increased with 555

training (compare RGL and LGL in Figure 10B and C). Comparatively, although EMG 556

rhythmic activity was still present in untrained rats at the same time point, the 557

electromyogram indicates that the normal pattern and organization was lost including 558

transient co-activations of antagonist muscles and consequently the failure of muscle 559

coordination (see RTA vs RGL and LTA vs LGL in Figure 10C). This uncoordinated 560

activity of the EMGs in untrained animals is coherent with our results on kinematics 561

(Figure 9B, C and D). This result reinforces the statement that our treadmill training 562

protocol positively interacts with the rats’ locomotor spinal network plasticity below the 563

complete lesion and allows the re-expression of near-normal EMG pattern. 564

Assessment of hindlimb muscle mass 565

Since the muscular properties could also interact with the locomotor performance, 566

and especially the muscle mass which is related to the muscle strength, contraction 567

25

velocity and endurance, we compared the weight of hindlimbs selected muscles in both 568

groups in order to perhaps correlate the differences measured in locomotor performances 569

with the potential differences in muscle properties that would be changed by training. At 570

the end of experiment rats were weighted then perfused, afterward the left ankle flexor 571

Tibialis Anterior (TA) and extensor Gastrocnemius (G) muscles were harvested and 572

weighted. The rats’ weight and the muscle mass of TA and G expressed in percent of the 573

total body weight were then averaged by group and depicted in figure 11. No significant 574

difference was measured in total body weight (Figure 11A), TA and G muscles mass 575

(Figure 11B) between trained and untrained rats at the end of experiment. These results 576

confirm that mass of muscles directly involved in hindlimb locomotor re-expression was 577

not changed by training in our protocol and consequently highlight the key role played by 578

changes in the spinal circuitry rather than changes in muscle mass. However no 579

histochemical analysis were performed. 580

581

26

Discussion 582

Since the first work on cats completely spinalized as adults, demonstrating that 583

two to three weeks of treadmill training could induce re-expression of involuntary 584

locomotion including weight support and plantar foot placement (Barbeau and Rossignol, 585

1987), several studies concluded that adult complete spinal rats could not walk with the 586

hindlimbs without invasive and/or complex therapeutic strategies (Slawinska et al., 2000; 587

Antri et al., 2002; Courtine et al., 2009). We hypothesized that sustained perineal 588

stimulation in rats could trigger the lumbar locomotor network to induce active treadmill 589

training in order to re-express locomotion in similar conditions than in adult spinal cats. 590

To address these points, we investigated the effects of daily treadmill locomotor training 591

elicited by perineal stimulation provided while the animal was held in a quadrupedal 592

posture on the functional plasticity of the lumbar spinal locomotor network in adult rats 593

completely spinalized at T7. In accordance with the previous studies in cats, our detailed 594

kinematic analysis shows strong evidences of intrinsic and sustained remodeling of the 595

spinal locomotor circuitry induced by our treadmill training protocol bringing back near-596

normal hindlimb locomotion, as demonstrated by the recovery of well-organized and 597

adaptive locomotion in trained rats while untrained ones showed poor locomotor 598

performance decreasing with time despite the fact that the untrained rats are recorded in 599

the same conditions as trained rats (i.e. perineal stimulation). 600

Non-invasive triggering of the spinal locomotor network 601

When the tail of an adult spinal rat is manually stimulated, it generally produces 602

short bouts of uncoordinated hindlimb flexions followed by prolonged extensions, 603

defective foot placements and abnormal swing phases (Slawinska et al., 2012). Our 604

27

results reflect the fact that the tonic stimulation on the perineal area is more efficient to 605

elicit robust hindlimb stepping allowing locomotor training in adult spinal rats. Although 606

this strategy is commonly used in adult spinal cats to induce locomotion, the underlying 607

mechanisms are not yet understood. Fictive locomotion recordings in rats clearly suggest 608

that the CPG below the lesion site can be spontaneously active after the intraspinal graft 609

of embryonic brainstem cells (Yakovleff et al., 1995) or activated by the stimulation of 610

mesencephalic locomotor region (Iles and Nicolopoulos-Stournaras, 1996; Canu et al., 611

2001). The spinal locomotor network is not created by these stimulations, it is revealed 612

(Rossignol et al., 2014) suggesting that such locomotor activity emerged in response to 613

the change of excitatory/inhibitory balance within the spinal cord. 614

Perineal stimulation seems to adequately modulate the excitability level of the 615

spinal locomotor circuitry and compensates for the decrease of supraspinal input in 616

incomplete SCI rat model (Alluin et al., 2011; Alluin et al., 2014) or for the complete 617

absence of brain input as in the present study. This triggering capability is probably 618

innate since it was effective as early as few days after SCI as demonstrated by the re-619

expression of alternated bilateral flexion-extension in untrained animals. However unlike 620

cats (Barbeau and Rossignol, 1987) none of the rats walked without perineal stimulation 621

even at the end of the post-SCI period. This suggest common mechanisms underlying 622

spinal cord modulation between species but different tuning giving less weight to 623

hindlimb cutaneous and proprioceptive information during spinal locomotion in rats 624

compared to cats. This predominant influence of the perineal afferents in our protocol 625

could be a double-edged sword because if their stimulation was sufficient to trigger the 626

CPG, it could also be responsible of the differences observed in the locomotor pattern of 627

28

trained compared to normal rats (for instance the overextension at the swing/stance 628

transition) given that a variety of sensory inputs can influence the spinal locomotor 629

network output (Rossignol and Frigon, 2011). Another hypothesis concerning this 630

overextension could be that complete thoracic SCI also affects the tone of abdominal 631

muscles favoring an anterior pelvic tilt and consequently a backward shift of the hindlimb 632

step cycles through an exaggerated extension at the end of the stance phase (Movie 1). 633

We have previously observed this also in incomplete spinal rats (Alluin et al., 2011; 634

Alluin et al., 2014). 635

From a clinical perspective, it is of primary interest to have a clearer idea of the 636

spinal cord functional potential devoid of supraspinal influence since the optimization of 637

the spinal locomotor circuitry will have a major impact on the feedback provided to the 638

remnant structures after partial spinal cord lesions. 639

Treadmill training as a key factor to positively drive the spinal plasticity 640

Different strategies were investigated to induce hindlimb locomotion in complete 641

spinal rats such as cell graft (Gimenez y Ribotta et al., 2000; Majczynski et al., 2005), 642

excitatory drugs (Feraboli-Lohnherr et al., 1999; Antri et al., 2002), epidural 643

electrostimulation and/or robotic assistance (Timoszyk et al., 2002; de Leon et al., 2002; 644

Cha et al., 2007), upright posture (Slawinska et al., 2012) or a combination between some 645

of them (Ichiyama et al., 2005; de Leon and Acosta, 2006; Gerasimenko et al., 2007; 646

Courtine et al., 2009; Hsieh and Giszter, 2011; Dominici et al., 2012). In some of these 647

studies, treadmill training had also been provided in combination with other strategies in 648

order to produce synergistic effects. When treadmill training was administered alone, its 649

beneficial effect on locomotor re-expression was very moderate (Zhang et al., 2007; Ilha 650

29

et al., 2011) or no beneficial effect was observed even in facilitating upright posture 651

(Ichiyama et al., 2008). 652

Our data clearly demonstrate that if the treadmill training is not passive but 653

triggered by perineal stimulation, it positively interacts with functional spinal plasticity 654

insomuch that, the locomotor parameters of trained rats return close to the normal values 655

while untrained rats had progressively lost the capability to walk on treadmill. For 656

instance, the amplitude of hip and MTP (Figure 4), the swing velocity (Figure 7), the 657

EMG activity of ankle flexors and extensors (Figure 10) and the variability of several 658

kinematic parameters (Figure 2, 5 and 9) returned to normal values only in trained rats. 659

We also have analyzed the interrelations of the different swing phase components (i.e. 660

duration, length, velocity and height; Figure 8) which are very representative of the 661

sensorimotor and adaptive control capacity of the spinal cord because the movements of 662

the limb in the air is devoid of direct environmental influence and only driven by the 663

locomotor network contrary to the stance phase which is directly driven by the belt. 664

These fine interrelations are also greatly restored by training while the absence of activity 665

induces a deterioration. Such impressive recovery was previously showed in spinal rats 666

trained on treadmill in upright posture with combination of 5-HT agonists and epidural 667

electrostimulation (Courtine et al., 2009). The locomotor effect of these agonists are 668

attributed to the activation of 5-HT2 and 5-HT7 receptors (Schmidt and Jordan, 2000; 669

Garraway and Hochman, 2001; Madriaga et al., 2004; Pearlstein et al., 2005; Liu et al., 670

2009; Dunbar et al., 2010). However, the mechanisms induced by epidural 671

electrostimulation in rats remain unknown although it has been suggested that such 672

stimulation activates spinal locomotor network through afferent fibers (Courtine et al., 673

30

2009). Although no spasticity was observed in the present study, it has also been shown 674

that spinal 5-HT receptors below the lesion site are upregulated after complete SCI 675

leading to serotonin denervation supersensitivity and the subsequent spasticity (Barbeau 676

and Bédard, 1981; Kong et al., 2010; Kong et al., 2011). Taken together, these 677

considerations suggests that the rehabilitative strategy used in the present study could acts 678

on spinal cord through similar mechanisms implicating afferent recruitment via the 679

perineum and concomitantly the neurochemical normalization of the spinal cord induced 680

by step training. Specific upregulation of BDNF in the lumbar spinal cord of trained rats 681

could also be implicated in the strong training effect observed in our results (Joseph et al., 682

2012). 683

Although Smith et al. have reported deleterious effect of exercise training starting 684

too early after SCI (Smith et al., 2009), the precocious locomotor improvements week 2 685

after SCI in our trained rats shows that starting it immediately after complete SCI have 686

positive effects on spinal plasticity and highlight the benefit to stimulate the spinal cord 687

even when the spinal circuits do not produce locomotor output in response to the 688

stimulation as we observed in the present study. The adaptive processes of the spinal cord 689

were also effective by early training as trained rats could adapt the locomotor pattern to 690

the belt velocity from week 2 (Figure 1) similar the observations made in spinal cats 691

(Barbeau and Rossignol, 1987). In addition, the duration of daily treadmill training 692

session previously used in complete spinal cats and rats was at least 20 min (Barbeau and 693

Rossignol, 1987; Lovely et al., 1990; de Leon et al., 1998; Courtine et al., 2009). The 694

robust results obtained in the present study with a daily exercise period as short as 10 min 695

raises the issue of the optimal duration and frequency of the treadmill training to induce 696

31

long-lasting plastic changes and the concomitant locomotor re-expression. Studies 697

involving a variety of treadmill training periods could be performed but would be very 698

demanding on human resources and would probably add little to the locomotor 699

performance seen here since the locomotor parameters return to values obtained in the 700

animals during the control period before spinalization. 701

Ultimately, the absence of difference between trained and untrained rats in terms 702

of hindlimb muscle mass (Figure 11) strengthens the idea that the very positive effect of 703

training on the detailed locomotor characteristics reported in the present study is largely 704

attributed to the activity-dependent plasticity occurring within the spinal circuitry 705

(Edgerton et al., 2004). On the other hand, the absence of activity after such drastic spinal 706

lesion induce a dramatic impairment of the spinal locomotor capability that degrade with 707

time as demonstrated previously (Courtine et al., 2009). 708

Conclusion and perspectives 709

The afferent inputs from the perineum and sacral area can access the rat’s spinal 710

locomotor circuitry (Etlin et al., 2010) to elicit step training and induce the activity-711

dependent spinal plasticity underlying the recovery of locomotion in our study. Treadmill 712

locomotor training in combination with perineal stimulation could appear as the key 713

components to easily induce locomotion in complete spinal rat and should question the 714

need to use more invasive central stimulations (i.e. pharmacology, epidural 715

electrostimulation, cell graft...) that should, at least, surpass the intrinsic capacity 716

highlighted in the present paper. Further neuroanatomical studies in adult spinal rats 717

submitted to training should improve our comprehensive understanding of intrinsic spinal 718

plasticity. In the clinical context mainly involving incomplete spinal injuries, it is of great 719

32

interest to exploit the intrinsic locomotor capacity of the spinal cord since the potentiation 720

of spinal locomotor network improves the functional interactions between afferent 721

feedback and remnant structures which are essential for the overall recovering of 722

locomotion. This latest point is crucial because safely strengthening the spared nervous 723

components after SCI should be the first priority before considering more invasive 724

therapeutic approaches. 725

726

33

References 727

Alluin O, Delivet-Mongrain H, Gauthier MK, Fehlings MG, Rossignol S, Karimi-728

Abdolrezaee S (2014) Examination of the Combined Effects of Chondroitinase ABC, 729

Growth Factors and Locomotor Training following Compressive Spinal Cord Injury on 730

Neuroanatomical Plasticity and Kinematics. PLoS One 9:e111072. 731

Alluin O, Karimi-Abdolrezaee S, Delivet-Mongrain H, Leblond H, Fehlings MG, 732

Rossignol S (2011) Kinematic study of locomotor recovery after spinal cord clip 733

compression injury in rats. J Neurotrauma 28:1963-1981. 734

Antri M, Barthe JY, Mouffle C, Orsal D (2005) Long-lasting recovery of locomotor 735

function in chronic spinal rat following chronic combined pharmacological stimulation of 736

serotonergic receptors with 8-OHDPAT and quipazine. Neurosci Lett 384:162-167. 737

Antri M, Orsal D, Barthe J-Y (2002) Locomotor recovery in the chronic spinal rat: effects 738

of long-term treatment with a 5-HT2 agonist. In: Spinal Cord Trauma: Regeneration, 739

Neural Repair and Functional Recovery (McKerracher L, Doucet G, Rossignol S, eds), 740

pp 467-476. Amsterdam: Elsevier Science. 741

Barbeau H, Bédard P (1981) Denervation supersensitivity to 5-HT in rats following 742

spinal transection and 5,7 dihydroxytryptamine injection. Neuropharmacology 20:611-743

616. 744

34

Barbeau H, Rossignol S (1987) Recovery of locomotion after chronic spinalization in the 745

adult cat. Brain Res 412:84-95. 746

Canu MH, Falempin M, Orsal D (2001) Fictive motor activity in rat after 14 days of 747

hindlimb unloading. Exp Brain Res 139:30-38. 748

Cha J, Heng C, Reinkensmeyer DJ, Roy RR, Edgerton VR, de Leon RD (2007) 749

Locomotor ability in spinal rats is dependent on the amount of activity imposed on the 750

hindlimbs during treadmill training. J Neurotrauma 24:1000-1012. 751

Courtine G, Gerasimenko Y, van den BR, Yew A, Musienko P, Zhong H, Song B, Ao Y, 752

Ichiyama RM, Lavrov I, Roy RR, Sofroniew MV, Edgerton VR (2009) Transformation 753

of nonfunctional spinal circuits into functional states after the loss of brain input. Nature 754

neuroscience 12:1333-1342. 755

de Leon RD, Acosta CN (2006) Effect of robotic-assisted treadmill training and chronic 756

quipazine treatment on hindlimb stepping in spinally transected rats. J Neurotrauma 757

23:1147-1163. 758

de Leon RD, Hodgson JA, Roy RR, Edgerton VR (1998) Full weight-bearing hindlimb 759

standing following stand training in the adult spinal cat. J Neurophysiol 80:83-91. 760

35

de Leon RD, Kubasak MD, Phelps PE, Timoszyk WK, Reinkensmeyer DJ, Roy RR, 761

Edgerton VR (2002) Using robotics to teach the spinal cord to walk. Brain Res Brain Res 762

Rev 40:267-273. 763

Dominici N, Keller U, Vallery H, Friedli L, Van den Brand R, Starkey ML, Musienko P, 764

Riener R, Courtine G (2012) Versatile robotic interface to evaluate, enable and train 765

locomotion and balance after neuromotor disorders. Nat Med 18:1142-1147. 766

Dunbar MJ, Tran MA, Whelan PJ (2010) Endogenous extracellular serotonin modulates 767

the spinal locomotor network of the neonatal mouse. J Physiol 588:139-156. 768

Edgerton VR, Tillakaratne NJ, Bigbee AJ, de Leon RD, Roy RR (2004) Plasticity of the 769

spinal neural circuitry after injury. Annu Rev Neurosci 27:145-167. 770

Etlin A, Blivis D, Ben-Zwi M, Lev-Tov A (2010) Long and short multifunicular 771

projections of sacral neurons are activated by sensory input to produce locomotor activity 772

in the absence of supraspinal control. J Neurosci 30:10324-10336. 773

Feraboli-Lohnherr D, Barthe J-Y, Orsal D (1999) Serotonin-induced activation of the 774

network for locomotion in adult spinal rats. J Neurosci Res 55:87-98. 775

Fong AJ, Cai LL, Otoshi CK, Reinkensmeyer DJ, Burdick JW, Roy RR, Edgerton VR 776

(2005) Spinal cord-transected mice learn to step in response to quipazine treatment and 777

robotic training. J Neurosci 25:11738-11747. 778

36

Garraway SM, Hochman S (2001) Pharmacological characterization of serotonin receptor 779

subtypes modulating primary afferent input to deep dorsal horn neurons in the neonatal 780

rat. Br J Pharmacol 132:1789-1798. 781

Gerasimenko YP, Ichiyama RM, Lavrov IA, Courtine G, Cai L, Zhong H, Roy RR, 782

Edgerton VR (2007) Epidural spinal cord stimulation plus quipazine administration 783

enable stepping in complete spinal adult rats. J Neurophysiol 98:2525-2536. 784

Gimenez y Ribotta M, Provencher J, Feraboli-Lohnherr D, Rossignol S, Privat A, Orsal 785

D (2000) Activation of locomotion in adult chronic spinal rats is achieved by 786

transplantation of embryonic raphe cells reinnervating a precise lumbar level. J Neurosci 787

20:5144-5152. 788

Harkema S, Gerasimenko Y, Hodes J, Burdick J, Angeli C, Chen Y, Ferreira C, Willhite 789

A, Rejc E, Grossman RG, Edgerton VR (2011) Effect of epidural stimulation of the 790

lumbosacral spinal cord on voluntary movement, standing, and assisted stepping after 791

motor complete paraplegia: a case study. Lancet. 792

Hsieh FH, Giszter SF (2011) Robot-driven spinal epidural stimulation compared with 793

conventional stimulation in adult spinalized rats. Conf Proc IEEE Eng Med Biol Soc 794

2011:5807-5810. 795

37

Ichiyama RM, Broman J, Roy RR, Zhong H, Edgerton VR, Havton LA (2011) 796

Locomotor training maintains normal inhibitory influence on both alpha- and gamma-797

motoneurons after neonatal spinal cord transection. J Neurosci 31:26-33. 798

Ichiyama RM, Courtine G, Gerasimenko YP, Yang GJ, Van den Brand R, Lazrov IA, 799

Zhong H, Roy RR, Edgerton VR (2008) Step Training reinforces specific spinal 800

locomotor circuitry in adult spinal rats. pp 7370-7375. 801

Ichiyama RM, Gerasimenko YP, Zhong H, Roy RR, Edgerton VR (2005) Hindlimb 802

stepping movements in complete spinal rats induced by epidural spinal cord stimulation. 803

Neurosci Lett 383:339-344. 804

Iles JF, Nicolopoulos-Stournaras S (1996) Fictive locomotion in the adult decerebrate rat. 805

Exp Brain Res 109:393-398. 806

Ilha J, Centenaro LA, Broetto CN, de Souza DF, Jaeger M, do Nascimento PS, Kolling J, 807

Ben J, Marcuzzo S, Wyse AT, Gottfried C, Achaval M (2011) The beneficial effects of 808

treadmill step training on activity-dependent synaptic and cellular plasticity markers after 809

complete spinal cord injury. Neurochem Res 36:1046-1055. 810

Joseph MS, Tillakaratne NJ, de Leon RD (2012) Treadmill training stimulates brain-811

derived neurotrophic factor mRNA expression in motor neurons of the lumbar spinal cord 812

in spinally transected rats. Neuroscience 224:135-144. 813

38

Kong XY, Wienecke J, Chen M, Hultborn H, Zhang M (2011) The time course of 814

serotonin 2A receptor expression after spinal transection of rats: an immunohistochemical 815

study. Neuroscience 177:114-126. 816

Kong XY, Wienecke J, Hultborn H, Zhang M (2010) Robust upregulation of serotonin 817

2A receptors after chronic spinal transection of rats: an immunohistochemical study. 818

Brain Res 1320:60-68. 819

Liu J, Akay T, Hedlund PB, Pearson KG, Jordan LM (2009) Spinal 5-HT7 receptors are 820

critical for alternating activity during locomotion: in vitro neonatal and in vivo adult 821

studies using 5-HT7 receptor knockout mice. J Neurophysiol 102:337-348. 822

Lovely RG, Gregor RJ, Roy RR, Edgerton VR (1990) Weight-bearing hindlimb stepping 823

in treadmill-exercised adult spinal cat. Brain Res 514:206-218. 824

Madriaga MA, McPhee LC, Chersa T, Christie KJ, Whelan PJ (2004) Modulation of 825

locomotor activity by multiple 5-HT and dopaminergic receptor subtypes in the neonatal 826

mouse spinal cord. J Neurophysiol 92:1566-1576. 827

Majczynski H, Maleszak K, Cabaj A, Slawinska U (2005) Serotonin-related enhancement 828

of recovery of hind limb motor functions in spinal rats after grafting of embryonic raphe 829

nuclei. J Neurotrauma 22:590-604. 830

39

Martinez M, Delivet-Mongrain H, Leblond H, Rossignol S (2012) Effect of locomotor 831

training in completely spinalized cats previously submitted to a spinal hemisection. J 832

Neurosci 32:10961-10970. 833

Musienko P, Van den Brand R, Marzendorfer O, Roy RR, Gerasimenko Y, Edgerton VR, 834

Courtine G (2011) Controlling Specific Locomotor Behaviors through Multidimensional 835

Monoaminergic Modulation of Spinal Circuitries. J Neurosci 31:9264-9278. 836

Orsal D, Barthe JY, Antri M, Feraboli-Lohnherr D, Yakovleff A, Ribotta M, Privat A, 837

Provencher J, Rossignol S (2002) Locomotor recovery in chronic spinal rat: long-term 838

pharmacological treatment or transplantation of embryonic neurons? Prog Brain Res 839

137:213-230. 840

Pearlstein E, Ben MF, Pflieger JF, Vinay L (2005) Serotonin refines the locomotor-841

related alternations in the in vitro neonatal rat spinal cord. Eur J Neurosci 21:1338-1346. 842

Philippson M (1905) L'autonomie et la centralisation dans le système nerveux des 843

animaux. Trav Lab Physiol Inst Solvay (Bruxelles ) 7:1-208. 844

Rossignol S (1996) Neural control of stereotypic limb movements. In: Handbook of 845

Physiology, Section 12. Exercise: Regulation and Integration of Multiple Systems 846

(Rowell LB, Sheperd JT, eds), pp 173-216. New York: Oxford University Press. 847

40

Rossignol S, Barriere G, Alluin O, Frigon A (2009) Re-expression of Locomotor 848

Function After Partial Spinal Cord Injury. Physiology (Bethesda ) 24:127-139. 849

Rossignol S, Dubuc R, Gossard JP (2006) Dynamic sensorimotor interactions in 850

locomotion. Physiol Rev 86:89-154. 851

Rossignol S, Frigon A (2011) Recovery of locomotion after spinal cord injury: some facts 852

and mechanisms. Annu Rev Neurosci 34:413-440. 853

Rossignol S, Giroux N, Chau C, Marcoux J, Brustein E, Reader TA (2001) 854

Pharmacological aids to locomotor training after spinal injury in the cat. J Physiol 855

533:65-74. 856

Rossignol S, Schmidt BJ, Jordan LM (2014) Spinal plasticity underlying the recovery of 857

locomotion after injury. In: Textbook of Neural Repair and Rehabilitation (Selzer ME, 858

Clarke S, Cohen LG, Kwakkel G, Miller RH, eds), pp 166-195. Cambridge: Cambridge 859

University Press. 860

Schmidt BJ, Jordan LM (2000) The role of serotonin in reflex modulation and locomotor 861

rhythm production in the mammalian spinal cord. Brain Res Bull 53:689-710. 862

Slawinska U, Majczynski H, Dai Y, Jordan LM (2012) The upright posture improves 863

plantar stepping and alters responses to serotonergic drugs in spinal rats. J Physiol 864

590:1721-1736. 865

41

Slawinska U, Majczynski H, Djavadian R (2000) Recovery of hindlimb motor functions 866

after spinal cord transection is enhanced by grafts of the embryonic raphe nuclei. Exp 867

Brain Res 132:27-38. 868

Slawinska U, Miazga K, Cabaj AM, Leszczynska A, Majczynski H, Nagy JI, Jordan LM 869

(2013) Grafting of fetal brainstem 5-HT neurons into the sublesional spinal cord of 870

paraplegic rats restores coordinated hindlimb locomotion. Exp Neurol 247:572-581. 871

Smith RR, Brown EH, Shum-Siu A, Whelan A, Burke DA, Benton RL, Magnuson DS 872

(2009) Swim training initiated acutely after spinal cord injury is ineffective and induces 873

extravasation in and around the epicenter. J Neurotrauma 26:1017-1027. 874

Tillakaratne NJ, Guu JJ, de Leon RD, Bigbee AJ, London NJ, Zhong H, Ziegler MD, 875

Joynes RL, Roy RR, Edgerton VR (2010) Functional recovery of stepping in rats after a 876

complete neonatal spinal cord transection is not due to regrowth across the lesion site. 877

Neuroscience 166:23-33. 878

Timoszyk WK, de Leon RD, London N, Roy RR, Edgerton VR, Reinkensmeyer DJ 879

(2002) The rat lumbosacral spinal cord adapts to robotic loading applied during stance. J 880

Neurophysiol 88:3108-3117. 881

Van den Brand R, Heutschi J, Barraud Q, DiGiovanna J, Bartholdi K, Huerlimann M, 882

Friedli L, Vollenweider I, Moraud EM, Duis S, Dominici N, Micera S, Musienko P, 883

42

Courtine G (2012) Restoring voluntary control of locomotion after paralyzing spinal cord 884

injury. Science 336:1182-1185. 885

Yakovleff A, Cabelguen JM, Orsal D, Ribotta M, Rajaofetra N, Drian MJ, Bussel B, 886

Privat A (1995) Fictive motor activities in adult chronic spinal rats transplanted with 887

embryonic brainstem neurons. Exp Brain Res 106:69-78. 888

Zhang Y, Ji SR, Wu CY, Fan XH, Zhou HJ, Liu GL (2007) Observation of locomotor 889

functional recovery in adult complete spinal rats with BWSTT using semiquantitative and 890

qualitative methods. Spinal cord 45:496-501. 891

892

893

894

43

Legends 895

Figure 1. General walking capacity throughout the recovery period. A, Proportion of rats 896

in trained (black, n = 13) and untrained (grey, n = 8) groups capable of walking with paw 897

placements (plantar or dorsal) on treadmill at 14 m.min-1 and expressed as % of total 898

number of animals in each group. B, same as A at 20 m.min-1. C, same as A and B at 26 899

m.min-1. 900

Figure 2. Recovery of the step cycle length, position and height after SCI. A, length 901

(millimetres) of the step cycle (from the foot contact on the treadmill belt to the next one) 902

before (baseline) and weekly after SCI in trained (black dots) and untrained (white dots) 903

rats. B, individual coefficient of variation (individual step to step variability averaged by 904

groups and expressed in percent) of data depicted in A. C, position (mm) of the stance 905

phase relative to the vertical projection of the great trochanter (vertical dash line 906

corresponding to 0) for the time points specified above. The left end of each bar 907

represents the contact position of the left foot with the treadmill belt (onset of stance 908

phase) while the right end represents the foot lift position (end of the stance phase) in 909

trained (black bars) and untrained (grey bars) rats. Sketches representing the foot contact 910

and lift position measurement are given above the graph. D, maximum height (mm) of 911

the foot during the swing phase of locomotion in trained (black dots) and untrained 912

(white dots) rats and for the time points specified above. Sketch depicting the foot height 913

measurement is given above the graph. Data were recorded at 14 m.min−1 and are 914

expressed as mean ± SEM. 915

916

44

Figure 3. Recovery of angular excursions. A, stick figures of stance and swing phases 917

from representative rats before (left panel) and week 10 after SCI in trained (middle 918

panel) and untrained (right panel) groups. The direction of movement is given (black 919

arrows). B, Angular excursions (in degrees) averaged by groups of hip, knee, ankle and 920

MTP before (left panel) and week 10 in trained (middle panel) and untrained (right panel) 921

groups. The step cycle duration (x axis) is normalized in % of the step cycle beginning 922

and finishing with a foot contact (down arrows at the top of each panel). The average 923

duration of the step cycle ± SEM (in milliseconds) is depicted above each panel. The 924

position of foot lift ± SD (up arrows at the top of each panel), the proportion of stance 925

and swing phases (bottom of each panel) and the Philippson’s subphases are also given. 926

The “Off period” on the right panel corresponds to the absence of active hindlimb 927

movement. Data were recorded at 14 m.min−1 and are expressed as mean ± SEM. 928

Figure 4. Recovery of hindlimb angle joints amplitude after SCI. A, averaged minimum, 929

maximum and amplitude of the hip joint angle (in degrees) during locomotion on 930

treadmill before and weekly after SCI in trained (black) and untrained (grey) rats. B, 931

same as A for the knee joint angle. C, same as A and B for the ankle joint angle. D, same 932

as A, B and C for the MTP joint angle. Black stars and White stars represent the 933

statistical comparison of minimum and maximum angle values with baseline in trained 934

and untrained groups respectively. Crosses below and above the sticks represent the 935

statistical comparison between both groups for minimum and maximum angle values 936

respectively at each time point. Signs in the bottom part of the graphs relate to the angle 937

amplitude (max – min illustrated by the length of the sticks) and follow the same rules 938

45

than describe above. Data were recorded at 14 m.min−1 and are expressed as mean ± 939

SEM. 940

Figure 5. Effect of treadmill training on the temporal characteristics of locomotion. A, 941

duration (milliseconds) of the swing phase of locomotion averaged by group in trained 942

(black dots) and untrained (white dots) rats before (baseline) and weekly throughout the 943

recovery period. Schematic representation of the swing phase of locomotion demarcated 944

by the foot lift and contact (up and down arrows respectively) and the movement 945

direction are given above the graph. B, duration of the F subphase of the swing (flexion 946

part) expressed in percent of the whole swing duration. C, duration of the E1 subphase of 947

the swing (extension part) expressed in percent of the whole swing duration. D duration 948

(ms) of the foot drag period on the treadmill belt during the swing phase of locomotion. 949

E, duration (ms) of the period with no movement of the hindlimb, named “Off period”, 950

before the onset of the swing phase. F, individual coefficient of variation (step to step 951

variability expressed in percent) of the whole step cycle duration and averaged in trained 952

(black bars) and untrained (grey bars) rats. Data were recorded at 14 m.min−1 and are 953

expressed as mean ± SEM. 954

Figure 6. Evolution of the hindlimb foot trajectory during the swing phase following 955

SCI. A, superimposed averaged trajectory of the third toe in normal rats during the swing 956

phase of locomotion on treadmill. The scale of the y axis (vertical foot displacements) is 957

increased (zoomed in) compared to the x for better depiction and comparison of the 958

vertical amplitude. The vertical line (labelled as Hip on the graph) represents the vertical 959

projection of the great trochanter and corresponds to the reference (0 value) on the x axis. 960

B, C and D, similar representations than A in trained rats week 2, 6 and 8 after SCI 961

46

respectively. E, F and G, similar representations than B, C and D in untrained rats at the 962

same time points. Data are expressed as averaged instant position relative to the hip (x 963

axis) and the treadmill belt (y axis) in millimetres. 964

Figure 7. Effect of treadmill training on the recovery of the swing phase velocity. A1-7, 965

averaged instant velocity of the left foot against normalized swing phase duration before 966

(A1) and Week 2, 6 and 8 after SCI in trained (A2-4) and untrained (A5-7) groups. The 967

height of the horizontal grey strips depicted on each panel reflects the range of normal 968

instant velocity (baseline) for direct comparison with the different epochs after SCI. The 969

subdivisions of the instant velocity (i.e. acceleration, deceleration and rebound) are 970

specified in each chart (noted , and respectively) and separated by the vertical 971

lines. B, C and D, averaged velocity of the left foot during the F, E1 subphases of the 972

swing and the overall swing phase respectively, in trained (filled circle) and untrained 973

(empty circle) groups. The corresponding phase and subphases of the step cycle are 974

indicated above each panel on the stick diagram. Data were recorded at 14 m.min−1 and 975

are expressed as mean ± SEM. 976

Figure 8. Recovery of hindlimbs duration, length, velocity and height relationship at the 977

end of experiment. A, fitted hyperplane of 3 independent variables (swing length [x], 978

swing velocity [z] and swing height [dots colour gradient]) predicting the swing duration 979

(dependant variable [y]) in normal animals. The hyperplane colour gradient (from orange 980

to dark blue) represent the z axis values to help for the spatial representation of the plane. 981

Each dot represents 1 step cycle and the swing height values are represented by the blue 982

to red colour gradient of the dots. R2 of the multiple correlation, the number of variables 983

correlated and the p value are given at the top of the graph. B, same representation than 984

47

A, week 10 after SCI in trained (dots and upper plane) and untrained (triangles and lower 985

plane) animals. Data are expressed as raw data. 986

Figure 9. Evolution of hindlimbs coordination after SCI. A1-7, circular representation of 987

temporal relationship of hindlimbs locomotor events (right contact, left lift, left contact 988

and right lift) before (baseline) and week 2, 6 and 8 after SCI in trained (circles at the 989

top) and untrained (circles at the bottom) groups during locomotion. Given that 990

continuous locomotion is a repeated sequence of events (left/right foot contact/lift), the 991

foot contact is both the beginning of a step cycle and the end of the previous one. 992

Consequently, in the present figure the relative duration of right step cycle (reference) is 993

represented between 0 and 1 which are at the same position on the graph (top of the 994

circles). Data are synchronized on the right contact (represented by 0 at the top of each 995

circle) and the sequence of events should be read in the clockwise direction. Dark grey 996

circular band between RC and RL represents the averaged normalized right stance phase 997

and the empty space between RL and RC represents the normalized average of the 998

concomitant swing phase, both relative to the right step cycle duration. Light grey 999

circular band between LC and LL and the empty space between LL and LC represent the 1000

averaged normalized left stance and swing phases relative to the right step cycle duration 1001

respectively. Limits of the dotted lines either side of the averaged position of each event 1002

represent the circular dispersion (equivalent of standard deviation in the circular 1003

mathematical model). B, C and D, evolution of individual left lift, left contact and right 1004

lift dispersion (step to step individual variability) respectively, averaged by group before 1005

and weekly throughout the recovery period. Data were recorded at 14 m.min−1 and 1006

48

expressed as circular mean ± circular dispersion in A1-7 and as standard arithmetic mean 1007

± SEM in B, C and D. 1008

Figure 10. Recovery of hindlimb EMG pattern during locomotion on treadmill. A, Raw 1009

EMG activity of right-left tibialis anterior (RTA and LTA respectively) and right-left 1010

gastrocnemius lateralis (RGL and LGL respectively) in typical normal rat walking on the 1011

treadmill at 14 m.min-1. B, similar representation than A in typical trained rat at the same 1012

velocity week 11 after SCI. C, similar representation than B in typical untrained rats at 1013

the same time point and same velocity. Data from each muscle were recorded 1014

transcutaneously. The graphs depict about 8 seconds of locomotion. Bar at the bottom of 1015

the top panel represents 1 second. 1016

Figure 11. Whole body weight, ankle flexor and extensor muscles weight at the end of 1017

experiment. A, averaged body weight (grams) in trained (black) and untrained (grey) 1018

groups. B, the average weight of the left tibialis anterior (TA) and gastrocnemius (G) 1019

muscles relative to the whole body weight week 11 after SCI is depicted in percent for 1020

trained (black) and untrained (grey) rats. Data are expressed as mean ± SEM. 1021

1022

Figure 1

ATrained U

ntrained

W 2W 3

W 4W 5

W 6W 8

W 10

Proportion of rats capable ofwalking with paw placement

at 14 m.min-1 (%)

++

+++

++

+++

++

+++

0 20 40 60 80

100

W 2W 3

W 4W 5

W 6W 8

W 10

B

+

++

+++

+++

+

++

++

20 m.min-1 (%)

0 20 40 60 80

100

W 2W 3

W 4W 5

W 6W 8

W 10

C

+

+++

+++

++

++

++

26 m.min-1 (%)

0 20 40 60 80

100

Figure 2

Baseline W 2 W 3 W 4 W 5 W 6 W 8

W 10

A

20406080

100120140160180

Trained Untrained

Step

cyc

le le

ngth

(mm

)

++++

+++

+++

+++

+++

+++

+

Trained Untrained

Distance relative to the hip (mm)-80-60-40-20020406080

BaselineWeek 2Week 3Week 4Week 5Week 6Week 8

Week 10

Hip

++++++++++++++++++

+

+

+++

++

C

0

5

10

15

20

25

30

Baseline W 2 W 3 W 4 W 5 W 6 W 8

W 10

CV

of s

tep

cycl

e le

ngth

(%)

+ +++

+ + + +++

B

LiftContact

0

2

4

6

8

10

12

14

Baseline W 2 W 3 W 4 W 5 W 6 W 8

W 10

Step

cyc

le h

eigh

t (m

m)

D

+++

+++

+++ ++

+++

+++ ++

+

Height

Figure 3

Ang

le (d

egre

es)

Baseline

Hip

110

40

Foot lift± SD

Kne

e

110

30

Ank

le

170

60

% of the step cycle0 20 40 60 80 100

MTP

260

140 Stance Swing

Trained Week 10

% of the step cycle0 20 40 60 80 100

110

40

110

30

170

60

260

140

Untrained Week 10

% of the step cycle0 20 40 60 80 100

110

40

110

30

170

60

260

140

Footcontact

Backward Forward

Stance Swing

A

B

E2 E3 F E1 E3 F E1 E3 FOff period

569.48 ms ± 13.71 404.38 ms ± 15.34 525.76 ms ± 109.31

Figure 4

Min-Max hip angle (degrees)

0 20 40 60 80

100

120

Trained U

ntrained

0 20 40 60 80100120140160180

A

0 20 40 60 80

100

120

Min-Max knee angle (degrees)

B

Min-Max ankle angle (degrees)

C

0 50

100

150

200

250

300

BaselineW 2

W 3W 4

W 5W 6

W 8W 10

BaselineW 2

W 3W 4

W 5W 6

W 8W 10

BaselineW 2

W 3W 4

W 5W 6

W 8W 10

BaselineW 2

W 3W 4

W 5W 6

W 8W 10

Min-Max MTP angle (degrees)

D

Max

(Extenstion)

Min

(Flexion)

Amplitude+

+

++

++++++++++++++++++++++

+

+

+++

+

+++++++++++

++++++++++++

+++++++++

++++++

++++++

++

+++++++++++++++++++++

Figure 5

Trained U

ntrained

A

80

100

120

140

160

180

200

Swing phase duration (ms)

BaselineW 2

W 3W 4

W 5W 6

W 8W 10

+++

Swing phase

BaselineW 2

W 3W 4

W 5W 6

W 8W 10

0 10 20 30 40 50 60Trained U

ntrained

++++

+ ++++++++

CV of the step cycle duration (%)

F

0

100

200

300

400

500

E

Off period duration (ms)

BaselineW 2

W 3W 4

W 5W 6

W 8W 10

Foot drag duration (ms)

0 20 40 60 80100120140160180200

D

BaselineW 2

W 3W 4

W 5W 6

W 8W 10 +++

+++++++++++++++

+

++++++

+++++++

Contact

Lift

BaselineW 2

W 3W 4

W 5W 6

W 8W 10

20 30 40 50 60 70 80 90100

F subphase duration in %of swing phase duration

B

+++

+++++

+++

+++

+++ F subphase

Foot dragO

ff period

BaselineW 2

W 3W 4

W 5W 6

W 8W 10

0 10 20 30 40 50 60 70 80

C

E1 subphase duration in %of swing phase duration

+++

++++++

++++++++

E1 subphase

Figure 6

Week 2 Week 6 Week 8

Swing phase

-60-40-2002040600

18

-600600

18

-600600

18

Trai

ned

Unt

rain

ed

-60-40-2002040600

18

B

-600600

18

C

-600600

18

D

A

Horizontal movement relative to the hip (mm)

-40-20020406080

Vert

ical

mov

emen

t (m

m)

02468

1012141618

Hip

Baseline

E F G

0

0.4

0.8

1.2

1.6

Week 2 Week 6 Week 8

Trai

ned

0 20 40 60 80 100 0 20 40 60 80 100

Normalized swing phase duration (%)

0 20 40 60 80 100

Unt

rain

ed

Baseline W 2 W 3 W 4 W 5 W 6 W 8

W 10

Ove

rall

swin

g ve

loci

ty (m

.s-1)

0

0.2

0.4

0.6

0.8

0.1

0.2

0.3

0.4

0.5

0.6

0.7

E1 s

ubph

ase

velo

city

(m.s

-1)

Baseline W 2 W 3 W 4 W 5 W 6 W 8

W 10

Trained Untrained

0

0.2

0.4

0.6

0.8

1

1.2

F su

bpha

se v

eloc

ity (m

.s-1)

Baseline W 2 W 3 W 4 W 5 W 6 W 8

W 10

0

0.4

0.8

1.2

1.6

0

0.4

0.8

1.2

1.6

0

0.4

0.8

1.2

1.6

0

0.4

0.8

1.2

1.6

0

0.4

0.8

1.2

1.61 2 3 1 2 3 1 2 3

1 2

3

1 2

3

1 2

3

Velo

city

(m.s

-1)

0

0.4

0.8

1.2

1.6

0 20 40 60 80 100

BaselineAccel. Deceleration Reb.

1 2 3

Normalized swing phaseduration (%)

A1

A2 A3 A4

A5 A6 A7

B C D

Figure 7

Swing phaseF subphase E1 subphase

+++

+++

+++

+++

+++

++++

+ +++ ++

+++

+++

+++

+++

+++

+++

Figure 8

A Baseline B Week 10 post-SCI

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0

20

4060

80100

z - S

win

g ve

loci

ty (m

.s-1)

x - S

win

g le

ngth

(mm

)

y - Swing duration (ms)50100150200250

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0

20

4060

80100

z - S

win

g ve

loci

ty (m

.s-1)

x - S

win

g le

ngth

(mm

)

y - Swing duration (ms)50100150200250

21.8

0.8

Swin

g he

ight

(mm

)

TrainedR2 = 0.784 variables correlated p < 0.05

UntrainedR2 = 0.783 variables correlated p < 0.05Swing height unrelated

R2 = 0.934 variables correlated p < 0.05

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

Figure 9

B

0

0.05

0.10

0.15

0.20

0

0.05

0.10

0.15

0.20

0

0.02

0.04

0.06

0.08

0.10

C D

Baseline W 2 W 3 W 4 W 5 W 6 W 8

W 10

Baseline W 2 W 3 W 4 W 5 W 6 W 8

W 10

Baseline W 2 W 3 W 4 W 5 W 6 W 8

W 10

Trained Untrained

Left

lift v

aria

bilit

y

Left

cont

act v

aria

bilit

y

Rig

ht li

ft va

riabi

lity

+++

+++

+++

+++

+++ ++

+++

+ +++

+++

++ ++ ++

+++

+++

+

Trai

ned

Unt

rain

ed

Week 2

0

0.25

0.5

0.75

0

0.25

0.5

0.75

Week 6

0

0.25

0.5

0.75

0

0.25

0.5

0.75

Week 8

0

0.25

0.5

0.75

0

0.25

0.5

0.75

Baseline

0

0.25

0.5

0.75

Tim

eRC

LL

LC

RL

A1

A2 A3 A4

A5 A6 A7

Left stanceRight stanceBilateral stance period

RC: Right contactLL: Left LiftLC: Left ContactRL: Right Lift

***

*** **

*

+

***

***

***

*** ***

***

Figure 10

A

B

RTA

RGL

LGL

Trained week 11

LTA

Untrained week 11

RTA

RGL

LGL

LTA

(1 s)

14 m.min-1

RTA

RGL

LGL

LTA

Baseline

C

Figure 11

Trained Untrained0

50

100

150

200

250

300

350

Bod

y w

eigh

t (g)

A

TA G0

0.1

0.2

0.3

0.4

0.5 Trained Untrained

Mus

cle

wei

ght r

elat

ive

to th

ebo

dy w

eigh

t (%

)Bns

ns

ns