This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

Author's personal copy

Research Report

Intraspinal cord graft of autologous activated Schwann cellsefficiently promotes axonal regeneration and functionalrecovery after rat's spinal cord injury

De-Xiang Bana,1, Xiao-Hong Kongb,1, Shi-Qing Fenga,⁎, Guang-Zhi Ninga,Jia-Tong Chenc, Shi-Fu Guoa

aDepartment of Orthopaedics, Tianjin Medical University General Hospital, Tianjin Heping District Anshan Road 154,Tianjin 300052, PR ChinabSchool of Medicine, Nankai University, Tianjin 300071, PR ChinacCollege of Life Science, NanKai University, Tianjin 300071, PR China

A R T I C L E I N F O A B S T R A C T

Article history:Accepted 26 November 2008Available online 10 December 2008

Basic research in spinal cord injury (SCI) has made great strides in recent years, and some newinsights and strategies have been applied in promoting effective axonal regrowth and sprouting.However, a relatively safe and efficient transplantation technique remains undetermined. Thisstudy, therefore, was aimed to address a question of how to graft Schwann cells to achieve thebest possible therapeutic effects. To clarify the issue, the ratswere subjected to spinal cord injuryat T10. Autologous activated Schwann cells (AASCs)were obtained byprior ligation of saphenousnerve and subsequently isolated and purified in vitro and then grafted into spinal cord-injuredrats via three different routes (group I: intravenous, group II: intrathecal and group III: intraspinalcord). Neurologic function was serially evaluated by Basso, Beattie, Bresnahan locomotor ratingscale and footprint analysis.We also evaluated themigration of the transplanted cells at 2weeksafter transplantation. Using biotinylated dextran amine (BDA) anterograde tracing, wedemonstrated that more regenerative axons of corticospinal tract (CST) surrounding theinjured cavity in group III than those in the other two groups, and we also confirmed it furtherby quantitative analysis. The microenvironment surrounding the injured spinal cord has beenimproved to the greatest extent in group III, as determined by immunohistological staining.Relatively complete myelin sheaths and more neurofilaments in axons were found in groups IIand III than those in group I under electronmicroscopy. The results showed that intraspinal cordinjection of AASCs promoted recovery of hindlimb locomotor function of injured rats moreefficiently than the other grafting routes. In addition, intact myelin sheaths and sufficientneurofilaments in axonswere not adequate for full functional recovery after SCI, suggesting thatreestablishment of normal synaptic connection is indispensable. The findings in this studystrongly suggest that transplantation of AASCs directly into the spinal cord may be one of thepromising candidates for potential scaffold for injured spinal cord, and such strategy oftransplantation of AASCs could be hopeful to treat patients with SCI.

© 2008 Elsevier B.V. All rights reserved.

Keywords:Spinal cord injuryAutologous activated Schwann cellFootprint analysisIntrathecalCorticospinal tract

B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1

⁎ Corresponding author.E-mail address: fengsq321@sina.com (S.-Q. Feng).

1 These authors have equal contribution to this research.

0006-8993/$ – see front matter © 2008 Elsevier B.V. All rights reserved.doi:10.1016/j.brainres.2008.11.098

ava i l ab l e a t www.sc i enced i rec t . com

www.e l sev i e r. com/ l oca te /b ra in res

Author's personal copy

1. Introduction

People suffering from SCI are at increased risk of developingparaplegia or quadriplegia due to the breakup of axons in theinjured spinal cord. In contrast to the peripheral nervoussystem (PNS), little structural and functional regeneration ofthe central nervous system (CNS) occurs spontaneouslyfollowing injury in adult mammals (Chopp et al., 1996; Li etal., 2006). Regeneration of injured adult neurons within theCNS is essentially abortive, attributable in part to lesion-induced inhibitors such as the chondroitin sulfate proteogly-cans (Iaci et al., 2007), themyelin inhibitors (Nogo-A,MAG, andOmgp), reactive axonal swelling, oxidative damage and scarformation etc (Cafferty et al., 2004). Post-traumatic syringo-myelia, cystic cavitation, extending rostrally or caudally fromthe area of original cord injury will occur in about 17–20% ofthe patients after SCI. This is a common cause of degeneratedmyelopathy (Squier and Lehr, 1994) and finally leads topermanent functional handicaps (Asher et al., 2000; Caroniand Schwab, 1988; Chen et al., 2000; McKeon et al., 1991;McKerracher et al., 1994; Snow et al., 1990).

The delivery of a suitable cell type to the specific target forthe repair after SCI has been intriguing over the past 5 years(Vroemen et al., 2003; Sheth et al., 2008). There are a variety ofcandidates for cell therapy in SCI. Olfactory ensheathing cells(OECs) could promote spinal cord synaptic plasticity (Barnettand Riddle, 2007). Grafted embryonic stem cells (ESCs) in thelesion cord could differentiate to astrocytes, oligodendrocytes,and neurons (McDonald et al., 1999; McDonald et al., 2004).Intravenously injected neural progenitor cells (NPCs) couldmigrate to the lesion site of rat's spinal cord and promotefunctional recovery (Fujiwara et al., 2004).

Schwann cells are the myelinating glial cells in the PNSand play a key role in Wallerian degeneration and subsequentregeneration. Central nervous system regeneration can beelicited by Schwann cell transplantation, which provides asuitable environment for regeneration (Dezawa, 2002).Grafted SCs in the injured spinal cord could reduce footexternal rotation of injured rats and enhanced forelimb–hindlimb coordination (Dinh et al., 2007). In response to PNSinjury, Schwann cells separate from the axons, proliferaterapidly, and align to create bands of Bunger (Lindberg et al.,1999). The activated Schwann cells synthesize and secretegrowth factors such as insulin-like growth factor-1 (IGF-1)(Hansson et al., 1986), ciliary neurotrophic factor (CNTF)(Rende et al., 1992), and brain-derived neurotrophic factor(BDNF) (Meyer et al., 1992), which guide and support theregenerating axons. In the absence of Schwann cells, thedistal stumpmay be unable to produce enough trophic factorsto reach the proximal stump (Eldridge et al., 1987). Intrathecaladministration of IGF-1 improved motor performance,delayed the onset of clinical disease, increased the expressionof phosphorylated Akt and ERK in spinal motor neurons, andpartially prevented motor neuron loss in mice sufferingamyotrophic lateralizing sclerosis (ALS) (Nagano et al., 2005).In addition to growth factors, Schwann cells also facilitateneuronal regrowth through the production of neurite promot-ing factors (NPFs) (Bolin et al., 1992), including extracellularmatrix (ECM) and cell adhesion molecule (CAM). Both ECMand CAM provide adequate “adhesion” for nerve growth,

promote the migration of SCs, and stimulate their divisiongrowth (Longo et al., 1984 Oudega et al., 2005; Zhou et al.,2003).

The optimal cell candidate for treating SCI should beobtained and collected easily in vitro and obtaining the cellsduring isolation procedure should be avoid of invasive to theSCI patient. Schwann cells were optimal option among thosecandidates for treatment of SCI. Their property of rich sourceand without immunological rejection after autologous trans-plantation made them applicable in numerous clinical fields.In this research, we compared three different grafting routesof autologous activated Schwann cells and finally demon-strated that the routes of intravenous and intrathecal cellgrafting were relatively safe in clinical application, but none ofthem was the most effective route.

2. Results

2.1. Animal observation

Immediately after SCI, all injured rats were paraplegic withincapability of hindlimb movement observed. Within oneweek after injury, one rat in group I and two in group III died ofcachexia caused by bedsore. Cell graft was performed at oneweek post injury. One week after cell grafting, the reaction ofall the rats was dull to the nociceptive stimulus. Their fur colorappeared light yellow and the edema of hindlimbs aggravatedin group I.While in group III, the fur color became lustrous andthe edema of hindlimbs attenuated gradually. After BDAinjection, two rats in group II were eliminated because ofepilepsy. The other rats were all alive until they weresacrificed.

2.2. The culture of AASCs

At 3 days post isolation of cells, AASCs crept out of the rimof the tissue lumps with radiant shape (Fig. 1A). Another5 days later, the cells proliferated and covered the wholebottom of the 6-well plates. The outline of the cells wasclear and bright and the cells arranged shoulder byshoulder. Some of the AASCs presented the whirlpoolshape. After purification with the method of differentialadhesion, the number of fibroblasts decreased evidently andthe morphous of AASCs appeared more unified (Figs. 1B, C).After five passages, AASCs expressed s-100 antigen stably,which showed that the cell cytoplasm was stained withgreen color. The nuclei of all the cells were stained withblue color by Hoechest33342 dye (Fig. 1D). The purity ofAASCs reached (97.34±1.45)% and their survival rate wasaround (95.66±2.56)% with trypan blue staining.

The doubling time of primary AASCs was around 2.07 d.The growth kinetics of AASCs showed that the cells grewquickly from the 1st d to the 8th d, the numbers of cellsincreased from about 1×105 to 5×105, and the growthkinetics reached the plateau up to the 10th d. The numberof cells maintained at the level of 5×105. This indicatedthat the AASCs grew slowly since the 8th day post isolation(Fig. 2).

150 B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1

Author's personal copy

2.3. Results of BBB locomotor score and footprint analysis

Preoperatively, the BBB scores of rats were 21 points in allthree groups. Immediately after SCI, all the rats demonstratedsignificant loss of motor function of hindlimbs and the BBB

score reduced to 0–1 points. Subsequently, BBB scoreincreased gradually in all groups and reached around 13points in average in group III until they were sacrificed. During3 weeks after injury, differences among three groups did notshow statistical significance (ANOVA, pN0.05). In contrast, onthe 4th week, the differences of BBB score among those threegroups were considered to be significant (ANOVA, p<0.05).After 3 months of injury, BBB score in group III (13.17±0.71,n=17) was evidently higher than that in group II (10.05±1.05,n=19) and group I (6.89±0.74, n=18) (Student's test, p<0.01)(Fig. 3).

Animals with a BBB score over 9 showed adequate weightsupported plantar stepping, which is crucial for footprintanalysis. As a result, only 15 among total 19 rats in group II and16 among total 17 rats in group III showed sufficient recoveryto perform footprint analyses from the 6th week after injury.The footprint patterns of uninjured rats showed a high degreeof coordination of forelimb and hindlimb foot placements anda normal angle of rotation. Karimi's study recently showedthat the interlimb coordination which was represented bycore-to-core distance and angle of rotation in normal rat arearound 1.49 cm and 8.4° respectively (Karimi-Abdolrezaee et

Fig. 1 – Culture and identification of AASCs. A: AASCs crept out of the rim of tissue lumps and showed radiant shape three daysafter isolation. B: The cells proliferated and reached the confluence of 90% after purification. C: An enlarged view from B wasused to show the paliform arrangement of AASCs with high power microscope. D: SCs expressed s-100 antigen and theircytoplasm was stained with green color after four passages. The amount of fibroblasts was rare (single stain) and the purity ofAASCs (double stain) was relatively high.

Fig. 2 – The growth kinetics of AASCs. The X axis indicateddifferent time points (day), The Y axis represented thenumber of cells with the scale of 1×105.

151B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1

Author's personal copy

al., 2006). In the current research, from the 5th week aftertransplantation (6 weeks after injury), the rats attendingfootprint analysis in group II and III consistently demon-strated the improvement in interlimb coordination and theangle of rotation (t-test; p<0.05) (Fig. 4) (Table 1).

2.4. Observation of cryostat sections

Two weeks after cell grafting, one rat in three groups wastaken out respectively to carry out 10 μm serial frozen section.It was shown that in group III more labeled cells aggregatedaround injured epicenter of the spinal cord than that in theother two groups (Fig. 5).

BDA positively stained corticospinal tracts (CST) should liein the dorsal funiculus of whitematter of spinal cord. No sparefibers were shown in the dorsal funiculus of rat's spinal cordthat received the impact of 10 g×50 mm weight drop (Fig. 6A).Axons were interrupted proximally to the injured cavity (Fig.6C). Some regenerative nerve fibers could present and past theinjured cavity in group III (Figs. 6B, D). While in group I, rege-nerative CST could rarely be found. The number of regene-rative CST in group III in cross section was obviously more

Fig. 3 – Assessment of motor function recovery by BBB score.The X axis indicated the different time points (days). The Yaxis represented BBB score and the scale was 2. All the datashown are derived from means in three groups. Asteriskindicated statistical difference among three groups (ANOVA,p<0.05).

Fig. 4 – The rat's footprints in group II and group III were shown in this picture. Interlimb coordination of rats was representedby core-to-core distance of the central pads of the ipsilateral paws. The angle of rotation of the injured rats was formed by line 1and line 2. A: the footprints in group II. B: the footprints in group III.

152 B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1

Author's personal copy

(from rostral (−1.8 mm) to caudal (+1.8 mm)) than that in theother two groups through statistical analysis (Student's test,p<0.05), which could be considered as an direct evidence toevaluate the effect of the cell grafting route (Fig. 7).

The observation of HE staining demonstrated that theinjured cavity was significantly smaller in group III than thatin group I. While in group I, injured cavity was large and anumber of gliacytes gathered to form scar (Figs. 8A–C). In theinjured spinal cord, the neurofilament protein was stainedwith NF200 dye and the microglial cells were stained with ED1dye. Three months after cell transplantation, quantitativeimaging analysis data acquired from the software of Image proplus 5.0 showed that the positive response area of NF200staining in group III was the largest among three groups

(ANOVA, p<0.01) (Figs. 8D–F) and that of ED1 staining was thesmallest (ANOVA, p<0.01) (Figs. 8G–I, 9).

2.5. Electron microscopic examination

In group I (Fig. 10A), the edema in the axoplasm was thecharacteristic of axonal change in the white matter. Neuro-filaments and vesicles in axons were sparse. Some of theaxonswith an empty appearancewere associatedwith axonaldisplacement. Some fibers lost their myelin cover andaxoplasmic organelles. Cristae usually disappeared in ede-matous mitochondria. The surrounding myelin sheath ofaxons was slender and some of them were cracked. Theneurons were apparently edematous with large vacuoles.

Table 1 – Comparison of footprint analysis between group II and group III

N Interlimb coordination(cm, mean±SD) Angle of rotation (degree, mean±SD)

6 w 8 w 10 w 12 w 6 w 8 w 10 w 12 w

Group II 15 2.89±0.12 2.63±0.22 2.36±0.24 2.28±0.19 26.07±1.28 23.4±3.36 21.2±1.15 19.53±2.9Group III 16 2.47±0.17 2.16±0.28 2.06±0.3 1.95±0.19 22.44±1.97 19.5±2.1 17.38±1.09 16.56±1.26t 7.93⁎ 5.21⁎ 3.01⁎ 4.75⁎ 6.05⁎ 3.91⁎ 9.53⁎ 3.74⁎

The interlimb coordination was represented by core-to-core distance. The difference between group II and group III was analyzed. p<0.05 wasconsidered significant. The asterisks meant p<0.05.

Fig. 5 – The localization of AASCs in three groups. A: AASCswere stainedwith Hoechst33342 in vitro. Their nuclei showed bluecolor and no apparent morphous change of the cells were found after 1 month. B: The localization of AASCs in group I. C: Thelocalization of AASCs in group II. D: The localization of AASCs in group III. The asterisk indicated injured epicenter.

153B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1

Author's personal copy

However, the results of the other two groups (Figs. 10B and C)showed very little changes likely or not any abnormalchanges at all, especially in group III. The density of

neurofilaments closed to the normal level (Fig. 10D) andmyelin laminae were regularly arranged in group III. Orga-nelles such as Golgi apparatus and rough endoplasmic

Fig. 6 – The results of BDA staining in coronal and cross sections. A: BDA staining in cross section in the injured epicenter in thetreatment-free group showed that no spare axons were found in dorsal funiculus of the injured spinal cord. B: BDA staining incross section in the injured epicenter in group III demonstrated that more regenerative axons presented in dorsal funiculus ofthe injured spinal cord than those in the other two experimental groups (not shown). C: BDA staining in coronal sectioncentered lesion site in the treatment-free group showed that no regenerative axons past the injury cavity. Axons wereinterrupted proximally to the injury cavity. D: BDA staining in coronal section centered lesion site in group III demonstrated thatmore regenerative axons past injury cavity compared with the other two experimental groups (not shown). The asteriskmeantthe injured epicenter and arrow represented the axons.

Fig. 7 – Comparison of the number of regenerative axons among three experimental groups. The X axis indicated the intervalbetween sections and the scale was 600 μm. The Y axis represented the percentage of total labeled CST axons. The asteriskmeant that the difference among three experimental groupswas statistically significant (ANOVA, p<0.05). Statistical differencein intergroups (group I vs group II, group I vs group III, and group II vs group III) was also confirmed by Student's test (p<0.05).

154 B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1

Author's personal copy

reticulum formation appeared in normal state and lessdegeneration in group III.

3. Discussion

Spinal cord injury (SCI) results in damage to ascending anddescending spinal pathways in addition to loss of interneu-rons, motoneurons and associated glial cells, all of which con-tribute to significant functional impairment (Armstrong et al.,2003; Jiang et al., 2008; Rossignol et al., 2007). A safe andeffective therapy for spinal cord injury is necessary for clinicalapplication (Chernykh et al., 2007; Jimenez Hamann et al.,2003). In the present research, we tested the three AASCsgrafting routes (intravenous, intrathecal, and intraspinal cord)and found that intraspinal cord injection of AASCs contributedmost efficiently towards axonal regeneration and functionalrecovery than the other two grafting routes. Injured spinal cordis capable of plasticity and its integrity can be reestablished,possibly through new intraspinal circuits (Bareyre et al., 2004).

Fig. 8 – The results of HE and immunohistochemistry staining. A: HE staining in group I. The injury cavity was the largest amongthree groups. B: HE staining in group II. C: HE staining in group III. D: NF200 staining in group I. E: NF200 staining in group II. F: NF200staining in group III. The positive response areawas the largest among three groups (ANOVA, p<0.01). G: ED1 staining in group I. H:ED1staining ingroupII. I: ED1staining ingroup III.Thepositive responseareawas thesmallestamongthreegroups (ANOVA,p<0.01).The asterisks indicated injury cavity where the glial scar was clearly seen. Arrow indicated positive staining in NF200 and ED1staining.

Fig. 9 – The comparison of positive response area of NF200and ED1 staining among three groups. The Y axisrepresented the positive response area (pixel unit). In ED1and NF200 staining, the difference among three groups wasconsidered significant (ANOVA, p<0.01). Statistical differencein intergroups (group I vs group II, group I vs group III, andgroup II vs group III) was also confirmed by Student's test(p<0.05).

155B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1

Author's personal copy

The cells grafted intravenously needs pass through thegeneral circulation. Most of the grafted cells will be removedby the immune system. It has shown that blood-spinal cordbarrier (BSCB) (Maikos and Shreiber, 2007) may stay open upto a week after SCI, which provides an opportunity for thegrafted cells to localize in the injured epicenter, althoughdistributed sparsely. The number of cells localizated in theinjured epicenter through intravenous grafting depends onthe amount of hemorrhage in the lesion site and the degree ofdamage (Khalatbary and Tiraihi, 2007). When we grafted cellsintrathecally, cerebrospinal fluid was expected to work ascarrier to finish cell localization around the lesion sitethrough circulation. Intrathecally grafted marrow stromalcells could be found in the spinal cord 4 weeks aftertransplantation and the capillary density in the ventral graymatter was significantly increased (Shi et al., 2007). In thisstudy, grafted cells were labeled with Hoechst33342 fortracing. It has been reported that Hoechst33342 can be takenup by living cells via membrane permeability and is nontoxic.This dye binds both specifically and quantitatively to DNAand, once bound, becomes strongly fluorescent (Lalande et al.,1981). Seen from the result, we found that cells accumulatedmainly around the injured area but less than that fromintraspinal cord grafting. The phagocytes in the cerebrospinalfluid and the edema around the injured epicenter might limitthe intrathecally-grafted cells in reaching their destination.

However, it was possible that the grafted cells would beuptaken in the spinal cord by host cells such as macrophageswhen grafted cells died. In the current research, we could notexclude this possibility that the cells observed in the cryostatsection were AASCs or host macrophages. However, thiseffect should be similar among three different groups.Therefore, in this research, the results displayed that thegrafted cells from different routes were indeed implantedaround the injured epicenter. Moreover, their repair effect onSCI shown in the “BBB score and footprint analysis” part inthis paper also provided a strong confirmation. Meanwhile,we are trying to elucidate this question in more detail in theongoing research.

BBB score and footprint analysis were adopted to evaluatethe recovery of hindlimbs of the injured rats in this study. BBBscore is an effective nerve functional assessment andproposed by Basso et al. (1995) in 1995. The footprint analysiswas modified from the method of de Medinaceli et al. (1982)and mainly used to assess the interlimb coordination of therats. At the first priority level, the assessment of BBB scoreenables the subdivision into groups of low and high locomotorability. The BBB score is reliable and most independent frommotivational aspects. It is easy and fast to perform and thus itcan serve as a ‘baseline’ before further tests are done (Bassoet al., 1996). However, because of partial overlap withmeasurements in BBB score, footprint analysis can serve to

Fig. 10 – Electron microscope examination in three experimental groups. A: Observation of injured epicenter in group I. Thethickness of axons was abnormal and some of the axons lost the cover of myelin sheath. B: Observation of injured epicenter ingroup II. C: Observation of injured epicenter in group III. The integrity of myelin sheathwas better in B and C than that in A. Thenumber of neurofilaments in axons in those two groups closed to normal. D: Observation of normal spinal cord at T10 in healthyrat. Arrow indicated the myelin sheath of axons. The asterisk represented the axoplasm, in which the neurofilaments andvesicle could be found in groups II and III, although sparsely in group I.

156 B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1

Author's personal copy

refine observations made in BBB tests for weight support,trunk stability, and foot placement. It has been proven thatfootprint analysis showed the highest correlation to the BBBscore (Metz et al., 2000). We also demonstrated that BBB scorein group III finally reached around 13 points, which meant therats had relatively fine interlimb coordination but no flexiblemovement. Footprint analysis supported the BBB score resultsrevealing that intraspinal cord cell grafting exertedmore effecton improving the interlimb coordination and reducing thehindlimb angle of rotation from 5th week after cell transplan-tation. This indicated that the route of intraspinal cord cellinjection promoted functional recovery of injured rats to thelargest extent when compared with the routes of intravenousand intrathecal grafting.

Observations using electronmicroscope showed, in group I,that some of the axons in the injured epicenter had relativelycomplete myelin sheath covers with fewer neurofilaments.While in groups II and III, more neurofilaments could beobserved in the axoplasm. Their presence approaches that ofthehealthy rat although theBBB score only reachedaround11–13. This phenomenon showed that intact myelin sheaths andsufficient neurofilaments in axons were insufficient for thefunctional recovery of rats after SCI, suggesting that reestab-lishment of normal synaptic connection is indispensable.

Activated Schwann cell is one of the optimal cell candi-dates for the repair after SCI. Cultures from adult mouseperipheral nerves in defined medium showed markedlyhigher purity and density of SCs when the nerve waspredegenerated in vivo for 7 days than when it was harvestedfreshly (Verdú et al., 2000). Many divisions of cells may inducethe formation of malignant cells, which however has not beenobserved following multiple divisions of SCs in vitro (Oudegaet al., 2005). The concentration of 2.5% FBS was suitable forboth SCs proliferation and fibroblasts suppression (Hedayat-pour et al., 2007). However, in this study, we did not get thesame result. Three days after cell culture in the mediumcontaining 2.5% FBS, lots of cell debris was found and the cellsurvival rate was less than 20% using trypan blue staining. Inthis research, we believed that 10% FBS promoted AASCs toproliferate rapidly and high-purified AASCs could be availablewith the method of differential adhesion. It has also beenproven that the purity of AASCs could reach over 95%with themethod of differential adhesion (Kreider et al., 1981).

Numerous neurotrophic factors (NTFs) secreted by AASCscan improve the microenvironment in the lesion site andpromote functional recovery of spinal cord injured rats (Fu andGordon, 1997; Feng et al., 2005, 2008). It has shown that brain-derived neurotrophic factor (BDNF) could enhance the growthability of axons and overcome MAG reaction by inhibitingphosphodiesterase activity in neurons and raising the level ofintracellular cAMP via the ErK-depended pathway (Behrens etal., 1999; Vavrek et al., 2006). Glial cell line-derived neuro-trophic factor (GDNF) guides SCs to migrate to the injuredepicenter and promote neoformative axons forming themyelin sheath (Korsching, 1993). SCs transfected with NGFgene survive from two weeks to one year in vivo, stablyexpressing NGFs, and promotes axonal regeneration com-bined with myelin sheath reestablishment (Tuszynski et al.,1998). Grafted SCs spontaneously aligned into regular spatialarrays within the cord, appropriately remyelinated axons that

regenerated into grafts (Weidner et al., 1999). In this research,the result of either NF200 or ED1 staining showed that graftedAASCs promoted axonal regeneration, reduced the number ofmicrogliacytes, and inhibited the scar formation to someextent. Intraspinal cord grafted AASCs improved the micro-environment in the injured epicenter better than the othergrafting routes.

In this study, AASCs were obtained by prior ligation ofsaphenous nerve of the rats with SCI and the operation wasminimal invasive to the rats. In addition, we also demon-strated that the 10g×50 mm weight drop with NUY impactormade all the CST in the dorsal funiculus of white matter ofspinal cord interrupted proximally to the injured cavity,indicating that the regeneration of axons in the injury cavitywere promoted by grafted cells. The grafted cells weresufficient for compensating the injured area through theroute of intraspinal cord cell grafting. In contrast, the cell graftvia route of intravenous or intrathecal did not show suchproperties as promoting regeneration of axons in injured areaand improvement of microenvironment in the lesion site,while those two routes of administration were relatively safefor the patients and easily to be operated.

4. Experimental procedures

4.1. Animal care

Sixty-six adult Wistar rats (female, 250 g±10 g, provided byRadiation Study Institute-Animal Center, Tianjin, China) wereused in this study.

4.2. Isolation and culture of autologous activated Schwanncells from rat

A Wistar rat was anaesthetized by intraperitoneal injectionwith 10% chloral hydrate (0.3 ml/100 g). Bilateral saphenousnerve of the rat was ligated proximally for oneweek to activateSCs. Then the nerve close to the ligation spotwas cut to releasethe major part of the nerve. After the epineurium of the nervetissue was peeled off, the remaining nerve tissue was cut intosmall pieces (about 0.5–1 mm3). The tissue lumps weredigested with 0.3% trypsin (provided by the laboratory inCollege of Life Sciences, Nankai University, Tianjin, China) andthen attached on the bottom of the 6-well culture platesprecoated with polylysine. SCs were propagated in DMEMsupplemented with 20% fetal bovine serum (FBS), 1 μg EGF/ml(Sigma Chemical Co, USA) and 1 μg bFGF/ml (Sigma ChemicalCo, USA). Then the cells were incubated at 37 °C, 5% CO2

incubator. Following an incubation of 3 days, AASCs crept outof the edge of tissue lumps. Following other incubation of5 days, AASCs reached confluence of 90%. The cells weredigested with 0.3% trypsin in the 37 °C incubator for 3 min andcompleteDMEMmediumwas added to stop the trypsinization.The suspension of cells was centrifuged at 800 rpm for 3 min.The supernatantwasdiscarded. 5mlDMEMwasadded into thecentrifuge tube to suspend the cells which were thenimplanted into a 25 ml culture flask. Two weeks later, thecells reach confluence of 90%. Then, AASCs were purified(reducing 20% FBS to 10%) by means of differential adhesion.

157B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1

Author's personal copy

After five passages, the cells were used in this experiment.Meanwhile, the viability of AASCswas assessed by trypan bluestaining.

4.3. Identification of autologous activated Schwann cells

AASCs were passaged and cultured on the cover slip for 48 hand washed three times with 0.01 M PBS at 37 °C and thenpostfixed with 4% paraformaldehyde (pH=7.4) for 4 h at roomtemperature. Then, the cell cover slip was treated with 0.1%TritonX-100 for 15 min at room temperature and blocked with5% goat serum albumin for 10 min to reduce non-specificstaining. Primary antibody (rabbit anti-rat S-100 IgG 1:50,sigma, USA) was added on the slides which was thenincubated at 37 °C for 30 min. Secondary antibody (goat anti-rabbit IgG/FITC 1:100, Sigma, USA) was the last step to finishthe staining. The slides were stored at 4 °C for laterobservation with fluorescence microscope.

4.4. The purity of AASCs and the cell growth kinetics

The number of S-100 positive cells and total cells were countedunder fluorescentmicroscope. AASC purity (%)=S-100 positivecells/ total cells×100%.

After five passages in vitro, AASCs were collected atdifferent time points of 1, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22,24, 26, and 28 days for growth kinetics. All the results werecalculated in triplicate as the following: the AASCs derivedfrom three wells were mixed, centrifuged, and resuspended tothe total volume of 1ml. The number of cells in the suspensionwas counted under the microscope with the aid of hemocyt-ometer. The medium in the other wells was replaced withfresh medium every three days until harvesting of cells.

4.5. Preparation for spinal cord injury model

Sixty-five healthy Wistar rats (female, 250±10 g) wereanesthetized by intraperitoneal injection with 10% chloralhydrate (0.3 ml/100 g). The surgical area was shaved anddisinfected with 70% ethanol and betadine. The spinousprocess and vertebral platewere removed usingmicro rongeurat T10 and the dura was opened. Then, the rat was placed onthe NYU impactor machine and the exposed spinal cord washit using 10 g×50 mm weight drop. It showed that twohindlimbs of the rat twitched involuntarily and tail waggled,which meant the rats were in accordance with the criteria ofSCI model. Sixty rats in total were randomly and evenlyassigned to three experimental groups. Another five rats wereset as treatment-free control which was not included in theexperiment groups in order to clarify whether completeinterruption of CST in the dorsal funiculus of white matterof spinal cord could be induced or not. After surgery, rats wereplaced in temperature and humidity controlled incubationchambers until they awoke. They were then transferred to thecages, and bladder evacuationwas applied daily until return ofreflexive bladder control.

4.6. Cell grafting via different routes

One week after injury, the rats were anesthetized again.AASCs labeled with Hochest33342 (1 μg/ml, Sigma, USA)

previously for 1 min were transplanted into the injured ratsby one of the three routes administered differently.

Route 1: intravenous (tail vein) AASCs graft (group I): Therat's tail was cleansed with 75% ethanol. 15 μl AASCssuspension with cell density of 2×105/ml was grafted fromone of the tail veins using microsyringe.Route 2: intrathecal AASCs graft (group II): 15 μl AASCssuspension with cell density of 2× 105/ml was grafted intosubarachnoid cavity usingmicrosyringe between the spaceof lumber 1–lumber 2 (L1–L2).Route 3: intraspinal cordAASCs graft (group III): 15 μl AASCssuspension with cell density of 2× 105/ml was injected intothe injured epicenter of spinal cord at 3 points (5 μl at eachpoint) with a needle angle of 45° rostral and caudal to theinjury site and of 90° at the center of the injury site.

All the AASCs suspension contained 0.5 mg cyclosporine A.Antibiotics (sodium ampicillin, 80 mg/kg body weight) wereinjected daily into the rats for a week after each graft.

4.7. Basso, Beattie, and Bresnahan open-field locomotorscale

The BBB is an open field locomotor test scored on a 21-pointscale, with 21 being normal and 0 indicating no movement ofthe hindlimbs. The rats were separately put in an open areawith a nonslippery surface. The BBB test was performed bytwo observers who were blinded to the animals' identity. Thetest period should not be less than 4 min and the rat thatstopped moving for more than 1 min was placed again in thecenter of the open field for retesting. BBB test started from the3rd day after injury and at weekly intervals until all the ratswere sacrificed.

4.8. Footprint analysis

The forelimbs and hindlimbs of the rats were labeled withgreen and red nontoxic dyes, respectively. Then the animalswere trained to walk on a paper-covered narrow runway (1 mlength and 7 cmwidth) in order to ensure that the rats walkedalong a straight path. A very bright box was put at thebeginning of the runway to prevent the rats from pausingwhile passing the track. A dark box with their food was put atthe end of the runway to encourage them to finish the test asfast as possible. The first and the last 15 cm of the print wereexcluded to avoid analyzing the beginning and end of thelimb movements. In order to get the consecutive footprint,the test was repeated if the rats stopped when they werepassing the track. Rats that dragged their hindlimbs andlacked adequate weight support were excluded from theexperiments. The angle of rotation was defined by the angleformed by the intersection of the line through the print of thethird digit and the print representing the metatarsophalan-geal joint and the line through the central pad parallel to thewalking direction. For interlimb coordination, the distancewas determined by measuring the core-to-core distance ofthe central pads of the ipsilateral paws. For the analyses, sixsteps or more from each side were measured for every animalin each group.

158 B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1

Author's personal copy

4.9. BDA anterograde tracing

Three months after injury, the rats were anesthetized againand placed on a stereotaxic frame. Two pieces of boneapproximately 6 mm×8 mm on each side of sutura sagittaliswere removed by using a mini-smooth-jaw electrosaw toreveal the dura matter. 0.5 μl of 10% biotinylated dextranamine (BDA) (Sigma Chemical Co, USA) solution was injectedslowly into the bilateral sensorimotor area cortex, first at0.5 mm depth and then at 1.0 mm depth, at 5 differentpositions on each side of the sutura sagittalis.

4.10. Histological analysis

Rats were sacrificed with an overdose of intraperitonealinjectionwith 10% chloral hydrate and perfused transcardiallywith 4% neutral paraformaldehyde (PFA) in 0.01 M PBS,pH=7.4. The spinal cords were subsequently postfixed in theperfusing solution plus 10% sucrose overnight at 4 °C. Then,the tissues were cryoprotected in 20% sucrose in PBS for 24–48 h at 4 °C. A 1.5 cm length of the spinal cord centered at theinjury site was separated and embedded in tissue embeddingmedium (HistoPrep; Fisher Scientific, Pittsburgh, PA) on dryice. Cryostat sections (10 μm) were cut and mounted ontogelatin-subbed slides and stored at −70 °C. For immunostain-ing, the frozen slides were air dried at room temperature for10 min and washed with PBS for 10 min. In this study, wepicked up one of the 15th–20th slices including the injured areafrom dorsal to ventro side of the spinal cord to reduce theinfluence of error.

For BDA developing, three coronal and three cross cryostatsections were selected from each group and rinsed twice with0.01 M PBS at room temperature. Avidin labeled FITC (1:50,Sigma Chemical Co, USA) was dripped on the frozen sectionwhich was then protected from light at room temperature for2 h. The total number of axons labeled in the CST in the T5cross-sections was counted and the number of labeled axonsin each cross section was counted at 10 points with a 600 μminterval between points (3 points caudal to, 1 point central to,and 6 points rostra to, the injury site). The percentage of nervefibers at each point was calculated by dividing the number ofnerve fibers at each level in cross sections by the total numberof labeled axons at the level of T5. The percentage of labeledaxons at different levels was calculated and the comparisonwas carried out among all three groups.

Two coronal sections in each group were used for HEstaining. Five cross sections in each group were stained forneurofilament (NF200, 1:500, Sigma Chemical Co, USA) andED1 (1:500, Sigma Chemical Co, USA) immunoractivitystaining respectively. Sections were rinsed with PBS forthree times and endogenous peroxidase was blocked withhydrogen peroxide in methanol for 1 h, and then treatedwith 10 mM sodium citrate buffer (pH 6.0) for 25 min at10 °C overnight. All sections were processed for immuno-histochemistry staining following the avidin–biotin perox-idase method (ABC elite kit, Vectastain, Vector, Burlingame,CA, USA) and lightly counterstained with haematoxylin.Positive response area of NF200 and ED1 staining (pixel unit)were analyzed among groups by using the software ofImage pro plus 5.0.

4.11. Electron microscope observation

Two millimeters of cross-sectional spinal cord lesion wereremoved for the ultrastructural examination. Tissue sampleswere prepared for electron microscopy as follows: the spinalcord segmentswere removed and fixed in 2.5% glutaraldehydefor 24 h. Then the tissue was post-fixed with OsO4 dehydratedin a graded alcohol series, and then embedded. Sections of 60–90 nm were stained with uranyl acetate and lead citrate,examined on an electron microscope (JEM 1200EX; Jeol, Tokyo,Japan) and photographed. The investigator who evaluated thesections was blinded to the group information.

4.12. Statistical analysis

All data were presented as mean±standard deviation (SD).The means of BBB score in three groups, regenerated CST,positive response areas (pixel unit) and footprint analysiswere analyzed by one-way ANOVA plus psot hoc Student's testand independent sample t test. p<0.05 was consideredsignificant.

Acknowledgments

This work was sponsored by National Natural ScienceFoundation of China (30872603), New Century ExcellentTalents Programme of Ministry of Education of China (NCET-06-0251), Applied Basic Research Programs of Science andTechnology Commission Foundation of Tianjin, China(07JCYBJC10200), and Science Foundation of Tianjin MedicalUniversity, China (2007ky28).

R E F E R E N C E S

Armstrong, J., Zhang, L., McClellan, A.D., 2003. Axonalregeneration of descending and ascending spinal projectionneurons in spinal cord-transected larval lamprey. Exp. Neurol.180, 156–166 [PubMed: 12684029].

Asher, R.A., Morgenstern, D.A., Fidler, P.S., Adcock, K.H., Oohira, A.,Braistead, J.E., Levine, J.M., Margolis, R.U., Rogers, J.H., Fawcett,J.W., 2000. Neurocan is upregulated in injured brain and incytokine-treated astrocytes. J. Neurosci. 20, 2427–2438[PubMed: 10729323].

Bareyre, F.M., Kerschensteiner, M., Raineteau, O., Mettenleiter,T.C., Weinmann, O., Schwab, M.E., 2004. The injured spinalcord spontaneously forms a new intraspinal circuit in adultrats. Nat. Neurosci. 7, 269–277 [PubMed: 14966523].

Barnett, S.C., Riddle, J.S., 2007. Olfactory ensheathing celltransplantation as a strategy for spinal cord repair—whatcan it achieve? Nat. Clin. Pract. Neurol. 3, 152–161[PubMed - 17342191].

Basso, D.M., Beattie, M.S., Bresnahan, J.C., 1995. A sensitive andreliable locomotor rating scale for open field testing in ratsJ. Neurotrauma 12, 112 [PubMed: 7783230].

Basso, D.M., Beattie, M.S., Bresnahan, J.C., 1996. Gradedhistological and locomotor outcomes after spinal cordcontusion using the NYU weightdrop device versustransaction. Exp. Neurol. 139, 244–256 [PubMed: 8654527].

Behrens, M.M., Strasser, U., Lobner, D., Dugan, L.L., 1999.Neurotrophin-mediated potentiation of neuronal injury.Microsc. Res. Tech. 45, 276–284 [PubMed: 10383120].

159B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1

Author's personal copy

Bolin, L.M., Iismaa, T.P., Shooter, E.M., 1992. Isolation of activatedadult Schwann cells and a spontaneously immortal Schwanncell clone. J. Neurosci. Res. 33, 231–238 [PubMed: 1280693].

Cafferty, W.B., Gardiner, N.J., Das, P., Qiu, J., McMahon, S.B.,Thompson, S.W., 2004. Conditioning injury-induced spinalaxon regeneration fails in interleukin-6 knock-out mice. J.Neurosci. 24, 4432–4443 [PubMed: 15128857].

Caroni, P., Schwab, M.E., 1988. Two membrane protein fractionsfrom rat central myelin with inhibitory properties for neuritegrowth and fibroblast spreading. J. Cell Biol. 106, 1281–1288[PubMed: 3360853].

Chen, M.S., Huber, A.B., van der Haar, M.E., Frank, M., Schnell, L.,Spillmann, A.A., Christ, F., Schwab, M.E., 2000. Nogo-A is amyelin-associated neurite outgrowth inhibitor and an antigenfor monoclonal antibody IN-1. Nature 403, 434–438 [PubMed:10667796].

Chernykh, E.R., Stupak, V.V., Muradov, G.M., Sizikov, M.Y.,Shevela, E.Y., Leplina, O.Y., Tikhonova, M.A., Kulagin, A.D.,Lisukov, I.A., Ostanin, A.A., Kozlov, V.A., 2007. Application ofautologous bone marrow stem cells in the therapy of spinalcord injury patients. Bull. Exp. Biol. Med. 143, 543–547 [PubMed:18214319].

Chopp, M., Chan, P.H., Hsu, C.Y., Cheung, M.E., Jacobs, T.P., 1996.DNA damage and repair in central nervous system injury:National Institute of Neurological Disorders and StrokeWorkshop Summary. Stroke 27, 363–369 [PubMed: 8610296].

de Medinaceli, L., Freed, W.J., Wyatt, R.J., 1982. An index of thefunctional condition of rat sciatic nerve based onmeasurements made from walking tracks. Exp. Neurol. 77,634–643 [PubMed: 7117467].

Dezawa, M., 2002. Central and peripheral nerve regeneration bytransplantation of Schwann cells and transdifferentiated bonemarrow stromal cells. Anat. Sci. Int. 77, 12–25 [PubMed: 12418080].

Dinh, P., Bhatia, N., Rasouli, A., Suryadevara, S., Cahill, K., Gupta,R., 2007. Transplantation of preconditioned Schwann cellsfollowing hemisection spinal cord injury. Spine 32, 943–949[PubMed: 17450067].

Eldridge, C.F., Bunge, M.B., Bunge, R.P., Wood, P.M., 1987.Differentiation of axon-related Schwann cells in vitro. I.Ascorbic acid regulates basal lamina assembly and myelinformation. J. Cell Biol. 105, 1023–1034 [PubMed: 3624305].

Feng, S.Q., Kong, X.H., Guo, S.F.,Wang, P., Li, L., Zhong, J.H., Zhou, X.F.,2005. Treatment of spinal cord injury with co-grafts of geneticallymodified Schwann cells and fetal spinal cord cell suspension inthe rat. Neurotox Res. 7, 169–177 [PubMed: 15639807].

Feng, S.Q., Zhou, X.F., Rush, R.A., Ferguson, I.A., 2008. Graft ofpre-injured sural nerve promotes regeneration of corticospinaltract and functional recovery in rats with chronic spinal cordinjury. Brain Res. 1209, 40–48 [PubMed: 18405884].

Fujiwara, Y., Tanaka, N., Ishida, O., Fujimoto, Y., Murakami, T.,Kajihara, H., Yasunaga, Y., Ochi, M., 2004. Intravenouslyinjected neural progenitor cells of transgenic rats can migrateto the injured spinal cord and differentiate into neurons,astrocytes and oligodendrocytes. Neurosci. Lett. 366, 287–291[PubMed: 15288436].

Fu, S.Y., Gordon, T., 1997. The cellular and molecular basis ofperipheral nerve regeneration. Mol. Neurobiol. 14, 67–116[PubMed: 9170101].

Hansson, H.A., Dahlin, L.B., Danielsen, N., Fryklund, L.,Nachemson, A.K., Polleryd, P., Rozell, B., Skottner, A., Stemme,S., Lundborg, G., 1986. Evidence indicating trophic importanceof IGF-I in regenerating peripheral nerves. Acta. Physiol. Scand.126, 609–614 PMID: [PubMed: 3521205].

Hedayatpour, A., Sobhani, A., Bayati, V., Abdolvahhabi, M.A.,Shokrgozar, M.A., Barbarestani, M., 2007. A method forisolation and cultivation of adult Schwann cells for nerveconduit. J. Arch. Iranian Med. 10, 474–480 [PubMed: 17903052].

Iaci, J.F., Vecchione, A.M., Zimber, M.P., Caggiano, A.O., 2007.Chondroitin sulfate proteoglycans in spinal cord contusion

injury and the effects of chondroitinase treatment.J. Neurotrauma 24, 1743–1759 [PubMed: 18001203].

Jiang, S., Ballerini, P., Buccella, S., Giuliani, P., Jiang, C., Huang,X., Rathbone, M.P., 2008. Remyelination after chronic spinalcord injury is associated with proliferation of endogenousadult progenitor cells after systemic administration ofguanosine. Purinergic. Signal. 4, 61–71. Epub 2008 Jan 8[PubMed: 18368534].

Jimenez Hamann, M.C., Tsai, E.C., Tator, C.H., Shoichet, M.S., 2003.Novel intrathecal delivery system for treatment of spinal cordinjury. Exp. Neurol. 182, 300–309 [PubMed: 12895441].

Karimi-Abdolrezaee, S., Eftekharpour, E., Wang, J., Morshead, C.M.,Fehlings, M.G., 2006. Delayed transplantation of adult neuralprecursor cells promotes remyelination and functionalneurological recovery after spinal cord injury. J. Neurosci. 26,3377–3389 [PubMed: 16571744].

Khalatbary, A.R., Tiraihi, T., 2007. Localization of bone marrowstromal cells in injured spinal cord treated by intravenousroute depends on the hemorrhagic lesions in traumatizedspinal tissues. Neurol. Res. 29, 21–26 [PubMed: 17427270].

Korsching, S., 1993. The neurotrophic factor concept: areexamination. Neuroscience 13, 2739–2748 [PubMed: 8331370].

Kreider, B.Q., Messing, A., Doan, H., Kim, S.U., Lisak, R.P., Pleasure,D.E., 1981. Enrichment of Schwann cell cultures from neonatalrat sciatic nerve by differential adhesion. Brain Res. 207,433–444 [PubMed: 7008901].

Lalande,M.E., Ling,V.,Miller, R.G., 1981.Hoechst 33342dyeuptakeasa probe ofmembrane permeability changes inmammalian cells.Proc. Natl. Acad. Sci. U. S. A. 78, 363–367 [PubMed: 6941251].

Li, M., Liu, J., Song, J., 2006. Nogo goes in the pure water: solutionstructure of Nogo-60 and design of the structured andbuffer-soluble Nogo-54 for enhancing CNS regenerationProtein Sci. 15, 1835–1841 [PubMed: 16877707].

Lindberg, R.L., Martini, R., Baumgartner, M., Erne, B., Borg, J.,Zielasek, J., Ricker, K., Steck, A., Toyka, K.V., Meyer, U.A., 1999.Motor neuropathy in porphobilinogen deaminase-deficientmice imitates the peripheral neuropathy of human acuteporphyria. J. Clin. Invest. 103, 1127–1134 PMID: [PubMed:10207164].

Longo, F.M., Hayman, E.G., Davis, G.E., Ruoslahti, E., Engvall, E.,Manthorpe, M., Varon, S., 1984. Neurite-promoting factors andextracellular matrix components accumulating in vivo withinnerve regeneration chambers. Brain Res. 309, 105–117[PubMed: 6488001].

Maikos, J.T., Shreiber, D.I., 2007. Immediate damage to theblood-spinal cord barrier due to mechanical trauma.J. Neurotrauma 24, 492–507 [PubMed: 17402855].

McDonald, J.W., Liu, X.Z., Qu, Y., Liu, S., Mickey, S.K., Turetsky,D., Gottlieb, D.I., Choi, D.W., 1999. Transplanted embryonicstem cells survive, differentiate and promote recovery ininjured rat spinal cord. Nat. Med. 5, 1410–1412 [PubMed:10581084].

McDonald, J.W., Becker, D., Holekamp, T.F., Howard, M., Liu, S., Lu,A., Lu, J., Platik, M.M., Qu, Y., Stewart, T., Vadivelu, S., 2004.Repair of the injured spinal cord and the potential ofembryonic stem cell transplantation. J. Neurotrauma 21,383–393 [PubMed; 15115588].

McKeon, R.J., Schreiber, R.C., Rudge, J.S., Silver, J., 1991. Reductionof neurite outgrowth in a model of glial scarring followingCNS injury is correlated with the expression of inhibitorymolecules on reactive astrocytes. J. Neurosci. 11, 3398–3411[PubMed: 1719160].

McKerracher, L., David, S., Jackson, D.L., Kottis, V., Dunn, R.J.,Braun, P.E., 1994. Identification of myelin-associatedglycoprotein as a major myelin-derived inhibitor of neuriteoutgrowth. Neuron 13, 805–811 [PubMed: 7524558].

Metz, G.A., Merkler, D., Dietz, V., Schwab, M.E., Fouad, K., 2000.Efficient testing of motor function in spinal cord injured rats.Brain Res. 883, 165–177 [PubMed - 11074045].

160 B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1

Author's personal copy

Meyer, M., Matsuoka, I., Wetmore, C., Olson, L., Thoenen, H., 1992.Enhanced synthesis of brain-derived neurotrophic factor in thelesioned peripheral nerve: different mechanisms areresponsible for the regulation of BDNF200 and NGF mRNA.J. Cell Biol. 119, 45–54 [PubMed: 1527172].

Nagano, I., Ilieva, H., Shiote, M., Murakami, T., Yokoyama, M.,Shoji, M., Abe, K., 2005. Therapeutic benefit of intrathecalinjection of insulin-like growth factor-1 in a mouse model ofAmyotrophic Lateral Sclerosis. J. Neurol. Sci. 235, 61–68[PubMed: 15990113].

Oudega, M., Moon, L.D.F., de Almeida Leme, R.J., 2005. Schwanncells for spinal cord repair. Braz. J. Med. Biol. Res. 38, 825–835[PubMed: 15933775].

Rende, M., Muir, D., Ruoslahti, E., Hagg, T., Varon, S., Manthorpe,M., 1992. Immunolocalization of ciliary neuronotrophic factorin adult rat sciatic nerve. Glia 5, 25–32 [PubMed: 1531807].

Rossignol, S., Schwab, M., Schwartz, M., Fehlings, M.G., 2007.Spinal cord injury: time to move? J. Neurosci. 27, 11782–11792[PubMed: 17978014].

Sheth, R.N., Manzano, G., Li, X., Levi, A.D., 2008. Transplantation ofhuman bone marrow-derived stromal cells into the contusedspinal cord of nude rats. J. Neurosurg. Spine 8, 153–162[PubMed: 18248287].

Shi, E., Kazui, T., Jiang, X., Washiyama, N., Yamashita, K., Terada,H., Bashar, A.H., 2007. Therapeutic benefit of intrathecalinjection of marrow stromal cells on ischemia-injured spinalcord. Ann. Thorac. Surg. 83, 1484–1490 [PubMed: 17383362].

Snow, D.M., Lemmon, V., Carrino, D.A., Caplan, A.I., Silver, J., 1990.Sulfated proteoglycans in astroglial barriers inhibit neurite

outgrowth in vitro. Exp. Neurol. 109, 111–130 [PubMed:2141574].

Squier, M.V., Lehr, R.P., 1994. Post-traumatic syringomyeliaJ. Neurol. Neurosurg. Psychiatry 57, 1095–1098 [PubMed:8089677].

Tuszynski, M.H., Weidner, N., McCormack, M., Miller, I., Powell, H.,Conner, J., 1998. Grafts of geneticallymodified Schwann cells tothe spinal cord: survival, axon growth, and myelination. CellTransplant. 7, 187–196 [PubMed: 9588600].

Vavrek, R., Girgis, J., Tetzlaff, W., Hiebert, G.W., Fouad, K., 2006.BDNF promotes connections of corticospinal neurons ontospared descending interneurons in spinal cord injured rats.Brain 129, 1534–1545 [PubMed: 16632552].

Verdú, E., Rodríguez, F.J., Gudiño-Cabrera, G., Nieto-Sampedro, M.,Navarro, X., 2000. Expansion of adult Schwann cells frommouse predegenerated peripheral nerves. J. Neurosci. Methods99, 111–117 [PubMed: 10936650].

Vroemen, M., Aigner, L., Winkler, J., Weidner, N., 2003. Adultneural progenitor cell grafts survive after acute spinal cordinjury and integrate along axonal pathways. Eur. J. Neurosci. 18(4), 743–751 PMID: 12925000 [PubMed - indexed for MEDLINE].

Weidner, N., Blesch, A., Grill, R.J., Tuszynski, M.H., 1999. Nervegrowth factor-hypersecreting Schwann cell grafts augmentand guide spinal cord axonal growth and remyelinate centralnervous system axons in a phenotypically appropriate mannerthat correlates with expression of L1. J. Comp. Neurol. 413,495–506 [PubMed: 10495438].

Zhou, F.Q., Zhong, J., Snider,W.D., 2003. Extracellular crosstalk: whenGDNF meets N-CAM. Cell. 113, 814–815 [PubMed: 12837237].

161B R A I N R E S E A R C H 1 2 5 6 ( 2 0 0 9 ) 1 4 9 – 1 6 1