Repetitive Intrathecal Catheter Delivery of Bone MarrowMesenchymal Stromal Cells Improves Functional

Recovery in a Rat Model of Contusive Spinal Cord Injury

Dasa Cizkova,1 Ivana Novotna,1 Lucia Slovinska,1 Ivo Vanicky,1 Stanislava Jergova,2

Jan Rosocha,3 and Jozef Radonak4

Abstract

Transplantation of bone marrow mesenchymal stromal cells (MSCs) has been shown to improve the functionalrecovery in various models of spinal cord injury (SCI). However, the issues of the optimal dose, timing, androute of MSC application are crucial factors in achieving beneficial therapeutic outcomes. The objective of thisstudy was to standardize the intrathecal (IT) catheter delivery of rat MSCs after SCI in adult rats. MSCs labeledwith PKH-67 were administered by IT delivery to rats at 3 or 7 days after SCI as one of the following treatmentregimens: (1) a single injection (5 · 105 MSCs/rat), or (2) as three daily injections (5 · 105 MSCs/rat/d for a totalof 1.5 · 106 MSCs/rat over 3 days, injected on days 3, 4, and 5, or days 7, 8, and 9 following SCI. The animalswere behaviorally tested for 4 weeks using the Basso, Beattie, and Bresnahan (BBB) locomotor rating scale, andhistologically assessed for MSC survival, distribution, and engraftment properties after 28 days. Rats treatedwith a single injection of MSCs at 3 or 7 days post-injury showed a modest, non-significant improvement infunction and low survival of grafted MSCs, which were found attached to the pia mater or accumulated aroundthe anterior spinal artery. In contrast, rats treated with three daily injections of MSCs at days 7, 8, and 9, but noton days 3, 4, and 5, showed significantly higher motor function recovery (BBB score 16.8 – 1.7) at 14–28 dayspost-injury. Transplanted PKH-67 MSCs were able to migrate and incorporate into the central lesion. However,only a limited number of surviving MSCs, ranging from 24,128 – 1170 to 116,258 – 8568 cells per graft, wereobserved within the damaged white matter. These results suggest that repetitive IT transplantation, whichimposes a minimal burden on the animals, may improve behavioral function when the dose, timing, andtargeted IT delivery of MSCs towards the lesion cavity are optimized.

Key words: neuroplasticity, regeneration, stem cells, traumatic spinal cord injury

Introduction

Spinal cord injury (SCI) represents a serious neuro-degenerative condition in which progressive cell death

and axonal degeneration results in the functional loss ofmultiple sensory, motor, and autonomic systems (Kakulas,1999). Patients who suffer from severe spinal cord traumashow limited functional recovery, which frequently leads toclinical signs of partial or complete paralysis (Marsala et al.,2005). The inability of spinal cord tissue to self-repair belowthe site of injury (Fawcett and Asher, 1999), together with theadverse inhibitory environment within the spinal cord(Schwab, 2004; Schwab and Bartholdi, 1996), and the dis-

continuity of long ascending (sensory) and descending (mo-tor) axonal projections caused by progressive tissue cavitation(Crowe et al., 1997), pose multifactorial obstacles that signif-icantly contribute to the lack of spinal cord regeneration.Surprisingly, despite these devastating events, some limitedspontaneous repair has been shown to occur after SCI (Horkyet al., 2006; Yang et al., 2006).

Therefore, because of these complexities, therapeutic in-terventions following SCI require a combination of at leastthree main strategies: (1) attenuation of the secondary injuryprocesses (such as inflammation and glutamate excitability),and inactivation of axonal growth inhibitors (Liu et al., 2008);(2) stimulation of endogenous spinal cord neurogenesis

1Institute of Neurobiology, Slovak Academy of Sciences, Kosice, Slovakia.2The Miami Project to Cure Paralysis, University of Miami, Miller School of Medicine, Miami, Florida.3Tissue Bank, Faculty of Medicine, and 4I Surgical Clinic, P.J. Safarik University, and L. Pasteur Faculty Hospital, Kosice, Slovakia.

JOURNAL OF NEUROTRAUMA 28:1951–1961 (September 2011)ª Mary Ann Liebert, Inc.DOI: 10.1089/neu.2010.1413

1951

(generating new spinal progenitors) through enhancedphysical activity, enrichment environment, and pharmaco-logical manipulation (Berrocal et al., 2007; Eaton et al., 2008);and (3) cell and gene therapy that aims to replace dead cells(neurons and oligodendrocytes), and induce a more favorableenvironment for axon regeneration (Cizkova et al., 2007;Coutts and Keirstead, 2008; Keirstead et al., 2005; Webberet al., 2007).

The impressive demonstration of regenerative medicineusing embryonic and adult stem cells to trigger healing andfunctional restoration of various tissues has focused attentionon the development of cellular therapies that can promoteregeneration of the damaged spinal cord as well (Conradet al., 2008; Garbuzova-Davis and Sanberg, 2009; Safinia andMinger, 2009). Several different types of cells, such as em-bryonic stem cells (Elisseeff, 2004), neural progenitors (Ciz-kova et al., 2007; Su et al., 2009), bone marrow-derivedmesenchymal stromal cells (BMSCs; Cizkova et al., 2006a),Schwann cells (Pearse et al., 2007), and olfactory ensheathingcells (OECs; Ruitenberg et al., 2005) have been explored in thesearch for efficient SCI therapies.

BMSCs or umbilical cord blood cells (hematopoietic andmesenchymal stromal cells), which have been primarily usedfor the treatment of serious and life-threatening blood-relateddiseases, have more recently been used clinically in broaderapplications for the treatment of various bone, cartilage, car-diac, and neurological diseases (Orlic, 2003; Sykova et al.,2006). However, in many cases, their beneficial effect is mostlikely due to the release of pro-regenerative growth factors orcytokines (Zacharek et al., 2007), or by indirect immuno-modulatory effects (Djouad et al., 2003), rather than by directcell replacement ( Jorgensen, 2009). Despite these findings, thetransplantation of autologous or allogeneic mesenchymalstromal cells has been considered as a potential therapeuticapproach to a wide variety of neurodegenerative diseases(Cizkova et al., 2006a; Hofstetter et al., 2002; Nandoe Tewarieet al., 2009; Sykova and Jendelova, 2005). However, one of thekey challenges for successful transplantation is to use the leastinvasive approach that is locally-targeted and minimizes cellloss and surgical procedure-induced tissue damage.

Therefore, in the present study we used MSCs derived fromrat bone marrow to standardize and optimize the intrathecal(IT) transplantation of stem cells after a moderate (12.5 lL)balloon-compression SCI in adult rats. Our results confirmthat IT delivery represents a non-invasive transplantationstrategy, which may improve post-injury locomotor function,and provides a promising strategy for additional testing ofembryonic stem cells or spinal oligodendroglial progenitortransplantation therapies in preclinical models with smallanimals or primates.

Methods

Isolation and culture of rat mesenchymal stromal cells

MSCs were isolated from the bone marrow of three adultmale Wistar rats (300 g), that had been collected from the longbones (femur and tibia), and were characterized as previouslydescribed (Cizkova et al., 2006b; Sykova and Jendelova, 2005).The isolated bone marrow was dissected into small pieces,gently homogenized, and filtered (70 lm) to remove bonefragments. Mononuclear cells (MNCs) were isolated by Ficolldensity gradient centrifugation (1.077 g/mL; Sigma-Aldrich,

Steinheim, Germany) at 400 g for 20 min. MNCs were col-lected from the interface, washed in Dulbecco’s modifiedEagle’s medium (DMEM), and centrifuged at 600 · g for10 min. The cell pellet was re-suspended in 1 mL of DMEM,the pooled cells were counted, and their viability was assessedusing the trypan blue dye exclusion method. MNCs weresubsequently resuspended in culture medium composed ofDMEM supplemented with 10% fetal calf serum (FCS; GIBCOLaboratories, Grand Island, NY), antibiotics (100 U/mL pen-icillin G and 100 lg/m streptomycin sulfate; Invitrogen,Carlsbad, CA), and plated at a density of 1 · 106 cells/25 cm2

in tissue culture flasks. The cells were incubated in a humid-ified atmosphere with 5% CO2 at 37�C. The medium wasexchanged every 3–4 days thereafter. When the culturesreached 80% of confluence, the MSCs were passaged with0.25% trypsin /0.53 mM EDTA (Invitrogen), and re-plated at adensity of 5000 cells/cm2. Cells at passage 3–4 were used forthe PKH-67 labeling procedure and thereafter for transplan-tation.

Labeling of mesenchymal stromal cellswith the cell linker dye PKH67

MSCs were labeled with the green fluorescent cell linkerPKH-67 according to the Sigma protocol (PKH67-GL) 24–48 hprior to transplantation. At passage 3–4, adherent MSCs weredetached with 0.25% trypsin/0.53 mM EDTA to obtain asingle cell suspension. After detachment, 1 · 107 cells werecentrifuged (600 · g for 10 min), washed, and then the finalpellet was resuspended in 1 mL of Diluent C (an iso-osmoticmannitol loading medium), with freshly prepared PKH-67 (2–4 lL) membrane linker (4 · 10 - 6 molar PKH-67). The samplewas then incubated at room temperature for 5 min followedby incubation at 4�C for 10 min. The staining reaction wasstopped by adding an equal volume of a-MEM with 5% FCS.The labeled MSCs were collected by centrifugation at 400 · gfor 10 min, and diluted in 500 lL of sterile saline, which wasused for IT transplantation (300 lL/15 rats). In addition, asample (* 200 lL) of labeled PKH-67 MSCs (1 · 105 cells) wasresuspended in DMEM supplemented with 10% FCS andantibiotics (100 U/mL penicillin G and 100 lg/mL strepto-mycin sulfate; Invitrogen), plated at a density of 5000 cells/cm2 in 6-well culture plates, and incubated in a humidifiedatmosphere with 5% CO2 at 37�C for two additional weeks inorder to confirm the efficacy of PKH-67 labeling in fully dif-ferentiated MSCs under in vitro conditions.

Evaluation of PKH-67 labeling efficacy

Prior to transplantation, the efficacy of PKH-67 MSC la-beling was determined using the cytocentrifuge method(StatSpin Cytofuge; Iris Sample Processing, Inc., Westwood,MA), in which centrifugal force is to used to deposit amonolayer of cells onto microscope slides. Four samples ofPKH-67-labeled MSCs (80–100 lL cell suspension) werecentrifuged at 27 · g for 6 min in disposable chambers. Sam-ples of labeled MSCs were concentrated in a monolayer thatcovered approximately a 7-mm-diameter area on the slidewith minimal cell loss to the filter. The cells were then fixedwith 4% paraformaldehyde (PFA), washed two times withphosphate-buffered saline (PBS), and 200 lL of 4¢,6-diamidino-2-phenylindole (DAPI) was incubated with the cells atroom temperature for 20 min in a dark humidified chamber.

1952 CIZKOVA ET AL.

Finally, the slides were washed two times in PBS and coveredwith mounting medium for analysis (Vectashield; VectorLaboratories, Burlingame, CA). The efficacy of the PKH-67staining procedure was investigated by fluorescence micros-copy (Nikon Eclipse Ti), with FITCI and DAPI filters. Thenumber of labeled cells (merged PKH-67 + DAPI) was ana-lyzed in a standard counting grid (40 lm · 40 lm frame), anda total of 100 squares/per slide were counted from 4 histo-logical slides (Fig. 1A–C).

Spinal cord trauma

All experimental procedures were approved by the insti-tutional ethical committee for animal research, and were inaccordance with the Slovak Law for Animal protection (no.115/1995). The SCI was induced using the modified balloon-compression technique in adult male Wistar rats (n = 63, n = 60for the transplantation experiments; n = 3 for the preliminaryIT dose and volume distribution study), weighing between300 and 320 g, according to our previous study (Vanicky et al.,2001). Manual bladder expression was required for 7–14 daysafter the injury until the bladder reflex was established. Noantibiotic treatment was used.

Intrathecal implant procedure

IT catheter implantation was performed according to ourpreviously published procedure (Hayes et al., 2003). The ratswere anesthetized with 1.5–2% halothane, and following aloss of responsiveness, the head was fixed in a stereotaxichead holder and a face mask was fixed over the rat’s muzzle tomaintain anesthesia administration. The posterior scalp wasincised (2–3 cm) at the midline and the muscles were gentlyseparated by blunt dissection with forceps. At the junction of

the occipital bone and atlas, the fascia overlying the cisternamagna was removed until the slightly transparent, white durawas visualized. The atlanto-occipital membrane was blotteddry to visualize the entire area, and a 3- to 4-mm incision wasmade through the dural midline using a 30-gauge needle withthe tip bent at a 45� angle. The PE-5 end (3.4 cm long) of the ITcatheter was inserted into the IT space and carefully pushedusing a slight rotating movement to the required depth withgentle pulling of the tail at the same time. The opposite PE-10catheter end was externalized on the forehead and the endwas plugged with 4–5 mm of stainless steel 28-gauge wire.The overlying muscles and skin were sutured using a silksuture to fix the catheter in place. All implanted IT catheterswere flushed with 10.0 lL of saline for 5 min to test theirtransition, and in the case of an obstruction, the catheter wasrepositioned or exchanged. After recovery, each rat wasevaluated for additional limb dysfunction, spinal asymmetry,pain, or adverse surgical effects, and sacrificed immediately ifany of these were observed.

Preliminary study: Evaluation of the intrathecal deliverydistribution pattern within the spinal cord by DiI delivery

To evaluate the extent and the penetration pattern of the ITcatheter delivery dose into the spinal cord tissue, three ratswere injected by IT delivery 3 days after the SCI with a single15-lL dose of DiI (4% in saline) flushed with 5 lL of sterilesaline over 15 min (Vybrant� DiI cell-labeling solution; In-vitrogen). After recovery, the rats were observed for possiblelocomotor impairments for an additional 4 h. Finally, all threerats were deeply anesthetized (IP injection of 50 mg/kg pen-tobarbital) and intracardially perfused with 0.1 M PBS (pH7.4) for 3 min, followed by cold 2% PFA in PBS for 10 min. Thespinal cord, brain, and externalized IT catheter were carefully

FIG. 1. (A) The efficacy of the in vitro PKH-67 mesenchymal stromal cell (MSC) labeling procedure reveals homogenousgreen autofluorescence in the majority of the MSCs. (B) Cells were double stained with 4–,6-diamidino-2-phenylindole(DAPI) and evaluated by applying a standard counting grid (C; 40 lm · 40 lm). (D) Stable PKH-67 labeling in fully-differ-entiated MSCs with processes (scale bars in A, B, and C = 50 lm; in D = 100 lm).

INTRATHECAL DELIVERY OF MESENCHYMAL STEM CELLS 1953

dissected (Fig. 2A and B). The position of the IT catheter fromits entrance (cisterna magna) throughout the entire spinalcord to the end of the tip (rostrally adjacent to the lesion site)was identified macroscopically (Fig. 2C and C¢). The distri-bution of DiI was visualized by the presence of pink-stainedspinal cord tissue. Samples taken 0.5–1.0 cm in each directionrostrally and caudally to the site of injury and from the centrallesion were embedded in embedding medium. Serial trans-verse cryostat spinal cord sections (20 lm thick) were then cutand mounted directly onto slides. Sections were taken at 500-to 1000-lm intervals, and 5–10 sections were taken from eachblock. Representative slides were used for analysis of DiIdistribution.

Transplantation of mesenchymal stromal cells

All SCI rats were randomly divided into the followinggroups: (1) low-dose (LD) group receiving a single injection(5 · 105 MSCs/20 lL/rat/d) at 3 days; n = 7; LD3D) or 7 days(n = 8; LD7D) after SCI; (2) high-dose (HD) group receivingthree daily injections of 5 · 105 MSCs/20 lL/rat/d for a total of1.5 · 106 MSCs/rat for 3 days, either on days 3, 4, and 5 (n = 7;HD3D), or on days 7, 8, and 9 (n = 8; HD7D) after SCI; (3) asaline LD group (SLD) receiving 20 lL of saline per rat startingat 3 days (n = 7; SLD3D) or 7 days (n = 8; SLD7D); and (4) a

saline HD group (SHD) receiving saline/20 lL/rat on days 3, 4,and 5 (n = 7; SHD3D) or on days 7, 8, and 9 (n = 8; SHD7D) afterSCI. A maximum dose of 15lL of MSCs resuspended in sterilesaline was flushed with 5 lL of saline over 15 min through acalibrated PE10 catheter (calibrated by 5-lL increments up to25 lL), which was connected to a 500-lL Hamilton syringeattached to a mechanical mini-pump. This procedure wasperformed in each rat of the LD3D, LD7D, HD3D, HD7D, SLD,and SHD groups. Implantation of the IT catheter was per-formed on the same day as transplantation, either 3 days or 7days after SCI. All animals were intracardially perfused 28days after SCI (21–25 days after transplantation).

Behavioral testing of motor function (BBB scoring)

Animals were evaluated before surgery and once a weekafter surgery for 4 weeks. Each rat was tested for 5 min by twoblinded examiners. The motor performance was assessedusing the Basso-Beattie-Bresnahan (BBB) 21-point open fieldlocomotor scale (Basso et al., 1995). BBB scores, which cate-gorize combinations of rat hindlimb movements, trunk posi-tion and stability, stepping, coordination, paw placement, toeclearance, and tail position, were analyzed. In this evaluation,0 represents no locomotion and 21 represents normal motorfunction.

FIG. 2. Macroscopic observation of the dissected brain and spinal cord with an implanted single-lumen catheter. Theinternalized PE-5 tubing with the lumen linked to the externalized PE-10 site is partially pulled out from the original insertionpoint (arrow); upper view (A), and side view (B). Note the junction of both catheter sites indicated by the ball structure(asterisk), which is normally on top of the cisterna magna. Dissemination of intrathecally (IT)-injected DiI identified as tissuemainly located over the ventral (C), but not the dorsal (D) aspect of the spinal cord lesion site. Note the tip of the catheterat the central lesion (two arrows in C), and the higher magnification area shown by the dashed line (C¢), that indicates theposition of the catheter tip (three arrows) at the site of central injury. The coronal cryostat spinal cord sections taken from therostral (E¢), central (E†), and caudal segments (E¢†) confirmed intense DiI autofluorescence located within the subarachnoidalspace of the spinal cord (pia mater and dura mater; arrowheads). At the central lesion and caudal segments, scatteredautofluorescence could be seen within damaged gray and white matter (E† and E¢†), indicated by the arrows. At the dorsalsegment, a cross-section of the IT catheter is shown indicated by two arrows in E¢ (scale bar in E¢† = 500 lm).

1954 CIZKOVA ET AL.

Quantification analysis

At the end of the study (28 days after SCI), all of the animalswere deeply anesthetized with a ketamine-xylazine cocktail(15 mg/kg of xylazine and 150 mg/kg of ketamine), andtranscardially perfused with 0.1 M PBS, followed by 2% para-formaldehyde in PBS. The spinal cord and externalized ITcatheter were removed, post-fixed in the same fixative solutionovernight at 4�C, and embedded in gelatin albumin substratein 2% paraformaldehyde in PBS. Frozen spinal cord sectionswere cut from a 1.5-cm-long spinal cord segment positioned onthe injury epicenter, embedded in embedding medium, dis-sected into three 0.5-cm-thick blocks (rostral, epicenter, andcaudal), and stored at - 20�C. Sections 10 lm thick were thenserially cut from the epicenter and mounted directly ontoslides. From the rostral and caudal regions, 40-lm-thick sec-tions were cut and immersed in PBS (floating sections). Sectionswere taken at 200-lm intervals and 20 sections per block wereobtained. Ten sections per block were stained with hematox-ylin and eosin (H&E) to assess tissue morphology and deter-mine the injury epicenter, and 10 sections per block were usedfor quantitative analysis of cellular PKH-67 distribution. Theslides were analyzed using a Nikon light or fluorescence mi-croscope (Nikon Eclipse), and captured with a Nikon digitalcamera DS Fi1 or DSQi1Mc. Quantitative analysis of trans-planted PKH-67 MSCs was performed according to our pre-viously published stereological technique (Cizkova et al., 2009;Marsala et al., 2004). PKH-67-positive fields were assessedfrom 10 sections per block under a 60 · oil immersion objective(optical images were made using a motorized z stage, 1 lm),and the number of DAPI/PKH-67-positive (optical imageswere made using a motorized z stage, 1 lm), and the totalnumbers of DAPI/PKH-67-positive cells were calculated usingstereological principles of systematic, random sampling, andthe stereological fractionator sampling formula N = Q · 1/hsf · 1/asf · 1/ssf, where N is the total number of positivenuclei, Q is the sum of the cells counted, hsf is the heightsampling fraction, asf is the area sampling fraction, and ssf isthe slice sampling fraction (Kakinohana et al., 2004). The slideswere analyzed using a Nikon light or fluorescence microscope,and the images were captured with a digital camera. The im-ages were prepared with Photoshop 7.0 and digitally stored forfurther processing.

Statistical analysis

Behavioral data determined from the BBB scores were av-eraged across the hindlimbs. Data from SLD animals at 3 daysand at 7 days were pooled together for the SLD group, and thedata from SHD animals at 3, 4, and 5 days, as well as 7, 8, and9 days, were pooled for the SHD group. The differences be-tween the experimental groups were analyzed using one-wayanalysis of variance (ANOVA) with a Tukey-Kramer multiplecomparisons test. All data are presented as mean val-ues – standard error of the mean (SEM). Differences betweengroups were considered statistically significant if p < 0.05 orp < 0.01.

Results

In vitro PKH-67 labeling

The viability of PKH-67-labeled MSCs was 91 – 3%, as as-sessed by the four independent calculations of the trypan blue

exclusion test, which indicated the low cytotoxity of the PKH-67-fluorescent dye. However, we found that successful la-beling of MSCs with PKH-67 required prolonging the labelingprocedure from the recommended 5 min up to 15 min (5 minat room temperature followed by a 10-min incubation at 4�C).This modification enhanced the staining procedure and en-abled more uniform labeling, which resulted in the majority ofMSCs being completely stained (Fig. 1A–D). Using this pro-tocol, we found that the efficacy of PKH-67 MSC labelingranged between 87–95%. Although these modifications en-hanced the staining of the cells, some variability betweenexperiments still occurred. In addition, we also confirmed thatPKH-67 labeling of MSCs was stable during differentiation(2–3 weeks) in vitro, where bright, PKH-67-positive, differ-entiated MSCs with processes were observed (Fig. 1D).

Intrathecal implantation/evaluation of DiI distribution

The dissemination of injected DiI was clearly visualized atthe lesion site of the dissected spinal cord. The pink-stainedtissue was mainly located over the ventral aspect of the spinalcord (total area of 1.5–2.0 cm; Fig. 2C and C¢), while on thedorsal surface that contained hemorrhages, almost no coloredtissue was observed macroscopically (Fig. 2D). The transversecryostat spinal cord sections taken from the rostral, central,and caudal segments (Fig. 2D) confirmed the DiI distribution,which was observed as intense autofluorescence locatedwithin the subarachnoidal space of the spinal cord (pia materand dura mater; Fig. 2E¢, E†, and E¢†). However, only scatteredautofluorescence could be seen within damaged gray andwhite matter at the central lesion and caudal segments (Fig.2E† and E¢†).

Improvement of motor locomotion

The LD group that received a single injection (5 · 105

MSCs/rat) at 3 days (LD3D) or 7 days (LD7D) after SCI didnot show any significant functional recovery, and the BBBscores only slightly increased, reaching a maximum BBB scoreof 12.9 – 1.7 after 28 days when the cells were administered at3 days, or 13.7 – 1.8 after 21 days when the cells were ad-ministered at 7 days, compared to the SLD group (BBB scores11.0 – 1.2 at 21 days and 10.9 – 1.4 at 28 days; Fig. 3A).

A non-significant locomotor improvement was seen in theHD3D group receiving three daily injections (a total of1.5 · 106 MSCs/rat) at 3, 4, and 5 days, which had a BBB scoreof 13.5 – 1.5 at 18 days. This score further decreased and clo-sely correlated with the LD and saline groups (Fig. 3B). Theonly significant improvement in motor function was detectedin the HD7D group, which received three daily injections (atotal of 1.5 · 106 MSCs/rat) at 7, 8, and 9 days after SCI, andhad final BBB scores of 15.1 – 1.3, 15.9 – 1.9, 16.8 – 1.7, and16.7 – 2.0, at 14, 18, 21, and 28 days, respectively. The gradualrecovery of hindlimb locomotion observed between the SHDand HD7D experimental groups was statistically significant at14 days ( p < 0.05), and at 18, 21, and 28 days ( p < 0.01; Fig. 3B).

Mesenchymal stromal cell survival and distribution

The LD group. After a single IT delivery at 3 or 7 days,transplanted PKH-67 MSCs were found scattered and sur-viving within the subarachnoidal space (Fig. 4A–D). Graftedcells showed limited migration into the central lesion,

INTRATHECAL DELIVERY OF MESENCHYMAL STEM CELLS 1955

although a few cells were found to be penetrating into thedorsal roots (Fig. 4A), as well as the ventral or lateral whitematter (Fig. 4B and C). However, the majority of transplantedcells were mainly found attached to the ventral pia mater asindividual or small clusters of cells surrounding the anteriorspinal artery (Fig. 4B). Furthermore, a clear, thin stream oftransplanted cells within the anterior median fissure wasdetected invading into the deeper ventral white matter (100–

200 lm; Fig. 4B). A similar cluster of PKH-67-grafted cells wasfound at the lateral spinal cord regions, but with no or mini-mal migration ( < 100 lm) within the white matter (Fig. 4C).

The HD group. In contrast to the LD group, animalstreated with three daily injections of MSCs at 7, 8, and 9 daysshowed engraftment of PKH-67 cells within the lesionedspinal cord tissue. Transplanted MSCs formed a migratingPKH-67 green fluorescence cell tract that continuously ex-tended from the outer white matter regions towards the lesioncavity site (Fig. 5A). The PKH-67 grafts were organized into anetwork-like structure, composed of tight, adjacent, or scat-tered cells filling all or part of the lesion cavity (Fig. 5A and B).Serial sectioning of the spinal cord at a depth of 200 lm en-abled us to distinguish almost complete grafts in 3 out of 8transplanted rats (Fig. 5A), or scattered grafted cells withindamaged tissue in 4 rats. One transplanted rat only had PKH-67 MSCs within the subdural space. Engrafted cells werefound mainly in the white matter (rostral or central lesion site)as round, undifferentiated, or occasionally branch-like indi-vidual cells occasionally surrounded with fine fibers (Fig. 5B).Dissemination of PKH-67 cells was also clearly located withinthe subarachnoidal space, attached to the pia mater, or evenlocated within the inner central canal area (Fig. 5C). In con-trast, early transplantation at 3, 4, and 5 days led to a completeabsence of PKH-67 MSCs in the damaged white matter. Inaddition, we were able to observe scattered PKH-67 MSCsaround the anterior spinal artery, blood vessels, and axons,similar to that observed in the LD group (Fig. 5D and E).Moreover, the engraftment of injected PKH-67 MSCs withinthe lesioned spinal cord tissue mainly correlated with theposition of the IT catheter (dorsal, lateral, or ventral) with bothapplication strategies (Fig. 5F).

Quantification of transplants

The majority of PKH-67-labeled cells were found within thedamaged white matter with maximal spreading of cells 2–3 mm rostrally, but not caudally, from the lesion in the HDgroup transplanted at 7, 8, and 9 days post-injury. In four rats,we identified complete grafts spreading into the lateral/ventrolateral white matter. These rats had 116,258 – 8568,81,024 – 3328, 111,878 – 10,631, and 92,064 – 2450 cells. Theaverage cell number from these rats was 100,306 – 5092 cellsper graft, which represented 6.85 – 0.41% of total transplantedPKH-67 MSCs (Fig. 6A and B; Table 1). In three rats, smallclusters (partial grafts) of engrafted MSCs were found at thevarious sites of damaged white matter, and revealed evenlower numbers of cells, with an average of 38,385 – 3556 cellsper graft, representing 2.5 – 0.23% of the total transplantedPKH-67 MSCs (Fig. 6D). Therefore, an average of 4.67 – 0.32%of grafted cells was found in the HD group. In one rat, thePKH-67 MSCs cells remained attached to the pia mater, whichwas similar to what was seen in the LD group.

Discussion

In the present study, we have provided evidence that PKH-67-labeled MSCs can be delivered through an intrathecalcatheter directly to the lesion site of rats that underwent amoderate SCI. We further showed that after three daily in-jections, MSCs migrated and integrated into the damagedspinal cord tissue when administered 7–9 days after SCI, but

FIG. 3. (A) Locomotor recovery of hindlimb function afterspinal cord injury (SCI) and mesenchymal stromal cell(MSC) transplantation at 3 or 7 days, evaluated in the low-dose 3-day group (LD3D; n = 7), low-dose 7-day group(LD7D; n = 8), and control saline low-dose group (SLD;n = 15) during the 28-day survival period. The highest Basso,Beattie, and Bresnahan (BBB) scores were 12.9 – 1.7 at 28 dayswhen the cells were administered 3 days post-injury, and13.7 – 1.8 at 21 days when the cells were administered 7 dayspost-injury. There were no significant differences in BBBscores between the groups during the survival period( p > 0.05). The arrows indicate the time of transplantation.Data are presented as means – standard error of the mean(SEM). (B) Locomotor recovery of hindlimb function afterSCI and MSC transplantation evaluated in the SHD (n = 15),HD3D (n = 7), and HD7D (n = 8) groups during the 28-daysurvival period. There were no significant differences in BBBscores in the HD 3- to 5-day group. The transplant groupsand HD 7- to 9-day group reached significantly greater lo-comotor recovery at 14 days (*p < 0.05), and at 18, 21, and 28days (**p < 0.01) after SCI compared to the control SHDgroup, with BBB scores of 15.1 – 1.3, 15.9 – 1.9, 16.8 – 1.7, and16.7 – 2.0, at 14, 18, 21, and 28 days, respectively. The col-umns indicate the time of transplantation. Data are presentedas means – SEM.

1956 CIZKOVA ET AL.

FIG. 4. Distribution of PKH-67 mesenchymal stromal cells (MSCs) in the thoracic (T9) transverse spinal cord sections after asingle intrathecal (IT) delivery in the low-dose (LD) 7-day group 28 days after spinal cord injury (SCI). MSCs were foundwithin the subarachnoidal space (A–D), although a few cells migrated through the pia mater into the host dorsal roots (A andA¢), ventral (B), or lateral white matter (C and C¢). The majority of transplanted cells were attached to the ventral pia mater asindividual or small clusters of cells surrounding the anterior spinal artery (thick arrow in B). Note the stream of transplantedcells within the anterior median fissure invading into the deeper ventral white matter (thin arrows in B). In the highmagnification images (A¢ and C¢) of the boxed areas in A and C, clusters of PKH-67-grafted cells can also be seen at the dorsaland lateral spinal cord regions (C), or attached to the dura mater (D; scale bars in A, B, C, and D = 100 lm; in A¢ andC¢ = 50 lm).

FIG. 5. Engrafted PKH-67 mesenchymal stromal cells (MSCs) in the thoracic (T9) serially cut transverse spinal cord sectionsafter intrathecal (IT) delivery in the high-dose (HD) 7- to 9-day group 28 days after spinal cord injury (SCI). (A) Variouslyassociated grafts show PKH-67 fluorescence cell tracts that extend from the outer white matter regions towards the lesioncavity site. The PKH-67 grafts are composed of tight, adjacent or scattered cells filling either all or part of the lesion cavity (Aand B). Engrafted cells were seen mainly in the white matter as round or branch-like (dashed arrow) individual cellssurrounded with scattered fine fibers (three arrows in B). (C) Dissemination of PKH-67 cells was also clearly located withinthe subarachnoidal space (2 arrows), around the anterior spinal artery, attached to the outer pia mater (dashed arrow), oreven located within the inner central canal area (asterisk). (D) In the high-dose (HD) 3- to 5-day group, scattered PKH-67MSCs were observed around the anterior spinal artery (asterisk), the anterior median fissure invading into the deeper ventralwhite matter (arrows), or around the blood vessels (arrows in E). Schematic drawings of the IT catheter position andengrafted PKH-67 MSCs in correlation to the observed spinal cord sections outlining the dorsal (FB), lateral (FC), or ventral(FD) injection sites (scale bars in A = 500 lm; in B, C, D, and E = 100 lm; and F = 500 lm).

INTRATHECAL DELIVERY OF MESENCHYMAL STEM CELLS 1957

not at earlier applications (3–5 days), or after a single injection,which most likely contributed to the functional recovery.Since MSCs secrete a wide variety of trophic factors (Wilkinset al., 2009) that support host axonal growth and functionalrecovery and easily target sites of tissue injury (Stappenbeckand Miyoshi, 2009), they were ideal candidates for our ITdelivery experiments. Visualization of MSC distributionwithin the recipient spinal cord tissue was made possible byusing the PKH-67 membrane linker, which provided highlyeffective, homogenous cellular labeling with low cytotoxicity.The highly-stable PKH-67 autofluorescence enabled us totrack the distribution of transplanted MSCs within the intra-thecal space and the spinal cord tissue directly on the cryostatsections.

The key focus of our experiments was to standardize andoptimize an IT application for the purpose of delivering PKH-67-labeled MSCs locally and as close as possible to the site ofSCI. To achieve this goal, we chose the single-lumen catheter

model that was developed primarily for experimental studiesof spinal opiate tolerance and spinal drug delivery (Yaksh,1999). With this IT technique, we were able to deliver repeti-tive doses of PKH-67 MSCs directly to the central lesionwithout the need for additional surgery or other stressfulmanipulations. The length of the intrathecal catheter (3.4 cm)matched the position of the primary lesion at the T9 segment,and therefore the tip of the catheter was located in closeproximity to the injury.

Before our main experimental study, we had to verify theposition of the catheter and expected volume of the dosedistribution in three separate rats with SCI by using IT injec-tion of the fluorescent dye DiI. A volume of 20 lL of DiI wasselected based on a previous study, which was less than 10%of the total cerebrospinal fluid (CSF) volume (250 lL) in the rat(Hua et al., 1999; Yaksh, 1999). The injected DiI tended togather primarily in the ventral portion of the central lesionsite, which may be related to the physiological body position

FIG. 6. (A–D) Cross-sections of thoracic spinal cord (T9) containing engrafted PKH-67 mesenchymal stromal cells (MSCs)from the high-dose (HD) group (7- to 9-days) co-stained with 4¢,6-diamidino-2-phenylindole (DAPI), and processed forquantification analysis using a standard stereological grid outlining the graft position (B and C). Optical section motorized zstage (1 lm) images reveal clear profiles of specific engrafted cells that express green autofluorescence. These cells were alsodistinguishable on DAPI co-stained sections within the observed 40-lm-thick section (C and D). A few grafted PKH-67 MSCsoccasionally showed a differentiation pattern with an overall bipolar shape and elongated multiple branching processes (Eand F; scale bars in A = 500 lm; in B and C = 100 lm; in D–F = 50 lm).

Table 1. Stereological Quantitative Analysis of PKH 67 MSCs and Graft Volume Determination

in the HD Group at 28 Days

Rat no.Targeted area of

white matter (WM)/segmentNumber of PKH-67

MSCs per graftVolume

of graft/mm3

1/compl Lateral WM/T8 116,258 – 8568 0.0964 – 0.00122/compl Ventrolateral WM/T9 81,024 – 3328 0.1051 – 0.01323/compl Lateral WM/T9 111,878 – 10631 0.1325 – 0.00204/compl Dorsolateral WM/T8 92,064 – 2450 0.0516 – 0.00165/part Ventral WM/T8 24,128 – 1170 0.0224 – 0.00116/part Dorsal WM/T9 53,854 – 4050 0.0351 – 0.00217/part Lateral WM/T9 37,173 – 5450 0.0281 – 0.00178/negat 0 0 0

HD, high-dose; MSC, mesenchymal stem cell; compl, complete graft; part, partial graft containing scattered PKH67 cells; negat, negative,no cell engrafted into the white matter.

1958 CIZKOVA ET AL.

of the rat and gravity. Furthermore, the fluorescent dye ac-cumulated along the central lesion within the range of 1–2 mmrostro-caudally, and entered the central white/gray matter ata very low density. One explanation for its low penetrationinto the lesioned tissue may be due to the lack of chemotacticsignaling, which is most likely involved in the migrationprocesses of stem cell transplantation, but is not applicable todye penetration (Greco and Rameshwar, 2008). Alternatively,a more reasonable explanation is that the low DiI incorpora-tion may be due to the short survival period of the rats afterdye application (4 h).

Once we verified the technical parameters related to ITdelivery, we focused our attention on defining two key pa-rameters for the present experiment: (1) the optimal dose oftransplanted PKH-67 MSCs, and (2) the ideal timing for cel-lular infusion. To determine the optimal dose of injectedMSCs, we compared the effect of a single dose (LD) with thatof three daily repetitive doses (HD). Our reasoning for therepetitive doses was not only to increase the number of de-livered PKH-67 MSCs, but to also provide the CSF with freshstem cells during three daily applications. As we expected,significant differences in the engraftment of PKH-67 MSCsbetween the two experimental groups were observed.

In the single-dose LD group, which received 5 · 105 ofPKH-67 MSCs at 3 or 7 days post-injury, we observed a ho-mogenous distribution of PKH-67-labeled MSCs that mainlyattached to the pia mater surrounding the central lesion. It iswell known that normal CSF produces trophic factors that arebeneficial for stem cell survival and differentiation, whereasduring pathological conditions, the altered substrate contentcauses apoptosis or other adverse effects (Oren et al., 2001).Since the majority of delivered cells were located within thesubarachnoidal space, we hypothesize that the CSF of theinjured spinal cord had no significant effect on MSC survivalin these experiments. This has also been confirmed by a pre-vious study, which showed high viability and proliferation ofIT-delivered neural progenitors in a traumatic SCI model (Baiet al., 2003). However, in this group we did not observe theexpected engraftment of transplants. The occasional invasionby injected PKH-67 MSCs into the ventral white matter oc-curred primarily along the anterior spinal artery. The fact thatsome MSCs were concentrated around major blood vessels inthe ventral white matter may be due to the rich nutrientsupply (Butler et al., 2010). Since many dose-response studieshave shown rapid and major cell loss shortly after trans-plantation, we hypothesize that the LD (5 · 105 of PKH-67MSCs) applied in the present experiment was too low, andtherefore limited the number of grafted cells that survived atsites of suitable substrates, attached to the pia mater, or ac-cumulated around blood vessels (Paul et al., 2009). Further-more, LD transplantation had no beneficial effect on motorfunction improvement at both treatment times, which is inagreement with the low cell distribution seen in the sub-arachnoidal space and the lack of engraftment into the le-sioned tissue.

A very similar distribution pattern of grafted MSCs withinthe recipient tissue was observed in the HD group injectedwith 3 daily doses (at days 3, 4, and 5, for a total of 1.5 · 106 ofPKH-67 MSCs) with early grafting 3 days after SCI. Althoughhigher numbers of surviving cells were attached to the piamater or distributed around blood vessels, the animalsshowed only partial, non-significant improvements of motor

function at the end of the study period. This may be caused byimproper early IT delivery due to the strong inflammatoryresponse that persisted within the damaged spinal cord pa-renchyma, which will be further discussed in more detail.

The most promising findings of this study were obtained inthe HD group that received later delivery of MSCs, starting at 7days post-injury (days 7, 8, and 9, for a total of 1.5 · 106 PKH-MSCs). In this group, the incorporation of transplanted cellsand the formation of grafts within the white matter of the in-jured spinal cord parenchyma was observed. On the serialcryostat sections, we were only able to identify three completegrafts that could be seen from the beginning of the implantationsite to the end. We noted a high variability in the numbers ofsurviving cells within the observed grafts, which ranged fromapproximately 24 · 103 - 116 · 103 cells, with an average of73 · 103 cells per graft. The reason for this engraftment differ-ence is not clear, but may be due to the incorrect position ortransition of the catheter during cellular infusion.

If the tip of the catheter is adjacent to non-lesioned piamater and white matter, penetration of cells is rather difficult,unlike when the catheter is in close proximity to the damagedtissue. Another complication may be related to catheter ob-struction with clusters of transplanted cells, macrophages, orother substances that impede infusion (Allen et al., 2006; Mieleet al., 2006). However, these limitations could easily beovercome in human trials by using radioimaging or ultra-sound techniques, which are critical for guiding the IT cath-eter insertion, and allow confirming the correct positioning, orrepositioning the catheter directly adjacent to the central le-sion (Hassenbusch et al., 2002; Hebl, 2008; Karmakar et al.,2009). Correct placement of the IT catheter is an essentialprerequisite for long-term use of the catheter and accurateengraftment of stem cells near damaged spinal cord paren-chyma. This point was also confirmed in the present study,since the IT catheters were located laterally and in closeproximity to the grafts in all of the successful engraftments.Furthermore, only the HD group at the late transplantationtime point (7–9 days) showed significant improvement inmotor function, which began at 14 days post-infusion andcontinued until sacrifice at day 28. Notably, in the case of theHD cellular infusion, we gradually injected 3 times more cellsthan in the LD transplant group. However, the goal of thisapplication strategy was not only to deliver an increasednumber of cells, but to feed the IT system with freshly pre-pared MSCs daily for 3 days. We believe that this approachprovides a new method for gradually replenishing stem cellswithin the damaged areas of nervous system tissue, whichallows for better survival and incorporation of the transplant,and stimulates endogenous regenerative processes. The ma-jority of the engrafted PKH-67 MSCs that survived had around, undifferentiated morphology and occasional branch-like individual cells. We did not identify the specific expres-sion of neuronal or glial markers, with the exception of nestin,which was only rarely found in PKH-67 MSCs (data notshown). Therefore, this study did not provide convincingevidence of a possible neural lineage for these cells.

Moreover, in both experimental groups, we sought to de-termine the ideal timing for transplantation. Our presentfindings are in agreement with the results from other previousstudies, which confirmed that early transplantation, either bysingle or multiple injections 3–5 days after SCI, leads to verylow cell survival, most likely due to an ongoing inflammatory

INTRATHECAL DELIVERY OF MESENCHYMAL STEM CELLS 1959

response (Karimi-Abdolrezaee et al., 2006). It seems likely thatafter 1 week, a partial attenuation of the inflammatory processoccurs, which allows for better survival of transplanted cells.In this study, we found that the optimal time for transplan-tation ranged between 7 and 9 days. In conclusion, we haveconfirmed the feasibility of delivering MSCs by IT catheterinto the contused spinal cord. Multiple deliveries of MSCsthrough the CSF may be improved by utilizing a more so-phisticated double-lumen catheter instead of a single-lumencatheter, which has a number of specific advantages for long-term transplantation studies, and allows for the injection ofstem cells concurrently with trophic factors, biomaterials, orother substrates. Furthermore, we suggest that a higher dose,more than 1.5 · 106 injected cells, which is two to three timeshigher than that used in the present study, may lead to higherefficacy of cell survival, more engraftment into the host tissue,and improvement of functional recovery.

Acknowledgments

The authors wish to thank Dr. Stephen Minger from King’sCollege for valuable comments and suggestions, and MarikaSpontakova for technical assistance. This study was sup-ported by grants VEGA 2-0114/11, VEGA 1-0674-09, MVTS-COST BM-1002, and APVV 51-002105.

Author Disclosure Statement

No competing financial interests exist.

References

Allen, J.W., Horais, K.A., Tozier, N.A., Wegner, K., Corbeil, J.A.,Mattrey, R.F., Rossi, S.S., and Yaksh, T.L. (2006). Time courseand role of morphine dose and concentration in intrathecalgranuloma formation in dogs: a combined magnetic resonanceimaging and histopathology investigation. Anesthesiology105, 581–589.

Bai, H., Suzuki, Y., Noda, T., Wu, S., Kataoka, K., Kitada, M.,Ohta, M., Chou, H., and Ide, C. (2003). Dissemination andproliferation of neural stem cells on the spinal cord by injec-tion into the fourth ventricle of the rat: a method for celltransplantation. J. Neurosci. Methods 124, 181–187.

Basso, D.M., Beattie, M.S., and Bresnahan, J.C. (1995). A sensitiveand reliable locomotor rating scale for open field testing inrats. J. Neurotrauma 12, 1–21.

Berrocal, Y., Pearse, D.D., Singh, A., Andrade, C.M., McBroom,J.S., Puentes, R., and Eaton, M.J. (2007). Social and environ-mental enrichment improves sensory and motor recovery aftersevere contusive spinal cord injury in the rat. J. Neurotrauma24, 1761–1772.

Butler, J.M., Nolan, D.J., Vertes, E.L., Varnum-Finney, B., Ko-bayashi, H., Hooper, A.T., Seandel, M., Shido, K., White, I.A.,Kobayashi, M., Witte, L., May, C., Shawber, C., Kimura, Y.,Kitajewski, J., Rosenwaks, Z., Bernstein, I.D., and Rafii, S.(2010). Endothelial cells are essential for the self-renewal andrepopulation of Notch-dependent hematopoietic stem cells.Cell Stem Cell 6, 251–264.

Cizkova, D., Kakinohana, O., Kucharova, K., Marsala, S., Johe, K.,Hazel, T., Hefferan, M.P., and Marsala, M. (2007). Functionalrecovery in rats with ischemic paraplegia after spinal grafting ofhuman spinal stem cells. Neuroscience 147, 546–560.

Cizkova, D., Nagyova, M., Slovinska, L., Novotna, I., Radonak,J., Cizek, M., Mechirova, E., Tomori, Z., Hlucilova, J., Motlik,

J., Sulla, I., Jr., and Vanicky, I. (2009). Response of ependymalprogenitors to spinal cord injury or enhanced physical activityin adult rat. Cell. Mol. Neurobiol. 29, 999–1013.

Cizkova, D., Rosocha, J., Vanicky, I., Jergova, S., and Cizek, M.(2006a). Transplants of human mesenchymal stem cells im-prove functional recovery after spinal cord injury in the rat.Cell Mol. Neurobiol. 26, 1165–1178.

Cizkova, D., Rosocha, J., Vanicky, I., Radonak, J., Galik, J., andCizek, M. (2006b). Induction of mesenchymal stem cells leadsto HSP72 synthesis and higher resistance to oxidative stress.Neurochem. Res. 31, 1011–1020.

Conrad, S., Renninger, M., Hennenlotter, J., Wiesner, T., Just, L.,Bonin, M., Aicher, W., Buhring, H.J., Mattheus, U., Mack, A.,Wagner, H.J., Minger, S., Matzkies, M., Reppel, M., Hescheler,J., Sievert, K.D., Stenzl, A., and Skutella, T. (2008). Generationof pluripotent stem cells from adult human testis. Nature 456,344–349.

Coutts, M., and Keirstead, H.S. (2008). Stem cells for the treat-ment of spinal cord injury. Exp. Neurol. 209, 368–377.

Crowe, M.J., Bresnahan, J.C., Shuman, S.L., Masters, J.N., andBeattie, M.S. (1997). Apoptosis and delayed degeneration afterspinal cord injury in rats and monkeys. Nat. Med. 3, 73–76.

Djouad, F., Plence, P., Bony, C., Tropel, P., Apparailly, F., Sany,J., Noel, D., and Jorgensen, C. (2003). Immunosuppressiveeffect of mesenchymal stem cells favors tumor growth in al-logeneic animals. Blood 102, 3837–3844.

Eaton, M.J., Pearse, D.D., McBroom, J.S., and Berrocal, Y.A.(2008). The combination of human neuronal serotonergic cellimplants and environmental enrichment after contusive SCIimproves motor recovery over each individual strategy. Be-hav. Brain Res. 194, 236–241.

Elisseeff, J.H. (2004). Embryonic stem cells: potential for moreimpact. Trends Biotechnol. 22, 155–156.

Fawcett, J.W., and Asher, R.A. (1999). The glial scar and centralnervous system repair. Brain Res. Bull. 49, 377–391.

Garbuzova-Davis, S., and Sanberg, P.R. (2009). Feasibility ofcell therapy for amyotrophic lateral sclerosis. Exp. Neurol.216, 3–6.

Greco, S.J., and Rameshwar, P. (2008). Microenvironmentalconsiderations in the application of human mesenchymal stemcells in regenerative therapies. Biologics 2, 699–705.

Hassenbusch, S., Burchiel, K., Coffey, R.J., Cousins, M.J., Deer,T., Hahn, M.B., Pen, S.D., Follett, K.A., Krames, E., Rogers,J.N., Sagher, O., Staats, P.S., Wallace, M., and Willis, K.D.(2002). Management of intrathecal catheter-tip inflammatorymasses: a consensus statement. Pain Med. 3, 313–323.

Hayes, C.S., Mulkmus, S.A., Cizkova, D., Yaksh, T.L., and Hua,X.Y. (2003). A double-lumen intrathecal catheter for studies ofmodulation of spinal opiate tolerance. J. Neurosci. Methods126, 165–173.

Hebl, J.R. (2008). Ultrasound-guided regional anesthesia and theprevention of neurologic injury: fact or fiction? Anesthesiology108, 186–188.

Hofstetter, C.P., Schwarz, E.J., Hess, D., Widenfalk, J., El Manira,A., Prockop, D.J., and Olson, L. (2002). Marrow stromal cellsform guiding strands in the injured spinal cord and promoterecovery. Proc. Natl. Acad. Sci. USA 99, 2199–2204.

Horky, L.L., Galimi, F., Gage, F.H., and Horner, P.J. (2006). Fateof endogenous stem/progenitor cells following spinal cordinjury. J. Comp. Neurol. 498, 525–538.

Hua, X.Y., Chen, P., Marsala, M., and Yaksh, T.L. (1999). In-trathecal substance P-induced thermal hyperalgesia and spinalrelease of prostaglandin E2 and amino acids. Neuroscience 89,525–534.

1960 CIZKOVA ET AL.

Jorgensen, C. (2009). Link between cancer stem cells and adultmesenchymal stromal cells: implications for cancer therapy.Regen. Med. 4, 149–152.

Kakinohana, O., Cizkova, D., Tomori, Z., Hedlund, E., Marsala,S., Isacson, O., and Marsala, M. (2004). Region-specific cellgrafting into cervical and lumbar spinal cord in rat: a quali-tative and quantitative stereological study. Exp. Neurol. 190,122–132.

Kakulas, B.A. (1999). A review of the neuropathology of humanspinal cord injury with emphasis on special features. J. SpinalCord Med. 22, 119–124.

Karimi-Abdolrezaee, S., Eftekharpour, E., Wang, J., Morshead,C.M., and Fehlings, M.G. (2006). Delayed transplantation ofadult neural precursor cells promotes remyelination andfunctional neurological recovery after spinal cord injury. J.Neurosci. 26, 3377–3389.

Karmakar, M.K., Li, X., Ho, A.M., Kwok, W.H., and Chui, P.T.(2009). Real-time ultrasound-guided paramedian epidural ac-cess: evaluation of a novel in-plane technique. Br. J. Anaesth.102, 845–854.

Keirstead, H.S., Nistor, G., Bernal, G., Totoiu, M., Cloutier, F.,Sharp, K., and Steward, O. (2005). Human embryonic stemcell-derived oligodendrocyte progenitor cell transplants re-myelinate and restore locomotion after spinal cord injury. J.Neurosci. 25, 4694–4705.

Liu, W.L., Lee, Y.H., Tsai, S.Y., Hsu, C.Y., Sun, Y.Y., Yang, L.Y.,Tsai, S.H., and Yang, W.C. (2008). Methylprednisolone inhibitsthe expression of glial fibrillary acidic protein and chondroitinsulfate proteoglycans in reactivated astrocytes. Glia 56, 1390–1400.

Marsala, M., Hefferan, M.P., Kakinohana, O., Nakamura, S.,Marsala, J., and Tomori, Z. (2005). Measurement of peripheralmuscle resistance in rats with chronic ischemia-inducedparaplegia or morphine-induced rigidity using a semi-automated computer-controlled muscle resistance meter.J. Neurotrauma 22, 1348–1361.

Marsala, M., Kakinohana, O., Yaksh, T.L., Tomori, Z., Marsala,S., and Cizkova, D. (2004). Spinal implantation of hNT neu-rons and neuronal precursors: graft survival and functionaleffects in rats with ischemic spastic paraplegia. Eur. J. Neu-rosci. 20, 2401–2414.

Miele, V.J., Price, K.O., Bloomfield, S., Hogg, J., and Bailes, J.E.(2006). A review of intrathecal morphine therapy relatedgranulomas. Eur. J. Pain 10, 251–261.

Nandoe Tewarie, R.D., Hurtado, A., Ritfeld, G.J., Rahiem, S.T.,Wendell, D.F., Barroso, M.M., Grotenhuis, J.A., and Oudega,M. (2009). Bone marrow stromal cells elicit tissue sparing afteracute but not delayed transplantation into the contused adultrat thoracic spinal cord. J. Neurotrauma 26, 2313–2322.

Oren, A., White, L.R., and Aasly, J. (2001). Apoptosis in neu-rones exposed to cerebrospinal fluid from patients with mul-tiple sclerosis or acute polyradiculoneuropathy. J. Neurol. Sci.186, 31–36.

Orlic, D. (2003). Adult bone marrow stem cells regeneratemyocardium in ischemic heart disease. Ann. NY Acad. Sci.996, 152–157.

Paul, C., Samdani, A.F., Betz, R.R., Fischer, I., and Neuhuber, B.(2009). Grafting of human bone marrow stromal cells intospinal cord injury: a comparison of delivery methods. Spine(Phila. Pa. 1976) 34, 328–334.

Pearse, D.D., Sanchez, A.R., Pereira, F.C., Andrade, C.M., Puzis,R., Pressman, Y., Golden, K., Kitay, B.M., Blits, B., Wood, P.M.,and Bunge, M.B. (2007). Transplantation of Schwann cellsand/or olfactory ensheathing glia into the contused spinal

cord: Survival, migration, axon association, and functionalrecovery. Glia 55, 976–1000.

Ruitenberg, M.J., Levison, D.B., Lee, S.V., Verhaagen, J., Harvey,A.R., and Plant, G.W. (2005). NT-3 expression from en-gineered olfactory ensheathing glia promotes spinal sparingand regeneration. Brain 128, 839–853.

Safinia, N., and Minger, S.L. (2009). Generation of hepatocytesfrom human embryonic stem cells. Methods Mol. Biol. 481,169–180.

Schwab, M.E., and Bartholdi, D. (1996). Degeneration and re-generation of axons in the lesioned spinal cord. Physiol. Rev.76, 319–370.

Schwab, M.E. (2004). Nogo and axon regeneration. Curr. Opin.Neurobiol. 14, 118–124.

Stappenbeck, T.S., and Miyoshi, H. (2009). The role of stromalstem cells in tissue regeneration and wound repair. Science324, 1666–1669.

Su, H., Zhang, W., Guo, J., Guo, A., Yuan, Q., and Wu, W. (2009).Neural progenitor cells enhance the survival and axonal regen-eration of injured motoneurons after transplantation into theavulsed ventral horn of adult rats. J. Neurotrauma 26, 67–80.

Sykova, E., Homola, A., Mazanec, R., Lachmann, H., Konrado-va, S.L., Kobylka, P., Padr, R., Neuwirth, J., Komrska, V.,Vavra, V., Stulik, J., and Bojar, M. (2006). Autologous bonemarrow transplantation in patients with subacute and chronicspinal cord injury. Cell Transplant. 15, 675–687.

Sykova, E., and Jendelova, P. (2005). Magnetic resonance track-ing of implanted adult and embryonic stem cells in injuredbrain and spinal cord. Ann. NY Acad. Sci. 1049, 146–160.

Vanicky, I., Urdzikova, L., Saganova, K., Cizkova, D., and Galik,J. (2001). A simple and reproducible model of spinal cord in-jury induced by epidural balloon inflation in the rat. J. Neu-rotrauma 18, 1399–1407.

Webber, D.J., Bradbury, E.J., McMahon, S.B., and Minger, S.L.(2007). Transplanted neural progenitor cells survive and dif-ferentiate but achieve limited functional recovery in the le-sioned adult rat spinal cord. Regen. Med. 2, 929–945.

Wilkins, A., Kemp, K., Ginty, M., Hares, K., Mallam, E., andScolding, N. (2009). Human bone marrow-derived mesen-chymal stem cells secrete brain-derived neurotrophic factorwhich promotes neuronal survival in vitro. Stem Cell Res.[Epub ahead of print].

Yaksh, T.L. (1999). Spinal systems and pain processing: devel-opment of novel analgesic drugs with mechanistically definedmodels. Trends Pharmacol. Sci. 20, 329–337.

Yang, H., Lu, P., McKay, H.M., Bernot, T., Keirstead, H., Steward,O., Gage, F.H., Edgerton, V.R., and Tuszynski, M.H. (2006).Endogenous neurogenesis replaces oligodendrocytes and astro-cytes after primate spinal cord injury. J. Neurosci. 26, 2157–2166.

Zacharek, A., Chen, J., Cui, X., Li, A., Li, Y., Roberts, C., Feng, Y.,Gao, Q., and Chopp, M. (2007). Angiopoietin1/Tie2 andVEGF/Flk1 induced by MSC treatment amplifies angiogenesisand vascular stabilization after stroke. J. Cereb. Blood FlowMetab. 27, 1684–1691.

Address correspondence to:Dasa Cizkova, D.V.M., Ph.D.

Institute of NeurobiologyCenter of Excellence for Brain Research

Slovak Academy of SciencesSoltesovej 4-6

04001 Kosice, Slovakia

E-mail: cizkova@saske.sk

INTRATHECAL DELIVERY OF MESENCHYMAL STEM CELLS 1961