ORIGINAL ARTICLE. functional recovery in completely transected adult rat spinal cord. of BMSC-SCs promotes axonal regeneration and

Transcription

1 JOBNAME: jnen 64# PAGE: 1 OUTPUT: Sat January 8 18:38: J Neuropathol Exp Neurol Copyright Ó 2005 by the American Association of Neuropathologists, Inc. Vol. 64, No. 1 January 2005 pp ORIGINAL ARTICLE Transplantation of Bone Marrow Stromal Cell-Derived Schwann Cells Promotes Axonal Regeneration and Functional Recovery after Complete Transection of Adult Rat Spinal Cord Takahito Kamada, MD, Masao Koda, MD, PhD, Mari Dezawa, MD, PhD, Katsunori Yoshinaga, MD, PhD, Masayuki Hashimoto, MD, PhD, Shuhei Koshizuka, MD, PhD, Yutaka Nishio, MD, Hideshige Moriya, MD, PhD, and Masashi Yamazaki, MD, PhD Abstract The aim of this study was to evaluate whether transplantation of Schwann cells derived from bone marrow stromal cells (BMSC-SCs) promotes axonal regeneration and functional recovery in completely transected spinal cord in adult rats. Bone marrow stromal cells (BMSCs) were induced to differentiate into Schwann cells in vitro. A 4-mm segment of rat spinal cord was removed completely at the T7 level. An ultra-filtration membrane tube, filled with a mixture of Matrigel (MG) and BMSC-SCs (BMSC-SC group) or Matrigel alone (MG group), was grafted into the gap. In the BMSC-SC group, the number of neurofilament- and tyrosine hydroxylase-immunoreactive nerve fibers was significantly higher compared to the MG group, although 5-hydroxytryptamine- or calcitonin gene-related peptideimmunoreactive fibers were rarely detectable in both groups. In the BMSC-SC group, significant recovery of the hindlimb function was recognized, which was abolished by retransection of the graft 6 weeks after transplantation. These results demonstrate that transplantation of BMSC-SCs promotes axonal regeneration of lesioned spinal cord, resulting in recovery of hindlimb function in rats. Transplantation of BMSC-SCs is a potentially useful treatment for spinal cord injury. Key Words: Axonal regeneration, Bone marrow stromal cell, Hindlimb function, Schwann cell, Spinal cord injury. INTRODUCTION It has been widely believed that axons in the lesioned adult mammalian spinal cord cannot regenerate. This failure of regeneration is attributed to the nonpermissive environment of the damaged adult mammalian spinal cord, the milieu of which From the Department of Orthopaedic Surgery (TK, MK, MH, SK, YN, HM, MY), Chiba University Graduate School of Medicine, Chiba, Japan; Department of Anatomy and Neurobiology (MD), Kyoto University Graduate School of Medicine, Kyoto, Japan; Division of Rehabilitation Medicine (KY), Chiba University Hospital, Chiba, Japan. Send correspondence and reprint requests to: Masashi Yamazaki, Department of Orthopaedic Surgery, Chiba University Graduate School of Medicine, Inohana, Chuo-ku, Chiba Japan; masashiy@ faculty.chiba-u.jp. This work was supported by grants-in-aid for Scientific Research from the Ministry of Education, Science, and Culture of Japan ( ). is formed of astrocyte-derived inhibitory molecules in the scar tissue, myelin components of oligodendrocytes interfering with the regeneration of axons, lack of trophic support for axotomized neurons, and to the intrinsic neuronal changes, including cell atrophy and death after axotomy (1 3). To date, only partial recovery can be obtained after spinal cord injury despite extensive laboratory and clinical investigations. In contrast to the central nervous system, the peripheral nervous system contains cellular and molecular components capable of eliciting axonal regrowth. Differences in the glial environment may be the key to the dissimilarity of the regenerating capacity between the central and peripheral nervous systems (2). Peripheral glia are composed of Schwann cells that express various types of neurotrophic factors and adhesion molecules supporting axonal regrowth, and can reconstruct the myelin sheath during the process of peripheral nervous system regeneration (4). Thus, Schwann cells may have the capacity to promote axonal regeneration of the central nervous system. It has been previously reported that Schwann cells can promote axonal regeneration of lesioned adult rat spinal cord (5 18). Dezawa et al have reported that Schwann cells could be induced from bone marrow stromal cells in vitro and that they effectively promoted regeneration of lesioned sciatic nerves (19). The aim of this study was to evaluate the efficacy of bone marrow stromal cell-derived Schwann cells (BMSC-SCs) in cell therapy for spinal cord injury. Here we show that transplantation of BMSC-SCs promotes axonal regeneration and functional recovery in completely transected adult rat spinal cord. MATERIALS AND METHODS Bone Marrow Stromal Cell Culture and BMSC-SC Induction In Vitro The culture of dorsal root ganglion (DRG)-derived Schwann cells and bone marrow stromal cells (BMSCs) was performed as previously described (19). In brief, Schwann cells were harvested from DRG of neonatal rats (19). Bone marrow cells were collected from the femurs of adult male Wistar rats (SLC, Hamamatsu, Japan), and cells adherent to culture dishes were isolated and cultured as BMSCs. We then J Neuropathol Exp Neurol Volume 64, Number 1, January

2 JOBNAME: jnen 64# PAGE: 2 OUTPUT: Sat January 8 18:38: Kamada et al J Neuropathol Exp Neurol Volume 64, Number 1, January 2005 FIGURE 1. (A) Defect of spinal cord after resection of the T8 spinal cord segment. Following laminectomy, the spinal cord was completely transected with microsurgical scissors at the T7/8 and T8/9 levels, and a 4-mm segment of the spinal cord at T8 was removed. A leftward arrow indicates the rostral side and a rightward arrow indicates the caudal side of the rat. Arrowheads indicate rostral and caudal stumps of transected spinal cord. (B) The graft. A 5-mm length of ultra-filtration membrane tube (Millipore, Bedford, MA) filled with a mixture of Matrigel and bone marrow stromal cell-derived Schwann cells (BMSC-SCs) ( /each tube; BMSC-SC group) or Matrigel alone (MG group). (C) Just after transplantation. The graft bridging both stumps of the transected spinal cord. A leftward arrow indicates the rostral side and a rightward arrow indicates the caudal side of the rat. Arrowhead indicates the graft. Bars = 5 mm in (A), (C); Bar = 2 mm in (B). performed the BMSC-SCs induction in vitro as previously described (19). Briefly, BMSCs were incubated with alphaminimum essential eagle medium (alpha-mem) containing 1 mm beta-mercaptoethanol for 24 hours. After washing with 0.1 M phosphate-buffered saline (PBS, ph 7.4), medium was replaced with alpha-mem containing 10% fetal bovine serum (FBS) and 35 ng/ml all-trans-retinoic acid (Sigma, St. Louis, MO) for 3 days. Cells were then transferred to alpha-mem containing 10% FBS, 5 mm forskolin (Calbiochem, La Jolla, CA), 10 ng/ml recombinant human basic fibroblast growth factor (Peprotech, London, UK), 5 ng/ml platelet derivedgrowth factor (Peprotech), and 200 ng/ml heregulin (R&D Systems, Minneapolis, MN) for 7 days. To characterize the induced BMSC-SCs in vitro, we compared the phenotype of BMSC-SCs with that of DRG-derived Schwann cells. We performed immunocytochemistry as previously described (19). Anti-S-100 rabbit polyclonal antibody (S-100, 1:100; DakoCytomation, Copenhagen, Denmark) and anti-p75 low affinity nerve growth factor receptor rabbit polyclonal antibody (p75, 1:200; Chemicon, Temecula, CA) were used as markers for Schwann cells. Cell nuclei were stained with TOTO-3 iodide (Molecular Probes, Eugene, OR). For the negative control, we performed immunocytochemistry by omitting the primary antibodies. Spinal Cord Injury and BMSC-SC Transplantation For experimental spinal cord injury (SCI), we used a total of 18 male Wistar rats aged 8 weeks (average weight 200 g; SLC). Animals were anesthetized with % halothane in 1.5 L/min oxygen. Laminectomies were performed at the T7 and T8 levels, leaving the dura intact. The T6 and T9 spinal processes were clamped to fix the spine. The spinal cord was completely transected with microsurgical scissors at the T7 and T8 levels, and the T7 spinal cord segment was removed (Fig. 1A). A 2-mm-diameter tube made of ultra-filtration membrane (nominal molecular weight cutoff: 10,000 Daltons, Millipore, Bedford, MA) was cut into 5-mm FIGURE 2. (A) Schema of sagittal sections of the spinal cord including the graft. Every fourth 30-mm sagittal serial section was examined with immunohistochemistry using 1 of 4 antibodies. At least 4 series of sections were inspected by immunohistochemistry. Thus, the central 480 mm portion of the graft (about half of the central width of graft; hatched area) was included in the immunohistochemical examination. (B) Schema of the nerve fiber count within the grafts. We counted the number of nerve fibers which traverse the lines perpendicular to the central axis of the grafts at rostral, middle, and caudal levels within the grafts. 38 q 2005 American Association of Neuropathologists, Inc.

3 JOBNAME: jnen 64# PAGE: 3 OUTPUT: Sat January 8 18:38: J Neuropathol Exp Neurol Volume 64, Number 1, January 2005 BMSC-SC for SCI FIGURE 3. (A) Phase-contrast microscopic image of Schwann cells derived from dorsal root ganglion in a neonatal rat 1 day old (cultured for 14 days). (B) Phase-contrast microscopic image of bone marrow stromal cell-derived Schwann cells (BMSC-SCs) 5 days after the induction from bone marrow stromal cells (BMSCs). Before the induction, GFP gene was introduced into the BMSCs using retroviral vector. The BMSC-SCs are morphologically and phenotypically similar to Schwann cells derived from dorsal root ganglion (C F). Immunofluorescence of Schwann cells for S-100 (C), BMSC-SCs for S-100 (D), Schwann cells for p75 (E), and BMSC-SCs for p75 (F). Cells are positive to S-100 or p75. Nuclei were stained with TOTO-3. Bars = 30 mm. lengths and filled with a mixture of Matrigel (BD Biosciences, Bedford, MA) and BMSC-SCs ( /each tube), (BMSC- SC group, n = 9) or Matrigel alone (MG group, n = 7) (Fig. 1B). These grafts were transplanted into the gap between the rostral and caudal spinal cord stumps. For identifying the transplanted cells, the green fluorescent protein (GFP) gene was introduced into BMSCs before the transplantation, using retroviral vector as previously described (19). After the transplantation, the T6 and T9 spinous processes were tied together tightly with a 4-0 nylon suture to prevent kyphosis and to obtain contact between the graft and spinal cord stumps (Fig. 1C). Muscles and skin were sutured layer to layer, and the rats placed in warm cages overnight. Food and water were provided ad libitum. Manual bladder expression was performed twice a day until recovery of the bladder reflex. The BMSC-SC transplanted animals showed no apparent abnormal behavior. All the experimental procedures were performed in compliance with the guidelines established by the Animal Care and Use Committee of Chiba University. Immunohistochemistry Animals were perfused transcardially with 4% paraformaldehyde in PBS (ph 7.4) after an overdose of pentobarbital anesthesia 6 weeks after transplantation. Three spinal cord segments (T6 8) including the graft were removed and postfixed in the same fixative overnight, stored in 20% sucrose in PBS at 4 C, and embedded in OCT (Sakura Finetechnical, Tokyo, Japan). Sagittal cryosections (30 mm) were mounted onto poly-l-lysine-coated slides (Matsunami, Tokyo, Japan). The following primary antibodies were used: antineurofilament 200 monoclonal antibody (NF, 1:400; Sigma), anti-tyrosine hydroxylase monoclonal antibody (TH, 1:400; Chemicon), anti-serotonin rabbit polyclonal antibody (5-HT, 1:5000; Sigma), and anti-calcitonin gene-related peptide rabbit polyclonal antibody (CGRP, 1:1000; AFFINITI, Exeter, UK). After incubation with primary antibodies, the sections were incubated with biotinylated secondary antibodies and reacted with peroxidase-conjugated streptavidin using the protocol from a HISTOFINE kit (Nichirei, Tokyo, Japan). Enzyme signals were detected with hydrogen peroxide and diaminobenzidine (WAKO, Osaka, Japan). q 2005 American Association of Neuropathologists, Inc. 39 Fig. 3 live 4/C

4 JOBNAME: jnen 64# PAGE: 4 OUTPUT: Sat January 8 18:38: Kamada et al J Neuropathol Exp Neurol Volume 64, Number 1, January 2005 FIGURE 4. Macroscopic examination of the graft. The grafts integrated well into the host spinal cords in the BMSC-SC group. A leftward arrow indicates the rostral side and a rightward arrow indicates the caudal side of the spinal cord. Arrowheads indicate the rostral (left) or caudal (right) interfaces between the graft and the host spinal cord. Bar = 5 mm. We used NF as a general marker for nerve fibers and the following markers for specific nerve fiber populations. THpositive nerve fibers are coerulo-spinal adrenergic nerve fibers and 5-HT-positive nerve fibers are raphe-spinal serotonergic fibers, both of which contribute to motor function (20 23). CGRP is a marker for sensory nerve fibers (14, 16, 24). To evaluate regeneration of NF-, TH-, 5-HT-, or CGRPpositive nerve fibers, every fourth section was reacted with a specific antibody and the number of immunoreactive (IR) fibers that traversed the lines perpendicular to the central axis of the grafts at rostral (1 mm caudal to the rostral graft-cord interface), middle (middle of the graft) and caudal (1 mm rostral to the caudal graft-cord interface) levels within the grafts were counted in at least 4 samples from each animal; thus the central 480 mm portion of the graft was evaluated by immunohistochemistry (Fig. 2A, B). The numbers of nerve fibers between the BMSC-SC and MG groups were compared. Statistical analysis was performed using the Mann-Whitney U test. To characterize transplanted cells, we performed a double immunofluorescence study for GFP and markers for Schwann cells 6 weeks after transplantation in the BMSC-SC group. Anti-GFP goat polyclonal antibody (GFP, 1:100; Santa- Cruz, Santa-Cruz, CA) was used as a marker for transplanted cells. Anti-S-100 rabbit polyclonal antibody, anti-p75 rabbit polyclonal antibody, and anti-protein zero mouse monoclonal antibody (P0, 1:300; provided by Dr. Juan J. Archelos, Department of Neurology, University of Graz, Graz, Austria) (25) were used as Schwann cell markers. After reaction with primary antibodies, sections were incubated with Alexa Fluor 488-conjugated anti-goat IgG (Molecular Probes), and Alexa Fluor 594-conjugated anti-mouse or anti-rabbit IgG (Molecular Probes). In a case of P0, the sections were incubated with biotinylated anti-mouse IgG (Nichirei) followed by a reaction with Texas Red-conjugated avidin (1:400, Vector Laboratories, Inc., Burlingame, CA). The fluorescent signals were observed by fluorescence microscopy (ECLIPSE E600; Nikon, Tokyo, Japan). Assessment of Locomotor Activity Hindlimb function of animals in both the BMSC-SC and MG groups was assessed using the Basso, Beattie, Bresnahan (BBB) locomotor scale (26) before injury and 1 day, 3 days, and 1 6 weeks (once a week) after injury. Statistical analysis of the 2 groups was performed using a repeated measure FIGURE 5. (A) NF-IR nerve fibers in the Matrigel (MG) group. Arrows indicate the rostral or caudal stumps in the host lesioned spinal cord. (B) High magnification of box B in (A). Several NF-IR nerve fibers (arrowheads) penetrated the graft. (C) High magnification of box C in (A). A few NF-IR nerve fibers were detected in the middle level of graft (arrowheads). (D) NF-IR nerve fibers in BMSC-SC group. Arrows indicate the rostral and caudal level of the graft. (E) High magnification of box E in (D). Several NF-IR nerve fibers were observed in the rostral part of the graft (arrowheads). (F) High magnification of box F in (D). Several NF-IR nerve fibers were detected in the middle level within the graft (arrowheads). A leftward arrow indicates the rostral side and a rightward arrow indicates the caudal side of the spinal cords. Bars = 5 mm in (A, D); Bars = 100 mm in(b, C, E, F). 40 q 2005 American Association of Neuropathologists, Inc.

5 JOBNAME: jnen 64# PAGE: 5 OUTPUT: Sat January 8 18:39: J Neuropathol Exp Neurol Volume 64, Number 1, January 2005 BMSC-SC for SCI FIGURE 6. Comparison of the average number of NF-IR nerve fibers at all 3 levels within the grafts between the Matrigel (MG) group(rostral ¼ , middle ¼ , caudal ¼ ) and thebmsc-sc group(rostral¼ , middle¼ , caudal ¼ ). There was a significant difference (*, p, 0.01) between the 2 groups at all 3 levels within the grafts. Closed columns, MG group; open columns, BMSC-SC group. Bars = 6SE. ANOVA. Two animals in the BMSC-SC group were retransected at the mid point of the graft 6 weeks after transplantation to evaluate whether the regenerated nerve fibers contributed to hindlimb functional recovery. RESULTS Induced BMSC-SCs were morphologically similar to the DRG-derived Schwann cells (Fig. 3A, B). Immunocytochemistry showed almost all of the induced BMSC-SCs were positive for S-100 and p75, both of which are known as markers for Schwann cells (Fig. 3C, D, E, F). Thus the induced BMSC-SCs were phenotypically similar to Schwann cells in vitro as previously described (19). Macroscopic examination revealed that the grafts had integrated well into the host spinal cords in the BMSC-SC group (Fig. 4). The average graft diameter was 0.78 mm in the BMSC-SC group and 0.51 mm in the MG group at mid point of the graft. However, there was no significant difference in average graft diameters between the two groups. Tumor formation was not observed in either group. To evaluate axonal regeneration, we performed immunohistochemistry for NF in the grafts. Because the T7 spinal cord segment was completely removed and replaced by the graft, all of the NF-IR nerve fibers within the graft can be considered as regenerated axons. Present results revealed that the number of NF-IR nerve fibers at all 3 levels within the grafts in the BMSC-SC group was significantly larger than those in the MG group (p, 0.01) (Fig. 5A F, Fig. 6). To reveal the character of regenerating axons, we performed immunohistochemistry for nerve fiber markers. The numbers of TH-IR nerve fibers at all 3 levels within the grafts in BMSC-SC group were significantly larger than those in the MG group (p, 0.01) (Fig. 7B, C, Fig. 8A). Avery small FIGURE 7. (A) Schema of sagittal section of spinal cord containing the graft in the BMSC-SC group. (B) TH-IR nerve fibers (arrowheads) at the rostral level (indicated as box B in [A]) within the graft in the BMSC-SC group. (C) TH-IR nerve fibers (arrowheads) at the middle level (indicated as box C in [A]) within the graft in the BMSC-SC group. (D) 5-HT-IR nerve fibers (arrowheads) at the rostral level (indicated as box D in [A]) within the graft in the BMSC-SC group. (E) 5-HT-IR nerve fibers (arrowheads) at the middle level (indicated as box E in [A]) within the graft in the BMSC-SC group. (F) CGRP-IR nerve fibers (arrowheads) at the rostral level (indicated as box F in [A]) within the graft in the BMSC-SC group. (G) CGRP-IR nerve fibers (arrowheads) at the middle level (indicated as box G in [A]) within the graft in the BMSC-SC group. A leftward arrow indicates the rostral side and the rightward arrow indicates the caudal side of the grafts. Bars = 100 mm. number of 5-HT-IR or CGRP-IR nerve fibers were observed within the grafts of both groups (Fig. 7D G), (Fig. 8B, C). There was no significant difference in the numbers of 5-HT-IR or CGRP-IR nerve fibers between both groups (Fig. 8B, C). Double immunofluorescence analyses revealed that GFP-IR transplanted BMSC-SCs were simultaneously positive for S-100, p75, or P0 6 weeks after transplantation, q 2005 American Association of Neuropathologists, Inc. 41

6 JOBNAME: jnen 64# PAGE: 6 OUTPUT: Sat January 8 18:39: Kamada et al J Neuropathol Exp Neurol Volume 64, Number 1, January 2005 and host spinal cord (Fig. 9A F.), suggesting that endogenous Schwann cells migrated into the graft. Hindlimb function recovered significantly in the BMSC-SC group from 4 weeks after transplantation, and a significant difference from the MG group was recognized up to 6 weeks after transplantation. The average recovery score in the BMSC-SC group 6 weeks after transplantation was 7.0 (minimum 5 to maximum 10), which indicates all 3 joints of hindlimbs had extensive movement (Fig. 10A). The best recovery score in the BMSC-SC group was 10, which indicates occasional weight supporting plantar steps, but no forelimb-hindlimb coordination. However, the average recovery score in the MG group was 3.6 (minimum 2 to maximum 5), indicating 2 joints of hindlimbs had extensive movement (Fig. 10A). We retransected the grafts at their mid-point in 2 of the rats from the BMSC-SC group 6 weeks after transplantation. Retransection abolished the recovered hindlimb function and no significant recovery was observed 4 weeks after retransection (Fig. 10B). FIGURE 8. (A) Quantification of TH-IR nerve fibers in the Matrigel (MG) and BMSC-SC groups. The numbers of TH-IR nerve fibers at all 3 levels within the grafts in the BMSC-SC group were significantly larger than those in the MG group. (*, p, 0.01) (B) The numbers of 5-HT-IR nerve fibers in both groups. (C) The numbers of CGRP-IR nerve fibers in both groups. There was no significant difference in the numbers of 5-HT-IR or CGRP-IR nerve fibers within the grafts of both groups. Closed columns, MG group; Open columns, BMSC-SC group. Bars = 6SE. indicating that transplanted BMSC-SCs maintained their specific phenotype in vivo for the long term. In addition, S-100-, p75-, or P0-positive cells without GFP-IR were observed in both rostral and caudal interfaces between graft DISCUSSION In the present study, transplantation of BMSC-SCs effectively promoted axonal regeneration, and resulted in functional recovery after complete transection of the spinal cord in adult rats. Retransection of the grafts completely abolished the recovered function in rats from the BMSC-SC group. Accordingly, these data exclude the possibility that transplanted cells enhance the activity of a locomotor pattern generator in the spinal cord, and emphasize that axonal regeneration contributes to functional recovery. The axonal growth-promoting effects of Schwann cells have been studied in various spinal cord injury models (2). Regrowth of axons is promoted by implants of cultured Schwann cells in a photochemical spinal cord injury model in adult rats (6). Schwann cell transplantation promotes axonal regeneration of CGRP-positive sensory fibers in contusive injury (9) and in complete transaction (13) of adult rat spinal cord. Schwann cell transplants also improve the number of remyelinated axons and hindlimb locomotor performance in the moderately contused adult rat thoracic spinal cord (18). These reports indicate that Schwann cells have the capability to promote axonal regeneration in the lesioned adult rat spinal cord. Thus, Schwann cells are strong candidates for cell transplantation therapy for spinal cord injury. In the present study, transplantation of BMSC-SCs significantly promoted regeneration of TH-IR nerve fibers, which constitute the coerulo-spinal tract one of the supraspinal tracts organizing motor function. The present results are consistent with previous reports that regeneration of THpositive nerve fibers is important for hindlimb functional recovery (20 23). In contrast to the present results, peripheral nerve-derived Schwann cells failed to promote regeneration of TH-positive fibers in complete transaction (8, 9, 13) or contusion injury (18), although they could promote THpositive fiber regrowth in a spinal cord hemisection model (16). 42 q 2005 American Association of Neuropathologists, Inc.

7 JOBNAME: jnen 64# PAGE: 7 OUTPUT: Sat January 8 18:39: J Neuropathol Exp Neurol Volume 64, Number 1, January 2005 BMSC-SC for SCI FIGURE 9. Double immunofluorescence study of bone marrow stromal cell-derived Schwann cells (BMSC-SCs) within the graft 6 weeks after transplantation. A large proportion of S-100-IR cells within the graft was also positive for GFP. (A) S-100, (B) GFP, (C) merged view; arrowheads indicate cells double-labeled for S-100 and GFP. A large proportion of p75-ir cells within the graft was also positive for GFP. (D) p75, (E) GFP, (F) merged view; arrowheads indicate cells double-labeled for p75 and GFP. A large proportion of P0-IR cells within the graft was also positive for GFP. (G) P0, (H) GFP, (I) merged view; arrowheads indicate cells double-labeled for P0 and GFP. Arrows in (C), (F), and (I) indicate cells that were positive for S-100, p75, or P0, but negative for GFP. Bars = 100 mm. We observed only a few 5-HT-IR raphe-spinal fibers and CGRP-IR sensory fibers within the grafts of BMSC-SCs transplanted rats. Transplantation of Schwann cells derived from peripheral nerve promotes regeneration of 5-HT-positive fibers (15, 16), and CGRP-positive sensory nerve fibers (9, 10, 12, 14, 15) in complete transection of adult rat spinal cord. Differences in axonal regrowth-promoting capability between peripheral nerve- and BMSC-derived Schwann cells may be because of their differential production of various chemicals, including neurotrophic factors and adhesion molecules (2). The precise mechanism by which BMSC-SCs promote the functional recovery after spinal cord injury is still unclear. Previous studies have shown that Schwann cells secrete various types of growth factors, including NGF, BDNF, and NT3 (2, 4, 10, 15), and they have a potential to promote regeneration of the injured axons and functional recovery (15, 18, 24). Thus, it is possible that BMSC-SCs also secrete some growth factors after the transplantation. These growth factors may contribute to the axonal regeneration and the migration of endogenous Schwann cells, though there has been no direct evidence that BMSC-SCs synthesize growth factors in vitro and in vivo. Transplantation of BMSC-SCs has some advantages compared to the transplantation of Schwann cells derived from peripheral nerves. Firstly, BMSCs are easily harvested because they can be obtained from the iliac crest or femur by bone marrow aspiration. In contrast, harvesting Schwann cells derived from peripheral nerves potentially causes complications, including anesthesia or abnormal pain at the harvesting site. Bone marrow aspiration is much less invasive, and can be performed in outpatients. Secondly, BMSCs are easily expanded in vitro because they can proliferate more vigorously than Schwann cells derived from peripheral nerves. These q 2005 American Association of Neuropathologists, Inc. 43 Fig. 9 live 4/C

8 JOBNAME: jnen 64# PAGE: 8 OUTPUT: Sat January 8 18:39: Kamada et al J Neuropathol Exp Neurol Volume 64, Number 1, January 2005 transplanted BMSCs might differentiate into aberrant cells in the spinal cord. Additionally, cell fusion has recently been proposed as an explanation of stem cell plasticity, raising questions about its mechanisms in adult bone marrow-derived cells (30, 31). Before transplantation, we manipulated BMSCs to differentiate to the Schwann cell phenotype in vitro. From this standpoint, our method has advantages compared to the transplantation of undifferentiated BMSCs. BMSC-SCs that have been induced to differentiate from BMSCs prior to transplantation have maintained the long term expression of specific markers for Schwann cells in vivo in the present study. In conclusion, transplantation of Schwann cells derived from bone marrow stromal cells is a potentially useful treatment for spinal cord injury. Although further exploration is needed to establish optimal conditions for more effective axonal regeneration and functional recovery, our study advances us toward using cell transplantation therapy for treatment of spinal cord injury. ACKNOWLEDGMENTS We are grateful to Dr. Juan J. Archelos (Department of Neurology, University of Graz, Graz, Austria) for providing us with mouse monoclonal P0 antibody. FIGURE 10. (A) Basso, Beattie, Bresnahan (BBB), score after transplantation, indicating hindlimb functional recovery. Hindlimb function recovered significantly in the BMSC-SC group from 4 weeks after transplantation, and a significant difference from the Matrigel (MG) group was maintained until 6 weeks after transplantation (*, p, 0.01). The average recovery score in the BMSC-SC group 6 weeks after transplantation was 7.0 (minimum 5 to maximum 10), which indicates all 3 joints of hindlimbs had extensive movement, whereas that in the MG group was 3.6 (minimum 2 to maximum 5), which indicates 2 joints of hindlimbs had extensive movement. (B) BBB score after retransection of the grafts in rats from the BMSC-SC group 6 weeks after transplantation. Retransection completely abolished the recovered hindlimb function and no significant recovery was observed after retransection for 4 weeks. j, MG group;, BMSC-SC group. Bars 6SE. advantages suggest that BMSC-SCs might be strong candidates for cell transplantation therapy for spinal cord injury. Several reports have indicated the effectiveness of BMSC transplantation for spinal cord injury. Transplantation of BMSCs significantly improved hindlimb function in a rat spinal cord contusion injury (27). BMSCs formed guiding strands in contused spinal cord and promoted hindlimb functional recovery (28). BMSCs enhanced the differentiation of a cocultured neurosphere in vitro and promoted regeneration of injured spinal cord (29). Although transplantation of BMSCs is effective, it is difficult to completely control the differentiation of transplanted BMSCs. One cannot exclude the possibility that REFERENCES 1. Schwab ME, Bartholdi D. Degeneration and regeneration of axons in the lesioned spinal cord. Physiol Rev 1996;76: Bunge MB. Bridging areas of injury in the spinal cord. Neuroscientist. 2001;7: Fawcett JW, Asher RA. The glial scar and central nervous system repair. Brain Res Bull 1999;49: Bolin LM, Shooter EM. Characterization of a Schwann cell neuritepromoting activity that directs motoneuron axon outgrowth. J Neurosci Res 1994;37: Duncan ID, Aguayo AJ, Bunge RP. Wood PM. Transplantation of rat Schwann cells grown in tissue culture into the mouse spinal cord. J Neurol Sci 1981;49: Paino CL, Fernandez-Valle C, Bates ML, Bunge MB. Regrowth of axons in lesioned adult rat spinal cord: Promotion by implants of cultured Schwann cells. J Neurocytol 1994;23: Li Y, Raisman G. Schwann cells induce sprouting in motor and sensory axons in the adult rat spinal cord. J Neurosci 1994;14: Xu XM, Guenard V, Kleitman N, Bunge MB. Axonal regeneration into Schwann cell-seeded guidance channels grafted into transected adult rat spinal cord. J Comp Neurol 1995;351: Xu XM, Guenard V, Kleitman N, Aebischer P, Bunge MB. A combination of BDNF and NT-3 promotes supraspinal axonal regeneration into Schwann cell grafts in adult rat thoracic spinal cord. Exp Neurol 1995;134: MartinD,RobeP,FranzenR,DelreeP,SchoenenJ,StevenaertA, Moonen G. Effects of Schwann cell transplantation in a contusion model of rat spinal cord injury. J Neurosci Res 1996;45: Montgomery CT, Tenaglia EA, Robson JA. Axonal growth into tubes implanted within lesions in the spinal cords of adult rats. Exp Neurol 1996;137: Chen A, Xu XM, Kleitman N, Bunge MB. Methylprednisolone administration improves axonal regeneration into Schwann cell grafts in transected adult rat thoracic spinal cord. Exp Neurol 1996;138: Xu XM, Chen A, Guenard V, Kleitman N, Bunge MB. Bridging Schwann cell transplants promote axonal regeneration from both the rostral and caudal stumps of transected adult rat spinal cord. J Neurocytol 1997; 26: Oudega M, Xu XM, Guenard V, Kleitman N, Bunge MB. A combination of insulin-like growth factor-i and platelet-derived growth factor enhances 44 q 2005 American Association of Neuropathologists, Inc.

9 JOBNAME: jnen 64# PAGE: 9 OUTPUT: Sat January 8 18:39: J Neuropathol Exp Neurol Volume 64, Number 1, January 2005 BMSC-SC for SCI myelination but diminishes axonal regeneration into Schwann cell grafts in the adult rat spinal cord. Glia 1997;19: Guest JD, Rao A, Olson L, Bunge MB, Bunge RP. The ability of human Schwann cell grafts to promote regeneration in the transected nude rat spinal cord. Exp Neurol 1997;148: Xu XM, Zhang SX, Li H, Aebischer P, Bunge MB. Regrowth of axons into the distal spinal cord through a Schwann-cell-seeded mini-channel implanted into hemisected adult rat spinal cord. Eur J Neurosci 1999;11: Imaizumi T, Lankford KL, Kocsis JD. Transplantation of olfactory ensheathing cells or Schwann cells restores rapid and secure conduction across the transected spinal cord. Brain Res 2000;854: Takami T, Oudega M, Bates ML, Wood PM, Kleitman N, Bunge MB. Schwann cell but not olfactory ensheathing glia transplants improve hindlimb locomotor performance in the moderately contused adult rat thoracic spinal cord. J Neurosci 2002;22: Dezawa M, Takahashi I, Esaki M, Takano M, Sawada H. Sciatic nerve regeneration in rats induced by transplantation of in vitro differentiated bone-marrow stromal cells. Eur J Neurosci 2001;14: Shapovalov AI. Neuronal organization and synaptic mechanisms of supraspinal motor control in vertebrates. Rev Physiol Biochem Pharmacol 1975;72: Villanueva L, Bernard JF, Le Bars D. Distribution of spinal cord projections from the medullary subnucleus reticularis dorsalis and the adjacent cuneate nucleus: A Phaseolus vulgaris-leucoagglutinin study in the rat. J Comp Neurol 1995;352: Bregman BS, Kunkel-Bagden E, Reier PJ, Dai HN, McAtee M, Gao D. Recovery of function after spinal cord injury: Mechanisms underlying transplant-mediated recovery of function differ after spinal cord injury in newborn and adult rats. Exp Neurol 1993;123: Grill R, Murai K, Blesch A, Gage FH, Tuszynski MH. Cellular delivery of neurotrophin-3 promotes corticospinal axonal growth and partial functional recovery after spinal cord injury. J Neurosci 1997;17: Jones LL, Oudega M, Bunge MB, Tuszynski MH. Neurotrophic factors, cellular bridges and gene therapy for spinal cord injury. J Physiol 2001; 533: Archelos JJ, Roggenbuck K, Schneider-Schaulies J, Linington C, Toyka KV, Hartung HP. Production and characterization of monoclonal antibodies to the extracellular domain of P0. J Neurosci Res 1993;35: Basso DM, Beattie MS, Bresnahan JC. A sensitive and reliable locomotor rating scale for open field testing in rats. J Neurotrauma. 1995;12: Chopp M, Zhang XH, Li Y, Wang L, Chen J, Lu D, Lu M, Rosenblum M. Spinal cord injury in rat: Treatment with bone marrow stromal cell transplantation. Neuroreport 2000;11: Hofstetter CP, Schwarz EJ, Hess D, Widenfalk J, El Manira A, Prockop DJ, Olson L. Marrow stromal cells form guiding strands in the injured spinal cord and promote recovery. Proc Natl Acad Sci USA. 2002;99: Wu S, Suzuki Y, Ejiri Y, Noda T, Bai H, Kitada M, Kataoka K, Ohta M, Chou H, Ide C. Bone marrow stromal cells enhance differentiation of cocultured neurosphere cells and promote regeneration of injured spinal cord. J Neurosci Res 2003;72: Terada N, Hamazaki T, Oka M, Hoki M, Mastalerz DM, Nakano Y, Meyer EM, Morel L, Petersen BE, Scott EW. Bone marrow cells adopt the phenotype of other cells by spontaneous cell fusion. Nature 2002;416: Alvarez-Dolado M, Pardal R, Garcia-Verdugo JM, Fike JR, Lee HO, Pfeffer K, Lois C, Morrison SJ, Alvarez-Buylla A. Fusion of bonemarrow-derived cells with Purkinje neurons, cardiomyocytes and hepatocytes. Nature 2003;425: q 2005 American Association of Neuropathologists, Inc. 45