Cytotherapy, 2010; 12: 841–845
Implantation of polymer scaffolds seeded with neural stem cells in a canine spinal cord injury model
BYUNG GON KIM1, YOUNG MI KANG1, JI HOON PHI3, YOON-HA KIM2, DONG HOON HWANG1, JUN YOUNG CHOI1, SUN RYU2, ALAA-ELDEN ELASTAL2, SUN HA PAEK3, KYU-CHANG WANG3, SEUNG-HOON LEE2, SEUNG U. KIM4,5 & BYUNG-WOO YOON2 1Brain
Disease Research Center, Institute for Medical Sciences and Department of Neurology, Ajou University School of Medicine, Suwon, Korea, 2Departments of Neurology and Neuroscience Research Center, Clinical Research Institute, Seoul National University Hospital, Seoul, Korea, 3Department of Neurosurgery, Seoul National University Hospital, Seoul, Korea, 4Department of Neurology, University of British Columbia,Vancouver, Canada, and 5Medical Research Institute, Chungang University School of Medicine, Seoul, Korea Abstract Background aims. Combinatorial approaches employing diverse therapeutic modalities are required for clinically relevant repair of injured spinal cord in human patients. Before translation into the clinic, the feasibility and therapeutic potential of such combinatorial strategies in larger animal species need to be examined. Methods. The present study tested the feasibility of implanting polymer scaffolds via neural stem cell (NSC) delivery in a canine spinal cord injury (SCI) model. The poly(lacticco-glycolic acid) (PLGA) scaffolds seeded with human NSC were implanted into hemisected canine spinal cord. Results. The PLGA scaffolds bridged tissue defects and were nicely integrated with residual canine spinal cord tissue. Grafted NSC survived the implantation procedure and showed migratory behavior to residual spinal cord tissue. Ectopic expression of a therapeutic neurotrophin-3 gene was also possible in NSC seeded within the PLGA scaffolds. Conclusions. Our description of a canine SCI model would be a valuable resource for pre-clinical trials of combinatorial strategies in larger animal models. Key Words: canine model, neural stem cells, neurotrophin-3, poly(lactic-co-glycolic acid) scaffold, spinal cord injury
Introduction and methods Spinal cord injury (SCI) causes damage not only to neural cells in the gray matter but also to axonal tracts and myelin sheaths in the white matter, which carry sensation and motor signals to and from the brain. These various pathologic events act in concert to induce severe neurologic dysfunctions at and below the injured spinal cord segment level. Intense research efforts over the last few decades have yielded several promising therapeutic approaches. Because of the multifactorial nature of the pathomechanisms induced by SCI, it is widely accepted that any single therapeutic strategy would not suffice to achieve clinically relevant functional improvements, and that future therapeutic strategies are likely to involve combinations of diverse therapeutic modalities (1,2). Cellular transplantation of neural stem or progenitor cells is regarded as a promising approach for SCI. Previous studies have reported enhanced functional
recovery in a rodent SCI model by transplantation of stem cell-based therapy (3,4). The development of combinatorial strategies centered on neural stem cell (NSC) utilization could potentiate the therapeutic efficacies of stem cell-based transplantation approaches. Secondary injuries often lead to the formation of cystic cavities after SCI, which severely impede the regeneration of spinal cord tissue (5). Various biocompatible polymer scaffolds have been employed in SCI repair strategies to bridge tissue defects. Combined implantation of polymer scaffolds with therapeutic cells can bridge lesion cavities, guide axonal growth and function as a means of delivering NSC (6,7). Several combinatorial strategies have proved to be effective in rodent SCI models (7–9). However, differences in size, anatomy and physiology of spinal cord between rodents and humans make it challenging to translate the results of rodent research to human patients (10). A recent survey of the SCI
Correspondence: Dr Byung-Woo Yoon, Department of Neurology, Seoul National University Hospital, 28 Yongon-dong, Jongno-gu, Seoul 110–744, Korea. E-mail:
[email protected] (Received 27 October 2009; accepted 11 June 2010) ISSN 1465-3249 print/ISSN 1477-2566 online © 2010 Informa Healthcare DOI: 10.3109/14653249.2010.501784
842
B. G. Kim et al.
research community revealed strong support for the demonstration of efficacy in large animal models, especially for an invasive cell transplantation therapy (11), warranting adoption of a canine SCI model to test the feasibility of a combinatorial repair strategy involving polymer scaffolds and cellular transplantation. The present study was of a combinatorial strategy comprising human NSC and a bridging biomaterial in a canine SCI model. We implanted polymer scaffolds seeded with human NSC after hemisection injury at the T11 level. We also attempted to overexpress the ectopic therapeutic gene (neurotrophin-3; NT3) in NSC to test the feasibility of combining diverse combinatorial strategies in this large animal model of SCI. The polymer scaffolds used in this study were purchased from REGEN Biotech (Seoul, Korea). Scaffolds were composed of 65:35 poly(lactic-co-glycolic acid) (PLGA) and were prepared using a gas-foaming/ salt-leaching method. Before cell seeding, PLGA scaffolds were sterilized by treating them with 70% ethanol in a –20°C freezer overnight. The next day, scaffolds were washed five times in distilled water containing antibiotics. To improve NSC penetration into the PLGA scaffolds, air bubbles were removed using a vacuum desiccator. The scaffolds were then washed with cell culture medium three times and incubated in the same medium overnight. The immortalized human NSC line (F3) was established and prepared as described previously (12). NT3 overexpressing human NSCs were produced by retroviral transduction of the F3 cells with human NT3 cDNA. Human NSCs were grown as attached in culture dishes, and dissociated into single cells by a brief trypsin treatment, and then suspended in medium at a density of 105 cells/μL. A total of 15 μL medium containing cells was slowly injected into the scaffold using a pipette tip. The PLGA scaffolds seeded with cells were transferred into a 12-well plate and incubated on a shaker at 37°C for at least 3 days before implantation. For scanning electron microscopy (SEM), PLGA scaffolds containing NSC were fixed with modified Karnovsky’s solution, dried using a critical point dryer, and imaged by scanning electromicroscope (SEM) (JSM-5410LV; JEOL, Tokyo, Japan). All animal procedures were approved by the Institutional Review Board of Seoul National University College of Medicine (Seoul, Korea). Female mongrel dogs (weighing 25–32 kg) were used in this study. Detailed surgical procedures and perioperative care can be found in the previous study (13). All animals had been appropriately vaccinated, and preoperative general health checkups were performed by attending veterinary doctors from the Department of Preclinical Experiments at the Clinical Research Institute, Seoul National University Hospital. Aseptic conditions at
the levels required for human surgery were enforced to minimize surgery-related infections. Animals were anesthetized with acepromazine hydrochloride (0.1 mg/kg) and ketamine hydrochloride (10 mg/kg) and maintained on a 1:1 mix nitrous oxide/oxygen containing halothane (0.5–0.7%). The dogs were intubated and mechanically ventilated to maintain normal respiration and acid–base balance. Body temperatures were monitored using an anal temperature probe and maintained using a heating pad. All animals received 0.9% normal saline solution, which was infused at 10 mL/kg/hour. All surgeries were performed by the same neurosurgeon (JHP). Total laminectomy was carried out to expose the T11 level of the thoracic spinal cord. The dural sac was opened at the midline and a 5-mm long longitudinal midline myelotomy was performed. Through a pial incision, a transverse left hemisection was made at both ends of the midline myelotomy, and a 5-mm long left hemicolumn of the spinal cord was removed via subpial dissection. Remaining spinal cord tissue was removed by vacuum suction until the vertebral body at the ventral side was visualized. Immediately after the hemicordectomy, a PLGA scaffold seeded with NSC was implanted into the tissue defect. Complete dural closure was performed and the paraspinal muscles and skin were closed in layers. On the day of surgery, dogs were injected with antibiotics and nonsteroidal anti-inflammatory drugs (NSAID) to prevent infections and reduce pain. Bowels and bladders were evacuated twice daily until reflex bladder function was re-established. After the operation, dogs were taken care of by a veterinary doctor according to the standards and guidelines set out by the Department of Animal Experiment and the Office of Laboratory Animal Welfare at Seoul National University. Immunosuppressant was not administered because previous studies had shown long-term survival of human NSC without any immunosuppressant (13,14). Detailed experimental methods are available in the supplementary methods section available on-line. Results and discussion In order to implant scaffolds that fitted hemisected cavities of canine spinal cord, the PLGA scaffold was tailored to have a half-cylinder form with the specified dimensions as described in Figure 1A. To seed NSC into the PLGA scaffold, NSC (total 1.5 ⫻ 106 cells) were slowly injected into the scaffolds and incubated for 3 days before implantation. With SEM, seeded NSC appeared healthy and were grown attached to the walls of PLGA micropores (Figure 1B). We also determined whether a therapeutic target gene transduced into the human NSC could be expressed inside the PLGA scaffold. We generated NT3 overexpressing
Combinatorial NSC therapy in a canine SCI model
843
Figure 1. Implantation of PLGA scaffolds seeded with NSC. (A) Specified dimensions of the PLGA scaffold that was tailored to fit hemisected lesion cavities of canine spinal cord. (B) SEM image of NSC within the PLGA scaffold. NSC appeared to grow attached to the walls of micropores. (C) The hemisected canine spinal cord was imaged immediately after removal of left-side hemicord. Arrows indicate the extent of lesion cavity. (D) Implantation of PLGA scaffold into a cavity created by hemisection. (E) A gross morphology of spinal cord tissue dissected at 12 weeks post-surgery. Arrows indicate the margin of hemisection lesion and remaining PLGA scaffold. A dotted line indicates a plane of transverse tissue section in (F). (F) Eriochrome cyanine RC/cresyl violet staining of transverse spinal cord tissue section. Arrows indicate the interface between remaining PLGA scaffold (PL) and residual spinal cord tissue. Scale bar ⫽ 1 mm. (G–L) Visualization of grafted human NSC on the basis of immunoreactivity against human-specific mitochondrial antigen in injured canine spinal cord. (G) A diagram of transverse spinal cord section depicting regions of the following images (H–L). Implanted PLGA scaffold is illustrated as a gray region. A dotted line indicates the margin of residual gray matter. The boxes with alphabets above indicate the regions where the corresponding figure panels were imaged. (H) NSC inside the PLGA scaffold at 2 weeks post-surgery. Insets are magnified images with DAPI counterstaining in the boxed areas. (I, J) Migration of parental NSC (I) or NT3 overexpressing NSC (J) at the interface between the PLGA scaffold and spinal cord tissue at 2 weeks post-surgery. Arrows indicate migrating cells detected in residual spinal cord tissue. Dotted lines indicate the interface between PLGA scaffold and residual spinal cord tissue. (K, L) Very few NSC were detected in the PLGA scaffold or in residual spinal cord tissue (K), whereas NT3 overexpressing NSC showed improved survival within the PLGA scaffolds (L). Scale bars in (H–L) indicate 100 μm.
human NSC by retroviral transduction of parental NSC, and measured the NT3 level in culture media by ELISA as an index of gene expression. The NT3 concentration was 450 ⫾ 11.0 pg/mL in culture media collected after incubating the PLGA scaffolds seeded with NT3 overexpressing NSC. In comparison, the mean concentration of NT3 from dissociated NT3 overexpressing NSC was 859 ⫾ 13.9 pg/mL. Parental NSC, whether dissociated or grown inside the PLGA scaffold, produced only barely detectable levels of NT3. These findings indicated that, although the extent of expression was diminished to some extent, it was possible to achieve a significant degree of transgene expression in the NSC seeded inside the PLGA scaffolds.
To produce a canine SCI model, left-side hemicord spanning approximately 5 mm was removed surgically at the T11 level (Figure 1C). Cultured PLGA scaffolds seeded with NSC were implanted into lesion cavities immediately after injury (Figure 1D). The predesigned PLGA scaffold of half-cylinder form was fitted snuggly into the hemisected cavity. At 12 weeks post-surgery, the tissue defect created by hemisection injury was found to be bridged by the implanted PLGA scaffolds (Figure 1E). The PLGA scaffolds seemed to be well apposed to lesion cavities, and showed varying degrees of biodegradation. Cross-sections of the spinal cord tissues at the center of the lesion cavity revealed an uninterrupted interface between scaffolds and residual spinal cord tissue
844
B. G. Kim et al.
without microcavities (Figure 1F). This finding indicated that the PLGA scaffolds were biocompatible with the remaining canine spinal cord tissue. Glial scar formation was evaluated on transverse spinal cord sections immunostained with an astroglial marker, glial fibrillary acidic protein (GFAP). GFAP-positive astrocytes were occasionally found inside the scaffolds but most reactive astrocytes were aligned in residual tissue along the interface. GFAP-staining intensities were slightly higher along the interface between scaffolds and spinal cord tissue, but no exaggerated astroglial reactions were noted (data not shown). Grafted NSC were readily detected in animals killed at 2 weeks post-surgery (Figure 1G–J). Surviving NSC were identified based on immunoreactivity against human-specific mitochondrial antigen. Clusters of grafted NSC were frequently observed inside the PLGA scaffolds at 2 weeks post-surgery (Figure 1H), indicating that NSC implanted inside the PLGA scaffolds survived the transplantation procedure. The average number of surviving NSC was 112,648 (7.5% of transplanted cells). At the interface, surviving NSC were observed to migrate into residual spinal cord tissue (Figure 1I). Some NSC were also detected at the lateral column of residual white matter, several millimeters away from scaffolds, suggesting that they migrated a considerable distance. We also implanted PLGA scaffolds seeded with NSC overexpressing a therapeutic target gene, NT3. NT3 overexpression slightly increased the number of surviving cells at this time point (139,238; 9.3% of transplanted cells). NT3 overexpressing NSC tended to migrate to the spinal cord tissue more frequently than parental NSC (Figure 1J). The percentage of migrating cells was more than two-fold higher for NT3 overexpressing NSC than for parental NSC (7.4% and 3.2% of the total number of surviving cells, respectively). At 12 weeks post-surgery, very few surviving parental NSC were detected either inside scaffolds or in residual spinal cord tissue (Figure 1K). Clusters of surviving cells were observed within the PLGA scaffold in three of nine animals that received the PLGA scaffolds with NT3 overexpressing NSC (Figure 1L). Migrating NSC in the contralateral white matter occasionally differentiated into oligodendrocytes, but not into astrocytic or neuronal lineage. Most NSC inside the scaffolds were immunoreactive for nestin (data not shown), a marker of immature NSC, suggesting that the grafted NSC residing within the PLGA scaffolds did not differentiate but maintained their immature characteristics. The present study tested the feasibility of therapeutic cell delivery using a polymer scaffold in a canine SCI model. Biocompatible polymer scaffolds have been employed to bridge lesion cavities in SCI. The utility of the polymer scaffolds has been extended
to functioning as a cell-delivery vehicle. Previous rodent studies have reported successful cell therapy in combination with various artificial polymer scaffolds (8,9,15). To the authors’ knowledge, however, there has not been any report on the use of polymer scaffolds for cell therapies in animals larger than rodents. Differences in the size of spinal cord between animal species may present a critical issue in designing therapeutic strategies utilizing scaffolds and cellular transplantation (16). For example, larger interface regions between the scaffolds and spinal cord tissue may render the scaffold integration more challenging in larger animal species. Furthermore, supplying blood and nutrients to the cells inside larger scaffolds may be more demanding. The optimal number of cells required to populate the spinal cord also depends on the size of the cord. Therefore, the number of cells needed for transplantation in human spinal cord should be determined using a larger animal model with a cord size similar to humans. In addition to the dimensional issues, the anatomical and physiologic differences in the supraspinal pathways warrant caution in interpreting any cell-therapy results in rodent studies (10). The major corticospinal spinal tract runs in the lateral funiculus in humans, but in the dorsal column in rodents (17). The difference in anatomical arrangement of the corticospinal tract may reflect the different physiologic significance of the corticospinal tract in voluntary motor function between the two species. In contrast, dogs have the corticospinal tract in the lateral column of the spinal cord (18), suggesting that the physiologic basis underlying a locomotor recovery in canine species is closer to humans than rodents. Therefore, the canine SCI model for cell transplantation therapy in combination with polymer scaffolds would be a valuable system for pre-clinical studies evaluating the efficacy of a newly developed combinatorial cell therapy. Our data demonstrate that the implantation of polymer scaffolds seeded with NSC is feasible in a larger animal SCI model. The dogs in this study weighed from 25 to 32 kg (approximately 100 times heavier than rats), and their spinal cord size was comparable to that of human spinal cord (Figure 1E). Implanted PLGA scaffolds bridged tissue defects created by surgical spinal cord lesion and did not show signs of incompatibility with the host canine spinal cord tissue. The astroglial reactions along the interface were not notably intense. The NSC seeded within the PLGA scaffolds were found to survive within the scaffolds at least until 2 weeks after surgery. They also migrated to the host spinal cord, indicating that the PLGA scaffolds functioned as a reservoir to provide NSC and possibly various secretory factors from them to the residual spinal cord (14,19). The poor long-term survival of grafted NSC
Combinatorial NSC therapy in a canine SCI model indicated that a complementary measure needs to be combined to improve survival of grafted cells (20). We also showed that ectopic expression of a therapeutic gene is possible in NSC seeded within the PLGA scaffolds. NT3 overexpressing NSC tended to migrate more frequently and improve long-term survival. Therefore, the present study shows that the polymer scaffolds can provide structural support and deliver therapeutic cells and genes simultaneously, and thereby play a central role in the integration of diverse combinatorial repair strategies in a canine SCI model (6). In summary, the present study shows that a therapeutic strategy involving polymer scaffolds with cellular therapy is feasible in a canine SCI model. Our description of a canine SCI model should be a valuable resource for subsequent trials of combinatorial strategies in larger animal models. Acknowledgments This study was supported by the 21st Century Frontier Research Fund of the Ministry of Science & Technology (SC-3111; BWY), Republic of Korea, and Ajou University Guwon Scholarship Fund (BGK). Disclosure of interest: All contributing authors have no conflict of interest for this study. References 1. Fouad K, Schnell L, Bunge MB, Schwab ME, Liebscher T, Pearse DD. Combining Schwann cell bridges and olfactoryensheathing glia grafts with chondroitinase promotes locomotor recovery after complete transection of the spinal cord. J Neurosci. 2005;25:1169–78. 2. Schwab JM, Brechtel K, Mueller CA, Failli V, Kaps HP, Tuli SK, et al. Experimental strategies to promote spinal cord regeneration: an integrative perspective. Prog Neurobiol. 2006;78:91–116. 3. Tetzlaff W, Okon EB, Karimi-Abdolrezaee S, Hill CE, Sparling JS, Plemel JR, et al. A systematic review of cellular transplantation therapies for spinal cord injury. J Neurotrauma 2010 April 20. [Epub ahead of print]. 4. Kim BG, Hwang DH, Lee SI, Kim EJ, Kim SU. Stem cellbased cell therapy for spinal cord injury. Cell Transplant. 2007;16:357–66. 5. Bunge MB. Bridging areas of injury in the spinal cord. Neuroscientist. 2001;7:325–39.
Supplementary material available online Supplementary methods
845
6. Nomura H, Tator CH, Shoichet MS. Bioengineered strategies for spinal cord repair. J Neurotrauma. 2006;23:496–507. 7. Hatami M, Mehrjardi NZ, Kiani S, Hemmesi K, Azizi H, Shahverdi A, et al. Human embryonic stem cell-derived neural precursor transplants in collagen scaffolds promote recovery in injured rat spinal cord. Cytotherapy. 2009;11:618–30. 8. Nomura H, Zahir T, Kim H, Katayama Y, Kulbatski I, Morshead CM, et al. Extramedullary chitosan channels promote survival of transplanted neural stem and progenitor cells and create a tissue bridge after complete spinal cord transection. Tissue Eng Part A. 2008;14:649–65. 9. Teng YD, Lavik EB, Qu X, Park KI, Ourednik J, Zurakowski D, et al. Functional recovery following traumatic spinal cord injury mediated by a unique polymer scaffold seeded with neural stem cells. Proc Natl Acad Sci USA. 2002;99: 3024–9. 10. Blesch A, Tuszynski MH. Spinal cord injury: plasticity, regeneration and the challenge of translational drug development. Trends Neurosci. 2009;32:41–7. 11. Kwon BK, Hillyer J, Tetzlaff W. Translational research in spinal cord injury: a survey of opinion from the sci community. J Neurotrauma. 2010;27:21–33. 12. Kim SU, Nagai A, Nakagawa E, Choi HB, Bang JH, Lee HJ, et al. Production and characterization of immortal human neural stem cell line with multipotent differentiation property. Methods Mol Biol. 2008;438:103–21. 13. Lee SH, Chung YN, Kim YH, Kim YJ, Park JP, Kwon DK, et al. Effects of human neural stem cell transplantation in canine spinal cord hemisection. Neurol Res. 2009;31:996–1002. 14. Lee HJ, Kim KS, Kim EJ, Choi HB, Lee KH, Park IH, et al. Brain transplantation of immortalized human neural stem cells promotes functional recovery in mouse intracerebral hemorrhage stroke model. Stem Cells. 2007;25:1204–12. 15. Sykova E, Jendelova P, Urdzikova L, Lesny P, Hejcl A. Bone marrow stem cells and polymer hydrogels: two strategies for spinal cord injury repair. Cell Mol Neurobiol. 2006; 26:1113–29. 16. Jeffery ND, Smith PM, Lakatos A, Ibanez C, Ito D, Franklin RJ. Clinical canine spinal cord injury provides an opportunity to examine the issues in translating laboratory techniques into practical therapy. Spinal Cord. 2006;44:584–93. 17. Terashima T, Ochiishi T Yamauchi T. Immunohistochemical detection of calcium/calmodulin-dependent protein kinase II in the spinal cord of the rat and monkey with special reference to the corticospinal tract. J Comp Neurol. 1994; 340:469–79. 18. King JL. The cortico-spinal tract of the rat. Anat Rec (Hoboken). 1910;4:245–52. 19. Lu P, Jones LL, Snyder EY, Tuszynski MH. Neural stem cells constitutively secrete neurotrophic factors and promote extensive host axonal growth after spinal cord injury. Exp Neurol. 2003;181:115–29. 20. Lee SI, Kim BG, Hwang DH, Kim HM, Kim SU. Overexpression of Bcl-XL in human neural stem cells promotes graft survival and functional recovery following transplantation in spinal cord injury. J Neurosci Res. 2009;87:3186–97.