SEARCH

SEARCH BY CITATION

Keywords:

  • Spinal cord injury;
  • Induced pluripotent stem cells;
  • Neural stem cells;
  • Transplantation;
  • Regenerative medicine

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Because of their ability to self-renew, to differentiate into multiple lineages, and to migrate toward a damaged site, neural stem cells (NSCs), which can be derived from various sources such as fetal tissues and embryonic stem cells, are currently considered to be promising components of cell replacement strategies aimed at treating injuries of the central nervous system, including the spinal cord. Despite their efficiency in promoting functional recovery, these NSCs are not homogeneous and possess variable characteristics depending on their derivation protocols. The advent of induced pluripotent stem (iPS) cells has provided new prospects for regenerative medicine. We used a recently developed robust and stable protocol for the generation of long-term, self-renewing, neuroepithelial-like stem cells from human iPS cells (hiPS-lt-NES cells), which can provide a homogeneous and well-defined population of NSCs for standardized analysis. Here, we show that transplanted hiPS-lt-NES cells differentiate into neural lineages in the mouse model of spinal cord injury (SCI) and promote functional recovery of hind limb motor function. Furthermore, using two different neuronal tracers and ablation of the transplanted cells, we revealed that transplanted hiPS-lt-NES cell-derived neurons, together with the surviving endogenous neurons, contributed to restored motor function. Both types of neurons reconstructed the corticospinal tract by forming synaptic connections and integrating neuronal circuits. Our findings indicate that hiPS-lt-NES transplantation represents a promising avenue for effective cell-based treatment of SCI. STEM CELLS2012;30:1163–1173


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Our current ability to reconstruct the damaged central nervous system, including spinal cord injury (SCI), is limited. SCI is one of the commonest causes of loss of movement and sensation below the level of the injured spinal cord. While partial spontaneous functional recovery has been observed in the case of moderate SCI, there is no adequate treatment to repair the injured spinal cord itself, and most patients have no therapeutic options except for general management and rehabilitation [1].

Transplantation of neural stem cells (NSCs) into the injured spinal cord has been shown to be an effective treatment [2] because of their competence to differentiate into neurons and oligodendrocytes and to secrete neurotrophic factors. Transplantation into injured spinal cords of several different types of human NS/progenitor cells, derived from human fetal tissue [3, 4] or from human embryonic stem cells (hESCs) [5, 6], promotes functional recovery in animal models. Nonetheless, the mechanism underlying such functional improvement remains to be fully elucidated.

The establishment of induced pluripotent stem cells (iPSCs) offers new prospects for regenerative therapies [7, 8]. Human iPSCs (hiPSCs) can be generated from cells in adult tissue, making it possible to create iPSCs from SCI patients themselves. From the viewpoint of ethics and host immune rejection, NSCs derived from human iPSCs (hiPS-NSCs) appear to be an ideal resource for transplantation therapy, and the efficacy of hiPS-NSC transplantation in SCI treatment has just begun to be investigated using the mouse model of SCI [9].

We used a recently developed protocol for the generation of long-term self-renewing neuroepithelial-like stem (lt-NES) cells, which satisfy the criteria to be defined as NSCs, from several different lines of hESCs and hiPSCs [10, 11]. hiPSC-derived lt-NES (hiPS-lt-NES) cells exhibit consistent characteristics such as continuous expandability, stable neuronal and glial differentiation ability, and the capacity to generate functional mature neurons in monolayer culture. Whereas neurosphere cultures display heterogeneous character and are sensitive to variation in methodological procedures [12], monolayer cultures offer a more homogeneous and robust cell generation [13, 14].

We report here that transplanted hiPS-lt-NES cells, derived from our robust and stable monolayer cultures, have a therapeutic potential comparable to that of NSCs from human fetal spinal cord (hsp-NSCs) for SCI in the nonobese diabetic-severe combined immunodeficient (NOD-SCID) mouse model. We further show that hiPS-lt-NES cell transplantation promotes recovery of hind limb motor function through the reconstruction of the corticospinal tract (CST), and restores disrupted neuronal circuitry in a relay manner as we have previously demonstrated in a study of mouse NSC transplantation [15]. Our results suggest that hiPS-lt-NES cells represent a promising cell source for transplantation into the injured spinal cord.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Cell Culture

hsp-NSCs and hiPS-lt-NES cells were established and maintained as described previously [10, 11, 16]. The hsp-NS cell line CB660sp and hiPS-lt-NES cell line AF22 were used in the present study. hsp-NSCs were plated onto 10 μg/ml laminin (Sigma, St. Louis, MO, http://www.sigmaaldrich.com)-coated plates in maintenance medium, consisting of Euromed-N medium (Euroclone, Milano, Italy, http://www.euroclonegroup.it), 2 mM L-glutamine, 0.1 mg/ml penicillin/streptomycin (Sigma), N2 supplement (1:100), 20 ml/l B27 (all from Invitrogen, Carlsbad, CA, http://www.invitrogen.com), 10 ng/ml fibroblast growth factor (FGF, R&D Systems, Minneapolis, MN, http://www.rndsystems. com)-2, and 10 ng/ml epidermal growth factor (EGF, R&D Systems). Cells were passaged at a ratio of 1:2 every second to third day using Accutase (Sigma). hiPS-lt-NES cells were plated onto 0.1 mg/ml poly-L-ornithine and 10 μg/ml laminin (O/L, Sigma)-coated plates in maintenance medium, consisting of Dulbecco's modified Eagle's medium/F12 (Invitrogen), 2 mM L-glutamine, 1.6 mg/ml glucose, 0.1 mg/ml penicillin/streptomycin, N2 supplement (1:100), 1 μl/ml B27, 10 ng/ml FGF2 and 10 ng/ml EGF. Cells were passaged at a ratio of 1:3 every second to third day using trypsin. To induce in vitro differentiation, hiPS-lt-NES cells were plated onto an O/L-coated 35-mm dish at a density of 5 × 105 cells per dish in culture medium without both EGF and FGF, containing 1% fetal bovine serum (FBS), and cultured for 4 weeks. Half of the medium was changed every 2 days and laminin (1:500) was added to the medium once a week to prevent detachment of the cells.

Lentivirus Production and Infection of hsp-NSCs and hiPS-Lt-NES Cells

Lentivirus production and infection of cells were performed as described previously [17, 18]. hsp-NSCs and hiPS-lt-NES cells were infected with lentiviruses harboring the luciferase and green fluorescent protein (GFP) genes connected by an internal ribosomal entry site (IRES), EFp (elongation factor promoter)- luciferase-IRES-GFP (Fig. 1B). hsp-NSCs and hiPS-lt-NES cells that had undergone more than 20 passages were infected. GFP-positive cells were collected by fluorescence activated cell sorting (FACS)-Aria II CellSorter (B.D. Biosciences, San Jose, CA, http://www.bdbiosciences.com) and used for transplantation.

thumbnail image

Figure 1. Characterization of hiPS-lt-NES cells. (A): Expansion culture. hiPS-lt-NES cells can be expanded continuously in the presence of both epidermal growth factor and fibroblast growth factor. hiPS-lt-NES cells were uniformly immunopositive for Sox2 (green) and Nestin (red). Hoechst (blue) shows nuclear staining. (B): The pCSII-EFp-luciferase-IRES2-GFP construct. (C): Flow cytometric analysis of GFP-positive cells in hiPS-lt-NES cells. Uninfected cells (left) and cells infected with lentiviruses expressing luciferase and GFP (middle) were subjected to flow cytometric analysis. GFP-positive cells in the infected cell population were sorted on the basis of GFP fluorescence and reanalyzed for GFP expression (right). (D): After 4 weeks in differentiation conditions, infected hiPS-lt-NES cells differentiated into Tuj1-positive neurons (red) and GFAP-positive astrocytes (blue). (E): Quantification of neural marker-positive cells after 4 weeks' differentiation. Lentiviral infection did not influence the differentiation tendency. Data are means ± SD (n = 3). Scale bars = 100 μm. Abbreviations: EFp, elongation factor promoter; GFP, green fluorescent protein; GFAP, glial fibrillary acidic protein; hiPS-lt-NES, human induced pluripotent stem cell-derived long-term self-renewing neuroepithelial-like stem; and IRES2, internal ribosomal entry site 2.

Download figure to PowerPoint

SCI Model and Cell Transplantation

All aspects of animal care and treatment were carried out according to the guidelines of the experimental animal care committee of Nara Institute of Science and Technology. We used female NOD-SCID mice (8-10 weeks old, weighing 18-20 g, Charles River, Osaka, Japan, http://www.crj.co.jp). Anesthetized (ketamine 50 mg/kg, xylazine 5 mg/kg, and sodium pentobarbital 20 mg/kg) mice received laminectomies and partial laminectomies at the ninth and 10th thoracic spinal vertebrae, respectively. The dorsal surface of the dura mater was exposed and SCI was induced using an SCI device (70 kdyn; Infinite Horizon impactor, Precision Systems & Instrumentation, Lexington, KY, http://www.presysin.com) as previously described [15]. The muscle and skin were closed in layers. All mice subcutaneously received gentamicin (8 mg/kg) daily. The mice underwent manual bladder evacuation once a day. Seven days after injury, mice were anesthetized and transplanted with hiPS-lt-NES cells or hsp-NSCs using a glass micropipette attached to a stereotaxic injector (Narishige, Tokyo, Japan, http://www.narishige.co.jp). The tip of the micropipette was inserted into the injury epicenter in the injured spinal cord, and 2 μl of culture medium lacking growth factors, with or without NSCs (5 × 105 μl−1), was injected at a rate of 1 μl/min.

Behavioral Testing and Electrophysiological Recordings

We evaluated the motor function of the hind limbs for up to 10 weeks after injury. Two individuals, blinded to the treatment of the mice, examined motor function using the Basso Mouse Scale (BMS) locomotor rating scale [19]. Hind limb movements of the mice were captured using a high-definition digital camcorder. We edited these movies and exported movie files using editing software.

To examine signal conduction in motor pathways after SCI, motor-evoked potentials (MEPs) at 12 weeks after SCI were measured. Mice were anesthetized with intraperitoneally injected ketamine (100 mg/kg) and their heads were fixed in a stereotactic frame. The skulls were tightly fixed to a stereotaxic apparatus (Narishige). Scalping and two small craniotomies were performed with a drill over the motor cortex area (Nakanishi, Tochigi, Japan; http://www.nsk-nakanishi.co.jp). Silver ball electrodes were placed epidurally via the holes, into which mineral oil was applied. The motor cortex was stimulated with 0.2-msecond square wave pulses at a constant current of 10 mA using an electrical stimulator (SEM-3301, Nihon Kohden, Tokyo, Japan; http://www.nihonkohden.co.jp). Recording needle electrodes were inserted into the hamstring. A subcutaneous ground electrode was placed in the tail. Electrophysiological recordings were made (MED64, Alpha MED Scientific, Osaka, Japan; http://www. amedsci.com) and band-pass filtered at 1-10 kHz. Amplitudes from onset to peak of the negative deflection were measured. Latency of MEP was measured as the time interval between the end of stimulation and the onset of the first wave. Indicated values are the average of five experiments from each mouse.

Antibodies

The following antibodies were used: rabbit anti-Sox2 (1:1,000, Chemicon-Millipore, Billerica, MA, http://www.millipore.com), mouse anti-Nestin (1:250, Chemicon), rabbit anti-brain lipid-binding protein (BLBP, 1:500, Abcam, Cambridge, U.K., http://www.abcam.com), mouse anti-β-tubulin isotype III (Tuj1; 1:500, Sigma), rabbit anti-β-tubulin isotype III (Tuj1; 1:1,000, Covance, Madison, WI, http://www.covance.com), rabbit anti-GFP (1:1,000, Molecular Probes, Carlsbad, CA, http://probes.invitrogen.com), chick anti-GFP (1:500, Aves Labs, Tigard, OR, http://www.aveslab.com), rabbit anti-glial fibrillary acidic protein (GFAP, 1:2,000, DAKO, Carpinteria, CA, http://www.dako.com), mouse anti-human-specific GFAP (hGFAP, 1:1,000, STEM123, StemCells Science, Newark, CA, http://www.stemcellsinc.com), mouse anti-MAP2ab (1:500, Sigma), chick anti-myelin basic protein (MBP) (1:200, Aves Labs), mouse anti-adenomatous polyposis coli CC-1 (APC, 1:200, Calbiochem-Merck, Darmstadt, Germany, http://www.merckgroup.com), mouse anti-synaptophysin (1:200, Chemicon), mouse anti-Bassoon (Bsn, 1:400, Stressgen-Enzo Life Sciences, Plymouth Meeting, PA, http://www.enzolifesciences.com), mouse anti-human-specific synaptophysin (hSyn, 1:200, Chemicon), and mouse anti-NeuN (1:500, Millipore).

Immunocytochemistry

Immunocytochemistry experiments were performed as described previously [20]. Cells were washed with phosphate buffered saline (PBS) and fixed with 4% paraformaldehyde (PFA) in PBS for 10 minutes, and then washed with PBS and incubated in blocking solution (PBS containing 10% FBS and 0.1% Triton X-100). Subsequently, cells were incubated overnight at 4°C with the primary antibodies described above. After three washes in PBS, cells were incubated for 1 hour with the following secondary antibodies: FITC-conjugated donkey anti-chick/rabbit, Cy3-conjugated donkey anti-chick/mouse/rabbit, Cy5-conjugated donkey anti-mouse/rabbit (1:500, Jackson ImmunoResearch, West Grove, PA, http://www.jacksonimmuno.com). After three washes with PBS, nuclei were stained with Hoechst (bisbenzimide H33258 fluorochrome trihydrochloride, Calbiochem-Merck). Samples were washed with PBS and mounted on glass slides with Immu-Mount (Thermo Scientific, Waltham, MA, http://www.thermoscientific.com). The cells were examined using a fluorescence microscope (Axiovert 200M, Zeiss, Jena, Germany, http://www.zeiss.com) equipped with the appropriate epifluorescence filters.

Immunohistochemistry

Animals were anesthetized and perfused with PBS followed by 4% PFA in 0.1 M PBS, pH 7.4. The spinal cords were dissected and postfixed overnight in the same fixative at 4°C. For cryosectioning, fixed tissues were cryoprotected in 10% sucrose in PBS overnight at 4°C, then in 20% sucrose in PBS overnight at 4°C, and embedded in optimal cutting temperature (OCT) compound (Tissue Tek, Sakura Finetek, Tokyo, Japan, http://www.sakura-finetek.com). Cryostat sections (20 μm) were cut and affixed to matsunami adhesive slide (MAS)-coated glass slides (Matsunami Glass, Osaka, Japan, http://www.matsunami-glass.co.jp). The sections were permeabilized in PBS-T (PBS containing 0.1% Triton X-100) for 10 minutes and blocked with 10% FBS in PBS-T for 1 hour, and then incubated overnight at 4°C with the primary antibodies described above. After three washes with PBS, they were incubated in a mixture of the secondary antibodies described above for 1 hour. After a final rinse with PBS, nuclei were stained using Hoechst. Sections were mounted and examined under a fluorescence microscope (Axiovert 200M, Zeiss) and a scanning laser confocal imaging system (LSM 710, Zeiss).

In Vivo Imaging of Transplanted Cells

In vivo imaging experiments were performed as described previously [18]. A Xenogen-IVIS 100 cooled CCD optical macroscopic imaging system (SC BioScience, Tokyo, Japan, www.scbio.co.jp) was used for bioluminescence imaging. Mice were given an intraperitoneal injection of D-luciferin (150 mg/kg) and serial images were acquired from 20 minutes after administration until the maximum intensity was obtained with the field-of-view set at 10 cm. All images were analyzed using Igor (WaveMetrics, Portland, OR, http://www.wavemetrics.com) and Xenogen Living Image software, and optical signal intensity was expressed as photon flux in units of photons/s per cm2/steradian. To quantify the measured light, we defined regions of interest (ROIs) over the cell-implanted area and examined all values within the same ROI. The obtained photon count intensity was expressed as a percentage of the initial value.

Anterograde Labeling of the CST

Twelve weeks after injury, descending CST fibers were labeled with biotinylated dextran amine (BDA) (MW 10,000, 10% in saline, 2 μl per cortex; Molecular Probes) [21, 22] by injection into the left and right motor cortices [23]. The skulls of anesthetized mice were tightly fixed to a stereotaxic apparatus (Narishige). Scalping and craniotomy over the motor cortex area were carried out using a micromotor system (Nakanishi). The injection site was 2.1 mm posterior to the bregma, 2 mm lateral to the bregma, and 0.7 mm in depth [23]. For pressure injections with a 20-μm outer diameter glass capillary attached to a microsyringe (Narishige), we used 0.1-μl steps per minute until the desired volume (1 μl per injection site) was injected. Two weeks later, the animals were perfused and their spinal cords were fixed as described above. Sections (30 μm) were cut and used for immunohistochemistry. To visualize the BDA, a tyramide signal amplification fluorescence system (Perkin Elmer, Waltham, MA, http://www.perkinelmer.com) was used.

Visualization of Multisynaptic Neural Pathways

To visualize selective and functional transsynaptic neural pathways, wheat germ agglutinin (WGA)-expressing recombinant adenoviruses were used [24, 25]. Twelve weeks after injury, 1 μl of saline containing WGA-expressing virus at a titer of 1 × 1011 pfu/ml was injected into the bilateral motor cortices (0.5 μl per injection site). The injection site was 2.1 mm posterior to the bregma, 2 mm lateral to the bregma, and 0.7 mm in depth. Two weeks later, animals were perfused and the spinal cords were fixed as described above. Sections (15 μm) were cut and used for immunohistochemistry. Rabbit anti-WGA polyclonal antibody (3 μg/ml, Sigma) was preabsorbed with 1% acetone powder of mouse brains in blocking solution overnight at 4°C.

Ablation of Transplant-Derived Cells

Cell ablation experiments were performed as described previously [26, 27]. Diphtheria toxin (DT) was purified from conditioned medium of the PW8 strain of Corynebacterium diphtheriae by diethylaminoethyl Sepharose column chromatography and diluted to an appropriate concentration with saline. Seven weeks after injury, DT solution (50 μg/kg per day × 2 days) was administered by intraperitoneal injection into seven hiPS-NES cell-transplanted mice.

Statistical Analysis

We performed statistical analysis with an unpaired two-tailed Student's t test for single comparisons. For &&&BMS and BDA fiber analysis, we used repeated-measures analysis of variance (ANOVA) with Tukey-Kramer multiple comparison test at each point (Prism, GraphPad, LA Jolla, CA, http://www.graphpad.com). p < .05 was considered significant.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Characterization of hiPS-Lt-NES Cells In Vitro

We have previously shown that hiPS-NES cells, which are reliably derived from different hiPS cell lines, exhibit consistent characteristics such as continuous expandability, stable neuronal and glial differentiation, and the capacity to generate functional mature human neurons [11]. hiPS-lt-NES cells can be expanded in the presence of FGF2 and EGF in monolayer culture, and express the NSC markers Sox2 and Nestin (Fig. 1A) and the radial glial marker BLBP (Supporting Information Fig. 1), but not the ES/iPS cell markers Oct3/4 or Nanog (not shown).

To visualize transplanted cells by both luminescence and fluorescence, hiPS-lt-NES cells were infected with lentiviruses engineered to express luciferase and GFP (Fig. 1B). Almost all hiPS-lt-NES cells were infected (Fig. 1C, middle), and we isolated only strongly GFP-expressing cells for use in subsequent in vitro and transplantation experiments (Fig. 1C, right).

We then examined whether lentiviral infection affected the differentiation potential of hiPS-lt-NES cells. Uninfected and infected hiPS-lt-NES cells were induced to differentiate by removal of growth factors and cultured in the presence of 1% FBS for 4 weeks. Both hiPS-lt-NES cell population differentiated into large and small numbers of Tuj1-positive neurons and GFAP-positive astrocytes, respectively, as observed in our previous study [11] (Fig. 1D). Quantitative analysis revealed no significant differences in the proportions of differentiated cells between uninfected and infected hiPS-lt-NES cells (Fig. 1E). Thus, we concluded that lentiviral infection did not influence the differentiation potential of hiPS-NES cells.

Transplantation of hiPS-Lt-NES Cells into the Injured Spinal Cord Improves Functional Recovery of Hind Limbs

In the present study, we used NOD-SCID mice, which are a suitable model for xenograft research because they are constitutively immunodeficient; their pathology and innate immune response after SCI are also similar to those of other mouse strains [4, 9, 28].

Previous studies have already shown that transplantation of human NSCs from fetal spinal cord or brain tissue improves locomotor functional recovery of SCI models [3, 4]. Nevertheless, only recently has the efficacy of hiPS-NSC transplantation into SCI begun to be investigated [9]. To determine whether our hiPS-lt-NES cells have a therapeutic capability, we first sought to compare the effects of these cells with those of human fetal spinal cord-derived NSCs (hsp-NSCs). Seventy-kilodyne contusive SCI was applied at the ninth thoracic vertebral level of mouse spinal cords, and 1 week later we injected medium alone (SCI control), or medium containing hsp-NSCs or hiPS-lt-NES cells, into the epicenter. We then monitored the animals' hind limb motor function using the BMS [19] for at least 8 weeks. All mice showed complete paralysis at 1-day after SCI (BMS scores were 0, Supporting Information Video 1). At 8 weeks after SCI, control mice could move their hind limbs but could not support their own weight (BMS scores were approximately 2, Supporting Information Video 2). In contrast, mice that received hsp-NSCs could touch the ground with their paws and/or support their body weight using their hind limbs (BMS scores were 3-4). The hiPS-lt-NES cell-transplanted group also showed functional recovery (BMS scores were 3-4, Supporting Information Video 3) compared to the control group, and there was no significant difference in BMS scores between the hsp-NSC- and hiPS-lt-NES cell-transplanted groups (Fig. 2A). These results indicate that hiPS-NES cells have a comparable therapeutic effect to hsp-NSCs on SCI.

thumbnail image

Figure 2. Transplantation of hiPS-lt-NES cells into the injured spinal cord improves functional recovery of hind limbs. (A): Time course of functional recovery of hind limbs after SCI. Data are means ± SEM (SCI control, n = 8; hsp-NSCs, n = 8; hiPS-lt-NES cells, n = 9). Mean BMS values of hsp-NSC- and hiPS-lt-NES cell-transplanted groups were compared with those of the SCI control group. *, p < .05. There was no significant difference between values in the hsp-NSC group and the hiPS-lt-NES cell group. (B): Representative MEP waves of intact, SCI control, and hiPS-lt-NES cell-treated mice at 12 weeks after injury. The motor cortices were stimulated and MEP amplitudes were recorded from hamstring muscles. Amplitudes from onset to peak of the negative deflection were measured. (C): Relative values of the mean MEP amplitudes. Values are expressed as percentages of those in intact mice. Mean relative MEP amplitude in the hiPS-lt-NES cell group was significantly higher than that in the SCI control group. *, p < .05. Data are means ± SD (n = 3). (D): Relative values of the mean MEP latency. Mean relative MEP latency in the hiPS-lt-NES cell group was significantly shorter than that in the SCI control group. *, p < .05. Data are means ± SD (n = 3). Abbreviations: BMS, Basso Mouse Scale; hiPS-lt-NES, human induced pluripotent stem cell-derived long-term self-renewing neuroepithelial-like stem; MEP, motor-evoked potential; hsp-NSC, human fetal spinal cord-derived neural stem cell; and SCI, spinal cord injury.

Download figure to PowerPoint

Next, to evaluate the recovery of descending pathways from the motor cortex to the hind limb motor neurons, we monitored MEPs at 12 weeks after SCI. We stimulated motor cortices electrically and recorded MEP amplitudes from the hamstring muscles. The MEP amplitudes in the hiPS-NES cell-transplanted group were significantly higher than those in the SCI control group (Fig. 2B, 2C). Furthermore, the latency of MEP response in the hiPS-lt-NES cell-transplanted group was significantly shorter than that in the SCI control group (Fig. 2D). These results suggest that transplantation of hiPS-lt-NES cells into the injured spinal cord stimulates the functional recovery of hind limbs.

Transplanted hiPS-Lt-NES Cells Survive and Differentiate in the Injured Spinal Cord of NOD-SCID Mice

We have established hiPS-lt-NES cells that express luciferase and GFP, enabling us to trace the survival of the transplanted cells with a bioluminescence imaging system which detects photon signals from living cells after administration of luciferin (a substrate of luciferase) to the mice [18]. The photon signals decreased gradually after transplantation (Fig. 3A) and approximately 20% of transplanted hiPS-lt-NES cells survived in the injured spinal cord at 5 weeks after SCI (4 weeks after transplantation), whereafter the photon signals remained stable (Fig. 3B).

thumbnail image

Figure 3. Transplanted human induced pluripotent stem cell-derived long-term self-renewing neuroepithelial-like stem (hiPS-lt-NES) cells survive and differentiate in the injured spinal cord of nonobese diabetic-severe combined immunodeficient mice. (A): Survival of transplanted cells was checked every week using a bioluminescence imaging system; left, 2 hours after transplantation; middle, 4 weeks after SCI; and right, 8 weeks after SCI. (B): Time course of transplanted hiPS-lt-NES cell survival in SCI model mice. Optical signal intensity was measured using the bioluminescence imaging system. Quantification of the photon intensity revealed that approximately 20% of the transplanted cells survived 5 weeks after SCI; thereafter, the photon signals remained stable. Data are means ± SEM (n = 6). (C): Sagittal sections of SCI model mice treated with hiPS-lt-NES cells at 8 weeks after transplantation. Sections were stained with anti-GFP antibody (green). The epicenter of the SCI is indicated (*). Higher-magnification images of the white dotted boxes (C-1 and C-2) show GFP-positive transplant-derived cells extending their processes into gray and white matter. (D): Immunostaining images, 8 weeks after injury, of hiPS-lt-NES cells grafted into spinal cord. Spinal cord sections were stained with anti-GFP (green), hGFAP (magenta) and Tuj1 (red) antibodies and with Hoechst (blue). (E): Confocal images, 8 weeks after injury, of hiPS-lt-NES cells transplanted into spinal cord, revealing transplanted cells which were double-positive for GFP and markers of neural lineages. (F): Quantitative analyses of Tuj1-positive neurons, hGFAP-positive astrocytes, and MBP-positive oligodendrocytes as in E. Data are means ± SD (n = 3). Scale bars = 500 μm in C, 100 μm in C-1, C-2, D, and E. Abbreviations: GFP, green fluorescent protein; hGFAP, anti-human-specific glial fibrillary acidic protein; MBP, and SCI, spinal cord injury.

Download figure to PowerPoint

Using immunohistochemistry, we found that GFP-positive transplanted cells survived and migrated to both rostral and caudal areas around the lesion site in the injured spinal cord (Fig. 3C). High-magnification images showed that transplanted hiPS-lt-NES cells extended processes into both gray and white matter (Fig. 3C-1, 3C-2).

We then examined the differentiation of the transplanted hiPS-lt-NES cells in the injured spinal cord and found a similar differentiation tendency to that observed in vitro (Fig. 1D, 1E), with many GFP-positive transplant-derived cells having become Tuj1-positive neurons (75%) at 8 weeks after SCI (Fig. 3D–3F). Twenty percent of GFP-positive transplanted cells differentiated into GFAP-positive astrocytes, while their differentiation into MBP- or APC-positive oligodendrocytes was less than 1% (Fig. 3E, 3F, Supporting Information Fig. 2). These data indicate that transplanted hiPS-lt-NES cells differentiate preferentially into neurons in the injured spinal cord.

hiPS-Lt-NES Cell Transplantation Does Not Promote CST Axon Re-Extension After SCI

Since transplanted NSCs have been reported to play a supportive role in the re-extension of injured axons [29], we examined the effect of the transplanted hiPS-lt-NES cells on CST first-order axonal regeneration. We injected BDA into the bilateral motor cortices and labeled CST first-order neuron axons by anterograde tracing [21, 22]. Because BDA is not transported from first- to second-order neurons across the synapse, only the axons of first-order neurons in the CST are visualized by this method.

In the region caudal to the injury site, >50% of the labeled fibers observed at 5 mm anterior to the lesion site were detected in the intact mice, whereas almost no BDA-traced CST fibers could be seen in the SCI control or hiPS-lt-NES cell-transplanted mice (Fig. 4A). Quantitative analysis revealed that there was no significant difference in the number of BDA-labeled fibers between the SCI control and hiPS-lt-NES cell-treated groups at any position up to 5 mm on either side of the lesion site (Fig. 4B). Although we cannot exclude possibilities such as the re-extension of descending raphespinal axons, which are also important for the motor functional recovery of hind limbs [30, 31], these results suggest that the regeneration of CST axons, if it occurs, cannot be a major contributing factor in the recovery induced by hiPS-lt-NES cell transplantation in our SCI model.

thumbnail image

Figure 4. hiPS-lt-NES cell transplantation does not promote corticospinal tract (CST) axon re-extension after SCI. (A): Representative pictures of BDA-labeled CST fibers (red) at 5 and 2 mm rostral and 2 mm caudal to the lesion site at 12 weeks after SCI. Hoechst (blue) shows nuclear staining. Scale bar = 50 μm. (B): Quantification of the labeled CST fibers in the spinal cords of untreated, SCI control, and hiPS-lt-NES cell-transplanted mice. The x-axis indicates specific locations along the rostro-caudal axis of the spinal cord, and the y-axis indicates the ratio of the mean number of BDA-labeled fibers at the indicated site to that at 5 mm rostral to the lesion site (thoracic vertebra [Th] 9). Intact mice were compared with SCI control mice. *, p < .05; **, p < .01. There was no significant difference in the number of BDA-labeled fibers between hiPS-lt-NES cell-treated (blue line) and SCI control groups (red line). Data are means ± SEM (n = 5). Abbreviations: BDA, biotinylated dextran amine; hiPS-lt-NES, human induced pluripotent stem cell-derived long-term self-renewing neuroepithelial-like stem; and SCI, spinal cord injury.

Download figure to PowerPoint

Local Neurons Reconstruct Disrupted CST Neuronal Circuits in a Relay Manner

Recent studies have revealed that local endogenous and transplant-derived neurons can form new neuronal circuits and make synaptic connections after SCI [32–34]. To determine whether disrupted CSTs were reconstructed by forming a local neuronal relay, we injected WGA-expressing adenoviruses into the motor cortex of the hind limb area at 12 weeks after SCI. WGA, a plant lectin, has been widely used as a tracer of neuronal pathways [24, 25] because it is transported in axons and dendrites, and across synapses, to second- and even third-order neurons.

After injection, we could detect WGA-immunoreactive intracellular granule-like structures in MAP2ab-positive neurons (Fig. 5A). Furthermore, we found that hiPS-lt-NES cell-treated mice had more WGA/Map2ab double-positive cells than SCI control mice in the caudal region below the injured site (Fig. 5A, 5B). Taking these data and the results of the BDA experiment into consideration, it is suggested that WGA was transferred to the caudal area through the lesion site via new synaptic connections, and that hiPS-lt-NES cell transplantation promoted the CST reconstruction without CST axonal re-extension. In support of this proposition, we could observe GFP-positive transplant-derived neurons adjacent to synaptophysin-positive patches (Fig. 5C-1 and 5C-1′). Furthermore, immunohistochemistry using species-specific antibodies for presynaptic markers revealed that transplant-derived neurons made synapses with endogenous neurons, suggesting that they reconstructed disrupted CST neuronal circuits in a relay fashion (Fig. 5C-2, 5C-2′, 5C-3, and 5C-3′).

thumbnail image

Figure 5. Local neurons reconstruct disrupted corticospinal tract neuronal circuits in a relay manner. (A): Representative pictures of WGA-labeled neuronal cell bodies located in the caudal area (Th11 to lumbar vertebra [L] 1) at 12 weeks after spinal cord injury (SCI; left, intact mice; middle, SCI control mice; and right, hiPS-lt-NES cell-treated mice). Spinal cord sections were stained with anti-WGA (red) and anti-MAP2ab (neuronal marker, blue) antibodies. WGA immunoreactivity was observed as intracellular granule-like structures. (B): WGA-positive cells/MAP2ab-positive neurons in the caudal area were quantified. Values are expressed as percentages of those in intact mice. The percentage of WGA-positive cells in hiPS-lt-NES cell-treated mice was significantly higher than that in SCI control mice. *, p < .05. Data are means ± SD (n = 3). (C): Spinal cord sections were stained with anti-GFP and Tuj1 antibodies, Hoechst (blue), and Syn, Bsn, or hSyn antibodies. (C-1) Representative confocal image showing that GFP (green) and Tuj1 (red) double-positive transplant-derived neurons were adjacent to synaptophysin-positive patches (white). (C-1′) High-magnification view of boxed area in (C-1). (C-2) Host and graft synapses were distinguished using a monoclonal antibody for the presynaptic marker Bsn that selectively recognizes rat and mouse, but not human, epitopes. Confocal image showing that GFP (green) and Tuj1 (red) double-positive transplant-derived neurons were in contact with Bsn-positive host synapses (white). (C-2′) High-magnification view of boxed area in (C-2). (C-3) Confocal images showing that GFP (green)-negative and Tuj1 (red)-positive endogenous neurons were in contact with hSyn-positive synapses (white). (C-3′) High-magnification view of boxed area in (C-3). Scale bars = 10 μm. Abbreviations: Bsn, anti-Bassoon; GFP, green fluorescent protein; hiPS-lt-NES, human induced pluripotent stem cell-derived long-term self-renewing neuroepithelial-like stem; hSyn, anti-human-specific synaptophysin; Syn, anti-synaptophysin; and WGA, wheat germ agglutinin.

Download figure to PowerPoint

hiPS-Lt-NES Cell Transplantation Promotes the Survival of Endogenous Neurons

Massive endogenous cell death is observed in the injured spinal cord, probably due to environmental changes such as increased levels of inflammatory cytokines [35, 36]. Previous studies have suggested that neurotrophic factors secreted from transplanted cells could alleviate cell death and thereby contribute to improved locomotor function after SCI [37]. We therefore wanted to examine whether transplanted hiPS-lt-NES cells influence the survival of endogenous neurons. We performed immunohistochemistry and counted the number of NeuN-positive/GFP-negative endogenous neurons around the lesion site (Fig. 6A) and found that the number of endogenous neurons in the hiPS-lt-NES cell-treated mice was significantly higher than that in the SCI control (Fig. 6B). These findings indicate that transplantation of hiPS-lt-NES cells alters the environment of the injured spinal cord and promotes the survival of endogenous neurons.

thumbnail image

Figure 6. Transplanted hiPS-lt-NES cells promote the survival of endogenous neurons. (A): Representative sagittal sections of SCI control and hiPS-lt-NES cell-transplanted mice at 12 weeks after SCI. Sections were stained with anti-GFAP (white), anti-NeuN (mature neuronal marker, red) and anti-GFP (green) antibodies. Two 500-μm regions, at the rostral and caudal edge of the lesioned site (double-headed arrows), were selected for the assessment. The epicenter of the SCI is indicated (*). Scale bars = 500 μm. (B): Quantification of NeuN-positive/GFP-negative endogenous neurons. Values are expressed as percentages of those in intact mice. The number of NeuN-positive/GFP-negative endogenous neurons in hiPS-lt-NES-treated mice was significantly higher than that in SCI control mice. *, p < .05. Data are means ± SD (n = 3). Abbreviations: GFAP, glial fibrillary acidic protein; GFP, green fluorescent protein; hiPS-lt-NES, human induced pluripotent stem cell-derived long-term self-renewing neuroepithelial-like stem; and SCI, spinal cord injury.

Download figure to PowerPoint

Transplanted Cells Contribute Directly to Functional Recovery of Hind Limb Movement in SCI Model Mice

Ablation of specific cells is a useful method to evaluate their functional contribution in animal models. Mouse and rat cells are less sensitive to DT than human cells, which express human heparin-binding epidermal growth factor-like growth factor (hHB-EGF) as a DT receptor [26, 27]. We administered DT to hiPS-lt-NES cell-transplanted mice at 7 weeks after SCI. Using an in vivo imaging system, we were able to trace the survival of transplanted cells while evaluating hind limb function. Almost all the transplanted hiPS-lt-NES cells were specifically ablated following DT administration (Fig. 7A). After ablation of the transplanted cells, the BMS score of hiPS-lt-NES cell-treated mice dropped to a level similar to that of SCI control mice, although the reduction was not large enough to make the score differ significantly from that attained by hiPS-lt-NES cell-treated mice without DT administration (Fig. 7B). This may be because transplanted hiPS-lt-NES cells promoted the survival of endogenous neurons (Fig. 6) that participated in the reconstruction of disrupted neuronal circuitry. Nevertheless, these results suggest that transplanted hiPS-lt-NES cells play direct and important roles in the functional recovery of the injured spinal cord.

thumbnail image

Figure 7. Transplanted cells contribute directly to the functional recovery of hind limb movement in SCI model mice. (A): Survival of transplanted cells was monitored using a bioluminescence imaging system. Seven weeks after SCI (6 weeks after transplantation), each mouse was administered DT. One week later, luciferase activity had completely disappeared in hiPS-lt-NES cell-transplanted mice. (B): Time course of functional recovery of hind limbs after SCI and DT administration. Data are means ± SEM (SCI control, n = 8; hiPS-lt-NES, n = 9; and hiPS-lt-NES+DT, n = 7). Hind limb function of hiPS-lt-NES cell-transplanted mice deteriorated after DT administration (red line). BMS values of the hiPS-lt-NES and hiPS-lt-NES+DT groups were compared with those of the SCI control group. *, p < .05. Abbreviations: BMS, Basso Mouse Scale; DT, diphtheria toxin; hiPS-lt-NES, human induced pluripotent stem cell-derived long-term self-renewing neuroepithelial-like stem; and SCI, spinal cord injury.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

The advent of cell reprogramming and the establishment of iPS cells have provided new prospects for stem cell transplantation therapy [7, 8], and a large number of different types of iPSCs have already been generated by various methods [38–40]. Transplantation of NSCs into the injured spinal cord has been shown to be effective, and with recent advances in stem cell biology, particularly in human ES/iPS cells, we anticipate further improvements in the treatment of SCI [41]. However, transplantation studies of hiPS-NSCs, initially using mouse models of SCI [9], are in their infancy.

We have previously shown that human lt-NES cells, derived from different hESC and iPSC lines in two independent laboratories, exhibit consistent characteristics and retain their ability to proliferate and differentiate even after several passages and long-term in vitro growth [11]. In the present study, we investigated the differentiation potential of transplanted hiPS-lt-NES cells and found that the majority of differentiated hiPS-lt-NES cells in the injured spinal cord were neurons (Fig. 3D–3F), even though endogenous or transplanted NSCs in the injured spinal cord have been reported by others to differentiate preferentially into the glial lineage [42, 43]. Recent studies have indicated that neurons derived from transplanted human NSCs can integrate into the injured spinal cord and form synaptic connections with host neurons [33, 34]. We therefore took advantage of the tendency of our hiPS-lt-NES cells to differentiate into neurons and tried to apply these cells to reconstruction of the injured spinal cord.

Multiple therapeutic effects of NSC transplantation have been reported, and many different types of stem cells have been grafted into the injured spinal cord, yielding improved functional recovery in animal models [44, 45]. However, the growth of axons through the lesion to reinnervate the side caudal to the injury has so far been very limited. Our experiments using anterograde labeling of CST fibers revealed that this was also the case after transplantation of hiPS-lt-NES cells (Fig. 4). We have previously shown that neurons derived from mouse NSCs could restore disrupted neuronal circuits [15]. Consistent with that observation, we demonstrated here that WGA-positive cell numbers in hiPS-lt-NES cell-treated mice were higher than those in SCI control mice in the area caudal to the SCI (Fig. 5A, 5B). Furthermore, immunohistochemistry using species-specific antibodies for presynaptic markers suggested that transplant-derived neurons made synapses with endogenous neurons (Fig. 5C-2, 5C-2′, 5C-3, and 5C-3′). These results suggest that WGA was transferred to the caudal area through the lesion site via new synaptic connections, and that hiPS-lt-NES cell transplantation promoted CST reconstruction in a relay fashion. Other studies have provided evidence that local neurons can form intraspinal neural circuits in the lesion site and make synaptic connections with descending-tract collaterals after SCI [32, 46], and human NSCs transplanted into the injured rat or mouse spinal cord differentiated into neurons that formed axons and synapses, and also established contacts with host neurons [4, 34].

As well as their roles in cell replacement, NSCs reportedly exert other effects through the secretion of neurotrophic factors [37], which, by means such as suppressing myelin inhibitors, prevent neuronal cell death, enhance remyelination and regenerate axons [47, 48]. We found that transplantation of hiPS-lt-NES cells supported the survival of endogenous neurons (Fig. 6). Conversely, ablation of transplanted cells decreased previously attained functional recovery (Fig. 7), suggesting that transplanted cells contribute directly to functional recovery of hind limb movement in SCI model mice. In summary, we suggest that not only neurons derived from transplanted hiPS-lt-NES cells but also surviving endogenous neurons contribute to functional improvement by forming multiple synaptic connections which restore disrupted neuronal circuits.

Following injury, demyelinated and reconstructed axons must be remyelinated to ensure proper functional recovery. Generally, endogenous or transplanted NSCs which differentiate into oligodendrocytes have been reported to contribute to remyelination of axons and thereby to help in recovery after SCI. Transplanted oligodendrocyte progenitor cells (OPCs) derived from human ESCs have been shown to differentiate into oligodendrocytes and enhance remyelination in the moderate rat model of contusion spinal cord injury [5, 6]. A further study reported that in complete spinal cord transection rats, locomotor recovery after transplantation with both OPCs and human ESC-derived motor neuron progenitor cells was significantly greater than after treatment with either cell type alone [49]. Given that the application of stem cells and their progenitors for transplantation is currently in its infancy, and that SCI pathology differs between contusion and transection [50], these findings indicate that neural cells in an appropriate lineage should be transplanted according to the degree and/or type of SCI.

When ES and iPS cells are used for transplantation treatment, a key consideration is the likelihood of tumor formation, since the transplanted cell population may include undifferentiated cells [51, 52]. Indeed, the sources and methods of induction of NSCs are critical for differentiation and tumor formation [33, 53]. Neurosphere cultures are heterogeneous and sensitive to variations in methodological procedures [12], whereas monolayer cell cultures give rise to more homogeneous population [13, 14]. In this study, we detected no tumor formation in more than 40 hiPS-lt-NES cell-treated mice up to 12 weeks after SCI. This is probably attributable to complete differentiation of hiPS cells into NES cells in our robust and stable monolayer cell cultures, since the NES cells were established by initially purifying neural rosette structures from differentiating cultures [11]. Furthermore, we transplanted hiPS-lt-NES cells which had been expanded in the presence of growth factors and passaged more than 20 times. Before human iPS cell-based therapies can be implemented for clinical application, the cells' proliferation and tumor formation after transplantation must be strictly evaluated.

CONCLUSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

In this study, we have demonstrated that hiPS-lt-NES cells survived and differentiated in the injured spinal cord of NOD-SCID mice, and promoted functional recovery of hind limbs. Moreover, we have shown that transplanted hiPS-lt-NES cells support the reconstruction of CST pathways, promote endogenous neuron survival, and directly contribute to improved hind limb movement. To achieve a more efficient treatment for SCI, detailed investigations of hiPS-lt-NES cells and SCI pathology will be necessary. Nevertheless, our study raises the possibility that hiPS-based therapy can be applied in the near future to SCI patients who currently have few or no therapeutic options.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

We thank M. Saito and K. Kohno for providing diphtheria toxin, H. Miyoshi, H. J. Okano, and H. Okano for lentiviral vectors, Y. Yoshihara for WGA-expressing adenovirus, N. Uchida for anti-human specific antibodies, and S. Nori and M. Nakamura for species-specific antibodies. We also thank Y. Bessho, T. Matsui, Y. Nakahata, T. Matsuda, S. Komai, and M. Arai for valuable discussions and technical advice, T. Matta for statistical analysis, I. Smith for editing the manuscript, and M. Tano for secretarial assistance. lt-NES® is a registered trademark of LIFE & BRAIN GmbH, Bonn, Germany.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. CONCLUSION
  8. Acknowledgements
  9. DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST
  10. REFERENCES
  11. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
SC_11-1000_sm_supplFigure1.tiff1549KSupplemental Figure 1 hiPS-lt-NES cells express the radial glial marker BLBP (green) and the NSC marker Nestin (red). Hoechst (blue) shows nuclear staining. Scale bars = 30 μm.
SC_11-1000_sm_supplFigure2.tiff1904KSupplemental Figure 2 Confocal images of transplanted hiPS-lt-NES cells (GFP-positive, green) that differentiated into APC-positive oligodendrocytes (red). Hoechst (blue) shows nuclear staining. Scale bars = 20 μm.
SC_11-1000_sm_supplVideo1.wmv1062KVideo 1 One day after injury. All mice showed complete paralysis.
SC_11-1000_sm_supplVideo2.wmv1031KVideo 2 Eight weeks after injury. SCI control mice could move their hind limbs, but could not support their own weight.
SC_11-1000_sm_supplVideo3.wmv968KVideo 3 Eight weeks after injury. hiPS-lt-NES cell-transplanted mice could touch the ground with their paws and/or support their body weight using their hind limbs.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.