- Email: [email protected]
In vivo tracking of bone marrow stromal cells transplanted into mice cerebral infarct by fluorescence optical imaging
Brain Research Protocols 13 (2004) 166 – 175 www.elsevier.com/locate/brainresprot
Protocols
In vivo tracking of bone marrow stromal cells transplanted into mice cerebral infarct by fluorescence optical imaging Hideo Shichinohe a, Satoshi Kuroda a,*, Jang-Bo Lee a, Goro Nishimura b, Shunsuke Yano a, Toshitaka Seki a, Jun Ikeda a, Mamoru Tamura b, Yoshinobu Iwasaki a a
Department of Neurosurgery, Hokkaido University Graduate School of Medicine, North 15 West 7, Kita, Sapporo 060-8638, Japan b Laboratory of Biophysics, Research Institute for Electronic Science, Hokkaido University, Japan Accepted 27 April 2004 Available online 15 June 2004
Abstract Recent experimental studies have indicated that bone marrow stromal cells (BMSC) improve neurological deficits when transplanted into the animal models of various neurological disorders, although precise mechanism still remains unclear. In this study, we developed a new in vivo fluorescence optical imaging protocol to sequentially track the transplanted into the brain of the living animals subjected to cerebral infarct. Mice BMSC were harvested from transgenic mice expressing green fluorescent protein (BMSC-GFP). They were stereotactically transplanted into the ipsilateral striatum of mice subjected to permanent middle cerebral artery occlusion after 7 days of ischemia (n = 12). During 12 weeks after transplantation, the skull was exposed and the green fluorescence emitted from the brain surface was sequentially observed, using in vivo fluorescence optical microscopy. As the results, regional green fluorescence was detected in the ipsilateral parietal region 4 – 12 weeks after transplantation in all animals and became more apparent over the time. The images obtained through the skull were very similar to those acquired by thinning or removing the skull. Immunohistochemistry evaluation revealed that the transplanted cells migrated towards the ischemic boundary zone and expressed the neuronal or astrocytic marker, supporting the findings on fluorescence optical images. Sequential visualization of the BMSC transplanted into the brain of living animals would be valuable for monitoring the migration, growth and differentiation of the transplanted cells to explore the fate and safety of stem cell transplantation for various neurological disorders. D 2004 Elsevier B.V. All rights reserved. Theme: Disorders of the nervous system Topic: Ischemia Keywords: Bone marrow stromal cell; Transplantation; Migration; Fluorescence optical imaging; Differentiation; Green fluorescence protein
1. Type of research Because of the limited regenerative capacity of neural tissue, the therapeutic potential of stem cell grafting has recently been studied in various pathological conditions of the central nervous system (CNS). Embryonic stem (ES) cells, neural stem cells and bone marrow stromal cells (BMSC) have been considered as candidates for transplantation therapy. There is increasing evidence that stem cells transplanted into the CNS extensively migrate into damaged tissue and differentiate into neural cells, improv* Corresponding author. Tel.: +81-11-706-5987; fax: +81-11-708-7737. E-mail address: [email protected] (S. Kuroda). 1385-299X/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.brainresprot.2004.04.004
ing neurological function. Of these, BMSC may have enormous therapeutic potential because they can be harvested from the patients themselves without ethical or immunological problems [2,8,10,16,22]. Most techniques for the study of stem cell transplantation in animal models require histological analysis to determine the fate and migration of cells. Thus, the number and location of BMSC transplanted into the central nervous system can only be estimated postmortem, permitting observation at only one time point. A technique for monitoring the expansion, migration and fate of the transplanted cells continuously and non-invasively is crucial to guide further advances in neurotransplantation research and its future clinical application.
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
We report here a new approach to detect and track transplanted BMSC non-invasively and repeatedly in living animals. For this purpose, BMSC were harvested from transgenic mice expressing green fluorescence protein (GFP) and were transplanted into mice brains subjected to permanent middle cerebral artery occlusion. We developed the technique by modifying the method described before by Yang et al. [23 – 25]. In vivo fluorescence optical imaging technique was employed to detect the green fluorescence emitted from the transplanted BMSC through the skull just after transplantation and 2 –12 weeks later. The transplanted GFP-BMSC selectively migrated toward the ischemic boundary area between 4 and 12 weeks in each animal. We confirmed the migration and differentiation of the transplanted cells using immunohistochemistry.
2. Time required Harvest of the mice BMSC: 90 min/animal to harvest the BMSC and start the culture. Culture of the BMSC: 3 –4 weeks for two or three passages. Mice middle cerebral artery (MCA) occlusion model: 30 min/animal. Transplantation: 30 min/animal. In vivo fluorescence imaging: 60 min/animal.
3. Materials
167
Jackson Laboratory, USA. Male Balb/c mice, weighing 24– 28 g were purchased from CLEA Japan, Tokyo, Japan. 3.2. Special equipment 3.2.1. Harvest of the GFP-BMSC Sterile clean bench system (VWP-1000, Nihon Ika, Osaka, Japan). General sterilized surgical equipment: scalpel, forceps and microscissors. Disposable pipettes. Universal swing rotor RS-4/6 and centrifuge (model 8800, Kubota, Tokyo, Japan). CO2 incubator (MCO-18AIC, Sanyo Electric Biomedical, Tokyo, Japan). Transmission fluorescence microscopy (Axiovert 100, Zeiss, Germany). 3.2.2. Permanent middle cerebral artery occlusion model Isoflurane anesthesia setup with induction chamber and mask. Surgical microscope (Zeiss, Germany). Bipolar forceps (ME50, Martin Medizintechnik, Tuttlingen, Germany). General sterilized surgical equipment: scalpel, forceps, microscissors and dental drill. 3.2.3. Transplantation Isoflurane anesthesia setup with induction chamber and mask. Sterile clean bench system (VWP-1000, Nihon Ika). Stereotactic apparatus for small experimental animals with atraumatic earpoles and a micromanipulator (Model DKI900, David Kopf Instrument, Tujunga, CA, USA). Hamilton microsyringe (10 Al). Automatic microinjection pump system (Model KDS-310, Muromachi Kikai, Tokyo, Japan).
3.1. Animals Transgenic mice expressing enhanced green fluorescent protein (EGFP, 4 –8-week-old) were purchased from The
3.2.4. In vivo fluorescence imaging Isoflurane anesthesia setup with induction chamber and mask. Fluorescence microscope equipped with a mercury
Fig. 1. A diagram of the in vivo fluorescence optical imaging system in this study. The animal is placed on the microscope table under anesthesia (1 and 7). The exposed skull is illuminated by the white light or by the excitation light passed through a D470/70 band-pass filter. The system contains a mercury lamp system (2), light fiber of excitation light (3), magnifying lenses and a long-pass filter GG510 (4), CCD camera (5) and image control system (6).
168
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
120-W lamp power supply (model VB-G25/S20/S21/L11, Keyence, Osaka, Japan). A D470/70 band-pass filter for excitation and a long-pass filter GG510 for green fluorescence collection were equipped on a CCD camera and image control system (model VB-6000/6010, Keyence) for digital recording of high-resolution images of 1434 1050 pixels. VH Analyzer VH-H1A5 Ver 2.2 (Keyence) and Adobe Photoshop Elements (Adobe Systems, USA) for processing of the captured images. A diagram of the in vivo fluorescence imaging system is shown in Fig. 1. 3.2.5. Fluorescence immunohistochemistry Fluorescence microscopy (BX51, Olympus). CCD camera and image control system (model VB-6000/6010, Keyence).
was induced by direct MCA occlusion as described by Majid et al. [12] and Shichinohe et al. [17] with minor modifications (n = 12). Briefly, under anesthesia, a 1.0-cm vertical skin incision was made between the right eye and ear on the right side. The temporal muscle was mobilized and the temporal bone was exposed. Under surgical microscope, a 2.0-mm burr hole was made just on the MCA, which was visible through the temporal bone. The main trunk of the MCA was directly occluded with a bipolar coagulator, and complete interruption of blood flow at the occlusion site was confirmed by severance of the occlusion site of MCA. The animals were allowed to awake from anesthesia, and their neurological signs including circling toward the left side were confirmed [3]. Then, they were housed until following transplantation surgery. Core temperature was maintained between 36.5 and 37.5 jC through the procedures.
4. Detailed procedure 4.3. Transplantation of bone marrow stromal cells 4.1. Isolation of bone marrow stromal cells from GFPexpressing mice All animal experiments were approved by the Animal Studies Ethical Committee at Hokkaido University Graduate School of Medicine. To harvest mice BMSC, femurs were aseptically dissected from EGFP transgenic mice. Both ends of the femurs were cut, and the marrow was extruded with 5 ml of Dulbecco’s modified Eagle’s medium (DMEM; Nissui, Japan) containing 10% fetal bovine serum, 2 mM L-glutamine, 100 units/ml penicillin G and 10% heparin, using a 2.5-ml syringe and a 21-gauge needle. Between 10 and 15 106 whole marrow cells were placed in 75-cm2 tissue culture flask that was coated with collagen I (Becton Dickinson Labware, UK), in DMEM/10% FBS. After 24 h, the non-adherent cells were removed by changing the medium. The culture medium was replaced three times a week. When the cells were grown to confluency, the cell were lifted by 0.25% trypsin and 0.02% EDTA in PBS. The cells were passed two to three times. As reported previously, FACS analysis revealed that the phenotype of the cultured BMSC was CD34 negative; BMSC expressed low levels of CD45, and high levels of CD90 and Sca-1 [11]. Using transmission fluorescence microscopy (Axiovert 100, Zeiss), the fluorescence emitted from GFP-BMSC was confirmed prior to transplantation and was digitally photographed using a CCD camera (model DXM1200, Nikon). 4.2. Permanent middle cerebral artery occlusion model of mice Balb/c mice were fasted overnight but were allowed free access to water. Anesthesia was induced by inhalation of 4.0% isoflurane in N2O/O2 (70:30); an operation was performed under spontaneous ventilation in 2.0% isoflurane in N2O/O2(70:30). Permanent focal cerebral ischemia
The BMSC were transplanted into the ipsilateral striatum of the adult Balb/c mice 7 days after the onset of middle cerebral artery occlusion (n = 12). The animals were anesthetized in the above-mentioned condition. Under aseptic condition on a clean bench, the animals were fixed to a stereotactic apparatus and the cranium was exposed through midline skin incision. A burr hole was made 2 mm right to the bregma, using a small dental drill. A Hamilton syringe was inserted 3 mm into the brain parenchyma from the surface of the dura mater, and 10 Al of cell suspension (2 105 cells) was introduced into the striatum over a period of 5 min, using an automatic microinjection pump. The syringe was removed in the striatum 5 min after the injection in order to minimize leakage of the cell suspension from the injection site. All animals were treated with 10-mg kg 1 of the immunosuppressant cyclosporin A subcutaneously every day for 4 weeks following transplantation [11]. 4.4. In vivo fluorescence optical imaging Using a fluorescence microscope equipped with a mercury 120-W lamp power supply using fluorescence microscopy (model VB-G25/S20/S21/L11, Keyence), in vivo fluorescence imaging was performed just after transplantation as well as 2, 4, 8 and 12 weeks after transplantation. The animals were anesthetized in the abovementioned condition, using a face mask and were placed on the microscope table. The skull was exposed through midline skin incision. Selective excitation of GFP was produced through a D470/70 band-pass filter, and the skull was illuminated by the excitation light. Emitted fluorescence was collected through a magnifying lenses and longpass filter GG510 equipped on a Keyence CCD camera, and high-resolution images of 1434 1050 pixels were digitally recorded by a image control system (model VB-
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
169
6000/6010). Captured images were processed for contrast and brightness and analyzed using VH Analyzer VHH1A5 Ver 2.2 (Keyence) and Adobe Photoshop Elements (Adobe Systems). A diagram of the system is shown in Fig. 1. 4.5. Fluorescence immunohistochemistry The animals were anesthetized with 4.0% isoflurane in N2O/O2 (70:30) and were transcardially perfused with 20 ml of heparinized saline, followed by 50 ml of 4% paraformaldehyde 4, 8 and 12 weeks after transplantation surgery. Then, the brains were removed for histological evaluation. The species were embedded in paraffin and 4Am-thick coronal sections were prepared for subsequent staining. The deparaffinized sections were processed through antigen retrieve for 2 min by pressure pot. Double fluorescence immunohistochemistry was employed to identify cells derived from BMSC and to assess their differentiation into neural cells. Briefly, each section was treated with the monoclonal antibody against glial fibrillary acid protein (GFAP; rabbit polyclonal, dilution 1:200, Chemicon, CA) or microtubule-associated protein 2 (MAP2; mouse monoclonal, dilution 1:200, Chemicon). GFAP was employed for astrocyte identification. MAP2 was employed for neuron identification. Goat anti-mouse or anti-rabbit rhodamine-conjugated secondary antibody (dilution 1:200, Chemicon) was added for immunoreactivity identification. Subsequently, the sections were treated with the monoclonal antibody against GFP (mouse monoclonal, dilution 1:100, Santa Cruz, CA), and Zenon Alexa Fluor 488 (Mouse IgG Labeling Kit, Molecular Probes, OR) was added as the secondary antibody to identify its immunoreactivity. To avoid non-specific immunostaining by antimouse secondary reagents, negative control staining was also performed using mouse IgG. The fluorescence emitted from the secondary antibody was observed through an appropriate filter using a fluorescence microscopy (BX51, Olympus) and was digitally photographed using a cooled CCD camera equipped to the microscopy (model VB-6000/ 6010, Keyence).
5. Results 5.1. Characterization of cultured GFP-expressing bone marrow stromal cells The BMSC harvested from GFP-transgenic mouse bone marrow were isolated by their adherence to plastic and grown for two to three passages. The GFP-BMSC became comparatively homogeneous in appearance as the cells were passed. As the cells approached confluency, they assumed relatively elongated or spindle-shaped cells (Fig. 2). The findings are identical to those obtained in the experiment using the BMSC harvested from the wild-type mouse [11].
Fig. 2. Microscopic view of the cultured GFP-BMSC. The BMSC harvested from a GFP-transgenic mouse bone marrow assume relatively elongated or spindle shape (A). Transmission fluorescence microscopy reveals that all GFP-BMSC express green fluorescence (B).
Fluorescence microscopy clearly showed that all cultured GFP-BMSC emitted sufficient fluorescence for microscopic observations (Fig. 2). 5.2. In vivo visualization of transplanted GFP-expressing bone marrow stromal cells In all animals (n = 12), in vivo fluorescence imaging enabled clear visualization of a subpopulation of the transplanted GFP-BMSC that seeped out from the brain surface at the site of injection just after transplantation. However, the majority of the GFP-BMSC injected into the striatum could not be identified (Figs. 3A,D and 4A,D). The result indicates that the fluorescence optical imaging system equipped in this study could not detect green fluorescence emitted from the GFP-expressing BMSC through the skull when located in 3 mm depth of the brain. The findings were almost identical 2 weeks after transplantation (data not shown), suggesting that the migration of the transplanted cells did not begin or was not sufficient for fluorescence detection through the skull, although it was not histologically confirmed in this study. Four weeks after transplantation, however, the green fluorescence emitted from the transplanted GFP-BMSC was detected in the right parietal region in 7 out of 12 animals, strongly suggesting that the transplanted GFPBMSC could reach the surface of the cerebral cortex in the ischemic boundary zone enough for fluorescence detection
170
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
Fig. 3. In vivo white-light (A, B, C) and fluorescence (D, E, F) images after transplantation of BMSC into the mouse subjected to permanent MCA occlusion. Just after transplantation, only some transplanted GFP-BMSC that seeped out from the brain surface at the site of injection was visualized (A and D arrow heads: a small drilled hole for transplantation). Four weeks after transplantation, green fluorescence (arrow) was identified in the borderzone of infarct through the skull (B and E). The image acquired through the skull closely matched the image under the open brain after skull removal (C and F). Thus, green fluorescence (double arrow) could be clearly identified green fluorescence after the removal of skull (asterisk). Schema: the small circle and pink area represent the injection point and infarct area, respectively. The square of the red line represents the field shown in the figures.
(Fig. 3B and E). The image acquired through the skull closely matched the one of the open brain after the skull was removed (Fig. 3C and F). Of these seven mice, four were sacrificed at 4 weeks for histological analysis. A series of in vivo fluorescence optical images of the transplanted GFP-BMSC in other three animals was obtained from 4 to 12 weeks after transplantation. When assessed qualitatively, the size of the positive fluorescence area did not show apparent change, although quantitative measurements were not performed. In the remaining five animals, no fluorescence was detected through the skull 4 weeks after transplantation. However, in vivo fluorescence optical imaging could clearly detect the green fluorescence emitted from the transplanted GFP-BMSC in the right parietal region 8 or 12 weeks after transplantation in these five mice (Fig. 4). Thus, green fluorescence was observed in the right parietal region through the skull (Fig. 4B and E). Thinning of the right parietal bone using a dental drill improved the detection sensitivity of the green fluorescence (Fig. 4C and F). The findings strongly indicate that the migration and proliferation of the transplanted BMSC was not enough for fluorescence optical imaging 4 weeks after transplantation and required 8 – 12 weeks in these five animals.
5.3. Migration and differentiation of transplanted GFPexpressing bone marrow stromal cells Histological evaluation was performed 4 (n = 4) or 12 weeks (n = 8) after stereotactic transplantation of the GFPBMSC into the ipsilateral striatum of the mice subjected to permanent middle cerebral artery occlusion. In vivo fluorescence imaging detected the green fluorescence through the skull in all four mice that was sacrificed 4 weeks after transplantation. When assessed 4 weeks after transplantation, the GFP-positive cells were easily identified under fluorescence microscopy. They were found in the striatum, white matter and neocortex of the ipsilateral hemisphere, but the majority was distributed in the striatum and neocortex adjacent to infarct. Thus, the distribution of the GFPpositive cells in the neocortex correlated well with the region where the green fluorescence was detected on in vivo fluorescence optical images (Fig. 5A – D). Most of the cells had an oval shape in the striatum (Fig. 5E –H), whereas some transplanted GFP-BMSC had a triangle shape associated with fine projections, simulating neuron-like shape (Fig. 5A – D). Double fluorescence immunohistochemistry of brain sections revealed that some GFP-positive cells in the cerebral cortex, but not in the striatum, were reactive for
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
171
Fig. 4. In vivo white-light (A, B, C) and fluorescence (D, E, F) images after transplantation of BMSC into the mouse subjected to permanent MCA occlusion. Just after transplantation, only some transplanted GFP-BMSC that seeped out from the brain surface at the site of injection was detected (A and D arrow heads: a small drilled hole for transplantation). Twelve weeks after transplantation, green fluorescence (arrows) was detected in the infarct periphery through the skull (B and E). Thinning of the skull using a dental drill (C) markedly improve the detection sensitivity of green fluorescence (arrows, F). Schema: the small circle and pink area represent the injection point and infarct area, respectively. The square of the red line represents the field shown in the figures.
the neuronal marker MAP2 (Fig. 5). Therefore, the results suggest that the GFP-BMSC transplanted into the ipsilateral striatum migrated into the ischemic boundary zone through the white matter within 4 weeks and that some of them simulated the neuron in the neocortex morphologically and biologically. Other eight mice were histologically evaluated 12 weeks after transplantation. The GFP-positive cells were widely distributed in the entire brain. Especially, a number of the GFP-positive cells were found in the neocortex adjacent to infarct, supporting the findings on in vivo fluorescence images. Furthermore, the GFP-positive cells were found in the bilateral olfactory bulb, hippocampus, cerebellum and neocortex (Fig. 6A –F). Some of these cells showed neuronlike appearance. Some of them mimicked the appearance of the microvasculature in the striatum and ipsilateral cerebral cortex (Fig. 6J –L). Double fluorescence immunohistochemistry of brain sections revealed that a significant number of GFP-positive cells in the cerebral cortex, olfactory bulb, hippocampus and cerebellum were reactive for the neuronal marker MAP2 (Fig. 6A – F and J– L), and that some of GFPpositive cells in the white matter were reactive for the astrocytic marker GFAP (Fig. 6G – I). The histological findings suggested that the transplanted GFP-BMSC migrated
into the entire brain between 4 and 12 weeks and simulated the neuron and astrocyte morphologically and biologically.
6. Discussion This study was designed to sequentially image the distribution of GFP-BMSC transplanted into the brain over 12 weeks in an attempt to establish the technique to monitor the cells transplanted into the brain. Transplanted BMSC had previously been shown to migrate throughout the CNS, to become normal constituents of the host cytoarchitecture, although precise route of their movements in the CNS still remains unclear. The BMSC used in the current study had been previously characterized to express low levels of CD45 and high levels of CD90 and Sca-1 on flowcytometry. Transplanted BMSC survive in the normal or injured central nervous system, including the brain and spinal cord. Prior study has also shown the ability of these BMSC to migrate toward cerebral infarct and spinal cord injury [11]. As described above, although other investigators have reported similar results, the magnitude and timing of this effect have not been quantified [2,4,5]. In this study, we demonstrate that the BMSC harvested from transgenic mice express GFP
172
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
Fig. 5. Fluorescent immunohistochemistry images in the boundary zone of infarct (A – D) and the ipsilateral striatum (E – H) from coronal brain sections stained with antibodies against GFP (A, B, E, F), MAP2 (C, G) and the merged images (D, H). Original magnification: A, 100; E, 200; B – D and F – H, 1000.
continuously, and that in vivo fluorescence optical imaging technique enables a detection and tracking of the GFP-BMSC that migrate into the cerebral cortex over 4 –12 weeks after stereotactic transplantation into the ipsilateral striatum of mice subjected to permanent middle cerebral artery occlusion. Histological evaluation indicated that the GFP-BMSC were mainly identified in the striatum, white matter, and infarct periphery 4 weeks after transplantation. In some cases, these cells were also found in the olfactory bulb, hippocampus, cerebellum and contralateral cerebral hemisphere 8 – 12
weeks later. Serial in vivo fluorescence optical images could identify the cells distributed in the cerebral cortex adjacent to infarct and showed real-time cell migration in a single animal, with clear time dependence. The imaging technique also clarified that the cell migration toward the infarct periphery reached saturation by 4 – 12 weeks, varying among the animals. The deviation may depend on the difference in tissue damage among the animals. In addition, using a double immunohistochemistry technique, we demonstrate that the transplanted BMSC migrate
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
173
Fig. 6. Fluorescent immunohistochemistry images in the internal granular layer of the ipsilateral olfactory bulb (A – C), the mid-layer of the contralateral neocortex (D – F), the ipsilateral white matter (G – I) and the boundary zone of infarct (J – L) from the coronal brain sections stained with antibodies against GFP (A, D, G, J), MAP2 (B, E, K), GFAP (H) and the merged images (C, F, I, L).
into the ipsilateral cerebral cortex and simulate the neuron morphologically and biologically by 4 weeks, although the cells that survived in the striatum showed no significant differentiation into the neural cells. Subsequently, they migrate into the entire brain and expressed the neuronal marker, MAP2, in the ipsilateral and contralateral cerebral cortex, hippocampus and olfactory bulb 8– 12 weeks after transplantation. They also simulated neuron-like appearance morphologically. They expressed the astrocytic marker, GFAP, when they came to the white matter including the corpus callosum. These findings strongly suggest that the differentiation of the transplanted BMSC depends on their
surrounding microenvironment [9,18]. Interestingly, the transplanted cells also formed luminal structures, simulating the appearance of capillary endothelium in the entire brain, as reported previously [5]. However, very recent studies have suggested another possibility that the transplanted stem cells may fuse with the host cells and simulate the ‘‘differentiation’’ into the host cells [1,19 – 21]. Both options are possible as the mechanism by which the transplanted BMSC simulated the neural cells morphologically and biologically in the current study. GFP is a fluorescence protein identified in Aequorea Victoria in 1962 and has widely been employed to label
174
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
various kinds of cells using transgene or transfection technique [15]. Yang et al. injected mouse melanoma cells expressing very high level of GFP into the tail vein or portal vein of mice. They showed that whole-body optical images could clearly visualize metastatic lesion in the brain, liver and bone in real time. A fluorescence images (5.5 mm in diameter and 0.8 mm in depth) were obtained externally through the scalp and skull. The size of the external image correlated well with that of the metastatic lesion [23]. In another study, they injected the adenoviral vector expressing enhanced GFP into the mice brain, liver, pancreas, prostate and bone marrow. As the result, the fluorescence of the expressed GFP in these organs became visible by wholebody optical imaging system. The images made externally to the animal appeared very similar to those of the exposed organs [24]. Recently, they also developed a reversible skinflap method and could increase the detection sensitivity through the skull many-fold. In their report, they concluded that green fluorescence emitted from a single tumor cell could be detected, although the detection sensitivity was rather lower and only regional green fluorescence could be visualized in the present study [25]. Therefore, using GFP as a probe, in vivo fluorescence imaging technique would be valuable to track the fate of the cells transplanted into the organs including the brain. As they also pointed out, however, the current imaging technique has limitations, primarily the relatively short wavelength of fluorescence emitted from GFP (520 nm). The emitted light is strongly scattered by the surrounding tissue including the scalp and skull. Thus, we had to expose the skull each time in vivo fluorescence images were taken. Immediately after transplantation, the GFP-BMSC injected into the striatum could not be visualized. Only some of them could be observed at the brain surface of the injection site. More clear fluorescence images were obtained when the skull was removed or thinned 4– 12 weeks after transplantation. In fact, Yang et al. [23] reported that they could detect the green fluorescence emitted from maximally 2.2 mm depth. However, powerful new techniques of using ultra-fast lasers, dual photon imaging and ballistic photon imaging may offer large gains in sensitivity, increased depth of detection, and spatial resolution. Developments of these technologies would allow us to obtain completely noninvasive images through the skin in the future. Recently, other imaging modalities have also been attempted to track the cells transplanted into the CNS [6,13,14,26]. MRI is a non-optical method. MRI can image intact, opaque organisms in three dimensions with good spatial resolution. However, MRI requires long imaging times and consequently slows data acquisition because of the low sensitivity. More importantly, magnetic nanoparticles that label the transplanted cells cannot be succeeded to all the cells during their proliferation. Alternatively, wholebody imaging technique using the luciferase reporter technique has been believed as useful method. However, the technique requires the injection of luciferin as the substrate
for photon production and long imaging time to collect the sufficient photon [7]. In conclusion, although the in vivo imaging technique used in the current study requires improvement, the technique is feasible and represents a method for non-invasively tracking the quantity and location of the transplanted cells. The technique offers a method to study the biological aspects of the transplanted cells and to develop the optimal protocols for delivery of the BMSC for CNS regeneration. Development of such imaging technique would also be useful to monitor various kinds of stem cells transplanted into the CNS when cell therapy is clinically applied for neurological disorders. 6.1. Troubleshooting 6.1.1. Graft preparation It is crucial to use a culture flask coated with collagen I to culture the bone marrow cells in order to improve the yield of the BMSC. Prior to transplantation, it should be confirmed that the cultured BMSC express enough green fluorescence using. 6.1.2. Permanent MCA occlusion model Animal strain is essential to produce focal cerebral ischemia model constantly [12,17]. The important factors for successful production of the model are to completely obliterate the flow of the middle cerebral artery under surgical microscope and to assess neurological deficits just after surgery. 6.1.3. Transplantation It is very important to slowly inject the cell suspension over 5 min, using a programmable, automatic microinjection system and to remove the needle at 5 min after the end of injection for certain transplantation. 6.1.4. In vivo fluorescence imaging Completely dark room is essential to obtain good visualization of the green fluorescence. Alternatively, the dark box covering the animal and whole imaging system may be useful.
7. Quick procedure
Three to four weeks prior to transplantation. Harvest and culture of the GFP-BMSC. Day 0: permanent MCA occlusion in the host. Day 7: transplantation of the GFP-BMSC and in vivo fluorescence imaging 1. Day 21: in vivo fluorescence imaging 2. Day 35: in vivo fluorescence imaging 3 and removing the brain for histological analysis. Day 63: in vivo fluorescence imaging 4. Day 91: in vivo fluorescence imaging 5 and removing the brain for histological analysis.
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
8. Essential literature references A similar in vivo fluorescence imaging technique using the GFP expressing cells is described by Hoffman and his colleagues [23 – 25], which served as a basis for the present study. The basic techniques for cell culture, stereotactic transplantation and fluorescence immunohistochemistry are described in detail by our previous report [11]. Mice permanent MCA occlusion model was produced according to the method describled by Majid et al. [12] and Shichinohe et al. [17] with minor modification.
Acknowledgements This study was supported by Grant-in-Aids from the Ministry of Education, Science and Culture of Japan (No. 14370424, Dr. Kuroda and No. 15390426, Dr. Iwasaki), and by a grant from Mitsubishi Pharma Research Foundation (Dr. Kuroda). The authors thank Dr. Hiroyuki Kobayashi for his helpful advice on histological findings and Ms. Yumiko Shinohe for her technical assistance in cell culture and immunohistochemistry.
References [1] M. Alvarez-Dolado, R. Pardal, J.M. Garcia-Verdugo, J.R. Fike, H.O. Lee, K. Pfeffer, C. Lois, S.J. Morrison, A. Alvarez-Buylla, Fusion of bone-marrow-derived cells with Purkinje neurons, cardiomyocytes and hepatocytes, Nature 425 (2003) 968 – 973. [2] S.A. Azizi, D. Stokes, B.J. Augelli, C. DiGirolamo, D.J. Prockop, Engraftment and migration of human bone marrow stromal cells implanted in the brains of albino rats—similarities to astrocyte grafts, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 3908 – 3913. [3] J.B. Bederson, L.H. Pitts, M. Tsuji, M.C. Nishimura, R.L. Davis, H. Bartkowski, Rat middle cerebral artery occlusion: evaluation of the model and development of a neurologic examination, Stroke 17 (1986) 472 – 476. [4] M. Chopp, X.H. Zhang, Y. Li, L. Wang, J. Chen, D. Lu, M. Lu, M. Rosenblum, Spinal cord injury in rat: treatment with bone marrow stromal cell transplantation, NeuroReport 11 (2000) 3001 – 3005. [5] D.C. Hess, W.D. Hill, A. Martin-Studdard, J. Carroll, J. Brailer, J. Carothers, Bone marrow as a source of endothelial cells and NeuNexpressing cells after stroke, Stroke 33 (2002) 1362 – 1368. [6] M. Hoehn, E. Kustermann, J. Blunk, D. Wiedermann, T. Trapp, S. Wecker, M. Focking, H. Arnold, J. Hescheler, B.K. Fleischmann, W. Schwindt, C. Buhrle, Monitoring of implanted stem cell migration in vivo: a highly resolved in vivo magnetic resonance imaging investigation of experimental stroke in rat, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 16267 – 16272. [7] R. Hoffman, Green fluorescent protein imaging of tumour growth, metastasis, and angiogenesis in mouse models, Lancet Oncol. 3 (2002) 546 – 556. [8] Y. Jiang, B.N. Jahagirdar, R.L. Reinhardt, R.E. Schwartz, C.D. Keene, X.R. Ortiz-Gonzalez, M. Reyes, T. Lenvik, T. Lund, M. Blackstad, J. Aldrich, S. Aldrich, A. Lisberg, W.C. Low, D.A. Largaespada, C.M. Verfaillie, Pluripotency of mesenchymal stem cells derived from adult marrow, Nature 418 (2002) 41 – 49. [9] C.B. Johansson, S. Momma, D.L. Clarke, M. Risling, U. Lendahl, J.
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20] [21]
[22]
[23]
[24]
[25]
[26]
175
Frisen, Identification of a neural stem cell in the adult mammalian central nervous system, Cell 96 (1999) 25 – 34. G.C. Kopen, D.J. Prockop, D.G. Phinney, Marrow stromal cells migrate throughout forebrain and cerebellum, and they differentiate into astrocytes after injection into neonatal mouse brains, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 10711 – 10716. J.B. Lee, S. Kuroda, H. Shichinohe, J. Ikeda, T. Seki, K. Hida, M. Tada, K. Sawada, Y. Iwasaki, Migration and differentiation of nuclear fluorescence-labeled bone marrow stromal cells after transplantation into cerebral infarct and spinal cord injury in mice, Neuropathology 23 (2003) 169 – 180. A. Majid, Y.Y. He, J.M. Gidday, S.S. Kaplan, E.R. Gonzales, T.S. Fenstermacher, J.D. Fenstermacher, L. Wei, D.W. Choi, C.Y. Hsu, Differences in vulnerability to permanent focal cerebral ischemia among 3 common mouse strains, Stroke 31 (2000) 2707 – 2714. M. Modo, D. Cash, K. Mellodew, S.C. Williams, S.E. Fraser, T.J. Meade, J. Price, H. Hodges, Tracking transplanted stem cell migration using bifunctional, contrast agent-enhanced, magnetic resonance imaging, NeuroImage 17 (2002) 803 – 811. M. Modo, K. Mellodew, D. Cash, S.E. Fraser, T.J. Meade, J. Price, S.C. Williams, Mapping transplanted stem cell migration after a stroke: a serial, in vivo magnetic resonance imaging study, NeuroImage 21 (2004) 311 – 317. M. Okabe, M. Ikawa, K. Kominami, T. Nakanishi, Y. Nishimune, ‘Green mice’ as a source of ubiquitous green cells, FEBS Lett. 407 (1997) 313 – 319. J. Sanchez-Ramos, S. Song, F. Cardozo-Pelaez, C. Hazzi, T. Stedeford, A. Willing, T.B. Freeman, S. Saporta, W. Janssen, N. Patel, D.R. Cooper, P.R. Sanberg, Adult bone marrow stromal cells differentiate into neural cells in vitro, Exp. Neurol. 164 (2000) 247 – 256. H. Shichinohe, S. Kuroda, T. Abumiya, T. Kobayashi, J. Ikeda, Y. Iwasaki, FK506 reduces infarct volume due to permanent focal ischemia by maintaining BAD turnover and mitochondrial integrity, Brain Res. 1001 (2004) 51 – 59. L.S. Shihabuddin, P.J. Horner, J. Ray, F.H. Gage, Adult spinal cord stem cells generate neurons after transplantation in the adult dentate gyrus, J. Neurosci. 20 (2000) 8727 – 8735. G. Vassilopoulos, D.W. Russell, Cell fusion: an alternative to stem cell plasticity and its therapeutic implications, Curr. Opin. Genet. Dev. 13 (2003) 480 – 485. G. Vassilopoulos, P.R. Wang, D.W. Russell, Transplanted bone marrow regenerates liver by cell fusion, Nature 422 (2003) 901 – 904. X. Wang, H. Willenbring, Y. Akkari, Y. Torimaru, M. Foster, M. AlDhalimy, E. Lagasse, M. Finegold, S. Olson, M. Grompe, Cell fusion is the principal source of bone-marrow-derived hepatocytes, Nature 422 (2003) 897 – 901. D. Woodbury, E.J. Schwarz, D.J. Prockop, I.B. Black, Adult rat and human bone marrow stromal cells differentiate into neurons, J. Neurosci. Res. 61 (2000) 364 – 370. M. Yang, E. Baranov, P. Jiang, F.X. Sun, X.M. Li, L. Li, S. Hasegawa, M. Bouvet, M. Al-Tuwaijri, T. Chishima, H. Shimada, A.R. Moossa, S. Penman, R.M. Hoffman, Whole-body optical imaging of green fluorescent protein-expressing tumors and metastases, Proc. Natl. Acad. Sci. U. S. A. 97 (2000) 1206 – 1211. M. Yang, E. Baranov, A.R. Moossa, S. Penman, R.M. Hoffman, Visualizing gene expression by whole-body fluorescence imaging, Proc. Natl. Acad. Sci. U. S. A. 97 (2000) 12278 – 12282. M. Yang, E. Baranov, J.W. Wang, P. Jiang, X. Wang, F.X. Sun, M. Bouvet, A.R. Moossa, S. Penman, R.M. Hoffman, Direct external imaging of nascent cancer, tumor progression, angiogenesis, and metastasis on internal organs in the fluorescent orthotopic model, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 3824 – 3829. Z.G. Zhang, Q. Jiang, R. Zhang, L. Zhang, L. Wang, P. Arniego, K.L. Ho, M. Chopp, Magnetic resonance imaging and neurosphere therapy of stroke in rat, Ann. Neurol. 53 (2003) 259 – 263.
Protocols
In vivo tracking of bone marrow stromal cells transplanted into mice cerebral infarct by fluorescence optical imaging Hideo Shichinohe a, Satoshi Kuroda a,*, Jang-Bo Lee a, Goro Nishimura b, Shunsuke Yano a, Toshitaka Seki a, Jun Ikeda a, Mamoru Tamura b, Yoshinobu Iwasaki a a
Department of Neurosurgery, Hokkaido University Graduate School of Medicine, North 15 West 7, Kita, Sapporo 060-8638, Japan b Laboratory of Biophysics, Research Institute for Electronic Science, Hokkaido University, Japan Accepted 27 April 2004 Available online 15 June 2004
Abstract Recent experimental studies have indicated that bone marrow stromal cells (BMSC) improve neurological deficits when transplanted into the animal models of various neurological disorders, although precise mechanism still remains unclear. In this study, we developed a new in vivo fluorescence optical imaging protocol to sequentially track the transplanted into the brain of the living animals subjected to cerebral infarct. Mice BMSC were harvested from transgenic mice expressing green fluorescent protein (BMSC-GFP). They were stereotactically transplanted into the ipsilateral striatum of mice subjected to permanent middle cerebral artery occlusion after 7 days of ischemia (n = 12). During 12 weeks after transplantation, the skull was exposed and the green fluorescence emitted from the brain surface was sequentially observed, using in vivo fluorescence optical microscopy. As the results, regional green fluorescence was detected in the ipsilateral parietal region 4 – 12 weeks after transplantation in all animals and became more apparent over the time. The images obtained through the skull were very similar to those acquired by thinning or removing the skull. Immunohistochemistry evaluation revealed that the transplanted cells migrated towards the ischemic boundary zone and expressed the neuronal or astrocytic marker, supporting the findings on fluorescence optical images. Sequential visualization of the BMSC transplanted into the brain of living animals would be valuable for monitoring the migration, growth and differentiation of the transplanted cells to explore the fate and safety of stem cell transplantation for various neurological disorders. D 2004 Elsevier B.V. All rights reserved. Theme: Disorders of the nervous system Topic: Ischemia Keywords: Bone marrow stromal cell; Transplantation; Migration; Fluorescence optical imaging; Differentiation; Green fluorescence protein
1. Type of research Because of the limited regenerative capacity of neural tissue, the therapeutic potential of stem cell grafting has recently been studied in various pathological conditions of the central nervous system (CNS). Embryonic stem (ES) cells, neural stem cells and bone marrow stromal cells (BMSC) have been considered as candidates for transplantation therapy. There is increasing evidence that stem cells transplanted into the CNS extensively migrate into damaged tissue and differentiate into neural cells, improv* Corresponding author. Tel.: +81-11-706-5987; fax: +81-11-708-7737. E-mail address: [email protected] (S. Kuroda). 1385-299X/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.brainresprot.2004.04.004
ing neurological function. Of these, BMSC may have enormous therapeutic potential because they can be harvested from the patients themselves without ethical or immunological problems [2,8,10,16,22]. Most techniques for the study of stem cell transplantation in animal models require histological analysis to determine the fate and migration of cells. Thus, the number and location of BMSC transplanted into the central nervous system can only be estimated postmortem, permitting observation at only one time point. A technique for monitoring the expansion, migration and fate of the transplanted cells continuously and non-invasively is crucial to guide further advances in neurotransplantation research and its future clinical application.
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
We report here a new approach to detect and track transplanted BMSC non-invasively and repeatedly in living animals. For this purpose, BMSC were harvested from transgenic mice expressing green fluorescence protein (GFP) and were transplanted into mice brains subjected to permanent middle cerebral artery occlusion. We developed the technique by modifying the method described before by Yang et al. [23 – 25]. In vivo fluorescence optical imaging technique was employed to detect the green fluorescence emitted from the transplanted BMSC through the skull just after transplantation and 2 –12 weeks later. The transplanted GFP-BMSC selectively migrated toward the ischemic boundary area between 4 and 12 weeks in each animal. We confirmed the migration and differentiation of the transplanted cells using immunohistochemistry.
2. Time required Harvest of the mice BMSC: 90 min/animal to harvest the BMSC and start the culture. Culture of the BMSC: 3 –4 weeks for two or three passages. Mice middle cerebral artery (MCA) occlusion model: 30 min/animal. Transplantation: 30 min/animal. In vivo fluorescence imaging: 60 min/animal.
3. Materials
167
Jackson Laboratory, USA. Male Balb/c mice, weighing 24– 28 g were purchased from CLEA Japan, Tokyo, Japan. 3.2. Special equipment 3.2.1. Harvest of the GFP-BMSC Sterile clean bench system (VWP-1000, Nihon Ika, Osaka, Japan). General sterilized surgical equipment: scalpel, forceps and microscissors. Disposable pipettes. Universal swing rotor RS-4/6 and centrifuge (model 8800, Kubota, Tokyo, Japan). CO2 incubator (MCO-18AIC, Sanyo Electric Biomedical, Tokyo, Japan). Transmission fluorescence microscopy (Axiovert 100, Zeiss, Germany). 3.2.2. Permanent middle cerebral artery occlusion model Isoflurane anesthesia setup with induction chamber and mask. Surgical microscope (Zeiss, Germany). Bipolar forceps (ME50, Martin Medizintechnik, Tuttlingen, Germany). General sterilized surgical equipment: scalpel, forceps, microscissors and dental drill. 3.2.3. Transplantation Isoflurane anesthesia setup with induction chamber and mask. Sterile clean bench system (VWP-1000, Nihon Ika). Stereotactic apparatus for small experimental animals with atraumatic earpoles and a micromanipulator (Model DKI900, David Kopf Instrument, Tujunga, CA, USA). Hamilton microsyringe (10 Al). Automatic microinjection pump system (Model KDS-310, Muromachi Kikai, Tokyo, Japan).
3.1. Animals Transgenic mice expressing enhanced green fluorescent protein (EGFP, 4 –8-week-old) were purchased from The
3.2.4. In vivo fluorescence imaging Isoflurane anesthesia setup with induction chamber and mask. Fluorescence microscope equipped with a mercury
Fig. 1. A diagram of the in vivo fluorescence optical imaging system in this study. The animal is placed on the microscope table under anesthesia (1 and 7). The exposed skull is illuminated by the white light or by the excitation light passed through a D470/70 band-pass filter. The system contains a mercury lamp system (2), light fiber of excitation light (3), magnifying lenses and a long-pass filter GG510 (4), CCD camera (5) and image control system (6).
168
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
120-W lamp power supply (model VB-G25/S20/S21/L11, Keyence, Osaka, Japan). A D470/70 band-pass filter for excitation and a long-pass filter GG510 for green fluorescence collection were equipped on a CCD camera and image control system (model VB-6000/6010, Keyence) for digital recording of high-resolution images of 1434 1050 pixels. VH Analyzer VH-H1A5 Ver 2.2 (Keyence) and Adobe Photoshop Elements (Adobe Systems, USA) for processing of the captured images. A diagram of the in vivo fluorescence imaging system is shown in Fig. 1. 3.2.5. Fluorescence immunohistochemistry Fluorescence microscopy (BX51, Olympus). CCD camera and image control system (model VB-6000/6010, Keyence).
was induced by direct MCA occlusion as described by Majid et al. [12] and Shichinohe et al. [17] with minor modifications (n = 12). Briefly, under anesthesia, a 1.0-cm vertical skin incision was made between the right eye and ear on the right side. The temporal muscle was mobilized and the temporal bone was exposed. Under surgical microscope, a 2.0-mm burr hole was made just on the MCA, which was visible through the temporal bone. The main trunk of the MCA was directly occluded with a bipolar coagulator, and complete interruption of blood flow at the occlusion site was confirmed by severance of the occlusion site of MCA. The animals were allowed to awake from anesthesia, and their neurological signs including circling toward the left side were confirmed [3]. Then, they were housed until following transplantation surgery. Core temperature was maintained between 36.5 and 37.5 jC through the procedures.
4. Detailed procedure 4.3. Transplantation of bone marrow stromal cells 4.1. Isolation of bone marrow stromal cells from GFPexpressing mice All animal experiments were approved by the Animal Studies Ethical Committee at Hokkaido University Graduate School of Medicine. To harvest mice BMSC, femurs were aseptically dissected from EGFP transgenic mice. Both ends of the femurs were cut, and the marrow was extruded with 5 ml of Dulbecco’s modified Eagle’s medium (DMEM; Nissui, Japan) containing 10% fetal bovine serum, 2 mM L-glutamine, 100 units/ml penicillin G and 10% heparin, using a 2.5-ml syringe and a 21-gauge needle. Between 10 and 15 106 whole marrow cells were placed in 75-cm2 tissue culture flask that was coated with collagen I (Becton Dickinson Labware, UK), in DMEM/10% FBS. After 24 h, the non-adherent cells were removed by changing the medium. The culture medium was replaced three times a week. When the cells were grown to confluency, the cell were lifted by 0.25% trypsin and 0.02% EDTA in PBS. The cells were passed two to three times. As reported previously, FACS analysis revealed that the phenotype of the cultured BMSC was CD34 negative; BMSC expressed low levels of CD45, and high levels of CD90 and Sca-1 [11]. Using transmission fluorescence microscopy (Axiovert 100, Zeiss), the fluorescence emitted from GFP-BMSC was confirmed prior to transplantation and was digitally photographed using a CCD camera (model DXM1200, Nikon). 4.2. Permanent middle cerebral artery occlusion model of mice Balb/c mice were fasted overnight but were allowed free access to water. Anesthesia was induced by inhalation of 4.0% isoflurane in N2O/O2 (70:30); an operation was performed under spontaneous ventilation in 2.0% isoflurane in N2O/O2(70:30). Permanent focal cerebral ischemia
The BMSC were transplanted into the ipsilateral striatum of the adult Balb/c mice 7 days after the onset of middle cerebral artery occlusion (n = 12). The animals were anesthetized in the above-mentioned condition. Under aseptic condition on a clean bench, the animals were fixed to a stereotactic apparatus and the cranium was exposed through midline skin incision. A burr hole was made 2 mm right to the bregma, using a small dental drill. A Hamilton syringe was inserted 3 mm into the brain parenchyma from the surface of the dura mater, and 10 Al of cell suspension (2 105 cells) was introduced into the striatum over a period of 5 min, using an automatic microinjection pump. The syringe was removed in the striatum 5 min after the injection in order to minimize leakage of the cell suspension from the injection site. All animals were treated with 10-mg kg 1 of the immunosuppressant cyclosporin A subcutaneously every day for 4 weeks following transplantation [11]. 4.4. In vivo fluorescence optical imaging Using a fluorescence microscope equipped with a mercury 120-W lamp power supply using fluorescence microscopy (model VB-G25/S20/S21/L11, Keyence), in vivo fluorescence imaging was performed just after transplantation as well as 2, 4, 8 and 12 weeks after transplantation. The animals were anesthetized in the abovementioned condition, using a face mask and were placed on the microscope table. The skull was exposed through midline skin incision. Selective excitation of GFP was produced through a D470/70 band-pass filter, and the skull was illuminated by the excitation light. Emitted fluorescence was collected through a magnifying lenses and longpass filter GG510 equipped on a Keyence CCD camera, and high-resolution images of 1434 1050 pixels were digitally recorded by a image control system (model VB-
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
169
6000/6010). Captured images were processed for contrast and brightness and analyzed using VH Analyzer VHH1A5 Ver 2.2 (Keyence) and Adobe Photoshop Elements (Adobe Systems). A diagram of the system is shown in Fig. 1. 4.5. Fluorescence immunohistochemistry The animals were anesthetized with 4.0% isoflurane in N2O/O2 (70:30) and were transcardially perfused with 20 ml of heparinized saline, followed by 50 ml of 4% paraformaldehyde 4, 8 and 12 weeks after transplantation surgery. Then, the brains were removed for histological evaluation. The species were embedded in paraffin and 4Am-thick coronal sections were prepared for subsequent staining. The deparaffinized sections were processed through antigen retrieve for 2 min by pressure pot. Double fluorescence immunohistochemistry was employed to identify cells derived from BMSC and to assess their differentiation into neural cells. Briefly, each section was treated with the monoclonal antibody against glial fibrillary acid protein (GFAP; rabbit polyclonal, dilution 1:200, Chemicon, CA) or microtubule-associated protein 2 (MAP2; mouse monoclonal, dilution 1:200, Chemicon). GFAP was employed for astrocyte identification. MAP2 was employed for neuron identification. Goat anti-mouse or anti-rabbit rhodamine-conjugated secondary antibody (dilution 1:200, Chemicon) was added for immunoreactivity identification. Subsequently, the sections were treated with the monoclonal antibody against GFP (mouse monoclonal, dilution 1:100, Santa Cruz, CA), and Zenon Alexa Fluor 488 (Mouse IgG Labeling Kit, Molecular Probes, OR) was added as the secondary antibody to identify its immunoreactivity. To avoid non-specific immunostaining by antimouse secondary reagents, negative control staining was also performed using mouse IgG. The fluorescence emitted from the secondary antibody was observed through an appropriate filter using a fluorescence microscopy (BX51, Olympus) and was digitally photographed using a cooled CCD camera equipped to the microscopy (model VB-6000/ 6010, Keyence).
5. Results 5.1. Characterization of cultured GFP-expressing bone marrow stromal cells The BMSC harvested from GFP-transgenic mouse bone marrow were isolated by their adherence to plastic and grown for two to three passages. The GFP-BMSC became comparatively homogeneous in appearance as the cells were passed. As the cells approached confluency, they assumed relatively elongated or spindle-shaped cells (Fig. 2). The findings are identical to those obtained in the experiment using the BMSC harvested from the wild-type mouse [11].
Fig. 2. Microscopic view of the cultured GFP-BMSC. The BMSC harvested from a GFP-transgenic mouse bone marrow assume relatively elongated or spindle shape (A). Transmission fluorescence microscopy reveals that all GFP-BMSC express green fluorescence (B).
Fluorescence microscopy clearly showed that all cultured GFP-BMSC emitted sufficient fluorescence for microscopic observations (Fig. 2). 5.2. In vivo visualization of transplanted GFP-expressing bone marrow stromal cells In all animals (n = 12), in vivo fluorescence imaging enabled clear visualization of a subpopulation of the transplanted GFP-BMSC that seeped out from the brain surface at the site of injection just after transplantation. However, the majority of the GFP-BMSC injected into the striatum could not be identified (Figs. 3A,D and 4A,D). The result indicates that the fluorescence optical imaging system equipped in this study could not detect green fluorescence emitted from the GFP-expressing BMSC through the skull when located in 3 mm depth of the brain. The findings were almost identical 2 weeks after transplantation (data not shown), suggesting that the migration of the transplanted cells did not begin or was not sufficient for fluorescence detection through the skull, although it was not histologically confirmed in this study. Four weeks after transplantation, however, the green fluorescence emitted from the transplanted GFP-BMSC was detected in the right parietal region in 7 out of 12 animals, strongly suggesting that the transplanted GFPBMSC could reach the surface of the cerebral cortex in the ischemic boundary zone enough for fluorescence detection
170
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
Fig. 3. In vivo white-light (A, B, C) and fluorescence (D, E, F) images after transplantation of BMSC into the mouse subjected to permanent MCA occlusion. Just after transplantation, only some transplanted GFP-BMSC that seeped out from the brain surface at the site of injection was visualized (A and D arrow heads: a small drilled hole for transplantation). Four weeks after transplantation, green fluorescence (arrow) was identified in the borderzone of infarct through the skull (B and E). The image acquired through the skull closely matched the image under the open brain after skull removal (C and F). Thus, green fluorescence (double arrow) could be clearly identified green fluorescence after the removal of skull (asterisk). Schema: the small circle and pink area represent the injection point and infarct area, respectively. The square of the red line represents the field shown in the figures.
(Fig. 3B and E). The image acquired through the skull closely matched the one of the open brain after the skull was removed (Fig. 3C and F). Of these seven mice, four were sacrificed at 4 weeks for histological analysis. A series of in vivo fluorescence optical images of the transplanted GFP-BMSC in other three animals was obtained from 4 to 12 weeks after transplantation. When assessed qualitatively, the size of the positive fluorescence area did not show apparent change, although quantitative measurements were not performed. In the remaining five animals, no fluorescence was detected through the skull 4 weeks after transplantation. However, in vivo fluorescence optical imaging could clearly detect the green fluorescence emitted from the transplanted GFP-BMSC in the right parietal region 8 or 12 weeks after transplantation in these five mice (Fig. 4). Thus, green fluorescence was observed in the right parietal region through the skull (Fig. 4B and E). Thinning of the right parietal bone using a dental drill improved the detection sensitivity of the green fluorescence (Fig. 4C and F). The findings strongly indicate that the migration and proliferation of the transplanted BMSC was not enough for fluorescence optical imaging 4 weeks after transplantation and required 8 – 12 weeks in these five animals.
5.3. Migration and differentiation of transplanted GFPexpressing bone marrow stromal cells Histological evaluation was performed 4 (n = 4) or 12 weeks (n = 8) after stereotactic transplantation of the GFPBMSC into the ipsilateral striatum of the mice subjected to permanent middle cerebral artery occlusion. In vivo fluorescence imaging detected the green fluorescence through the skull in all four mice that was sacrificed 4 weeks after transplantation. When assessed 4 weeks after transplantation, the GFP-positive cells were easily identified under fluorescence microscopy. They were found in the striatum, white matter and neocortex of the ipsilateral hemisphere, but the majority was distributed in the striatum and neocortex adjacent to infarct. Thus, the distribution of the GFPpositive cells in the neocortex correlated well with the region where the green fluorescence was detected on in vivo fluorescence optical images (Fig. 5A – D). Most of the cells had an oval shape in the striatum (Fig. 5E –H), whereas some transplanted GFP-BMSC had a triangle shape associated with fine projections, simulating neuron-like shape (Fig. 5A – D). Double fluorescence immunohistochemistry of brain sections revealed that some GFP-positive cells in the cerebral cortex, but not in the striatum, were reactive for
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
171
Fig. 4. In vivo white-light (A, B, C) and fluorescence (D, E, F) images after transplantation of BMSC into the mouse subjected to permanent MCA occlusion. Just after transplantation, only some transplanted GFP-BMSC that seeped out from the brain surface at the site of injection was detected (A and D arrow heads: a small drilled hole for transplantation). Twelve weeks after transplantation, green fluorescence (arrows) was detected in the infarct periphery through the skull (B and E). Thinning of the skull using a dental drill (C) markedly improve the detection sensitivity of green fluorescence (arrows, F). Schema: the small circle and pink area represent the injection point and infarct area, respectively. The square of the red line represents the field shown in the figures.
the neuronal marker MAP2 (Fig. 5). Therefore, the results suggest that the GFP-BMSC transplanted into the ipsilateral striatum migrated into the ischemic boundary zone through the white matter within 4 weeks and that some of them simulated the neuron in the neocortex morphologically and biologically. Other eight mice were histologically evaluated 12 weeks after transplantation. The GFP-positive cells were widely distributed in the entire brain. Especially, a number of the GFP-positive cells were found in the neocortex adjacent to infarct, supporting the findings on in vivo fluorescence images. Furthermore, the GFP-positive cells were found in the bilateral olfactory bulb, hippocampus, cerebellum and neocortex (Fig. 6A –F). Some of these cells showed neuronlike appearance. Some of them mimicked the appearance of the microvasculature in the striatum and ipsilateral cerebral cortex (Fig. 6J –L). Double fluorescence immunohistochemistry of brain sections revealed that a significant number of GFP-positive cells in the cerebral cortex, olfactory bulb, hippocampus and cerebellum were reactive for the neuronal marker MAP2 (Fig. 6A – F and J– L), and that some of GFPpositive cells in the white matter were reactive for the astrocytic marker GFAP (Fig. 6G – I). The histological findings suggested that the transplanted GFP-BMSC migrated
into the entire brain between 4 and 12 weeks and simulated the neuron and astrocyte morphologically and biologically.
6. Discussion This study was designed to sequentially image the distribution of GFP-BMSC transplanted into the brain over 12 weeks in an attempt to establish the technique to monitor the cells transplanted into the brain. Transplanted BMSC had previously been shown to migrate throughout the CNS, to become normal constituents of the host cytoarchitecture, although precise route of their movements in the CNS still remains unclear. The BMSC used in the current study had been previously characterized to express low levels of CD45 and high levels of CD90 and Sca-1 on flowcytometry. Transplanted BMSC survive in the normal or injured central nervous system, including the brain and spinal cord. Prior study has also shown the ability of these BMSC to migrate toward cerebral infarct and spinal cord injury [11]. As described above, although other investigators have reported similar results, the magnitude and timing of this effect have not been quantified [2,4,5]. In this study, we demonstrate that the BMSC harvested from transgenic mice express GFP
172
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
Fig. 5. Fluorescent immunohistochemistry images in the boundary zone of infarct (A – D) and the ipsilateral striatum (E – H) from coronal brain sections stained with antibodies against GFP (A, B, E, F), MAP2 (C, G) and the merged images (D, H). Original magnification: A, 100; E, 200; B – D and F – H, 1000.
continuously, and that in vivo fluorescence optical imaging technique enables a detection and tracking of the GFP-BMSC that migrate into the cerebral cortex over 4 –12 weeks after stereotactic transplantation into the ipsilateral striatum of mice subjected to permanent middle cerebral artery occlusion. Histological evaluation indicated that the GFP-BMSC were mainly identified in the striatum, white matter, and infarct periphery 4 weeks after transplantation. In some cases, these cells were also found in the olfactory bulb, hippocampus, cerebellum and contralateral cerebral hemisphere 8 – 12
weeks later. Serial in vivo fluorescence optical images could identify the cells distributed in the cerebral cortex adjacent to infarct and showed real-time cell migration in a single animal, with clear time dependence. The imaging technique also clarified that the cell migration toward the infarct periphery reached saturation by 4 – 12 weeks, varying among the animals. The deviation may depend on the difference in tissue damage among the animals. In addition, using a double immunohistochemistry technique, we demonstrate that the transplanted BMSC migrate
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
173
Fig. 6. Fluorescent immunohistochemistry images in the internal granular layer of the ipsilateral olfactory bulb (A – C), the mid-layer of the contralateral neocortex (D – F), the ipsilateral white matter (G – I) and the boundary zone of infarct (J – L) from the coronal brain sections stained with antibodies against GFP (A, D, G, J), MAP2 (B, E, K), GFAP (H) and the merged images (C, F, I, L).
into the ipsilateral cerebral cortex and simulate the neuron morphologically and biologically by 4 weeks, although the cells that survived in the striatum showed no significant differentiation into the neural cells. Subsequently, they migrate into the entire brain and expressed the neuronal marker, MAP2, in the ipsilateral and contralateral cerebral cortex, hippocampus and olfactory bulb 8– 12 weeks after transplantation. They also simulated neuron-like appearance morphologically. They expressed the astrocytic marker, GFAP, when they came to the white matter including the corpus callosum. These findings strongly suggest that the differentiation of the transplanted BMSC depends on their
surrounding microenvironment [9,18]. Interestingly, the transplanted cells also formed luminal structures, simulating the appearance of capillary endothelium in the entire brain, as reported previously [5]. However, very recent studies have suggested another possibility that the transplanted stem cells may fuse with the host cells and simulate the ‘‘differentiation’’ into the host cells [1,19 – 21]. Both options are possible as the mechanism by which the transplanted BMSC simulated the neural cells morphologically and biologically in the current study. GFP is a fluorescence protein identified in Aequorea Victoria in 1962 and has widely been employed to label
174
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
various kinds of cells using transgene or transfection technique [15]. Yang et al. injected mouse melanoma cells expressing very high level of GFP into the tail vein or portal vein of mice. They showed that whole-body optical images could clearly visualize metastatic lesion in the brain, liver and bone in real time. A fluorescence images (5.5 mm in diameter and 0.8 mm in depth) were obtained externally through the scalp and skull. The size of the external image correlated well with that of the metastatic lesion [23]. In another study, they injected the adenoviral vector expressing enhanced GFP into the mice brain, liver, pancreas, prostate and bone marrow. As the result, the fluorescence of the expressed GFP in these organs became visible by wholebody optical imaging system. The images made externally to the animal appeared very similar to those of the exposed organs [24]. Recently, they also developed a reversible skinflap method and could increase the detection sensitivity through the skull many-fold. In their report, they concluded that green fluorescence emitted from a single tumor cell could be detected, although the detection sensitivity was rather lower and only regional green fluorescence could be visualized in the present study [25]. Therefore, using GFP as a probe, in vivo fluorescence imaging technique would be valuable to track the fate of the cells transplanted into the organs including the brain. As they also pointed out, however, the current imaging technique has limitations, primarily the relatively short wavelength of fluorescence emitted from GFP (520 nm). The emitted light is strongly scattered by the surrounding tissue including the scalp and skull. Thus, we had to expose the skull each time in vivo fluorescence images were taken. Immediately after transplantation, the GFP-BMSC injected into the striatum could not be visualized. Only some of them could be observed at the brain surface of the injection site. More clear fluorescence images were obtained when the skull was removed or thinned 4– 12 weeks after transplantation. In fact, Yang et al. [23] reported that they could detect the green fluorescence emitted from maximally 2.2 mm depth. However, powerful new techniques of using ultra-fast lasers, dual photon imaging and ballistic photon imaging may offer large gains in sensitivity, increased depth of detection, and spatial resolution. Developments of these technologies would allow us to obtain completely noninvasive images through the skin in the future. Recently, other imaging modalities have also been attempted to track the cells transplanted into the CNS [6,13,14,26]. MRI is a non-optical method. MRI can image intact, opaque organisms in three dimensions with good spatial resolution. However, MRI requires long imaging times and consequently slows data acquisition because of the low sensitivity. More importantly, magnetic nanoparticles that label the transplanted cells cannot be succeeded to all the cells during their proliferation. Alternatively, wholebody imaging technique using the luciferase reporter technique has been believed as useful method. However, the technique requires the injection of luciferin as the substrate
for photon production and long imaging time to collect the sufficient photon [7]. In conclusion, although the in vivo imaging technique used in the current study requires improvement, the technique is feasible and represents a method for non-invasively tracking the quantity and location of the transplanted cells. The technique offers a method to study the biological aspects of the transplanted cells and to develop the optimal protocols for delivery of the BMSC for CNS regeneration. Development of such imaging technique would also be useful to monitor various kinds of stem cells transplanted into the CNS when cell therapy is clinically applied for neurological disorders. 6.1. Troubleshooting 6.1.1. Graft preparation It is crucial to use a culture flask coated with collagen I to culture the bone marrow cells in order to improve the yield of the BMSC. Prior to transplantation, it should be confirmed that the cultured BMSC express enough green fluorescence using. 6.1.2. Permanent MCA occlusion model Animal strain is essential to produce focal cerebral ischemia model constantly [12,17]. The important factors for successful production of the model are to completely obliterate the flow of the middle cerebral artery under surgical microscope and to assess neurological deficits just after surgery. 6.1.3. Transplantation It is very important to slowly inject the cell suspension over 5 min, using a programmable, automatic microinjection system and to remove the needle at 5 min after the end of injection for certain transplantation. 6.1.4. In vivo fluorescence imaging Completely dark room is essential to obtain good visualization of the green fluorescence. Alternatively, the dark box covering the animal and whole imaging system may be useful.
7. Quick procedure
Three to four weeks prior to transplantation. Harvest and culture of the GFP-BMSC. Day 0: permanent MCA occlusion in the host. Day 7: transplantation of the GFP-BMSC and in vivo fluorescence imaging 1. Day 21: in vivo fluorescence imaging 2. Day 35: in vivo fluorescence imaging 3 and removing the brain for histological analysis. Day 63: in vivo fluorescence imaging 4. Day 91: in vivo fluorescence imaging 5 and removing the brain for histological analysis.
H. Shichinohe et al. / Brain Research Protocols 13 (2004) 166–175
8. Essential literature references A similar in vivo fluorescence imaging technique using the GFP expressing cells is described by Hoffman and his colleagues [23 – 25], which served as a basis for the present study. The basic techniques for cell culture, stereotactic transplantation and fluorescence immunohistochemistry are described in detail by our previous report [11]. Mice permanent MCA occlusion model was produced according to the method describled by Majid et al. [12] and Shichinohe et al. [17] with minor modification.
Acknowledgements This study was supported by Grant-in-Aids from the Ministry of Education, Science and Culture of Japan (No. 14370424, Dr. Kuroda and No. 15390426, Dr. Iwasaki), and by a grant from Mitsubishi Pharma Research Foundation (Dr. Kuroda). The authors thank Dr. Hiroyuki Kobayashi for his helpful advice on histological findings and Ms. Yumiko Shinohe for her technical assistance in cell culture and immunohistochemistry.
References [1] M. Alvarez-Dolado, R. Pardal, J.M. Garcia-Verdugo, J.R. Fike, H.O. Lee, K. Pfeffer, C. Lois, S.J. Morrison, A. Alvarez-Buylla, Fusion of bone-marrow-derived cells with Purkinje neurons, cardiomyocytes and hepatocytes, Nature 425 (2003) 968 – 973. [2] S.A. Azizi, D. Stokes, B.J. Augelli, C. DiGirolamo, D.J. Prockop, Engraftment and migration of human bone marrow stromal cells implanted in the brains of albino rats—similarities to astrocyte grafts, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 3908 – 3913. [3] J.B. Bederson, L.H. Pitts, M. Tsuji, M.C. Nishimura, R.L. Davis, H. Bartkowski, Rat middle cerebral artery occlusion: evaluation of the model and development of a neurologic examination, Stroke 17 (1986) 472 – 476. [4] M. Chopp, X.H. Zhang, Y. Li, L. Wang, J. Chen, D. Lu, M. Lu, M. Rosenblum, Spinal cord injury in rat: treatment with bone marrow stromal cell transplantation, NeuroReport 11 (2000) 3001 – 3005. [5] D.C. Hess, W.D. Hill, A. Martin-Studdard, J. Carroll, J. Brailer, J. Carothers, Bone marrow as a source of endothelial cells and NeuNexpressing cells after stroke, Stroke 33 (2002) 1362 – 1368. [6] M. Hoehn, E. Kustermann, J. Blunk, D. Wiedermann, T. Trapp, S. Wecker, M. Focking, H. Arnold, J. Hescheler, B.K. Fleischmann, W. Schwindt, C. Buhrle, Monitoring of implanted stem cell migration in vivo: a highly resolved in vivo magnetic resonance imaging investigation of experimental stroke in rat, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 16267 – 16272. [7] R. Hoffman, Green fluorescent protein imaging of tumour growth, metastasis, and angiogenesis in mouse models, Lancet Oncol. 3 (2002) 546 – 556. [8] Y. Jiang, B.N. Jahagirdar, R.L. Reinhardt, R.E. Schwartz, C.D. Keene, X.R. Ortiz-Gonzalez, M. Reyes, T. Lenvik, T. Lund, M. Blackstad, J. Aldrich, S. Aldrich, A. Lisberg, W.C. Low, D.A. Largaespada, C.M. Verfaillie, Pluripotency of mesenchymal stem cells derived from adult marrow, Nature 418 (2002) 41 – 49. [9] C.B. Johansson, S. Momma, D.L. Clarke, M. Risling, U. Lendahl, J.
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20] [21]
[22]
[23]
[24]
[25]
[26]
175
Frisen, Identification of a neural stem cell in the adult mammalian central nervous system, Cell 96 (1999) 25 – 34. G.C. Kopen, D.J. Prockop, D.G. Phinney, Marrow stromal cells migrate throughout forebrain and cerebellum, and they differentiate into astrocytes after injection into neonatal mouse brains, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 10711 – 10716. J.B. Lee, S. Kuroda, H. Shichinohe, J. Ikeda, T. Seki, K. Hida, M. Tada, K. Sawada, Y. Iwasaki, Migration and differentiation of nuclear fluorescence-labeled bone marrow stromal cells after transplantation into cerebral infarct and spinal cord injury in mice, Neuropathology 23 (2003) 169 – 180. A. Majid, Y.Y. He, J.M. Gidday, S.S. Kaplan, E.R. Gonzales, T.S. Fenstermacher, J.D. Fenstermacher, L. Wei, D.W. Choi, C.Y. Hsu, Differences in vulnerability to permanent focal cerebral ischemia among 3 common mouse strains, Stroke 31 (2000) 2707 – 2714. M. Modo, D. Cash, K. Mellodew, S.C. Williams, S.E. Fraser, T.J. Meade, J. Price, H. Hodges, Tracking transplanted stem cell migration using bifunctional, contrast agent-enhanced, magnetic resonance imaging, NeuroImage 17 (2002) 803 – 811. M. Modo, K. Mellodew, D. Cash, S.E. Fraser, T.J. Meade, J. Price, S.C. Williams, Mapping transplanted stem cell migration after a stroke: a serial, in vivo magnetic resonance imaging study, NeuroImage 21 (2004) 311 – 317. M. Okabe, M. Ikawa, K. Kominami, T. Nakanishi, Y. Nishimune, ‘Green mice’ as a source of ubiquitous green cells, FEBS Lett. 407 (1997) 313 – 319. J. Sanchez-Ramos, S. Song, F. Cardozo-Pelaez, C. Hazzi, T. Stedeford, A. Willing, T.B. Freeman, S. Saporta, W. Janssen, N. Patel, D.R. Cooper, P.R. Sanberg, Adult bone marrow stromal cells differentiate into neural cells in vitro, Exp. Neurol. 164 (2000) 247 – 256. H. Shichinohe, S. Kuroda, T. Abumiya, T. Kobayashi, J. Ikeda, Y. Iwasaki, FK506 reduces infarct volume due to permanent focal ischemia by maintaining BAD turnover and mitochondrial integrity, Brain Res. 1001 (2004) 51 – 59. L.S. Shihabuddin, P.J. Horner, J. Ray, F.H. Gage, Adult spinal cord stem cells generate neurons after transplantation in the adult dentate gyrus, J. Neurosci. 20 (2000) 8727 – 8735. G. Vassilopoulos, D.W. Russell, Cell fusion: an alternative to stem cell plasticity and its therapeutic implications, Curr. Opin. Genet. Dev. 13 (2003) 480 – 485. G. Vassilopoulos, P.R. Wang, D.W. Russell, Transplanted bone marrow regenerates liver by cell fusion, Nature 422 (2003) 901 – 904. X. Wang, H. Willenbring, Y. Akkari, Y. Torimaru, M. Foster, M. AlDhalimy, E. Lagasse, M. Finegold, S. Olson, M. Grompe, Cell fusion is the principal source of bone-marrow-derived hepatocytes, Nature 422 (2003) 897 – 901. D. Woodbury, E.J. Schwarz, D.J. Prockop, I.B. Black, Adult rat and human bone marrow stromal cells differentiate into neurons, J. Neurosci. Res. 61 (2000) 364 – 370. M. Yang, E. Baranov, P. Jiang, F.X. Sun, X.M. Li, L. Li, S. Hasegawa, M. Bouvet, M. Al-Tuwaijri, T. Chishima, H. Shimada, A.R. Moossa, S. Penman, R.M. Hoffman, Whole-body optical imaging of green fluorescent protein-expressing tumors and metastases, Proc. Natl. Acad. Sci. U. S. A. 97 (2000) 1206 – 1211. M. Yang, E. Baranov, A.R. Moossa, S. Penman, R.M. Hoffman, Visualizing gene expression by whole-body fluorescence imaging, Proc. Natl. Acad. Sci. U. S. A. 97 (2000) 12278 – 12282. M. Yang, E. Baranov, J.W. Wang, P. Jiang, X. Wang, F.X. Sun, M. Bouvet, A.R. Moossa, S. Penman, R.M. Hoffman, Direct external imaging of nascent cancer, tumor progression, angiogenesis, and metastasis on internal organs in the fluorescent orthotopic model, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 3824 – 3829. Z.G. Zhang, Q. Jiang, R. Zhang, L. Zhang, L. Wang, P. Arniego, K.L. Ho, M. Chopp, Magnetic resonance imaging and neurosphere therapy of stroke in rat, Ann. Neurol. 53 (2003) 259 – 263.