|Home | About | Journals | Submit | Contact Us | Français|
Cellular transplantation in the form of bone marrow has been one of the primary treatments of many lysosomal storage diseases (LSDs). Although bone marrow transplantation can help central nervous system manifestations in some cases, it has little impact in many LSD patients. Canine models of neurogenetic LSDs provide the opportunity for modeling central nervous system transplantation strategies in brains that more closely approximate the size and architectural complexity of the brains of children. Canine olfactory bulb-derived neural progenitor cells (NPCs) isolated from dog brains were expanded ex vivo and implanted into the caudate nucleus/thalamus or cortex of allogeneic dogs. Canine olfactory bulb-derived NPCs labeled with micron-sized superparamagnetic iron oxide particles were detected by magnetic resonance imaging both in vivo and postmortem. Grafts expressed markers of NPCs (i.e. nestin and glial fibrillary acidic protein), but not the neuronal markers Map2ab or β-tubulin III. The NPCs were from dogs with the LSD mucopolysaccharidosis VII, which is caused by a deficiency of β-glucuronidase. When mucopolysaccharidosis VII canine olfactory bulb-NPCs that were genetically corrected with a lentivirus vector ex vivo were transplanted into mucopolysaccharidosis VII recipient brains, they were detected histologically by β-glucuronidase expression in areas identified by antemortem magnetic resonance imaging tracking. These results demonstrate the potential for ex vivo stem cell-based gene therapy and noninvasive tracking of therapeutic grafts in vivo.
The isolation of neural progenitor cells (NPCs) that can be expanded ex vivo while maintaining their ability to self-renew and generate mature progeny has opened the way to transplantation therapy for central nervous system diseases (1–7). The first transplantation studies of ex vivo expanded NPCs showed that NPCs were capable of integrating into the existing neuroarchitecture (3, 6, 8–14). Moreover, NPCs could be expanded several logs in vitro, thereby providing a large supply of cells for transplantation (5, 15). The discovery that multipotent NPCs existed within the postnatal brain (1, 8, 16–20) broached the concept of autologous cell transplantation that would preclude the need for the immunosuppressive therapy accompanying transplantation procedures.
Neurodegenerative diseases are mediated by a broad spectrum of genetic and acquired processes that might benefit from central nervous system transplantation therapy. For example, lysosomal storage diseases (LSDs), which occur in as many as 1 in 7,000 births (21), are genetic diseases caused by a single lysosomal enzyme deficiency, and many of them are characterized by prominent neurodegeneration (22, 23). The principal neurological manifestations of LSDs usually are mental retardation or progressive dementia, but may also include seizures, ataxia, spasticity, and psychiatric disorders such as psychosis (23). Although the exact mechanisms of neuronal dysfunction and death remain elusive, the pathological features of LSDs include ectopic neuron dendrites, axonal enlargement, myelinopathy, astrocytosis, and microglial activation (24).
Mucopolysaccharidosis VII (MPS VII) is a neurodegenerative LSD that is caused by β-glucuronidase (GUSB) enzyme deficiency and which first manifests in early childhood (25). The existence of both small and large animal homologues of human MPS VII (26–30) makes this disease a useful model for transplantation therapy for LSDs. Cellular transplantation into the brains of MPS VII-affected mice for gene therapy has been performed using both neural and nonneural cell types (10, 31–35). In studies using NPCs for therapeutic transplantation into the MPS VII mouse model, widespread engraftment throughout the brain and therapeutic levels of enzyme expression that result in clearance of cell storage material in the regions of engraftment have been reported (10, 32, 34). These studies provide proof of principle that NPC transplantation is effective for the treatment of the central nervous system manifestations of MPS VII. Brains of children are, however, architecturally more complex, and nearly 2,000-fold larger than brains of mice. By contrast, because there is only a 10-fold size difference between dog and human brains, modeling cell transplantation therapy in the dog brain would more closely address the architectural and technical complexities inherent in cell transplantation into the human brain.
Previous cell transplantation studies of LSDs have used embryonic NPCs, whereas postnatally derived NPCs afford the potential for autologous transplantation and are more readily available as a transplant source than embryonic NPCs. Magnetic resonance imaging (MRI) is an effective method for detecting cell grafts in living animals (36), and in vivo graft monitoring will be essential for translating experimental stem cell-based therapies to the clinic. Here, we have modeled ex vivo gene therapy and in vivo graft monitoring of primary postnatal MPS VII-affected NPCs in a canine model of brain transplantation.
Mixed breed dogs were raised in the Animal Models Core of the W. F. Goodman Center for Comparative Medical Genetics, University of Pennsylvania School of Veterinary Medicine, according to National Institutes of Health and U.S. Department of Agriculture guidelines for the use of animals in research. All protocols were approved by the Institutional Animal Care and Use Committee of the University of Pennsylvania.
Screening for MPS VII dogs was performed by evaluating Wright-Giemsa-stained blood films for the presence of storage granules in white blood cells. Normal, carrier, or MPS VII genotype was then established by polymerase chain reaction for the mutation followed by restriction enzyme digestion (37). This method introduces an allele-specific restriction enzyme recognition site only in the polymerase chain reaction products from wild-type, not MPS VII alleles; samples from clinically normal carriers that are heterozygous for the mutation show partial digestion.
A 23-day-old MPS VII-affected dog was humanely killed by intravenous injection of a barbiturate solution. The brain was removed, placed into a balanced salt solution, and dissected, and the olfactory bulbs (OBs) isolated. The OB tissue was minced and then digested in 0.25% trypsin (Worthington Biochemical, Lakewood, NJ) in a 37°C water bath for 45 minutes to 1 hour. The enzymatic digestion was stopped by the addition of fetal bovine serum (Hyclone, Logan, UT). The tissue was then incubated with DNAse I (Sigma, St Louis, MO) for 15 minutes in a 37°C water bath and triturated to a single cell suspension with successively smaller-diameter pipettes, ending with a flame-polished Pasteur pipette. The cell suspension was centrifuged at 700 revolutions per minute at 4°C for 8 minutes, resuspended in 10% fetal bovine serum plating medium (see later), and triturated. The total number of viable cells was determined by manual count on a hemacytometer; cell viability was assessed using trypan blue exclusion (0.4%; Sigma).
Newly isolated canine OB-NPCs (OB-cNPCs) were plated into 25-cm2 tissue culture flasks (Corning, Acton, MA) coated with 10 μg/mL poly-D-lysine (Sigma) and expanded, as reported previously (38). Briefly, cells were plated at a concentration of 4 × 104/cm2 in 10% serum-containing medium consisting of Dulbecco modified Eagle medium: F12 (1:1 ratio; GibcoBRL Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, 1% N2 supplement (GibcoBRL), 1% penicillin/streptomycin/fungizone (100 U/mL penicillin, 100 μg/mL streptomycin, 0.25 μg/mL amphotericin B; GibcoBRL), and 1% L-glutamine (2 mM; GibcoBRL) with 20 ng/mL epidermal growth factor (recombinant murine; Roche, Nutley, NJ), 20 ng/mL basic fibroblast growth factor (recombinant human; Promega, Madison, WI), and heparin (5 μg/mL; Sigma). After 24 to 48 hours, the medium was changed to a serum-free feeding medium consisting of Dulbecco modified Eagle medium: F12 supplemented with 1% N2 supplement, 1% PSF, and 1% L-glutamine with 20 ng/mL epidermal growth factor, 20 ng/mL basic fibroblast growth factor, and 5 μg/mL heparin. The OB-cNPC cultures were maintained at 37°C in humidified 5% carbon dioxide tissue culture incubators. Cultures were fed every 3 to 5 days by changing half of the medium and adding fresh growth factors. Cultures were passaged by trypsinizing (0.05% trypsin-EDTA; GibcoBRL).
Canine OB-NPCs were transduced with a viral vector containing the human GUSB gene driven by the human GUSB promoter. The vector was a third-generation self-inactivating human immunodeficiency virus vector (39). The self-inactivating vectors are replication deficient, and the promoter ability of the long terminal repeat (LTR) is abolished because of a deletion in the 3′ U3 region of the LTR. Transgene expression is driven solely by an internal promoter. The cultures were transduced by incubating them with the viral supernatant supplemented with 8 μg/mL hexadimethrinebromide (polybrene; Sigma) in conditioned NPC medium at a multiplicity of infection of at least 10. After 4 to 6 hours, fresh medium was added overnight. The transduction efficiency was determined to be 59% on the basis of histochemical staining for GUSB activity (636 GUSB+/1087 total cells counted).
One batch of OB-cNPCs was labeled with super-paramagnetic iron oxide particles (MPIOs) for imaging experiments. Superparamagnetic iron oxide particles are microspheres composed of an inert polymer matrix of which magnetite iron oxide and a fluorescein isothiocyanate analog (Dragon Green) are components (Bangs Laboratories, Fishers, IN). Microspheres averaged 0.9 μm in diameter. The MPIOs were added to the OB-cNPC culture medium at a concentration of 1.5 × 108 particles/mL (40) and incubated overnight in the tissue culture incubator. Cells were harvested the following day for transplantation.
Before transplantation, MPIO-labeled and unlabeled OB-cNPCs were trypsinized and washed twice in PBS by centrifuging at 700 revolutions per minute for 8 minutes at 4°C. After the first wash, the cells were resuspended in 1 mL of PBS, and the total number of viable cells was determined by manual count on a hemacytometer; cell viability was assessed using trypan blue exclusion. The cells were then centrifuged and resuspended in a volume of PBS to yield approximately 5 × 104 cells/μL. The suspension was placed on ice until transplantation.
Eight MPS VII-affected dogs were injected with cNPCs. For the injections, all dogs were anesthetized with intravenous propofol (4 mg/kg), intubated, and maintained on isoflurane inhalant anesthesia. The dogs were divided into 2 groups by age: 23-day-old dogs (n = 4) and dogs aged 5 months (n = 4). For the younger group, a single bolus injection into the caudal portion of the left caudal caudate nucleus/rostral thalamus was performed using ultrasound guidance (GE logiq9, 8 MHz microcurved transducer, General Electric Medical Systems, Milwaukee, WI). After a midline skin incision, the skull was broached manually with an 18-gauge needle, and 5 × 104 GUSB-transduced OB-cNPCs were injected into the parenchyma with a 26-gauge Hamilton syringe over the course of 5 minutes. All dogs were injected with a total volume of 1 μL; 2 dogs received MPIO-labeled GUSB-OB-cNPCs, and the other two received unlabeled GUSB-OB-cNPCs. Posttransplantation surgery, all dogs received 10 mg of cyclosporine (Atopica, Novartis, East Hanover, NJ) divided every 12 hours until the day of killing and 13.75 mg/kg of amoxicillin/clavulanate (Clavamox, Pfizer, Exton, PA) twice daily for 10 days.
The older group of dogs received a series of injections into the right parietal cortex via a stereotaxic apparatus. A midline incision in the skin was made, and the skull was broached using a burr drill. Untransduced MPIO-labeled OB-cNPCs from an MPS VII dog (5 × 104 cells/μL) were injected with a 26-gauge Hamilton syringe using the following stereotaxic coordinates: 10 mm right of midline, 25 mm ventral to the pial surface, 23 mm rostral to the interaural line. The cells were deposited up the injection tract in 1-μL aliquots starting from the most ventral point (25 mm ventral to the pial surface) and every 2 mm dorsally. The aliquots were injected over 1 minute, and there was a 1-minute delay between each injection; at the end of the injection series (10–12 injections), there was a 5-minute delay before the needle was extracted. Posttransplantation surgery, the dogs received 50 mg of cyclosporine (Atopica) divided every 12 hours until the day the animals were killed and 13.75 mg/kg of amoxicillin/clavulanate (Clavamox) twice daily for 10 days.
Two animals from the younger group that received MPIO-labeled GUSB-OB-cNPCs were imaged using MRI. Monitoring of grafted cells was performed using a 4.7 T horizontal bore magnet equipped with a 12-cm 25 G/cm gradient insert 50 cm horizontal bore magnet with a Varian imaging console (Varian, Inc, Palo Alto, CA). Dogs were anesthetized with propofol and maintained with 1% isoflurane in air during the MRI experiment. A 12-cm linear polarized transmit-receive Litz MRI coil (Doty Scientific, Inc, Columbia, SC) was used for these studies. Core body temperature and electrocardiogram were monitored during the examination using an MRI-compatible unit (SA instruments, Inc, Stony Brook, NY). The body temperature of animals was maintained at 37°C ± 1°C by blowing warm air through the magnet bore, which was regulated by a feedback loop. In vivo 3-dimensional gradient echo imaging was performed with the following parameters: time to repitition/time echo (TR/TE) 50/4.5 milliseconds, 4 scans, field of view of 10 cm3, 128 × 128 × 128 matrix, slab thickness of 15 cm leading to an isotropic resolution of 781 μm, and acquisition time of about 1 hour.
At selected time points, 1 dog from the younger group and 2 from the older group were humanely killed with an intravenous injection of a barbiturate solution. Immediately before death, the dogs were anesthetized and given intravenous heparin (1,000 U/mL). After death, intracardiac perfusion with cold 0.9% saline followed by 4% paraformaldehyde solution was performed. The brains were removed, fixed in 4% paraformaldehyde for 24 hours, and kept at 4°C until ex vivo MRI. Imaging of perfused fixed brains was performed using the same magnet as in the in vivo imaging studies. Dog brains were immersed in a container with 30 mL Fomblin (Ausimount, Thorofare, NJ) and imaged using a 70-mm transmit-receive transverse electromagnetic MRI coil (InsightMRI, Worcester, MA). Fomblin does not contain hydrogen protons and provides a completely dark background on an MR image (41). Imaging was performed using a 3-dimensional gradient echo pulse sequence with the following parameters: TR/TE 10/4.5 milliseconds, 32 scans, FOV approximately 13 × 6.5 × 6.5 cm3, 768 × 384 × 384 matrix, slab thickness of 7 cm, with a final isotropic resolution of 169 μm and acquisition time of approximately 11 hours.
The analysis of in vivo and ex vivo 3-dimensional images was performed using the “Image Browser” program, which is a part of the standard Varian software or PC version of ImageJ program.
Dogs were humanely killed and transcardially perfused with cold 0.9% physiological saline followed by 4% paraformaldehyde. Brains were removed, postfixed in 4% paraformaldehyde for 24 hours, and then divided into right and left hemispheres. The injected half of the brain was sectioned at the rostral and caudal borders of the lateral ventricle and placed in 30% sucrose solution for 24 to 48 hours. The sections were then frozen in OCT (Tissue-Tek, Sakura Finetek, Torrance, CA) for cryosectioning in a sagittal plane.
Frozen tissue sections were assayed for enzymatic activity by staining with a naphthol-AS-BI-β-D-glucuronide substrate as reported previously (42). Briefly, slides were thawed for 5 minutes at room temperature and treated with chloral-formal-acetone fixative for 30 minutes at 4°C. The slides were washed 3 times with 0.05 M sodium acetate buffer (pH 4.5). The third wash in 0.05 M sodium acetate was performed at 65°C for all sections. Heat-inactivation serves to distinguish between the human and canine GUSB proteins; canine GUSB is inactivated at 65°C, whereas the human protein is stable. Next, 0.25 mM naphthol-AS-BI-β-D-glucuronide (pH 4.5) was added, and the slides were incubated at 4°C for 4 hours, at which point the solution was removed. Equal volumes of 4% pararosaniline solution in 2N hydrochloric acid and 4% sodium nitrite solution in deionized water were mixed. This solution was diluted 1:500 in 0.25 mM naphthol-AS-BI-β-D-glucuronide (pH 5.2), placed on slides, and incubated at 37°C overnight.
The frozen brain regions were sectioned into 20-μm slices on a cryotome and subjected to immunohistochemical staining. The polyclonal primary antibodies used were: rabbit polyclonal anti-nestin, 1:20 dilution (rabbit 130; kind gift of R. McKay, National Institutes of Health), and rabbit polyclonal anti-glial fibrillary acidic protein (GFAP), 1:100 dilution (Chemicon/Millipore, Temecula, CA). Primary mouse monoclonal antibodies were: anti-β tubulin III, 1:100 dilution (immunoglobulin G [IgG]; Chemicon); and anti-Map2ab, 1:100 dilution (IgG; Chemicon). Secondary fluorescent antibodies used were goat anti-rabbit IgG 594 Alexafluor, 1:300 dilution (Molecular Probes, Eugene, OR), and goat anti-mouse IgG 594 Alexafluor, 1:300 dilution (Molecular Probes).
Tissue sections were thawed, then blocked for 1 hour in 10% goat serum (GibcoBRL) with 0.2% Triton X-100 (Sigma) in PBS. Primary antibody incubation was performed for 2 hours at room temperature or overnight at 4°C in PBS with 2% goat serum and 0.2% Triton X-100. The sections were washed 3 times in PBS and then incubated with secondary antibody in PBS for 1 hour at room temperature or overnight at 4°C. After another 3 PBS washes, the slides were mounted in Vectashield containing 4′,6-diamidino-2-phenylindole (Vector Laboratories, Burlingame, VT).
Immunolabeled sections were scanned with a Leica DM IRE2 HC fluo TCS 1-B-UV microscope coupled to a Leica TCS SP2 spectral confocal system/UV (Leica, Bannockburn, IL). The 3 fluorochromes were then sequentially scanned.
Canine OB-derived NPCs were isolated from an MPS VII-affected dog, expanded, and transduced ex vivo with a lentiviral vector containing the human GUSB gene. The OB-cNPCs were approximately 60% GUSB+ before transplantation. Because the recipient dogs were GUSB deficient, expression of GUSB served as a reporter gene for graft identification on histological sections. To visualize engrafted cells in vivo, cells were labeled preimplantation with the MRI contrast agent. The contrast agent, micron-sized MPIOs, permits identification of cells by MRI and contains a fluorescein isothiocyanate-like dye for visualizing engrafted cells in the event of reporter gene silencing (40, 43). Micron-sized MPIOs are taken up by cells via endocytosis; they have been used previously to label nonneural stem cells with no adverse effects (43, 44). All OB-cNPCs labeled with fluorescently tagged MPIOs (OB-cNPC-MPIO) immediately before transplantation contained one or more MPIOs (Fig. 1).
Two groups of 4 dogs each were transplanted at 23 days and 5 months of age, respectively (Table). For the younger group of dogs, transduced OB-cNPCs (GUSB-OB-cNPCs) were injected as a single bolus into the caudal portion of the left caudal caudate nucleus/rostral thalamus. At 3 weeks of age, the canine fontanelle is incompletely calcified, thereby permitting the brain to be visualized via ultrasound. Using ultrasound guidance, 5 × 104 GUSB-OB-cNPCs were transplanted in each dog (Fig. 2). Two dogs received GUSB-OB-cNPCs labeled with MPIOs (GUSB-OB-cNPC-MPIOs), whereas the other 2 received unlabeled GUSB-OB-cNPCs. To determine whether we could monitor the position and extent of cell grafts noninvasively, the 2 dogs that received MPIO-labeled GUSB-OB-cNPCs were subjected to MRI. The animals were imaged in vivo 8 days after transplantation. Figures 3A and and5A5A show representative brain MR images of both 31-day-old dogs. A hypointense area representing the implanted cells is easily detected 8 days after transplantation in each dog. Immediately after in vivo MRI, one of the dogs that received MPIO-labeled GUSB-OB-cNPCs died of anesthetic complications; the remaining dog was killed at 4 weeks posttransplantation. The graft remained detectable at 4 weeks after implantation when high-resolution ex vivo MRI was performed (Fig. 3).
The 5-month-old group of dogs received a series of injections into the right parietal cortex using a stereotaxic apparatus. The OB-cNPCs were injected along a track at 2-mm intervals in 10 to 12 deposits at 5 × 104 cells/μL. The graft tract could be detected by ex vivo MRI in the postmortem brains 8 weeks postimplantation (Figs. 3E, F). Optical imaging of frozen brain sections demonstrated the presence of labeled cells detected by in vivo and ex vivo MRI in all dogs evaluated (Figs. 3C, D, G, and Figs. 5B–D). All dogs received cyclosporine for immune suppression post-transplantation. In the fifth week of treatment, 2 dogs developed progressive gingival hyperplasia, which is a reported side effect of cyclosporine and which did not affect feeding behavior.
To assess the immunophenotype of transplanted cells, graft sections from 2 dogs implanted with untransduced OB-cNPC-MPIOs were probed with antibodies that recognize markers for neural progenitors, astrocytes, and neurons. The phenotypic markers were coregistered with MPIO-labeled cells using confocal microscopy. At 2 weeks postimplantation, the grafts contained cells immunopositive for nestin and GFAP (Figs. 4A, C). When identifying immunopositive cells containing MPIOs, care was taken to localize the MPIOs immediately adjacent to the nucleus. Immunoreactivity for the neuronal markers Map2ab and β-tubulin III was not seen within the grafts (data not shown).
To determine whether genetically modified OB-cNPCs could be tracked in a developing brain, a group of 23-day-old dogs was injected with a bolus of GUSB-OB-cNPCs. These recipients were affected by MPS VII, thus, donor MPS VII NPCs expressing the GUSB enzyme from a lentivirus vector could be detected against the negative background of the host brain (45). All dogs in this group received cyclosporine after transplantation to suppress the immune system; no deleterious effects were detected. The graft site was located postmortem in 3 of 4 transplanted dogs that received transduced OB-cNPCs: in both of the dogs that received MPIO-labeled GUSB-OB-cNPCs and in 1 dog that received only GUSB-OB-cNPCs (Table). The GUSB-positive cells were present in 2 of 3 engrafted dogs evaluated (Figs. 5B–G). One of these dogs received NPCs labeled with MPIOs, which hampered visualization of the red GUSB staining (Figs. 5B–D); in the other dog that received MPIO-labeled GUSB-OB-cNPCs, the graft did not stain for GUSB (Fig. 3C). The lack of GUSB staining in the third graft possibly occurred because this brain was subjected to ex vivo MRI, where the tissue was immersed in Fomblin. We have found that ex vivo MRI interferes with the enzymatic reaction needed to detect GUSB (unpublished data).
Canine NPCs possess characteristics that are advantageous for therapeutic transplantation. They can be expanded ex vivo manyfold and are able to differentiate in vitro into neurons and glia (38, 46). Moreover, the use of cNPCs derived from the postnatal OB affords the opportunity for autologous transplantation. There is a large body of literature concerning NPC transplantation, but a relative paucity of data on the use of postnatal NPCs and very few studies with NPCs derived from regions that would permit autologous sampling (11, 13, 47–52).
Although the distribution of lesions associated with LSD is nearly global (22, 23, 53, 54) because of the nature of lysosomal enzymes, the LSDs represent a unique therapeutic opportunity. The acid hydrolases targeted for cellular lysosomes are posttranslationally tagged with mannose-6-phosphate residues that are recognized by surface receptors on neighboring cells (55). Thus, lysosomal enzymes released into the extracellular milieu can be taken up by adjacent cells. This process of cross correction is the basis of therapeutic strategies for the LSDs. Cross correction permits the establishment of overlapping “spheres of correction” produced from the expression of a lysosomal enzyme in foci of transplanted or endogenously transduced cells (35, 56, 57); this obviates the need to transduce every cell in an affected individual. Thus, for transplantation therapy of LSDs, 3 criteria are necessary: 1) efficient transduction of NPCs, 2) successful engraftment, and 3) sustained expression of the deficient enzyme. Our transplantation experiments demonstrate that cNPCs were efficiently transduced using a lentiviral vector that codes for the human GUSB gene, grafts could be identified in 3 of 4 dogs studied, and gene expression was detected at the latest time point evaluated, that is, 4 weeks postimplantation.
Canine OB-derived NPCs can differentiate into cells that express neuronal and glial antigens in vitro (38, 46), but we found no evidence of neuronal differentiation after transplantation into the canine cortex. These results are consistent with previous transplantation studies in which human or rodent NPCs grafted into nonneurogenic regions differentiated into glia (11, 13, 51, 58) or remained immature (59). For ex vivo gene therapy to work in the LSDs, however, NPC differentiation is not essential as long as enzyme delivery is accomplished.
The OB-cNPC grafts labeled with MPIOs were readily detectable in vivo by MRI and in histological sections. Studies using MPIO labeling have shown no adverse effects of the particles when used with human hematopoietic stem cells or when injected into murine embryos (43, 44). Studies with smaller superparamagnetic iron oxide particles, however, have shown subtle effects of smaller superparamagnetic iron oxide particles labeling on the differentiation of mesenchymal stem cells (60). The use of MPIO-labeled cells for transplantation allows serial in vivo graft monitoring that more closely models what would occur in human patients. In the canine cortical grafts, the nestin+ and GFAP+ immunophenotype of cells containing MPIOs suggests that a portion of the cells containing MPIOs are donor derived. Although some of the MRI signal may arise from endogenous phagocytic cells that have engulfed nonviable donor cells or free MPIOs, the marker studies indicate that some MRI-detectable cells are undifferentiated cNPCs and/or cNPCs that have differentiated into mature astrocytes. Others have reported a similar problem with the poor correlation between iron signal and stem cell survival (61). In mice, we found that although iron-positive microglia were present in some areas, they were located in regions with the engrafted donor cells, as indicated by reporter gene labeling (Sergey Magnitsky, PhD, unpublished data). Although the use of a transcription-dependent contrast agent (62) would ensure that MRI signal corresponds to viable graft rather than endogenous cells, MPIO and smaller superparamagnetic iron oxide particle labeling does allow grafts to be located and even mapped (63). Thus, we were able to detect a difference in brain engraftment patterns between primary and immortalized NPCs after neonatal intraventricular injection (Sergey Magnitsky, PhD, unpublished data).
In summary, these studies establish a model for ex vivo gene therapy and therapeutic transplantation in a large animal LSD model. Studying changes in storage pathology will require larger grafts that cover greater volumes of brain, as well as longer periods. The serial in vivo monitoring of neural grafts will be essential for translation from an animal model to human therapy because of the need for noninvasive methods to follow the treatments longitudinally in individuals. The use of MPIOs permitted the in vivo identification of posttransplantation grafts.
The authors thank P. O’Donnell and Dr M. Haskins of the Animal Models Core (RR02512) for expert assistance and A. Polesky and E. Cabacungan for expert technical assistance.
This work was supported by Grant Nos. NS56243 (to John Wolfe), and HD48582 (to Harish Poptani) from the National Institutes of Health. This work was also supported by Training Grant No. RR07063 from the National Center for Research Resources (to Raquel Walton).