Cellular repair in the parkinsonian nonhuman primate brain

Abstract

Parkinson disease (PD) is a neurodegenerative disorder that provides a useful model for testing cell replacement strategies to rejuvenate the affected dopaminergic neural systems, which have been destroyed by aging and the disease. We first showed that grafts of fetal dopaminergic neurons can reverse parkinsonian motor deficits induced by the toxin, 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP), validating the feasibility of cellular repair in a primate nervous system. Subsequent clinical trials in Parkinson patients showed encouraging results, including long-term improvement of neurological signs and reduction of medications in some patients. However, many experienced little therapeutic benefit, and some recipients experienced dyskinesias, suggesting a lack of regulated control of the grafts. We have since attempted to improve cell replacements by placing grafts in their correct anatomical location in the substantia nigra and using strategies such as co-grafting fetal striatal tissue or growth factors into the physiologic striatal targets. Moreover, the use of fetal cells depends on a variable supply of donor material, making it difficult to standardize cell quality and quantity. Therefore, we have also explored possibilities of using human neural stem cells (hNSCs) to ameliorate parkinsonism in nonhuman primates with encouraging results. hNSCs implanted into the striatum showed a remarkable migratory ability and were found in the substantia nigra, where a small number appeared to differentiate into dopamine neurons. The majority became growth factor-producing glia that could provide beneficial effects on host dopamine neurons. Studies to determine the optimum stage of differentiation from embryonic stem cells and to derive useful cells from somatic cell sources are in progress.

FIG. 1.

(A and B) Co-grafts of fetal ventral mesencephalon (asterisks) and striatum (arrows) implanted at 2.5 mm (A) and 5.0 mm (B) apart show survival of dopaminergic neurons (arrows), seen to advantage in C, and outgrowth of neurites (arrows in D) preferentially to the striatal grafts. Similar extension of dopamine positive neurites resulting in dense patterns of terminal fibers was not seen as extensively when the grafts were separated by 7.5 mm (not shown). (E) Injection of adeno-associated virus/glial-derived neurotrophic factor (AAV2-GDNF) directly into the host striatum (arrows) resulted in enhanced survival and neuritic outgrowth from grafts of dopaminergic neurons (GR) with a prominent polarity of the neurites extending toward the region of the vector as seen in detail in rectangles A and B. (F–G) Images of a grafted dopaminergic neuron that shows granules (arrows) of Fluoro-Gold transported in a retrograde direction from the site of injection in the target region, i.e., the striatum, to the graft in the mesencephalon. F is stained for tyrosine hydroxylase alone, G is fluorescence of Fluoro-Gold alone, and G is a combined view of E and F. (A–D are adapted from Sladek et al. 2008; E is reprinted with permission from Elsworth et al. 2008, and F–H are unpublished images from Redmond et al. 2009).

FIG. 2.

Over the course of three studies, spontaneous human neural stem cell (hNSC) differentiation (A–C), hNSC migration (D–E), and the mitigating effects of hNSC on 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced effects were studied. MPTP-treated monkeys were implanted bilaterally with undifferentiated hNSCs in the caudate and unilaterally in the substantia nigra (SN). hNSCs were labeled with nuclear bromodeoxyuridine (BrdU) prior to implantation. To evaluate spontaneous differentiation, confocal microscopic analysis of TH and BrdU staining in the SN revealed host nigral neurons that were TH-ir (tyrosine hydrolase-immunoreactive) in the cytoplasm but negative for BrdU in the nucleus (*)(A). A subpopulation of TH-ir cells were BrdU-ir (* red nuclei)(B). Red and blue lines indicate corresponding points in the orthogonal planes, confirming localization of the label within the pictured cell after the summation of serial optical sections. (C) Some donor-derived BrdU-ir cells in this region were also immunoreactive for a secondary marker of dopamine neurons, dopamine transporter (DAT) (closed arrow). These cells were juxtaposed with host DAT-ir neurons, indicated by the lack of black BrdU staining in the nucleus (open arrow). Migration of hNSCs was indicated by the fact that very few BrdU-ir cells were found in the caudate nucleus, which was bilaterally implanted 4 or 7 months prior to analyses (D). There were significantly more BrdU-ir cells found along the nigrostriatal pathway (ST (striatal) end and SN end) than in the caudate nucleus, an area specifically implanted with hNSCs. As many cells were found in the thalamus, an unimplanted site, as were found in the caudate nucleus. (*) Significantly greater than the same region in the 4-month-old animals; (^) significantly different from the other brain areas. Significance level p < 0.05. Parasagittal section of a monkey brain, stained for TH, with boxes depicting the four areas in which counts of BrdU⁺ cells were made (E). The caudate nucleus was implanted bilaterally with BrdU prelabeled hNSCs, whereas the SN was implanted unilaterally only. Most BrdU-positive cells appeared in the areas between the caudate and the SN, along the nigrostriatal pathway (boxes St end and SN end). The thalamus also was included as a control area. Cx, Cerebral cortex; Cd, caudate; ac, anterior commissure; Th, thalamus; SN, substantia nigra. Many apparently undifferentiated BrdU-ir hNSCs (circle indicates some of these black cells) were found intermingled with host Th-ir nigral neurons (brown cells, arrows) in both the implanted and unimplanted sides of the SN (F). After MPTP lesioning, TH-ir neurons found in the caudate nucleus increase in number (G), a compensatory but abnormal change induced by MPTP. TH-ir neurons in the caudate nucleus are typically small bipolar cells with long varicose processes. In MPTP-exposed brains implanted with hNSCs, the number of TH-ir cells in the caudate nucleus decrease to near normal control parameters, even though the hNSCs migrated away from the caudate nucleus (H). (A, B, C, and H are modified and reproduced, with permission, from Redmond et al. D and E are modified and reproduced, with permission, from Bjugstad et al. F and G are previously unpublished.)

FIG. 3.

Dopaminergic neurons differentiated from human embryonic stem (hES) cells release dopamine into medium. Briefly, human embryonic stem cells (H1 from Wicell) (passages P44–P54) were cultured in an undifferentiated state on feeder-free and serum-free conditions. They were differentiated into neural precursor cells (NPCs) by culturing them as floating cell aggregates (embryonic bodies) for 3 weeks in a medium supplemented with recombinant human Noggin and basic fibroblast growth factor (bFGF). The NPCs exhibited columnar morphology, formed neural rosettes (arrows, A) and expressed Pax6, immature neuronal marker (E). The selected neural structures were then cultured in suspension for 1 week to generate neurospheres (B). The majority of cells in the spheres stained positively for the neural precursor marker nestin (F). The spheres were then differentiatied into dopaminergic neurons according to the method of Yang et al. (C). The number of TH (green)/βIII-tubulin (red)–positive neurons (arrowheads) increased from 1 week (G) to 4 weeks (H) of the differentiation protocol. To explore whether the hESC-derived TH⁺ neurons were truly dopaminergic, hESC-derived dopaminergic neurons were analyzed for the release of dopamine (D). Cultures containing dopamine-differentiated cells were incubated in neural differentiation medium (NDM, control conditions) or in the same medium supplemented with KCl (which causes activity-dependent dopamine release) for 30 min. The media were then collected and dopamine levels were assayed by reverse-phase high-performance liquid chromatography (HPLC). As the percentage of TH⁺ cells in each plate varied somewhat (20–50%), examples of the potassium-stimulated release into the media from the cells are shown (D). In control conditions, 60 and 300 pg·mL⁻¹ dopamine were detected. The dopamine levels were elevated when the neurons were depolarized with potassium (72 and 450 pg·mL⁻¹, respectively).