SEARCH

SEARCH BY CITATION

Keywords:

  • dopamine neurons;
  • human embryonic stem cells;
  • in vitro differentiation;
  • Parkinson's disease;
  • transplantation

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

Human embryonic stem (hES) cells, due to their capacity of multipotency and self-renewal, may serve as a valuable experimental tool for human developmental biology and may provide an unlimited cell source for cell replacement therapy. The purpose of this study was to assess the developmental potential of hES cells to replace the selectively lost midbrain dopamine (DA) neurons in Parkinson's disease. Here, we report the development of an in vitro differentiation protocol to derive an enriched population of midbrain DA neurons from hES cells. Neural induction of hES cells co-cultured with stromal cells, followed by expansion of the resulting neural precursor cells, efficiently generated DA neurons with concomitant expression of transcriptional factors related to midbrain DA development, such as Pax2, En1 (Engrailed-1), Nurr1, and Lmx1b. Using our procedure, the majority of differentiated hES cells (> 95%) contained neuronal or neural precursor markers and a high percentage (> 40%) of TuJ1+ neurons was tyrosine hydroxylase (TH)+, while none of them expressed the undifferentiated ES cell marker, Oct 3/4. Furthermore, hES cell-derived DA neurons demonstrated functionality in vitro, releasing DA in response to KCl-induced depolarization and reuptake of DA. Finally, transplantation of hES-derived DA neurons into the striatum of hemi-parkinsonian rats failed to result in improvement of their behavioral deficits as determined by amphetamine-induced rotation and step-adjustment. Immunohistochemical analyses of grafted brains revealed that abundant hES-derived cells (human nuclei+ cells) survived in the grafts, but none of them were TH+. Therefore, unlike those from mouse ES cells, hES cell-derived DA neurons either do not survive or their DA phenotype is unstable when grafted into rodent brains.

Abbreviations used
BDNF

brain-derived neurotrophic factor

bFGF

basic fibroblast growth factor

DA

dopamine

DAT

dopamine transporter

En1

engrailed-1

ES

embryonic stem

FGF8

fibroblast growth factor 8

GDNF

glial-derived neurotrophic factor

GFAP

glial fibrillary acidic protein

hES

human embryonic stem

HN

human nuclei

KSR

Knockout serum replacement

MAP2

Microtubule associated protein 2

NT3

neurotrophin-3

MEF

Mouse embryonic fibroblasts

6-OHDA

6-hydroxydopamine

PCNA

Proliferating cell nuclear antigen

PD

Parkinson's disease

SHH

sonic hedgehog

TH

tyrosine hydroxylase

TuJ1

neuron-specific class III beta-tublin

Cell replacement therapy aims at grafting therapeutically relevant cells to impaired tissues and has been proposed for future therapies of intractable neurodegenerative disorders. Parkinson's disease (PD), characterized by progressive and selective loss of dopamine (DA) neurons in the midbrain substantia nigra, is a prime target for cell replacement therapy, given over a decade of successful clinical experiences with fetal ventral mesencephalic cell transplantation in PD patients (Piccini et al. 1999). However, fetal cell transplantation has significant technical, ethical, and practical limitations, partly due to limited availability and variable outcomes (Freed et al. 2001). Due to their self-renewal capacity and multi-lineage developmental potential, stem cells could be ideal cell sources for cell replacement therapy.

Embryonic stem (ES) cells, derived from the inner cell mass of pre-implantation embryo, are capable of unlimited cell expansion in vitro while maintaining their pluripotency. When exposed to appropriate culture conditions and/or genetic manipulation, ES cells can differentiate into multi-lineage cell types that are clinically relevant. Furthermore, recent progress in establishing somatic cell nuclear transfer ES cells (Hwang et al. 2004), referred to as ‘therapeutic cloning’, raises the possibility of using ES cell-based cell replacement therapy without immune rejection.

Directed differentiation to a specific cell type is the first step towards exploiting the potential of ES cells for cell replacement therapies. Several protocols including embryoid body-based lineage selection (Lee et al. 2000a), and co-culture with stromal cells (Kawasaki et al. 2000) have been introduced to direct mouse ES cells to differentiate towards midbrain DA neurons. After grafting into the striatum of Parkinsonian rats, mouse ES cell-derived DA neurons survived, integrated into the host striatum and provided functional improvements (Bjorklund et al. 2002; Kim et al. 2002; Shim et al. 2004). While these findings support ES cells usage for cell replacement therapy for the treatment of PD, it is imperative to test the therapeutic potential of human ES (hES) cells. Toward this long-term goal, we here report that midbrain-like DA neurons can be efficiently derived from hES cells. Furthermore, we assessed in vitro functionality of hES cell-derived DA neurons and performed their in vivo functional assays using experimental Parkinsonian rats grafted with these hES-derived DA cells.

Culture for maintaining undifferentiated hES cells

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

Human ES cell lines, HSF-6 (established at University of California, San Francisco, XX, passages 40–60), SNU-hES-3 [Seoul National University (SNU) Hospital, Seoul, Korea, XY, passages 70–85], and Miz-hES-1 (MizMedi Hospital-SNU, Seoul, Korea, XY, passages 35–50) were maintained as described previously (Reubinoff et al. 2000; Park et al. 2003). Briefly, undifferentiated hES cells were propagated on a feeder layer of γ-irradiated mouse (CF1, Charles River Kingston, Kingston, NY, USA) embryonic fibroblasts (MEF) in ES-medium [DMEM/F12 (Invitrogen, Grand Island, NY, USA) supplemented with 20% (v/v) knockout serum replacements (KSR, Invitrogen, Carlsbad, CA, USA), penicillin (100 IU/mL, Invitrogen) and streptomycin (100 μg/mL, Invitrogen), 0.1 mm nonessential amino acids (Invitrogen), 0.1 mm mercaptoethanol (Sigma-Aldrich, St. Louis, MO, USA), and 4 ng/mL basic fibroblast growth factor (bFGF, R & D Systems, Minneapolis, MN, USA)]. Medium was changed daily. For the maintenance of undifferentiated hES cells, cultures were passaged about once every week by mechanically dissecting and transferring hES colonies onto freshly prepared MEF feeder.

Differentiation of hES cells

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

Two different protocols were investigated for neural induction of hES cells; one using direct differentiation without feeders (Fig. 1a) and the other consisting of co-culture on a feeder layer of PA6 stromal cells (Kawasaki et al. 2000) or PA6 cells stably overexpressing sonic hedgehog (PA6-SHH) (Fig. 2a). PA6 stromal cells were maintained and the stromal feeders were prepared as described previously (Kawasaki et al. 2000). The PA6-SHH stable cell line has been established by viral transduction using a retroviral construct expressing human SHH N-terminus (generous gift from Dr Suh-Kim at Ajou University, Korea) followed by blasticidin (5 µg/mL, Invitrogen) selection. Undifferentiated hES colonies were detached from MEF feeders by incubation with 200 U/mL collagenase IV (Invitrogen) for 60 min at 37°C, followed by gentle dissociation into small clusters and then cells were resuspended in serum-free insulin/transferrin/selenium (ITS) medium (Okabe et al. 1996) with 0.2 mm ascorbic acid (AA, Sigma-Aldrich). Clusters were plated on intact culture dish surface (for the direct differentiation method) or on a layer of PA6 stromal cell feeder (for the co-culture method). In the direct differentiation protocol without feeder, rosette-like structures were formed by 2–3 weeks of culture in ITS + AA. Only the clusters of rosette structures were mechanically dissected and were grown as floating spheres in N2 (Johe et al. 1996) supplemented with bFGF (20 ng/mL) and AA (0.2 mm). After 1 week in culture, intact spheres were transferred onto coverslips (Bellco, Vineland, NJ, USA) precoated with poly l-ornithine (PLO, 15 µg/mL, Sigma-Aldrich)/fibronectin (FN, 1 µg/mL, Sigma-Aldrich) and maintained in N2 for cell differentiation. The other approach consisted of dissociating the spheres into single cells by incubation in Ca+2/Mg+2-free Hank's balanced salt solution (HBSS) for 1 h followed by plating on coated coverslips at 1–2 × 105 cells/cm2 in N2 + bFGF. After an added 4–6 days in culture, cell differentiation was induced by withdrawal of bFGF.

image

Figure 1. Neural induction of hES cells by eliminating factors required for maintenance of undifferentiated hES cells. (a) Schematic drawing of the in vitro differentiation protocol. (b) A representative colony of undifferentiated hES cell line SNU-hES-3. (c) Neural rosette structures (arrowheads) appear in the center of differentiating cell clusters during neural induction. The clusters of rosette structures were mechanically dissected, and cultured in suspension in the presence of bFGF to form floating cell spheres (d). (e–g) Antigenic properties of the floating spheres. Floating spheres cultured in suspension were embedded in paraffin and sectioned at 4 µm. Images e–g were obtained from neighboring sections. Sections were counter-stained with hematoxylin/eosin. The majority of cells in the spheres stained positive for neural precursor markers nestin (e) and vimentin (f). A small number of cells at the outer layer of a sphere (arrow) were not immunoreactive for these neural precursor markers. (g) Subsets of cells negative for neuronal precursor markers were stained with α-fetoprotein, an endodermal marker (arrow). (h) Acquisition of neuronal marker TuJ1 in hES-derived neurospheres. The spheres in (d) were transferred directly onto FN-coated surface and cultured in absence of bFGF. (i) Representative image of TH/TuJ1-positve neurons derived from hES cell line SNU-hES-3. Human ES-derived spheres in (d) were dissociated into single cells and transferred onto FN-coated dishes. Expansion of single cell isolated precursors was followed by cell differentiation in the absence of bFGF. At day 2 of post bFGF withdrawal, immunofluorescence analyses were carried out. The scale bar is 20 µm.

Download figure to PowerPoint

image

Figure 2. Stromal feeder-induced derivation of midbrain DA neurons from hES cells. (a) General scheme for the co-culture protocol used for DA differentiation of hES cells. (b) Representative phase-contrast image of differentiated hES colony after neural induction on a layer of PA6 stromal feeder. Human ES cell line HSF-6 was cultured on a layer of PA6 stromal feeder (stage I). After 2 weeks of co-culture, differentiated hES colonies were transferred into FN-coated plates and cultured in the presence of bFGF (stage IIa). The image in (b) was taken 1 day after plating. (c–d) TuJ1/TH+ cell clusters at day 1 (c) and day 7 (d) of stage IIa. (e–g) Antigenic properties of cell clusters differentiated from hES cells were further characterized by immunostaining for HN/TH (e), Pax2/TH (f), and Oct3/4 (g). Inset shows Oct3/4 staining of undifferentiated hES cells at stage 0 as a positive control (g). The scale bar is 20 µm. (h) Expression of genes specific to midbrain development during in vitro differentiation of hES cells.

Download figure to PowerPoint

In the co-culture method, cell clusters, cultured for 1 week on the PA6 feeder (stage Ia in Fig. 2a), were passaged on freshly prepared feeder of PA6 or PA6-SHH, and further cultured for 1 week (stage Ib). At the end of the co-culture, clusters of cells were harvested from the stromal feeder, gently triturated by pipeting into clusters of 20–300 cells in N2 + AA supplemented with bFGF and then replated on FN-coated dishes. After 1 week of culture in N2 + AA + bFGF for the expansion of neural precursor cells (stage IIa), cells were transferred in clusters or single cell dissociates on glass coverslips. An additional precursor expansion of 3–4 days in N2 + AA + bFGF (stage IIb), was followed by culture in ITS supplemented with AA (stage III). In certain experiments, cells were treated with cytokines brain-derived neurotrophic factor (BDNF) (20 ng/mL), glial cell-line derived neurotrophic factor (GDNF 20 ng/mL), neurotrophin-3 (NT3) (20 ng/mL), SHH (100–500 ng/mL) and fibroblast growth factor 8 (FGF8; 100 ng/mL), all from R & D Systems, or conditioned medium prepared from neuron-enriched or astrocyte-enriched cultures as previously described (Chang et al. 2003).

Immunostaining on cultured cells and brain slices

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

Floating spheres cultured in suspension were fixed overnight in 10% neutral-buffered formalin, dehydrated in a series of alcohol gradients (70–100%), embedded in paraffin and sectioned at 4 µm. Sections were counter-stained with hematoxylin/eosin. Perfused brain tissues were soaked in 30% sucrose overnight, frozen in Tissue-Tek® (Sakura Finetek USA, Torrance, CA, USA) solution and cut on cryostat at 35 µm. Cultured cells or cryosectioned brain slides were fixed in 4% paraformaldehyde/0.15% picric acid in phosphate buffered saline (PBS) [for γ-aminobutyric acid (GABA) immunostaining, 0.2% glutaraldehyde (Sigma-Aldrich) was included in the fixative]. The following primary antibodies were used: rabbit polyclonal Igs included nestin #130 1 : 50 (Martha Marvin and Ron McKay, National Institute of Heath, Bethesda, MD, USA), tyrosine hydroxylase (TH) 1 : 250 (Pel-Freez, Rogers, AR, USA), and GABA 1 : 700 (Sigma-Aldrich), neuron-specific class III beta-tubulin (TuJ1) 1 : 2000 (Covance, Richmond, CA, USA), Pax 21 : 400 (Covance) and calbindin-D28K 1 : 250 (Chemicon, Temecula, CA, USA). Mouse monoclonal IgG included TH 1 : 1000, CNPase 1 : 500, microtubule associated protein 2 (MAP2) 1 : 200 (Sigma-Aldrich), dopamine β-hydroxylase 1 : 100 (Chemicon), O4 1 : 100 (Chemicon), human nuclei (HN) 1 : 1000 (Chemicon), TuJ1 1 : 500 (Covance), glial fibrillary acidic protein (GFAP) 1 : 100 (ICN Biochem., Aurora, OH, USA), Ki67 1 : 100 (Novocastra, Newcastle, UK), N-CAM 1 : 100, Oct 3/4 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and proliferating cell nuclear antigen (PCNA) 1 : 40 (Upstate, Charlottesville, VA). Appropriate fluorescence-tagged (Jackson Immunoresearch Laboratories, West Grove, PA, USA) or biotinylated (Vector Laboratories, Burlingame, CA, USA) antibodies were used for visualization. Cells were mounted in VECTASHIELD® containing 4′,6-diamidino-2 phenylindole (DAPI; Vector Laboratories) and analyzed under an epifluorescent microscope (Nikon, Tokyo, Japan).

RT-PCR analysis

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

Total cellular RNA was isolated using TRI REAGENT (Molecular Research Center, Inc. Cincinnati, OH, USA) and cDNA was synthesized from 5 µg of total RNA in a 20 µL reaction using the Superscript kit (Invitrogen). Optimal PCR conditions for each primer set were determined by varying MgCl2 concentrations and annealing temperatures and cycle numbers to determine linear amplification range. Primer sequences, cycle numbers and annealing temperatures were as follows: G3PDH (5′-GCTCAGACACCATGGGGAAGGT-3′, 5′-GTGGTGCAGGAGGCATTGCTGA-3′, 55°C, 35 cycle, 474 bp); Oct3/4 (5′-CTTGCTGCAGAAGTGGGTGGAGGAA-3′, 5′-CTGCAGTGTGGGTTTCGGGCA-3′, 55°C, 35 cycle, 168 bp); TH (5′-GAGTACACCGCCGAGGAGATTG-3′, 5′-GCGGATATACTGGGTGCACTGG-3′, 62°C, 35 cycle, 279 bp); Nurr1 (5′-TTCTCCTTTAAGCAATCGCCC-3′, 5′-AAGCCTTTGCAGCCCTCACAG-3′, 60°C, 35 cycle, 332 bp); Pax2 (5′-GTACTACGAGACCGGCAGCATC-3′, 5′-CGTTTCCTCTTCTCACCATTGG-3′, 60°C, 35 cycle, 396 bp); Engrailed-1 (En1) (5′-GCAACCCGGCTATCCTACTTATG-3′, 5′-ATGTAGCGGTTTGCCTGGAAC-3′, 60°C, 35 cycle, 247 bp); Lmx1b (5′-ACGAGGAGTGTTTGCAGTGCG-3′, 5′-CCCTCCTTGAGCACGAATTCG-3′, 60°C, 30 cycle, 253 bp).

DA uptake assay

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

DA uptake assays were conducted according to the methods described previously with modifications (Lee et al. 2000b). Cells were washed with PBS and incubated with 50 nm[3H]DA in PBS (51 Ci/mmol, Amersham Co., Buckinghamshire, UK) without or with 10 μm nomifensine (RBI, Natick, MA, USA), a dopamine transporter (DAT) blocker, to determine non-specific uptake. After incubation for 10 min at 37°C, the uptake reactions were terminated by aspiration of the reaction solution and washing twice with ice-cold PBS. Cells were lyzed in 0.5 m NaOH and the radioactivity was measured by liquid scintillation counting (MicroBeta® TriLux ver. 4.4 Wallac, Turku, Finland). Specific DA uptake was calculated by subtracting non-specific uptake (with nomifensine) from uptake value without nomifensine.

Electrophysiology

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

Cells grown on a coverslip were transferred to the recording chamber and superfused continuously with artificial cerebrospinal fluid (1.5–2 mL/min) containing 125 mm NaCl, 2.5 mm KCl, 2 mm CaCl2, 1 mm MgSO4, 1.25 mm KH2PO4, 25 mm NaHCO3, and 25 mm d-glucose, bubbled with 95% O2/5% CO2. All recordings were performed at 32–33°C. Standard whole-cell patch-clamp technique with EPC-8 amplifier (Axon Instruments, Union City, CA, USA) was used to record ionic current and membrane potential. The patch electrodes (3–5 MΩ) were filled with a pipette solution containing 130 mm K gluconate, 10 mm KCl, 3 mm MgATP, 10 mm phosphocreatine, 0.3 mm GTP, 10 mm HEPES, 0.2 mm EGTA, 50 U/mL creatine phosphokinase (pH 7.3 with KOH). Whole-cell configuration was obtained after obtaining a tight seal of > 2 GΩ. Membrane current was recorded in voltage-clamp mode with 60% series resistance compensation and membrane potential in current-clamp mode, which was switched during the experiment.

DA determination by HPLC

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

DA release in the medium conditioned by the differentiated hES cells (stage III of the co-culture method) was determined using reverse phase-HPLC. Differentiated hES cells in 24-well plates were incubated in 200 µL ITS + AA or ITS + AA medium supplemented with 56 mm KCl (evoked) for 30 min. The media were then collected and stabilized with 0.1 mm EDTA and analyzed for DA. Samples (100 µL) were injected with a Rheodyne injector and separated with a reverse phase µ-Bondapak C18 column (150 × 3.0 mm, Eicom, Japan) maintained at 32°C with a column heater (Waters, Cotland, NY, USA). The mobile-phase consisted of 0.05 m citric acid, 0.05 m disodium phosphate (pH 3.1), 3.2 mm 1-octanesulfonic acid (sodium salt), 0.3 mm EDTA and 12% methanol, and was pumped at a flow rate of 0.5 mL/min using Waters' solvent delivery system (Waters, Milford, MA, USA). Electroactive compounds were analyzed at +650 mV using an analytical cell and an amperometric detector (Eicom, Model ECD-300, Japan). DA levels were calculated using external DA standard injected immediately before and after each experiment.

Transplantation, in vivo analysis of grafted cells, and behavioral tests

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

Animals were housed and treated following National Institutes of Health guidelines. Under phenobarbital anesthesia (50 mg/kg, i.p.), adult male Sprague–Dawley rats (220–250 g) were given unilateral sterotaxic injections of 4 µL of 6-hydroxydopamine (6-OHDA, 3 µg/µL; Sigma-Aldrich) into the substantia nigra [co-ordinates: anteroposterior (AP), −4.8 mm; mediolateral (ML), 1.5 mm; dorsoventral (V), 8.2 mm] and the median forebrain bundle (AP −1.8 mm, ML 1.8 mm, V 8.0 mm). Incisor bar was set at −3.5 mm, AP and ML coordinates are given relative to bregma (Paxinos and Watson, 1982).

For transplantation, differentiated HSF-6 hES cells (day 3 of stage II or III) were harvested and dissociated into single cell suspensions or in cell clusters by incubating in HBSS or collagenase IV, respectively, as described in ‘Differentiation of hES cells’. Single cells were resuspended in PBS at a concentration of 105 cells/µL. Using a 22-gauge needle, 5 µL of cell suspension was injected over a 5-min period into the ipsilateral striatum (AP +0.2 mm, ML 3.0 mm, V 5.5 mm, incisor bar set at 3.5 mm). The needle was left in place for 3–5 min following the completion of each injection. In other cases, harvested cell clusters were further cultured in ITS + AA + bFGF in ultra low binding culture dishes (Corning, Corning, NY, USA) to form solid floating cell aggregates (diameter: 0.5–1 mm). Eight to 10 of the cell aggregates were injected into the striatum using a 19-gauge spinal needle. Sham-operation was performed on control animals. Rats received daily injections of cyclosporine A (10 mg/kg, i.p.) starting 24 h prior to grafting and continuing for 3 weeks followed by a reduced dose of 5 mg/kg for the remaining in vivo period.

Three weeks after 6-OHDA lesioning, animals were tested for rotational asymmetry after i.p. injection of 3 mg/kg d-amphetamine sulfate (Sigma-Aldrich). Animals with an average of ≥ 5 turns/min over a 1 hr interval were selected and randomly assigned to treatment or control groups. Forelimb akinesia was assessed by the ‘stepping test’ (Olsson et al. 1995). Animals were adapted to the test conditions 5 days preceding the actual test. The investigator fixed both hindlimbs as well as one forelimb while the unrestrained forepaw was touching the table. The number of adjusting steps was counted while the rat was moved sideways along the table surface (90 cm in 10 s). Each step test consisted of three trials for each forepaw, alternating between forepaws. In all experiments, the average of three trials for each forepaw was used for analysis. The results were expressed as a percentage of steps performed with the lesioned side as compared with the intact side.

In the estimation of rotation and stepping scores 1 day before cell transplantation, no significant differences in the behavioral tests were observed among the groups assigned randomly. The absolute number of rotation for an hour in each group of animals before transplantation were 360.5 ± 22.1 (sham-operated), 343.3 ± 25.5 (clusters-grafted), and 344.2 ± 19.4 (single cells-grafted). Adjusting steps in each group of animals before transplantation were: intact paw: 18.7 ± 2.1 (sham-operated), 18.3 ± 1.8 (clusters-grafted), and 19.1 ± 1.5 (single cells-grafted); lesioned paw: 1.7 ± 0.2 (sham-operated), 1.3 ± 0.3 (clusters-grafted), and 1.4 ± 0.2 (single cells-grafted). The behavioral tests were performed weekly for 6 weeks after cell transplantation.

Two weeks after transplantation, animals were anesthetized with phenobarbital and perfused transcardially with 4% paraformaldehyde in PBS. Brains were equilibrated with 30% sucrose in PBS and sliced on a freezing microtome (CM 1850, Leica, Wetzlar, Germany). Free-floating brain sections (35 µm thick) were subjected to immunohistochemistry as described above.

Direct neural induction of hES cells in the absence of hES cell maintenance factors

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

It has been suggested that neural lineage defaults from mouse ES cells in the absence of signals for maintaining self-renewal and undifferentiated properties of these cells (Hitoshi et al. 2004). To initiate differentiation, colonies of hES cells (SNU-hES-3, Miz-hES-1 or HSF-6), grown on a feeder layer of MEF, were transferred and cultured in the absence of feeder and bFGF, both of which are necessary for the maintenance of undifferentiated hES cell properties. Data shown in this section were obtained from the hES cell line SNU-hES-3, unless specified otherwise, while similar data were obtained with Miz-hES-1 and HSF-6. Under differentiation conditions, the shape of hES colonies was strikingly changed to multilayered clusters of an increasing number of small, elongated cells in the center surrounded by flattened cells. By 2–3 weeks of culture in ITS medium, rosette-like structures, resembling early neural tube, were formed in the clusters (Fig. 1c). The clusters including rosettes were isolated mechanically under microscopic examination and were grown as free-floating suspension culture of cell aggregates in N2 supplemented with bFGF, a specific mitogen for neural precursor cells, for 5–7 days (Fig. 1d). Immunocytochemical analyses revealed that the majority of cells (> 90% of total cells) in the floating aggregates (spheres) were positive for neural precursor cell markers nestin (Fig. 1e) and vimentin (Fig. 1f). None of the cells were positive for desmin, a mesodermal marker (data not shown). The endodermal markers alpha-fetoprotein (Fig. 1g) and PECAM colocalized in 6.9 and 5.8% of the cells, respectively. The spheres were plated directly on FN-coated coverslips and cultured in N2. Five to 7 days after plating, processes emanating from the spheres had formed prominent fiber bundles. Immunofluorescence analyses revealed that the vast majority of the cells in the outgrowth areas expressed the neuronal marker TuJ1 (Fig. 1h). However, the DA neuronal marker TH colocalized only with a minority (< 10 cells per coverslip) of TuJ1+ cells. To increase the yield of TH/TuJ1+ cells, the floating spheres were dissociated into single cells and then plated on FN-coated dishes and cultured in N2 + bFGF. This protocol was based on our assumption that DA neuron precursors assembled inside of aggregates might require to be exposed to the culture environment for their in vitro terminal differentiation towards the DA phenotype. Dissociated cells were proliferated in response to bFGF. After 4 days of cell expansion (at 60–80% cell confluency), bFGF was withdrawn from the culture medium to induce cell differentiation. However, under differentiation conditions cell viability plummeted and we were unable to sustain the cultures for longer than 3 days after bFGF withdrawal. Thus, cultures were fixed and cells phenotyped 2 days after bFGF withdrawal. Cells (28.8 ± 3.8%) were positive for the neuronal marker TuJ1 by immunofluorecence detection (Fig. 1h), and 21.6 ± 3.3% of cells positive for TuJ1 expressed TH (Fig. 1i). The vast majority of cells negative for TuJ1 or TH expressed nestin, an intermediate filament specific to neural precursors (50.3 ± 3.1% of total cells), suggesting insufficient differentiation of neural precursor cells. Cell survival of neural precursors derived from hES is described below.

Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

In the protocol described above, mass production of DA neurons was hampered by the laborious mechanical dissection of rosettes. It has been reported that stromal cells can direct neural induction of mouse ES (Kawasaki et al. 2000) as well as primate ES cells (Kawasaki et al. 2002). Thus, we cultured hES cells on a layer of PA6 stromal cells (stage I, Fig. 2a). A striking difference in neural induction of hES cells among the cell lines was observed on PA6 stromal feeder. Immunocytochemical analyses at the end of stage I, revealed that 92.3 ± 2.5% of HSF-6 colonies were positive for TuJ1. In these colonies, the majority of cells stained positive for the neural precursor marker nestin. In contrast, only 7.7 ± 1.7% and 7.1 ± 2.4% of colonies acquired expression of the neuronal marker TuJ1 in SNU-hES-3 and Miz-hES-1 cells, respectively.

To eliminate the PA6 feeder cells as well as to obtain an increased yield of cells of neuronal lineages, HSF-6 colonies, harvested 2 weeks after co-culture, were disrupted into small clusters and then plated on FN-coated surfaces in N2 + AA supplemented with bFGF (stage IIa). Cell phenotypes were not largely altered after cell passage. At this stage, the majority of cells were positive for nestin. After 1 week of culture with bFGF, total cell numbers increased by 3–5 folds (estimated by viable cell counting after dissociating the clusters into single cells) suggesting an expansion of neural precursor cells at this stage. The expanded clusters were disrupted into smaller clusters and were passaged again for an additional precursor expansion (stage IIb), followed by precursor differentiation (stage III). Interestingly, cell phenotypes (stage IIb) were not significantly altered by bFGF withdrawal (stage III), when the cultures were maintained as cell clusters (see the result section ‘Further differentiation of nestin + clusters derived from hES cells’ for a detailed description). Cells in the clusters were uniform with small and elongated cell morphology (Fig. 2b). The vast majority of cells were positive for HN, suggesting an efficient elimination of mouse stromal feeder cells (Fig. 2e). Virtually all of the cell clusters contained TH/TuJ1+ cells (Fig. 2c). Cell clusters grew in size in the presence of bFGF and fused together (Fig. 2d). TH/TuJ1+ cells with neuronal shape constituted a major cell population in cultures (Figs 2c and d). Dopamine β-hydroxylase, a noradrenergic or adrenergic neuronal marker, did not colocalized with TH+ cells, suggesting a DA neuronal identity of TH+ cells (data not shown). None of the cells were positive for Oct3/4, an undifferentiated ES cell marker (Fig. 2g). These findings, collectively, suggest an efficient generation of DA neurons from HSF-6 hES cells.

A subset of cells in the clusters was immunoreactive for Pax2, a transcriptional factor specific to midbrain development (Fig. 2f). Semi-quantitative RT-PCR analyses revealed that expression of the markers characteristic of midbrain DA development, such as En1, Nurr1 and Lmx1b, was temporally induced during in vitro differentiation of hES cells (Fig. 2h). These findings suggest the midbrain DA nature of TH+ cells derived from hES cells. When passaged in the form of cell clusters, the phenotype of cultures including TH+ cells was not significantly altered after at least three passages (> 3 weeks). Furthermore, cells at stage IIa could be stored at − 80°C and reused by simple freeze/thawing in ITS + AA + bFGF containing 10% dimetylsulfoxide (DMSO) without affecting their potential to differentiate into the DA neuronal fate (data not shown).

SHH and FGF8 effects on midbrain DA differentiation of hES cells

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

Previous work established SHH and FGF8 as crucial factors in the specification of midbrain DA neurons in explant culture (Ye et al. 1998) and for mouse ES cell differentiation in culture (Lee et al. 2000a). Cells were exposed to SHH and FGF8 during the last half of the PA6 co-culture period (stage Ib) by transferring cell clusters, co-cultured with PA6 feeders for 1 week (stage Ia), onto freshly prepared feeders consisting of PA6-SHH plus FGF8 cytokine treatment (100 ng/mL) (Fig. 2a). Conditioned medium from confluent cultures of PA6-SHH had a comparable effect as 200–500 ng/mL of SHH in increasing cell number in the cultures for precursors isolated from embryonic day 14 rat cortices, confirming the release of SHH from PA6-SHH (data not shown). Total cell numbers at the end of stage Ib was increased by 2–3 folds in cultures exposed to SHH and FGF8, compared to those unexposed, suggesting survival and/or proliferation effects of these cytokines on differentiating hES cells. The effect of SHH on cell survival/proliferation of neural precursors has been demonstrated previously (Kenney and Rowitch 2000; Lai et al. 2003; Machold et al. 2003; Thibert et al. 2003). The effects of SHH and FGF8 on hES cell differentiation were analyzed at the first day of stage IIb. TuJ1+ neuronal numbers in the cultures pre-exposed to SHH and FGF8 were not significantly different from unexposed control cultures (30.5 ± 3.7 vs. 26.1 ± 1.9% of total cells, n = 20, Figs 3a, b and e). However, the percentage of TH+ cells was significantly increased in cultures pre-exposed to SHH and FGF8 (Figs 3a, b and e). Among the TuJ1+ neurons out of total cells, 41.1 ± 3.7% of the cells treated with SHH + FGF8 expressed TH compared to 25.9 ± 3.2% for untreated cells. Another striking effect of SHH and FGF8 treatment was reflected by the percentage of cells positive for Pax2, a specific transcriptional factor of midbrain neuronal development (16.9 ± 3.8 vs. 2.9 ± 1.0% of Pax2+ cells, Figs 3c, d and e). Furthermore, the exposure of SHH and FGF8 led to an enhanced expression of genes specific for midbrain DA neuronal markers such as En1, Nurr1, and Lmx1b as examined by semiquantitative RT-PCR analyses (Fig. 3f). These results suggest a role for SHH and FGF8 on midbrain DA neuronal development during in vitro differentiation of hES cells. Subsequent data were obtained with cells co-cultured with PA6-SHH.

image

Figure 3. Effects of SHH and FGF8 on in vitro DA derivation from hES cells. For exposure to SHH and FGF8, differentiating hES cells at stage Ia were passaged and cultured on a layer of PA6-SHH in ITS + AA medium supplemented with 100 ng/mL FGF8 (stage Ib). Cells were passaged in the form of cell clusters and the effect of SHH and FGF8 were examined at day 1 of stage IIb. (a and b) Representative images of TH/TuJ1+ clusters of untreated control (a) and cells treated with SHH and FGF8 (b). (c and d) Pax2+ cells in cultures of untreated control (c) and treated with SHH and FGF8 (d). Insets, DAPI nuclear staining of the same field. The scale bar is 20 µm. (e) Percentage of immunoreactive cells. Cells positive for TuJ1 (TuJ1/DAPI), TH (TH/DAPI) and Pax2 (Pax2/DAPI) out of total cells. TH+ cells out of TuJ1+ cells (TH/TuJ1). *Significantly different from control at p < 0.01. (f) RT-PCR analysis of genes involved in midbrain DA development. Note that expression of midbrain-specific genes Pax2, En1, Nurr1 and Lmx1b was enhanced by SHH and FGF8 treatment.

Download figure to PowerPoint

Further differentiation of nestin+ clusters derived from hES cells

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

In addition to TuJ1+ neurons, another major population of cells in the differentiated hES cell clusters consisted of proliferating neural precursors that are positive for both the neural precursor cell marker nestin and the proliferating cell marker Ki67 (nestin/Ki67+ cells, 52.6 ± 3.9% of total cells in stage IIb). The proportion of these cells was not significantly altered by further induction of cell differentiation in the absence of bFGF, suggesting an inadequate differentiation of nestin+ neural precursors assembled in the cell clusters. It has been suggested that clusters of neural precursor cells are proliferating while single isolated precursors take on their differentiation phenotype (Temple and Davis 1994). Thus we assumed that differentiation of neural precursors assembled in the clusters is prevented by cell–cell contacts. For adequate cell differentiation of nestin+ cells, stage IIa clusters were dissociated into single cells, and passaged onto FN-coated coverslips. Passaging cells in the form of cell clusters resulted in excellent cell recovery at stage IIb. In contrast, considerable cell loss was observed in cultures passaged by single cell dissociation: < 30% of the total cells plated were viable after being plated for 1 day. However, surviving cells had high proliferation activity in presence of bFGF reaching 70–90% confluency after 4 days in culture. At this stage, the vast majority of cells (> 95%) expressed nestin (Fig. 4a) but only 3.6 ± 0.1% of the cells were positive for TuJ1, suggesting a selective survival and proliferation of nestin+ precursors in cultures originating from single cell dissociation.

image

Figure 4. Further differentiation of hES-derived neural precursor cells. Cell clusters at stage IIa of the co-culture protocol were dissociated into single cells, and plated on FN-coated surface. After 3–4 days of bFGF-expansion, the vast majority of cells expressed nestin (neural precursor cell marker) and the proliferating cell marker Ki67 colocalized with nestin+ cells. (a) Representative image of nestin/Ki67+ cells at day 3 of cell expansion. (b–e) Phenotypes of the cultures differentiated from hES-derived neural precursor cells. Cell differentiation of nestin+ precursors was induced for 7 days in the absence of bFGF. Immunofluorescence analyses were performed for TuJ1/TH (b), MAP2/TH (c), Calbindin/TH (d), GABA/TH (e), and GFAP/TuJ1 (f). The scale bar is 20 µm.

Download figure to PowerPoint

Differentiation of nestin+ precursors was induced by withdrawal of bFGF. Similar to the direct differentiation protocol described above, cells expanded from single cell dissociation exhibited poor viability in N2 or N2 + AA medium without bFGF. Therefore, it was impossible to induce differentiation for more than 2 days. Cell survival was not significantly enhanced by treatment with neurotrophic cytokines (BDNF, GDNF, and NT3) and supplementation with conditioned media from neuron- or astrocyte-enriched cultures. Interestingly, hES-derived precursors cultured in ITS + AA were viable for more than 7 days in the absence of bFGF. After 7 days of differentiation in ITS + AA, 65.6 ± 1.3% of total cells were TuJ1+ neurons, suggesting an efficient differentiation of hES-derived neural precursor cells. Among the TuJ1-positve neuronal population 22.1 ± 1.7% of cells expressed TH (Fig. 4b). The other neuronal markers MAP2 and N-CAM colocalized with TH+ cells (Fig. 4c and data not shown). Less than 0.1% of TH+ cells expressed calbindin, a calcium-binding protein specifically expressed in midbrain DA neurons and which increases resistance to cell death in PD (Yamada et al. 1990; Gaspar et al. 1994; Damier et al. 1999; Fig. 4d). In addition to TH+ neurons, 1.4 ± 0.24% of TuJ1+ neurons were positive for GABA (Fig. 4e). Only a few GFAP+ astrocytes (< 10 cells per coverslip) were detected in the differentiated cultures (Fig. 4f). None of the cells was positive for CNPase and O4, markers for the oligodenrocytic lineage. The number of GFAP+ astrocytes was increased by extension of in vitro differentiation, but oligodendrocyte was still not observed after 16 days of differentiation. These findings suggest that nestin+ cells derived from hES cells in the present study may represent neural precursor cells of the early developmental stage, given that neural precursors sequentially yield neuron, astrocytes and oligodendrocytes in the developing brain (Bayer and Altman 1991; Jacobson 1991; Qian et al. 2000).

In vitro function of hES-derived TH cells

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

Cells showing long multiple processes in the peripheral region of stage III cell clusters were chosen for electrophysiological recording (n = 8, Fig. 5a). Membrane potential (−22.8 ± 4.3 mV) and whole cell capacitance (86.4 ± 9.5 pF) were measured immediately after membrane perforation. Total membrane current was recorded with the holding potential at −70 mV. All the cells showed fast inward currents and delayed outward currents with slight inactivation (Fig. 5b). Rapidly inactivating A-type currents, which disappeared while holding the potential at −40 mV, was also observed in some cells (5 out of 8 cells). Membrane response to this depolarizing current was recorded in 4 cells under the current-clamp mode. All 4 cells showed generation of action potentials (Fig. 5c). These results showed the presence of well developed sodium and potassium channels, generating action potentials.

image

Figure 5. In vitro characterization of differentiated hES cells. (a) Distinct cells located in periphery of neurospheres were chosen for the recordings. (b) Current traces under the voltage-clamp mode. Total membrane currents were measured without any blockers. Step voltage activation (10 mV) from holding at −70 mV evoked inward and outward currents. Inset shows the inward current at the beginning of the step voltage command. (c) Recording of action potentials. Membrane voltage was recorded under the current-clamp mode in the same cell. Action potentials were generated with positive step current injection (10 pA step). (d) HPLC determination of DA levels. Representative HPLC chromatograms for basal DA release (blue line: exposure to ITS + AA medium for 30 min) and DA release after 30 min of KCl-evoked depolarization (red line). The yellow line represents DA standard. (e) DA uptake. The graph depict the specific DA uptake of stage 0 (undifferentiated, white bar, n = 4) and stage III (differentiated, black bar, n = 4) hES cells. Specific DA uptake was calculated by subtracting non-specific uptake (with nomifensine) from uptake value without nomifensine. *Significantly different from control at p < 0.01.

Download figure to PowerPoint

An important physiological aspect of authentic DA neuron phenotypes is the ability to synthesize DA and release it in response to membrane depolarization. We measured levels of DA released from hES-derived TH+ cells in the medium conditioned by cultures of stage III HSF-6 cells for 30 min. Without KCl-induced depolarization stimuli, only small amounts of DA were detected in the medium (15.6 ± 3.8 pg/mL of medium, n = 4). However, after treatment with 56 mm KCl, the released DA in the media was greatly enhanced (1643 ± 276 pg/mL, n = 4), demonstrating depolarization-induced DA release of hES-derived TH+ neurons (Fig. 5d).

In addition to depolarization-induced release of DA, high affinity reuptake of the transmitter by DAT is a crucial process for presynaptic DA homeostasis. Specific DA uptake was scarcely observed in undifferentiated hES cultures. In contrast, stage III cells displayed avid DA uptake (27.8 ± 1.65 fmol/min/well, n = 4, Fig. 5e).

In vivo transplantation of DA neurons derived from hES cells

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

We investigated if hES-derived DA neurons can elicit in vivo function as measured by amelioration of Parkinsonian motor deficits. Differentiated HSF-6 cells (stage II or III) were prepared in the form of single cells or cell aggregates, and were injected into the striatum of hemi-parkinsonian rats. No functional improvement estimated by the amphetamine-induced rotation test (Fig. 6a) and the stepping test (Fig. 6b), was observed in animals grafted with hES-derived DA cells regardless of cell preparation.

image

Figure 6. In vivo survival and functions of hES-derived DA neurons. (a and b) Behavioral analysis. Differentiated HSF-6 cells at stage II of the co-culture protocol were harvested and injected in the form of cell clusters (n = 12) or single cell dissociates (n = 8) into the striatum of hemi-Parkinsonian rats. For negative control, 10 animals were sham-operated under identical schedule as the cell-grafted animals. (a) Amphetamine-induced rotation response. Data are given as mean ± SEM of changes in rotation scores for each animal as compared to pretransplantation values. (b) Stepping test. The results are expressed as a percentage of the lesioned side relative to the number of steps with the non-lesioned paw. (c–e) Immunohistochemical analyses on brain slices of grafted animals. Two weeks after transplantation, grafted animals were killed and immunostaining was performed. Images (c–e) are representatives for HN (c), HN/TH (d), and PCNA (e) staining of rat brains grafted with differentiated hES cell clusters.

Download figure to PowerPoint

In animals grafted with single cell dissociates, histological analysis revealed only a few HN+ cells along the needle tracts (data not shown). In contrast, a striking number of cells were positive for HN in grafts of animals transplanted with cell aggregates (Fig. 6c). Some TH immunoreactivity with unclear neuronal morphology was observed, but none of them colocalized with HN (Fig. 6d). Cells positive for HN (53.3 ± 4.9%) were positive for the proliferating cell marker PCNA (Fig. 6e), raising the possibility of teratoma formation from these proliferative cells. However, Oct3/4+ cells were not seen in any brain slices and no teratomas were observed in animals receiving grafts of hES-derived DA cells. No difference in in vivo cell survival and functions was observed between stage III and stage II cells (data not shown).

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References

The efficient derivation of midbrain DA neurons from hES cells is a prerequisite not only for the developmental study of human midbrain DA neurons, but also for realistic cell replacement therapy of PD and novel drug screening. The present study has demonstrated that a highly enriched population of midbrain DA neurons was generated from hES cells. Co-culture of the HSF-6 hES cell line with PA6 stromal cells effectively yielded a high number of DA neurons (up to 41% of TuJ1+ cells), expressing known midbrain DA neuronal markers. Similar to our findings, Zeng et al. (2004) reported recently a highly efficient derivation of DA neurons from the BG01 hES cell line, using the same PA6 co-culture method. In contrast to the efficient neural induction of HSF-6 (Fig. 2) and BG01 (Zeng et al. 2004) hES cell lines, our study showed that the co-culture method was not effective to direct SNU-hES-3 and Miz-hES-1 cells to differentiate towards neural lineages. The observed difference in their response to the co-culture system may reflect inherent differences of each human embryonic cell line and/or the underlying genetics of the embryos from which the lines were derived. Consistent with this possibility, difference in gene expression patterns among the hES cell lines and the unique gene expression signatures of independently derived hES cell lines have recently been reported (Abeyta et al. 2004).

In the present study, we have compared several different in vitro differentiation methods including those based on rosette formation, PA6 co-culture, and co-culture with Shh-expressing PA6 cells (PA6-SHH). While we need further investigation to optimize the in vitro culture conditions of hES cells, we propose an efficient procedure is as follows (Fig. 2a). Briefly, it consists of neural induction of hES cells on PA6 and PA6-SHH (stage I), bFGF-induced expansion of nestin+ neural precursors (stage II) and terminal differentiation of the nestin+ cells (stage III). Differentiated neurons almost always appeared in the periphery of large cell clusters during in vitro differentiation of hES cells, suggesting disruption of close cell–cell contacts inside the cell clusters might be required to facilitate differentiation of hES cells to the neuronal fate. Thus, our protocol consisted of dissociating cell clusters into smaller pieces or all the way to single cells in passage procedures. Similar to our study, Perrier et al. (2004) showed recently the efficient hES-derived DA neuronal generation based on co-culture with MS5 and S2 stromal cells. While both approaches gave rise to efficient generation of DA neurons, the conditions in our study allowed for faster (28 days vs. 50 days) and more efficient derivation of neural phenotypes (> 95% of cells were nestin+ precursors or TuJ1+ neurons). However, the proportion of TH/TuJ1+ was higher (60–80%) in the study by Perrier et al. (2004) than ours (41%), probably due to the different stromal cells used or the replating method. Notably, our co-culture protocol does not require the mechanical dissection procedure for rosette structures described in the study of Perrier et al. (2004), making the in vitro mass production of DA neuron more feasible. During neural induction on PA6 stromal cells layer (stage I), two- to threefold increases in total cell number were usually observed. After stage I, hES-derived neural precursor cells could be expanded for 3 weeks without significant loss of DA yield, resulting in about 64-fold increase in total cell number. Numerically, 10–20 TH+ cells are harvested for every undifferentiated hES cells. The other major concern of guided differentiation of ES cells is the efficient elimination of unwanted cells, especially undifferentiated ES cells. None of the cells generated in our procedure was positive for the undifferentiated ES cell marker Oct3/4 at stage IIb and III of hES cultures and in brain sections of animals grafted with differentiated hES cells. Consistent with this, no teratoma was formed in our grafting experiments.

Major functional components of presynaptic DA neurons consist of synthesis of DA neurotransmitter, depolarization-induced release of the transmitter and high affinity reuptake of transmitter by a sodium dependent (DA) transporter (DAT). DAT function is particularly important for regulating DA homeostasis because absence of this transporter could lead to either excess, unregulated dopaminergic transmission or to premature loss of intrasynaptic stores of the transmitter. We demonstrated the presence of electrophysiologically active sodium and potassium channels and generation of action potentials with positive stepped-current injection in hES cell-derived neurons. DA release was evoked by KCl-induced depolarization, suggesting an activity-dependent DA neurotransmission of the hES-derived DA cells. Furthermore we observed a substantial uptake of DA in differentiated hES cell cultures. Taken together, these findings demonstrate that functional DA neurons are efficiently derived from hES cells using our in vitro differentiation methods.

The in vivo function of hES-derived DA neurons was tested by intrastriatal transplantation into hemi-parkinsonian rats. We did not observe any significant functional improvements in parkinsonian rats grafted with hES-derived DA cells. As none of the postgrafted cells was positive for TH, lack of behavioral recovery in the grafted animal is attributable to the disappearance of DA neurons after grafting. We observed that TH immunoreactivity gradually decreased in vitro without significant loss of cell number during the precursor differentiation period (stage III, data not shown), suggesting that hES-derived DA neurons might only temporarily express their dopaminergic properties and thereby be responsible for the loss of TH immunoreactivity after grafting. However, it is also possible that hES-derived DA neurons selectively undergo apoptotic cell death in the host striatum. Dissimilar to our result of the in vivo TH immunostaining, Zeng et al. (2004) demonstrated detection of hES-derived cells positive for TH in the rats grafted, even though the number of the TH+ cells was quite few (< 8.8 TH+ cells per 10 brain sections). This discrepancy might be attributable to the differences in the hES cell lines used, cell preparation procedures used for the transplantation (e.g. such as repeated dissociation of cell clusters during the differentiation procedures in our study), and/or the transplantation/immunization protocols. While there is significant discrepancy in TH immunoreactive cells in the grafted animals, both studies demonstrated that survival and behaviors of hES-derived TH+ cells in host striatum after transplantation is quite different from those derived from mouse ES cells (Chung et al. 2002;Kim et al. 2002; Shim et al. 2004) and mesencephalic precursors (Studer et al. 1998; Kim et al. 2003) which could efficiently integrate, survive, and function to improve motor dysfunctions in rodent models of Parkinson's disease.

Survival and function of donor DA neurons are highly dependent on the host environment; trophic support (Rosenblad et al. 1996; Zawada et al. 1998) and immunologic factors (Larsson et al. 2000). In addition it has been demonstrated that intrinsic factors, in midbrain DA neurons, such as En1 (Simon et al. 2001), Nurr1 (Le et al. 1999; Perlmann and Wallen-Mackenzie 2004) and Lmx1b (Smidt et al. 2000) are crucial for survival and sustained expression of DA properties in midbrain DA cells. Further studies, taking in consideration of these extrinsic and intrinsic factors, should be addressed to achieve in vivo survival and function of hES-derived DA neurons.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Culture for maintaining undifferentiated hES cells
  5. Differentiation of hES cells
  6. Immunostaining on cultured cells and brain slices
  7. RT-PCR analysis
  8. DA uptake assay
  9. Electrophysiology
  10. DA determination by HPLC
  11. Transplantation, in vivo analysis of grafted cells, and behavioral tests
  12. Results
  13. Direct neural induction of hES cells in the absence of hES cell maintenance factors
  14. Efficient derivation of midbrain DA cells from the HSF-6 cell line by coculturing with stromal feeder cells
  15. SHH and FGF8 effects on midbrain DA differentiation of hES cells
  16. Further differentiation of nestin+ clusters derived from hES cells
  17. In vitro function of hES-derived TH cells
  18. In vivo transplantation of DA neurons derived from hES cells
  19. Discussion
  20. Acknowledgements
  21. References