• dopamine neurons;
  • human embryonic stem cells;
  • neural precursor cell;
  • Parkinson’s disease


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Human embryonic stem (hES) cells can be guided to differentiate into ventral midbrain-type neural precursor (NP) cells that proliferate in vitro by specific mitogens. We investigated the potential of these NP cells derived from hES cells (hES-NP) for the large-scale generation of human dopamine (DA) neurons for functional analyses and therapeutic applications. To address this, hES-NP cells were expanded in vitro for 1.5 months with six passages, and their proliferation and differentiation properties determined over the NP passages. Interestingly, the total hES-NP cell number was increased by > 2 × 104-folds over the in vitro period without alteration of phenotypic gene expression. They also sustained their differentiation capacity toward neuronal cells, exhibiting in vitro pre-synaptic DA neuronal functionality. Furthermore, the hES-NP cells can be cryopreserved without losing their proliferative and developmental potential. Upon transplantation into a Parkinson’s disease rat model, the multi-passaged hES-NP cells survived, integrated into the host striatum, and differentiated toward the neuronal cells expressing DA phenotypes. A significant reduction in the amphetamine-induced rotation score of Parkinson’s disease rats was observed by the cell transplantation. Taken together, these findings indicate that hES-NP cell expansion is exploitable for a large-scale generation of experimental and transplantable DA neurons of human-origin.

Abbreviations used

ascorbic acid


basic fibroblast growth factor




DA transporter




glial fibrillary acidic protein


G protein-gated inwardly rectifying K+ channel


human embryonic stem


human nuclei antigen


human neural cell adhesion molecule




Lewis X


microtubule-associated protein 2


mouse embryonic stem


neural precursor


Parkinson’s disease


population doubling level


sonic hedgehog


tyrosine hydroxylase


vesicular monoamine transporter 2

Human embryonic stem (hES) cells, derived from the inner cell mass of pre-implanted blastocysts, are multipotent cells, which have the capacity to differentiate into most cell types from the three germ layers. They are expandable in vitro and their multipotency can be retained even after multiple passages (Amit et al. 2000; Reubinoff et al. 2000; Thomson et al. 1998). These properties of hES cells are thought to offer promise of an unlimited source of many cell types, which could be used in basic and translational studies of cell replacement strategies. However, the potential applications of undifferentiated hES cells (especially for the purpose of therapeutic transplantation) are largely hampered by their genomic instability characterized by the high frequency of chromosomal anomalies during in vitro culture (Draper et al. 2004; Imreh et al. 2006; Maitra et al. 2005). Furthermore, in vitro maintenance and expansion of undifferentiated hES cells is fastidious requiring stringent experimental conditions. In addition, we herein additionally demonstrate the loss of developmental potential of hES cells toward neuronal progeny after maintenance in long-term culture.

Recent studies successfully derived midbrain-type dopamine (DA) neurons from hES cells. These findings support the idea of using hES cells as a renewable source of human DA cells for experimental or therapeutic purposes in Parkinson’s disease (PD), a disease characterized by the selective loss of midbrain DA neurons. In contrast to the direct differentiation of hES cells into DA neurons (Brederlau et al. 2006; Schulz et al. 2004; Zeng et al. 2004), we (Park et al. 2005) and Perrier et al. (2004) have presented protocols sequentially differentiating hES cells toward DA cells through the intermediary stage of embryonic midbrain-type neural precursor (NP) cells. NP cells derived from hES cells (hES-NP) proliferate in vitro with specific mitogen supplementation and differentiate into DA neurons which function in vitro as mature pre-synaptic DA neurons (Park et al. 2005; Perrier et al. 2004). These findings prompted us to test if the proliferative and DA neurogenic potentials are maintained after prolonged NP expansion in vitro, thus supporting the idea of exploiting NP cell proliferation (instead of the problematic expansion of undifferentiated ES cells) for large-scale generation of functional human DA neurons. Therefore, we expanded hES-NP cells in vitro for 1.5 months over six cell passages and assessed their developmental properties, survival, and functions in vitro and in vivo through the improvement of motor dysfunction upon transplantation in rat model of PD.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Culture and differentiation of hES cells

The maintenance and differentiation of the undifferentiated hES cell lines HSF-6 (established at University of California, San Francisco, CA, USA; passages 53–82) and H9 (WiCell, Madison, WI, USA; passages 35–38) were performed as previously described (Park et al. 2005). Briefly, undifferentiated hES cells were grown on γ-irradiated mouse (CF1) embryonic fibroblasts in ES-medium (Park et al. 2005). To induce neural differentiation, hES colonies were transferred onto a feeder layer of γ-irradiated PA6 or MS5 stromal cells and cultured in insulin/transferrin/selenium (ITS) medium (Okabe et al. 1996) with 0.2 mmol/L ascorbic acid (AA; Sigma, St Louis, MO, USA). Cultures were passaged onto freshly prepared γ-irradiated PA6 or MS5 feeder cells every 7 days, which continued for a total of three to eight passages until a desirable level of neural induction (> 85% hES colonies are differentiated colony with primitive neural structures; hES-NP colonies, see Results) was achieved. For the final round of subculturing, hES colonies were co-cultured on γ-irradiated PA6 or MA5 cells stably over-expressing sonic hedgehog (SHH; PA6-SHH or MS5-SHH) (Park et al. 2005). At the end of the co-culture, hES-NP colonies were harvested from the stromal feeder, gently triturated by pipetting into clusters of 50–500 cells in ITS + AA supplemented with basic fibroblast growth factor (bFGF, 20 ng/mL; R&D Systems, Minneopolis, MN, USA), and replated onto poly l-ornithine (15 μg/mL; Sigma)/fibronectin (FN, 1 μg/mL; Sigma)-coated culture dishes. After culture for 7–9 days, hES-NP clusters were dissociated into single cells after exposure to Ca2+/Mg2+-free Hank’s balanced salt solution for 1 h at 37°C. These were then replated at 2.5 × 105 cells/cm2 in ITS + AA + bFGF (NP passage 1, NP-P1). Cells were passaged every 7 days for 42 days (six passages) in medium supplemented with bFGF. For phenotypic determinations, hES-NP cells were plated on FN-coated glass coverslips (12-mm diameter; Bellco, Vineland, NJ, USA). Terminal differentiation of hES-NP cells was induced in the absence of bFGF but in the presence of brain-derived neurotrophic factor (20 ng/mL; R&D Systems), glial cell line-derived neurotrophic factor (20 ng/mL; R&D Systems), dibutyryl cAMP (0.5 mmol/L; Sigma).

Population doubling level

The population doubling level (PDL) of hES-NP cells at each NP passage was measured. The formula used for the calculation of population doubling in a single passage is: PDL = log(N/No)/log2, where N is the number of cells in the plate at the end of a period of growth and No is the number of cells plated in the plate (2.5 × 105 cells/cm2). Cumulative (total) PDL is total PDL (tPDL) = ∑(PDL)n, where n is the hES-NP passage number.

Immunostaining of cultured cells and brain slices

Cultured cells and cryosectioned brain slices were immunostained after fixation with 4%p-formaldehyde in phosphate-buffered saline (PBS), except in the case of Lewis X (LeX) immunostaining. The following primary antibodies were applied overnight at 4°C: rabbit polyclonal antibodies including nestin #130 1 : 50 (Dr Martha Marvin and Dr Ron McKay, National Institute of Heath, Bethesda, MD, USA), tyrosine hydroxylase (TH) 1 : 250 (Pel-Freez, Rogers, AR, USA), vesicular monoamine transporter 2 (VMAT2) 1 : 500 (Pel-Freez), doublecortin 1 : 500 (kindly gifted from Dr Woong Sun, Korea University, Seoul, South Korea), neuron-specific class III beta-tubulin (TuJ1) 1 : 2000 (Covance, Richmond, CA, USA), Pax2 1 : 100 (Covance), Serotonin 1 : 4000 (Sigma), γ-aminobutyric acid 1 : 700 (Sigma), calbindin-D28K (calbindin) 1 : 250 (Chemicon, Temecula, CA, USA), G protein-gated inwardly rectifying K+ channel (Girk2) 1 : 200 (Sigma), 1 : 80 (Alomone Labs, Jerusalem, Israel), and mouse monoclonal antibodies including TuJ1 1 : 500 (Covance), TH 1 : 1000 (Sigma and Immunostar, Hudson, WI, USA), glial fibrillary acidic protein (GFAP) 1 : 100 (ICN Biochemicals, Aurora, OH, USA), CNPase 1 : 500 (Sigma), O4 1 : 100 (Chemicon), Engrailed-1 (En1) 1 : 50 (Developmental Studies Hybridoma Bank, Iowa City, IA, USA), Pax5 1 : 50 (BD Biosciences, Franklin Lakes, NJ, USA), microtubule-associated protein 2 1 : 200 (Sigma), HuC/D 1 : 100 (Chemicon), synapsin 1 : 1000 (BD Biosciences), Ki67 1 : 100 (Novocastra, Newcastle, UK), human neural cell adhesion molecule (hNCAM) 1 : 100 (Santa Cruz Biotechnology, Santa Cruz, CA, USA), dopamine β-hydroxylase 1 : 100 (BD Biosciences), and human nuclei antigen (HN) 1 : 1000 (Chemicon). Appropriate fluorescence-tagged secondary antibodies (Jackson Immunoresearch Laboratories, West Grove, PA, USA) were used for visualization. Cells and tissue sections were mounted in VECTASHIELD® containing 4′,6 diamidino-2-phenylindole (Vector Laboratories Inc., Burlingame, CA, USA) and analyzed under an epifluorescence microscope (Nikon, Tokyo, Japan) or a Carl Zeiss LMS 510 confocal system (Carl Zeiss, Jena, Germany). For LeX immunostaining, live cells were incubated with a LeX antibody (clone MMA, CD15, 1 : 200, anti-mouse; BD Biosciences) and visualized with a Cy3-conjugated secondary antibody. Coverslips were fixed and mounted for microscopic examination.

In vivo transplantation and histological procedure

Animals were housed and treated following guidelines from the National Institutes of Health. Female Sprague–Dawley rats (200–250 g) were lesioned by unilateral stereotactic injection of 6-hydroxydopamine (Sigma) into the substantia nigra and the median forebrain bundle as described previously (Bing et al. 1988). Four weeks later, the animals were tested for drug-induced rotational asymmetry (amphetamine 3 mg/kg i.p.; Sigma). Rotation scores were monitored for 60 min in an automated rotameter system (Med Associate Inc., St Albans, VT, USA). Animals with ≥ 4 turns/min over a 1 h ipsilateral to the lesion were selected for the in vivo study and randomly assigned to the control (sham-operated) and transplantation groups. For transplantation, hES-NP cells expanded with bFGF were harvested and dissociated into single cells. Three microliters of cell suspension (2.5 × 105 cells/μL in PBS) per site were deposited at two sites of the striatum (coordinates in AP, ML, and V relative to bregma and dura: (1) 0.07, −0.30, 0.55; (2) −0.10, −0.40, −0.50; incisor bar set at 3.5 mm using a 22-gauge needle. The needle was left in place for 5 min following each injection. Control sham-operated rats were injected with a vehicle PBS solution. The rats received daily injections of cyclosporin A (10 mg/kg, i.p.) starting 24 h prior to grafting and continuing for 3 weeks, followed by a reduced dose (5 mg/kg) for the remaining time. At 2, 4, 6, and 8 weeks after transplantation, amphetamine-induced rotations were determined.

Eight weeks after transplantation, the animals were anesthetized with phenobarbital and perfused transcardially with 4%p-formaldehyde 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. The total numbers of cells positive for TH or HN in the graft were estimated as previously described with the Abercrombie correction factor (Shim et al. 2004).

Hematoxylin and eosin (H&E) staining was carried out with conventional method.


Total RNA preparation, cDNA synthesis, and RT-PCR reactions were performed as described previously (Park et al. 2005). Primer information was provided in supplemental data Table S1.

Dopamine uptake assay

Dopamine uptake assays were conducted as described previously (Park et al. 2005). Cells were washed with PBS and incubated with 50 nmol/L [3H]DA in PBS (51 Ci/mmol; Amersham Co., Buckinghamshire, UK) with or without 10 μmol/L nomifensine (RBI, Natick, MA, USA), a DA 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 lysed with 0.5 mol/L 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 the uptake value without nomifensine.

Detection of dopamine release by HPLC

HPLC analysis for DA levels was performed as previously described (Shim et al. 2004) with some modifications. To determine in vitro DA release from cultured cells, differentiated precursor cells in 24-well plates were incubated in 500 μL ITS + AA medium with 56 mmol/L KCl for 15 min (evoked release). The media was then collected and stabilized in 0.1 N perchloric acid containing 0.1 mmol/L EDTA, followed by extraction by aluminum adsorption. DA was separated with a reverse phase μ-Bondapak C18 column (150 × 3.0 mm; Eicom, Kyoto, Japan) at a flow rate of 0.5 mL/min. Electroactive compounds were analyzed at +750 mV using an analytical cell and an amperometric detector (Model ECD-300; Eicom). DA levels were calculated using an internal standard (50 nmol/L methyl-DOPA) and catecholamine standard mixtures, including 1–50 nmol/L DA (external standard) injected immediately before and after each experiment.

Cell counting and statistical analysis

Cell counting was performed in uniform microscopic fields that were randomly chosen across area where cells were grown using an eyepiece grid at a final magnification of 200 or 400×. Cells were counted in 15–30 microscopic fields on each coverslip, and two to six culture wells were analyzed in each experiment. Data are expressed as mean ± SEM. Statistical comparisons were made by anova with Tukey post hoc analysis (SPSS 11.0; SPSS Inc., Chicago, IL, USA) when two or more groups were involved.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Gradual and synchronous neural differentiation of hES cells by co-culturing with MS5 stromal feeder

Dopaminergic differentiation of hES cells is induced by co-culture on a feeder layer of PA6 or MS5 stromal cells (Brederlau et al. 2006; Park et al. 2005; Perrier et al. 2004; Zeng et al. 2004). hES cell colonies on the stromal cells differentiated into colonies with primitive neuroepithelial cell structures (supplemental data Fig. S1a and c; Park et al. 2005; Perrier et al. 2004). We repeatedly subcultured the differentiating hES colonies by splitting them into small colonies (50–500 cells/colony) and replating onto freshly prepared MS5 cells during the co-culture step. The percentage of differentiated colonies with primitive neural structures (NP colonies) gradually increased at each cell passages. Specifically, 15%, 39%, and 87% of hES colonies included NP colonies after the third, fifth, and sixth passages on MS5 cell feeder, respectively (1000–3000 colonies from two to three independent experiments were examined for each value). Although similar hES differentiation was achieved on a feeder of PA6 cells, a subpopulation of differentiated hES colonies on PA6 (but not on MS5 cells) contained cells with fully mature neuronal morphology (supplemental data Figs S1b and d; Brederlau et al. 2006; Zeng et al. 2004). This finding indicates that more synchronous hES cell differentiation, with homogenous populations of cells at the same developmental stage, is likely achieved with MS5 cells than with PA6 cells. All of the following results, therefore, were obtained with hES differentiation on MS5 feeder cells.

Multi-passaged undifferentiated hES cells lose their differentiation potential

We previously demonstrated that the efficiency of hES differentiation into neural lineages is affected by hES cell lines (Park et al. 2005). In addition, we observed that hES differentiation was also dependent on various experimental conditions, including the size of hES cell colonies, cell viability, and general conditions of the feeder cells (data not shown). In particular, the degree of hES differentiation greatly varied depending on the passage number of undifferentiated hES cells, even in the same HSF-6 hES cell line. In hES cells with an extended passage number (maintained longer on the mouse embryonic fibroblasts maintenance feeder), it took longer to achieve hES differentiation (in which > 80% of hES colonies differentiated into NP colonies), and such a level of differentiation could not be obtained with hES cells maintained over 75 passages (ES P75; Fig. 1a). Chromosomal analysis of the multi-passaged HSF-6 hES cells with lack of in vitro neural differentiation demonstrated multiple chromosomal aberrations (Fig. 1b), indicating that the decreased differentiation capacity of late-passage hES cells is probably associated with chromosomal alterations. These findings collectively suggest that long-term expansion of hES cells with multiple ES passages is problematic to be applied in large-scale generation of hES-derived neural lineage cells.


Figure 1.  Loss of differentiation potential and chromosomal anomaly of undifferentiated human embryonic stem (hES) cells after long-term culture. (a) Neural differentiation potential of hES cells is lost after long-term culture for maintenance and propagation of undifferentiated hES cells (HSF-6). The graph depicts the number of days required for hES differentiation (> 80% of hES colonies are differentiated neuroepithelial cell colonies). (b) Karyotype of the undifferentiated HSF-6 hES cells with ES passage 82. Variable structural and numerical chromosome abnormalities are shown.

Download figure to PowerPoint

Sustained neural precursor cell properties of hES-NP after extensive NP cell expansion in vitro

When cultures contained > 85% NP colonies, they were transferred as small clusters into FN-coated plates in ITS + AA supplemented with bFGF and were cultured for 7–9 days. The NP clusters were then dissociated into single cells, plated, and cultured in ITS + AA + bFGF (NP passage 1). After 7 days of bFGF-proliferation, over 90% of the total cells were positive for nestin, an intermediate filament specific to NPs (Park et al. 2005; data not shown). Only an extreme minority of cells were negative for HN antigen, suggesting the effective elimination of the feeder cells through mechanical passage followed by selective survival and proliferation of NPs. These finding suggest the derivation of an enriched population of NP cells from hES cells. Uniform population of NP cells was similarly derived from the hES cell line H9 by co-culturing with MS5 [(Perrier et al. 2004) and supplemental data Fig. S2a–d].

Human embryonic stem-NP cells were continuously expanded in the presence of bFGF for 42 days (a total of six passages), with the total PDL being 14.3 (2.0 × 104-fold increase in total cell number; doubling time, 2.94 days; Fig. 2a). The cell expandability at each passage did not vary significantly during the NP propagation period: for instance, the PDL of passage 6 (P6) NP cell culture was similar to that of the P1 NP culture (2.22 in P6 vs. 1.74 in P1, Fig. 2b). Normal karyotype was maintained in hES-NP cells during the NP cell expansion (Fig. 2c).


Figure 2. In vitro expansion of human embryonic stem (hES) -derived neural precursor (NP) cells. (a) Cumulative total population doubling level (PDL). (b) PDL of each NP culture at different NP passage numbers. NP cells derived from hES cells (hES-NP) were cultured with basic fibroblast growth factor and passaged every 7 days for 42 days (total of 6 passages). The PDL was calculated as described in Materials and methods. (c) Normal karyotype of hES-derived NP cells was maintained after five times of NP passages.

Download figure to PowerPoint

Immunocytochemical analyses revealed that 92.5 ± 1.8% and 84.6 ± 3.1% out of total cells were positive for nestin and Ki67, a proliferating cell marker, at the NP passage 2, respectively (Fig. 3a and g). Most of the nestin+ cells co-expressed the proliferating cell marker Ki67 (nestin+/Ki67+ out of total nestin+ cells: 93.2%), and nestin+ cell populations were maintained at 90.2–93.1% during the 6-NP passage period. Comparable results were obtained with the passaged NP cells derived from H9 hES cells (supplemental data Fig. S2a–d). LeX, known as SSEA-1 or CD15, is an extracellular matrix carbohydrate expressed by mouse embryonic stem (mES) cells (Bird and Kimber 1984). Recent studies have demonstrated a specific expression of LeX in adult and embryonic rodent neural stem or precursor cells (Capela and Temple 2002, 2006). LeX expression was detected in 47.6 ± 3.3% of cells at this NP stage of hES cell differentiation (Fig. 3d and g). Such the lower population of LeX+ cells, compared with nestin+ cell population, might be attributable to a matter of sensitivity of LeX antibody used in this study, in the immunocytochemical detection of the expressing cells. However, it may also suggest that LeX expression in human NP cells is restricted to a certain NP cell subpopulation, while nestin is expressed in a wide range of NP cells including early NP, definite NP and committed progenitor cells. The percentages of LeX-expressing cells were also not significantly altered during the six times of NP passages (P2: 47.6 ± 3.3% and P6: 47.9 ± 1.6%, Fig. 3d–g).


Figure 3.  Maintenance of neural precursor (NP) cell properties of human embryonic stem (hES) -NP cells during NP cell expansion. HSF-6 hES-NP cells at different NP passage numbers were proliferated in vitro in the presence of basic fibroblast growth factor, and samples for immunocytochemical (a–g) and RT-PCR (h) analyses were prepared on the last day (day 7) of basic fibroblast growth factor-proliferation. (a–f) Immunocytochemical analyses for nestin (NP cell marker)/Ki67 (proliferating cell marker; a–c) and LeX (surface marker for NP cells; d–f). Each panel in (a–f) is a representative image for nestin+, Ki67+, and LeX+ cells in the cultures for hES-NP cell with NP passage P2, P4, and P6. Insets: 4′,6 diamidino-2-phenylindole staining of the same field. Graph (g) depicts the percentage of cells immunoreactive for nestin, Ki67, and LeX of the total (4′,6 diamidino-2-phenylindole positive) cells. Note that the proportion of proliferating NP cells in the hES-NP cultures did not significantly alter after NP cell passages. Scale bar: 40 μm. (h) Semi-quantitative RT-PCR analyses for genes specific to NPs (nestin, Sox1, and Pax6), midbrain development (En1 and Pax5), midbrain dopaminergic markers (Nurr1 and Lmx1b) and forebrain and midbrain development (Otx2). The expression patterns shown are representatives of three to six experiments with two RNA samples independently prepared.

Download figure to PowerPoint

We did not observe significant nestin, Sox1, and Pax6 mRNAs level differences by semi-quantitative RT-PCR among hES-NPs with different NP passage numbers (Fig. 3h). Similarly, no obvious NP passage number-dependent changes in Nurr1, Lmx1b (markers for midbrain DA neuron), and Engrailed-1 (En1), Pax5 (midbrain development), Otx2 (forebrain and midbrain development) could be detected.

Sustained neuron/astrocyte yields from late hES-NP cells

To examine whether the differentiation potential of hES-NP cells is altered during in vitro expansion, terminal differentiation of hES-NP cells at each NP passage was induced for 6–8 days. NP cells derived from HSF-6 and H9 hES cells at NP passage 2 exhibited high neuronal differentiation with neuronal marker neuron-specific class III beta-tubulin (TuJ1) -expressing cells being 59.7 ± 2.8% (HSF-6) and 55.4 ± 0.5% (H9) of total cells in the differentiated cultures, whereas only 8.6 ± 1.5% (HSF-6) and 8.8 ± 3.1% (H9) of cells expressed the astrocyte-specific protein GFAP (Fig. 4a and g and supplemental data Fig. S2e and h). None of cells was positive for CNPase or O4, oligodendrocytic markers tested. These findings suggest an early developmental property of the hES-derived NP cells, given that NPs sequentially yield neuron, astrocytes, and oligodendrocytes in the developing brain (Bayer and Altman 1991; Jacobson 1991; Qian et al. 2000). TuJ1+ neuronal and GFAP+ astrocytic yields were not significantly altered after multiple NP passages (Fig. 4a–c and g and supplemental data Fig. S2e–h): for instance, percentages of TuJ1+ cells out of total (4′,6 diamidino-2-phenylindole positive) cells: 59.7 ± 2.8% (NP-P2), 61.2 ± 2.6% (P4), 64.3 ± 3.7% (P6); GFAP+ cells: 8.6 ± 1.5% (P2), 10.4 ± 2.0% (P4), and 10.8 ± 3.7% (P6) in the differentiated HSF-6 hES-NP cultures. This result is in a contrast to the differentiation of NP cells isolated from rodent embryonic brain demonstrating rapid neuron-to-astrocyte switch during their in vitro cell expansion (Chang et al. 2004).


Figure 4.  Differentiation phenotypes of human embryonic stem-neural precursor (NP) cells with different NP passages. After 6–8 days of in vitro differentiation, phenotypes of the differentiated cultures were assessed by immunocytochemical (a–g) and RT-PCR (h) analyses. (a–f) Representative images for cells positive for TuJ1 (neuronal)/glial fibrillary acidic protein (GFAP; astrocytic marker, a–c) and TuJ1/tyrosine hydroxylase (TH) [dopamine (DA) neuronal; d–f] of differentiated HSF-6 human embryonic stem-NP cells at P2, P4, and P6. Scale bar: 40 μm. Graph (g) depicts the percentage of cells immunoreactive for TuJ1 and GFAP of the total (4′,6 diamidino-2-phenylindole positive) cells or TH of the total TuJ1+ cells. (h) Semi-quantitative RT-PCR analyses for the genes specific to DA neuron [TH and DA transporter (DAT)], midbrain DA (En1, Ptx3, and Lmx1b), A9 midbrain DA (G protein-gated inwardly rectifying K+ channel; Girk2), neuron (TuJ1), and astrocyte (GFAP).

Download figure to PowerPoint

Dopamine neuron subtype differentiation

This study focused on hES-NP differentiation toward DA neurons, a neuronal subtype clinically relevant to psychiatric or neurodegenerative disorders such as PD. Consistent to previous studies (Park et al. 2005; Perrier et al. 2004), large population (HSF-6: 34.0 ± 2.9%, H9: 44.5 ± 6.6% at P2) of TuJ1+ neurons co-expressed TH, a DA neuronal marker. Dopamine β-hydroxylase, a noradrenergic or adrenergic neuronal marker, did not colocalize with TH+ cells (data not shown), suggesting a DA neuronal identity of the hES-derived TH+ cells. It has been demonstrated that NP-derived DA neuron yield was nearly completely lost after one to two times cell passages in the cultures derived from rodent embryonic ventral midbrain (Yan et al. 2001; Chung et al. 2006). In a clear contrast, hES-NP cells efficiently yielded TH+ neurons after four to six times of NP cell passages. A large difference was not observed in TH+ cell yield depending on NP cell passage. Instead, the percentage of TH+ cells out of the TuJ1+ neurons was slightly greater in late NP cells (36.6 ± 3.6% at P4 and 38.0 ± 4.5% at P6, compared with 34.0 ± 2.9% at P2). Semi-quantitative RT-PCR analyses for TuJ1, GFAP, and TH further supported unaltered neuron/astrocyte and DA neuronal differentiation over the hES-NP cell passages (Fig. 4h). Furthermore, mRNA expressions of the genes specific for other DA neuronal (DAT), midbrain dopaminergic (En1, Ptx3, and Lmx1b), A9 midbrain-specific dopaminergic (Girk2) markers were not greatly varied in the cultures at different NP cell passages (Fig. 4h). Maintenance of high DA neuron yields after prolonged NP cell expansion has been similarly shown in the differentiation of H9 hES-NP cells (Supplemental data Fig. S2i–l).

In addition to co-expression of TH/TuJ1 (Fig. 4d–f), human-specific neuronal (hNCAM and HuC/D; Fig. 5b and c) and mature neuronal (microtubule-associated protein 2; Fig. 5d) markers were expressed in virtually all TH+ cells derived from multi-passaged hES-NP cell cultures (P6, HSF-6 hES-NP). Synapse formation and neurotransmitter homeostasis are definite properties acquired in mature functional pre-synaptic neurons. Synapsin, a synaptic vesicle-specific protein, was co-localized in TH+ fibers in the differentiated hES-NP cultures (Fig. 5e). In addition to DAT mRNA expression shown in Fig. 4h, TH+ cells in the multi-passaged hES-NP cultures expressed another DA neurotransmitter homeostasis protein VMAT2 (Fig. 5f). These findings indicate well-differentiated mature DA neuronal properties of hES-derived TH+ cells.


Figure 5.  (a–l) Immunocytochemical phenotypes of multi-passaged human embryonic stem-dopamine (DA) cells. HSF-6 human embryonic stem-neural precursor cells at neural precursor passage 6 were differentiated for 6–8 days and subjected to double immunocytochemical analyses for TH/HN (a), TH/hNCAM (b), TH/HuC/D (c), TH/MAP2 (d), TH/Synapsin (e), TH/VMAT2 (f), TH/EN1 (g), TH/Pax2 (h), TH/Pax5 (i), TH/Calbindin (j), TH/Serotonin (k), and TH/GABA (l). Right and left images in (a–l) were taken from two isolated microscopic fields. Scale bar: 20 μm. Co-labeled cells were indicated as arrows. Note that all or most of TH+ cells express HN, hNCAM (human-specific neuronal), HuC/D, MAP2 (mature neuronal), and VMAT2 (DA homeostasis). By contrast, midbrain-specific markers Pax2 and Pax5 expressions are separated from TH (h–i), whereas En1 co-localized in subsets of TH+ cells (g). Insets of (k and l), the high-powered images. Note that cells immunoreactive for TH and GABA are well separated, of which co-localization is a characteristic feature of olfactory (forebrain) or developing striatal DA neurons. hNCAM, human neural cell adhesion molecule; HN, human nuclei antigen; MAP2, microtubule-associated protein 2; TH, tyrosine hydroxylase; VMAT2, vesicular monoamine transporter 2.

Download figure to PowerPoint

Figures 3h and 4h demonstrated unaltered expressions of the mRNAs specific to midbrain developmental and midbrain DA neuronal genes over NP cell passages in hES-NP-derived cultures. Consistently, cells immunoreactive for the midbrain-specific markers En1, Pax2, and Pax5 were readily detected in the differentiated cultures for the multi-passaged hES-NP cells (P6, HSF-6 hES-NP). Consistent to the previous studies (Perrier et al. 2004), En1 was localized in a subpopulation of TH+ cells (Fig. 5g), whereas Pax2- and Pax5-expressing cells were negative for TH (Fig. 5h and i). Calbindin is a calcium-binding protein, of which expression in midbrain DA neurons is related to increased resistance to cell death in PD (Yamada et al. 1990; Gaspar et al. 1994; Damier et al. 1999). Another line of studies has demonstrated a specific expression of calbindin in DA neurons of the ventral tegmental area (A10), but Girk2 in DA neurons of the substantia nigra (A9; Mendez et al. 2005; Thompson et al. 2005; McRitchie et al. 1996). Consistent with previous studies (Park et al. 2005; Roy et al. 2006), only a minor proportion (<1%) of TH+ cells expressed calbindin (Fig. 5j). Girk2 mRNA expression in differentiated hES-NP cell cultures was evident (Fig. 4h), but none of the cells differentiated from hES-NP was immunoreactive to Girk2. This finding is inconsistent with the study of Roy et al. (2006) manifesting Girk2 expression in a major population of TH+ cells in differentiated hES cell cultures. DA and serotonin neurons in mouse brain development are simultaneously generated in the ventral parts of the midbrain and hindbrain, respectively, by common combinational actions of SHH and FGF8 (Ye et al. 1998). The co-generation of these two neurotransmitter neuronal subtypes was also shown in mES differentiation in vitro (Lee et al. 2000). Consistently, serotonin+ neurons were another major neuronal population in the differentiated hES-NP cultures (21–25% of TuJ1 neurons), and the percentage of serotonin neurons was not significantly altered after multiple NP cell passages (Fig. 5k and data not shown). In accordance with previous studies for hES cell differentiation, the GABA+ neuronal population was relatively minor (8–10%, Fig. 5l). No cell passage-dependent variation in GABA+ neuron numbers was observed (data not shown). GABA and TH co-expression is a characteristic feature of olfactory DA neurons (Gall et al. 1987) and transiently observed in striatal GABAergic neurons (Max et al. 1996). No co-expression of GABA and TH was observed, further supporting midbrain dopaminergic differentiation of hES-NP cells.

In vitro functions as pre-synaptic DA neurons were examined by the assays for DA neurotransmitter release and uptake. HPLC analyses revealed that differentiated P2, P4 and P6 hES-NP cells released considerable amounts of DA upon KCl-induced depolarization: 7.25 ± 0.46 (P2); 6.86 ± 0.73 (P4); 5.93 ± 0.80 (P6) pg/15 min/TH+ cell (Fig. 6a and b). DAT-mediated high affinity reuptakes of DA were 0.23 ± 0.05 at P2, 0.26 ± 0.01 at P4, and 0.27 ± 0.06 fmol/min/TH+ cell at P6 (Fig. 6c). The DA release and uptake values were not significantly different in the cultures at different passages (p values shown in Fig. 6b and c). These findings collectively indicate maintenance of developmental potential of hES-NP cells toward mature functional midbrain-type DA neurons after multiple NP cell passages.


Figure 6. In vitro pre-synaptic neuronal functions of dopamine (DA) cells differentiated from human embryonic stem- neural precursor (NP) cells at NP passage 2, 4, and 6. (a) Typical HPLC chromatogram for DA released from differentiated NP-P6 (red) is shown with that for mixture of standards (NE, norepinephrine; EP, epinephrine; DA, blue). IS represents internal standard (3,4-dihydroxybenzylamine) used for quantification of DA concentrations. The graph (b) shows DA levels in the medium supplemented with 56 mmol/L KCl. (c) DA transporter-mediated specific DA uptake. DA uptake was calculated by subtracting non-specific uptake (with nomifensine) from the total uptake (without nomifensine). The level of DA release and uptake per tyrosine hydroxylase (TH+) cell was calculated by dividing each value by the TH+ cell number obtained from cultures conducted in parallel to the DA release and uptake assays. The boxes and bars in graph (b and c) represent means and standard errors of DA releases (n = 8) and DA uptakes (n = 6). p values indicated were obtained in one-way anova by comparing with the DA release and uptake values of P2 cultures.

Download figure to PowerPoint

Storage of hES-NP cells in liquid nitrogen

Human embryonic stem-NP cells at every passage were stored in bFGF-supplemented medium containing 10% dimethylsulfoxide in liquid N2 and re-cultured by a simple freezing/thawing method. When liquid N2-stored cells from NP passage 5 and 6 were thawed for re-culture, 78.6 ± 2.5% and 77.5 ± 3.3% of cells were viable 5 h after plating, respectively (n = 16 for each value). The result was in clear contrast to the poor viability of undifferentiated hES cells when re-cultured from liquid N2-storage (data not shown). The proliferative and DA neurogenic potentials of hES-NP cells were not significantly altered after several freeze–thaw cycles (Supplemental data Fig. S3 and data not shown). Furthermore, long-term storage (at least up to 20 months) did not abrogate NP cell properties (data not shown). These findings provide another substantial benefit of hES-NP cells serving as a continuous, on-demand source of human NP cells or DA neurons.

In vivo survival, differentiation, and function of hES-NP cells in PD rats

The final study explored the in vivo survival, differentiation, and function of the multi-passaged hES-NP cells (HSF-6 hES-derived NP-P4) after transplantation into the striatum of PD model rats. hES-NP cells survived in the host striatum and formed tubular mass of grafts along the needle tracts. The average graft volume in the animals grafted was 1.2 ± 0.4 mm3 at 8 weeks post-transplantation (n = 6). The total number of HN+ cells per graft was 301 327 ± 51 467 cells, resulting in a density of HN+ cells in the graft of 244 779 ± 20 398 cells per mm3. A large proportion of cells in the grafts were positive for hNCAM (Fig. 7d and f) and doublecortin (marker for migrating and differentiating neuron; Fig. 7e), suggesting an efficient neuronal differentiation from the NP cells grafted. However, only an extremely small proportion of the grafted cells expressed DA neuronal marker TH (Fig. 7a and d), and the average number of TH+ cells was 201 ± 86 cells per animal. The human origin of the TH+ cells was confirmed by co-localization of HN (Fig. 7a–c) and hNCAM (Fig. 7d) in the TH+ cells. Neuronal identity of the TH+ cells was unambiguous based on authentic neuronal morphology of the TH+ cells with multiple neurites (Fig. 7b) and co-expression of neuronal marker hNCAM. Such the low percentage of DA neurons is consistent with previous reports where very few (Ben-Hur et al. 2004; Brederlau et al. 2006; Martinat et al. 2006; Zeng et al. 2004) or no TH-positive cells (Park et al. 2005) were found in hES-derived grafts. Larger proportion of hNCAM+ cells in the grafts, however, expressed VMAT2, another DA neuronal marker (Fig. 7f). Transplantation of the passaged hES-NP cells generated uniform grafts with no evidence of teratoma formation or neoplastic change (Fig. 7g).


Figure 7. In vivo survival, differentiation, and function of hES-NP cells. In vitro proliferated HSF-6 hES-NP cells (NP-P4) were prepared and transplanted into Parkinson’s disease (PD) rats as described in Materials and methods. The animals were killed 8 weeks after transplantation. Immunostaining for TH/HN, TH/hNCAM, DCX/HN, and VMAT2/hNCAM was performed, and the images shown in (a–f) were obtained by stacking z-series through the thickness of section (35 μm). Image (b) is high magnification view of the cells indicated as arrow in (a). Scale bar: 20 μm. (g) Representative microscopic image of hematoxylin–eosin-stained graft of hES-NP P4 cells; h, host tissue and g, graft area, scale bar: 40 μm. (h) Amphetamine-induced rotation test. For the negative control, five animals were injected with phosphate-buffered saline under the identical schedule as the cell-grafted animals. Data are given as the mean ± SEM of changes in rotation scores for each animal compared with pre-transplantation values [n = 6 for transplanted and n = 5 for the sham control (phosphate-buffered saline injected]. Analysis was by anova, followed by the Newman–Keuls test. *p < 0.05 from the respective control values. hNCAM, human neural cell adhesion molecule; DCX, doublecortin; hES, human embryonic stem; NP, neural precursor; HN, human nuclei antigen; TH, tyrosine hydroxylase; VMAT2, vesicular monoamine transporter 2.

Download figure to PowerPoint

We next performed a behavioral assessment in the hemi-Parkinsonian rats grafted using an amphetamine-induced rotation test (Fig. 7h). None of the sham-operated rats (PBS-injected, n = 5) exhibited a reduction in amphetamine-induced rotation 8 weeks post-transplantation. By contrast, three out of six rats grafted exhibited > 50% reduction in rotation scores. The average rotation scores at 8 weeks of post-transplantation, compared with the pre-transplantation values, were 716 ± 82 in the sham-controls (n = 5) versus 427 ± 190 in the animals grafted (n = 6).


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

The present study demonstrates that NP cells derived from hES cells were expanded 2 × 104-fold during six passages over > 1 month, without losing their differentiation potentials toward an enriched population of DA cells, which exhibited functionalities as a pre-synaptic DA neuron. Furthermore, we further passaged HSF-6 hES-NP cells up to 12 NP passages and observed that the proliferative and enriched DA neuron yields of the longer-passaged hES-NP cells were, at least, unaltered (Supplemental data Fig. S4a–c). Thus it is undetermined yet how much longer the cellular properties of hES-NP could be maintained by in vitro culture. Such long-term maintenance of proliferative and developmental capacity of cultured precursors is quite exceptional, considering that culture period-dependent alteration in cellular properties is a rather general finding in cultures of stem or precursor cells derived from specific brain tissues. Specifically, NP cells derived from the ventral mesencephalons of rat embryonic brains were expanded only 1000- (Chung et al. 2006) or 5400-fold (Yan et al. 2001) in cell numbers and almost completely lost their potential to differentiate into TH+ neurons after 4 weeks of in vitro culture (Chung et al. 2006; Ko et al. 2005; Yan et al. 2001). In addition, the neurogenic differentiation potential of NPs derived from embryonic brain switches to an astrocytic one during in vitro cultures (Chang et al. 2004). Thus, maintenance of the ratio of neuron/astrocyte cell number in differentiated cultures for hES-NP cells (Fig. 4g) is another contrasting finding between hES-NP cells and embryonic brain-derived NP cells. Similar to our findings, a recent study demonstrated that NP cells derived from mES cells efficiently yielded DA cells after 4 weeks of in vitro expansion (Chung et al. 2006). Thus, in vitro maintenance of precursor cell properties seems to be specific to NP cells derived from ES cells. As a possible explanation for the ES-derived NP cell properties, it is suggested that NP cells generated from ES cells in vitro are early in the developmental process (such as primitive neuroepithelial cells), and the maturation process of the early NP cells into later cells is retarded under in vitro culture conditions (Hitoshi et al. 2004). This is probably because of the lack of in vivo environmental factors required for the maturation process, such as paracrine signals derived from blood vessels of endothelial cells (Shen et al. 2004).

Unlike the high proportion of TH+ DA neurons in the cultures of differentiated hES-NP cells, very few cells were positive for TH in the striatum of PD rats grafted with hES-NP cells. This is consistent with previous reports of hES cell transplantation in PD rats (Ben-Hur et al. 2004; Brederlau et al. 2006; Martinat et al. 2006; Park et al. 2006; Zeng et al. 2004). By contrast, abundant number of cells was positive to general neuronal marker hNCAM in the hES-derived graft. In addition, the serotonin neuronal population, another monoamine neuron subtype, was not greatly reduced after grafting, and much higher proportion of cells was serotonin+ neurons in the striatum of PD rats (data not shown). Thus, the great loss of the hES-derived neuronal subtype after grafting is likely to be the phenomenon that specifically occurred in TH+ DA neuron population. There are a number of possibilities to explain the lack of survival (or differentiation) of hES-derived TH cells in an in vivo host striatum. First, the low number of TH+ cells in the grafts is probably attributable to the possibility that the genuine identity of a major population of hES-derived DA cells is not A9 nigral midbrain dopaminergic. Previous studies have demonstrated that DA neurons with non-midbrain origins, such as those from adrenal medulla (Bing et al. 1988; Bohn et al. 1987) and carotid bodies (Espejo et al. 1998), survived poorly, whereas the midbrain-types, especially A9 nigral DA neurons, survived and functionally integrated with the host striatum (Thompson et al. 2005). Although other studies and ours have exhibited midbrain-specific marker expressions in differentiated hES cultures, these data do not absolutely indicate that the major population of the hES-DA cells is of the authentic midbrain type. After the consecutive reports of poor engraftment of hES-DA cells in PD rats, astonishingly, a recent study presented abundant DA neuron engraftments in PD rat striatum by transplanting hES cells which had undergone co-culture procedure with telomerase-immortalized human mesencephalic astrocytes (Roy et al. 2006). This is actually only one study demonstrating enriched hES-derived TH+ cells in PD rat striatum, even though the promising results in the study were accompanied by sharp diminution of TH+ neurons over a month in vivo and by neoplastic change in the center cores of the grafts. Importantly, the study by Roy et al. (2006) manifested that the co-cultures with midbrain-specific human astrocytes potentiated hES cells to differentiate toward A9-specific midbrain DA neurons, which could efficiently engraft and competent to ameliorate the behavioral deficits of striatal dopaminergic denervation. It remains to be identified whether the co-culture procedure or acquisition of A9 phenotype is really critical for enriched engraftment and in vivo functions of hES-derived DA cells. Second, the micro-environment of the rat host striatum might not allow the survival of human DA cells or the differentiation of hES-NP toward DA neurons in vivo. By contrast, DA cells derived from mES cells, regardless of differentiation protocols and preparation methods, efficiently survived and integrated into rat brain (Bjorklund et al. 2002; Kim et al. 2002; Shim et al. 2004). Finally, a possibility cannot be excluded that the number of DA neurons in the grafts is underestimated. This assumption is based on the previous reports demonstrating human DA neurons are not well equipped with detoxifying system to remove oxygen free radicals generated by TH activity, and thereby TH expression in human DA cells may be self-controlled at a low level (Liste et al. 2004a,b). Therefore, TH antibodies used in this and previous other studies are not likely to be sensitive enough to detect those DA cells expressing low level of TH in vivo. This possibility is further supported by the finding demonstrating much higher proportion of cells expressing VMAT2, another specific marker for DA neuron, in the grafts (Fig. 7f). However, it is also possible that the subpopulation of the VMAT2+ cells represents serotonergic neuron population.

A significant restoration of rotational behavior was observed in the PD rats grafted with multi-passaged hES-NP cells. Considering that the minimal number of DA neurons for inducing behavioral improvement in rodents is 100–200 (Galpern et al. 1996), behavioral restoration observed in this study is likely attributable to the dopaminergic replacement by the surviving TH+ cells (201 ± 86 cells per animal). At the same time, it is also possible that paracrine factors secreted from grafted hES-originated cells may have trophic effects on dying host DA neurons, thereby contributing to the behavioral recovery. Similar to our data, functional improvement in PD rat behaviors with similarly few TH+ cells in grafts has been observed in hES cell transplantation in several studies (Ben-Hur et al. 2004; Martinat et al. 2006). By contrast, a study by Brederlau et al. (2006) recently demonstrated no reversal of motor deficits in the PD rat when grafted with hES cells. Additional extensive studies are required to confirm the in vivo functionality of hES-derived NP cells.

Teratoma formation is the most serious concern in ES cell transplantation. We, here, found no teratoma formation by transplanting muti-passaged hES-NP cells, although it may be necessary to evaluate this more extensively for an even longer period of post-transplantation and with a larger population of animals. High incidence of teratoma formations has been observed in PD rats grafted with partially differentiated mES cells (Bjorklund et al. 2002). Similarly, it has been reported that the in vitro period of hES cell differentiation is inversely correlated with the development of teratoma in an in vivo hES-derived grafts (Brederlau et al. 2006). On the basis of those findings, a relatively longer period of in vitro differentiation in our protocol [ca 38–42 days vs. 16 days in the study by Brederlau et al. (2006)] to the point of yielding a uniform population of neural-lineage cells, likely contributes to no incidence of teratoma formation in the present study. Furthermore, further elongated in vitro cultures for NP cell expansion with multiple passages in the state of single cell dissociates may have served to reduce teratoma incidence.

In conclusion, this study demonstrates an in vitro expansion of NP cells derived from hES cells without significant alteration of their proliferative and developmental properties toward high yield of DA neurons exhibiting in vitro and in vivo functions. These observations highly implicate that hES-NP could be a useful, on-demand cell source for the developmental studies as well as for therapeutic approaches in PD.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

We are grateful to Dr Kwang-Soo Kim (McLean Hospital, Harvard University, MA, USA) for his critical comments on this study. This work was supported by SC2150, SC2130 (Stem Cell Research Center of the 21st Century Frontier Research Program) funded by the Ministry of Science and Technology, Republic of Korea.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information
  • Amit M., Carpenter M. K., Inokuma M. S., Chiu C. P., Harris C. P., Waknitz M. A., Itskovitz-Eldor J. and Thomson J. A. (2000) Clonally derived human embryonic stem cell lines maintain pluripotency and proliferative potential for prolonged periods of culture. Dev. Biol. 227, 271278.
  • Bayer S. A. and Altman J. (1991) Neurocortical Development. Raven Press, New York.
  • Ben-Hur T., Idelson M., Khaner H., Pera M., Reinhartz E., Itzik A. and Reubinoff B. E. (2004) Transplantation of human embryonic stem cell-derived neural progenitors improves behavioral deficit in Parkinsonian rats. Stem Cells 22, 12461255.
  • Bing G. Y., Notter M. F., Hansen J. T. and Gash D. M. (1988) Comparison of adrenal medullary, carotid body and PC12 cell grafts in 6-OHDA lesioned rats. Brain Res. Bull. 20, 399406.
  • Bird J. M and Kimber S. J. (1984) Oligosaccharides containing fucose linked alpha(1-3) and alpha(1-4) to N-acetylglucosamine cause decompaction of mouse morulae. Dev. Biol. 104, 449460.
  • Bjorklund L. M., Sanchez-Pernaute R., Chung S. et al. (2002) Embryonic stem cells develop into functional dopaminergic neurons after transplantation in a Parkinson rat model. Proc. Natl Acad. Sci. USA 99, 23442349.
  • Bohn M. C., Cupit L., Marciano F. and Gash D. M. (1987) Adrenal medulla grafts enhance recovery of striatal dopaminergic fibers. Science 237, 913916.
  • Brederlau A., Correia A. S., Anisimov S. V. et al. (2006) Transplantation of human embryonic stem cell-derived cells to a rat model of Parkinson’s disease: effect of in vitro differentiation on graft survival and teratoma formation. Stem Cells 24, 14331440.
  • Capela A. and Temple S. (2002) LeX/ssea-1 is expressed by adult mouse CNS stem cells, identifying them as noependymal. Neuron 35, 865875.
  • Capela A. and Temple S. (2006) LeX is expressed by principle progenitor cells in the embryonic nervous system, is secreted into their environment and binds Wnt-1. Dev. Biol. 291, 300313.
  • Chang M. Y., Park C. H., Lee S. Y and Lee S. H. (2004) Properties of cortical precursor cells cultured long term are similar to those of precursors at later developmental stages. Brain Res. Dev. Brain Res. 153, 8996.
  • Chung S., Shin B. S., Hwang M., Lardaro T., Kang U. J., Isacson O. and Kim K. S. (2006) Neural precursors derived from embryonic stem cells, but not those from fetal ventral mesencephalon, maintain the potential to differentiate into dopaminergic neurons after expansion in vitro. Stem Cells 24, 15831593.
  • Damier P., Hirsch E. C., Agid Y. and Graybiel A. M. (1999) The substantia nigra of the human brain. I. Nigrosomes and the nigral matrix, a compartmental organization based on calbindin D(28K) immunohistochemistry. Brain 122, 14211436.
  • Draper J. S., Smith K., Gokhale P., Moore H. D., Maltby E., Johnson J., Meisner L., Zwaka T. P., Thomson J. A. and Andrews P. W. (2004) Recurrent gain of chromosomes 17q and 12 in cultured human embryonic stem cells. Nat. Biotechnol. 22, 5354.
  • Espejo E. F., Montoro R. J., Armengol J. A. and Lopez-Barneo J. (1998) Cellular and functional recovery of Parkinsonian rats after intrastriatal transplantation of carotid body cell aggregates. Neuron 20, 197206.
  • Gall C. M., Hendry S. H., Seroogy K. B., Jones E. G. and Haycock J. W. (1987) Evidence for coexistence of GABA and dopamine in neurons of the rat olfactory bulb. J. Comp. Neurol. 266, 307318.
  • Galpern W. R., Burns L. H., Deacon T. W, Dinsmore J. and Isacson O. (1996) Xenotransplantation of porcine fetal ventral mesencephalon in a rat model of Parkinson’s disease: functional recovery and graft morphology. Exp. Neurol. 140, 113.
  • Gaspar P., Ben Jelloun N. and Febvret A. (1994) Sparing of the dopaminergic neurons containing calbindin-D28k and of the dopaminergic mesocortical projections in weaver mutant mice. J. Neurosci. 61, 293305.
  • Hitoshi S., Seaberg R. M., Koscik C., Alexson T., Kusunoki S., Kanazawa I., Tsuji S. and Van Der Kooy D. (2004) Primitive neural stem cells from the mammalian epiblast differentiate to definitive neural stem cells under the control of Notch signaling. Genes Dev. 18, 18061811.
  • Imreh M. P., Gertow K., Cedervall J. et al. (2006) In vitro culture conditions favoring selection of chromosomal abnormalities in human ES cells. J. Cell. Biochem. 99, 508516.
  • Jacobson M. (1991). Developmental Neurobiology. Plenum Press, New York.
  • Kim J. H., Auerbach J. M., Rodrı′guez-Go′mez J. A. et al. (2002) Dopamine neurons derived from embryonic stem cells function in an animal model of Parkinson’s disease. Nature 418, 5056.
  • Ko J. Y., Lee J. Y., Park C. H. and Lee S. H. (2005) Effect of cell-density on in vitro dopaminergic differentiation of mesencephalic precursor cells. Neuroreport 16, 499503.
  • Lee S. H., Lumelsky N., Studer L., Auerbach J. M. and McKay R. D. (2000) Efficient generation of midbrain and hindbrain neurons from mouse embryonic stem cells. Nat. Biotechnol. 18, 675679.
  • Liste I., Garcia-Garcia E. and Martinez-Serrano A. (2004a) The generation of dopaminergic neurons by human neural stem cells is enhanced by Bcl-XL, both in vitro and in vivo. J. Neurosci. 24, 1078610795.
  • Liste I., Navarro B., Johansen J., Bueno C., Villa A., Johansen T. E. and Martinez-Serrano A. (2004b) Low-level tyrosine hydroxylase (TH) expression allows for the generation of stable TH+ cell lines of human neural stem cells. Hum. Gene Ther. 15, 1320.
  • Maitra A., Arking D. E., Shivapurkar N. et al. (2005) Genomic alterations in cultured human embryonic stem cells. Nat. Genet. 37, 10991103.
  • Martinat C., Bacci J. J., Leete T., Kim J., Vanti W. B., Newman A. H., Cha J. H., Gether U., Wang H. and Abeliovich A. (2006) Cooperative transcription activation by Nurr1 and Pitx3 induces embryonic stem cell maturation to the midbrain dopamine neuron phenotype. Proc. Natl Acad. Sci. USA 103, 28742879.
  • Max S. R., Bossio A. and Iacovitti L. (1996) Co-expression of tyrosine hydroxylase and glutamic acid decarboxylase in dopamine differentiation factor-treated striatal neurons in culture. Brain Res. Dev. Brain Res. 91, 140142.
  • McRitchie D. A., Hardman C. D. and Halliday G. M. (1996) Cytoarchitectural distribution of calcium binding proteins in midbrain dopaminergic regions of rats and humans. J. Comp. Neurol. 364, 121150.
  • Mendez I., Sanchez-Pernaute R., Cooper O., Viñuela A., Ferrari D., Björklund L., Dagher A. and Isacson O. (2005) Cell type analysis of functional fetal dopamine cell suspension transplants in the striatum and substantia nigra of patients with Parkinson’s disease. Brain 128, 14981510.
  • Okabe S., Forsberg-Nilsson K., Spiro A. C., Segal M. and McKay R. D. (1996) Development of neuronal precursor cells and functional postmitotic neurons from embryonic stem cells in vitro. Mech. Dev. 59, 89102.
  • Park C. H., Minn Y. K., Lee J. Y. et al. (2005) In vitro and in vivo analyses of human embryonic stem cell-derived dopamine neurons. J. Neurochem. 92, 12651276.
  • Park C. H., Kang J. S., Kim J. S. et al. (2006) Differential actions of the proneural genes Mash1 and neurogenins in Nurr1-induced dopamine neuron differentiation. J. Cell Sci. 119, 23102320.
  • Perrier A. L., Tabar V., Barberi T., Rubio M. E., Bruses J., Topf N., Harrison N. L. and Studer L. (2004) Derivation of midbrain dopamine neurons from human embryonic stem cells. Proc. Natl Acad. Sci. USA 101, 1254312548.
  • Qian X., Shen O., Goderia S. K., He W., Capela A., Davis A. A. and Temple S. (2000) Timing of CNS cell generation: a programmed sequence of neuron and glial cell production from isolated murine cortical stem cells. Neuron 28, 6980.
  • Reubinoff B. E., Pera M. F., Fong C. Y., Trounson A. and Bongso A. (2000) Embryonic stem cell lines from human blastocysts: somatic differentiation in vitro. Nat. Biotechnol. 18, 399404.
  • Roy N. S., Cleren C., Singh S. K., Yang L., Beal M. F. and Goldman S. A. (2006) Functional engraftment of human ES cell-derived dopaminergic neurons enriched by coculture with telomerase-immortalized midbrain astrocytes. Nat. Med. 12, 12591268.
  • Schulz T. C., Noggle S. A., Palmarini G. M., Weiler D. A., Lyons I. G., Pensa K. A., Meedeniya A. C., Davidson B. P., Lambert N. A. and Condie B. G. (2004) Differentiation of human embryonic stem cells to dopaminergic neurons in serum-free suspension culture. Stem Cells 22, 12181238.
  • Shen Q., Goderie S. K., Jin L., Karanth N., Sun Y., Abramova N., Vincent P., Pumiglia K. and Temple S. (2004) Endothelial cells stimulate self-renewal and expand neurogenesis of neural stem cells. Science 304, 13381340.
  • Shim J. W., Koh H. C., Chang M. Y., Roh E., Choi C. Y., Oh Y. J., Son H., Lee Y. S., Studer L. and Lee S. H. (2004) Enhanced in vitro midbrain dopamine neuron differentiation, dopaminergic function, neurite outgrowth, and 1-methyl-4-phenylpyridium resistance in mouse embryonic stem cells overexpressing Bcl-XL. J. Neurosci. 24, 843852.
  • Thompson L., Barraud P., Andersson E., Kirik D. and Bjorklund A. (2005) Identification of dopaminergic neurons of nigral and ventral tegmental area subtypes in grafts of fetal ventral mesencephalon based on cell morphology, protein expression, and efferent projections. J. Neurosci. 25, 64676477.
  • Thomson J. A., Itskovitz-Eldor J., Shapiro S. S., Waknitz M. A., Swiergiel J. J., Marshall V. S. and Jones J. M. (1998) Embryonic stem cell lines derived from human blastocysts. Science 282, 11451147.
  • Yamada T., McGeer P. L., Baimbridge K. G. and McGeer E. G. (1990) Relative sparing in Parkinson’s disease of substantia nigra dopamine neurons containing calbindin-D28K. Brain Res. 256, 303307.
  • Yan J., Studer L. and McKay R. D. (2001) Ascorbic acid increases the yield of dopaminergic neurons derived from basic fibroblast growth factor expanded mesencephalic precursors. J. Neurochem. 76, 307311.
  • Ye W., Shimamura K., Rubenstein J. L., Hynes M. A. and Rosenthal A. (1998) FGF and Shh signals control dopaminergic and serotonergic cell fate in the anterior neural plate. Cell 93, 755766.
  • Zeng X., Cai J., Chen J. et al. (2004) Dopaminergic differentiation of human embryonic stem cells. Stem Cells 22, 925940.

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Fig. S1. Differentiated hES colonies with neural structures on the stromal feeder MS5 (a and c) and PA6 (b and d). Images c and d are high-magnification views of the boxed regions in a and b, respectively. Cultures of hES cells on MS5 stromal feeders are uniformly comprise of colonies with primitive neuroepithelial structure (NP colony) that are compactly arranged with small cells with abundant cytoplasm and a barely visible nucleus (a, c). Arrows indicate NP colonies. By contrast, a subpopulation of colonies on PA6 cells is assembled by fully differentiated neuronal cells with multiple extensive processes (b, d). Scale bar, 40 μm.

Fig. S2. Unaltered phenotypic and differentiation properties of H9 hES-derived NP cells after multiple NP passages. H9 hES-NP cells at different NP passage numbers were proliferated in vitro in the presence of bFGF as described in �Materials and Methods�, and immunocytochemical analyses for nestin (neural precursor cell marker)/Ki67 (proliferating cell marker) were performed on the last day (day 7) of bFGF-proliferation. To determine the differentiation phenotypes of the NP cells, terminal differentiation of H9 hES-NP cells at each NP passage was induced for 6�8 days. Shown are representative images for cells positive for nestin/ki67 (A-C) after NP expansion, and TuJ1/GFAP (E-G), TH/TuJ1 (I-K) after NP differentiation. Scale bar, 40 μm. Graph D, H, and L depict the percent of cells positive for nestin/ki67, TuJ1/GFAP, and TH/TuJ1, respectively.

Fig. S3. Unaltered DA neuronal yield of hES-NP cells after freeze-and-thaw cycle. After 4 NP cell passages, the hES-NP cells continued to be cultured or frozen in liquid nitrogen (LN2). One week after the LN2 storage, cells were thawed and re-cultured. TuJ1+ neuronal and TH+/TuJ1+ DA neuronal yields of the LN2-stored cells at P6 were compared with those directly cultured without frozen in LN2 at the same passage number. No significant differences in TuJ1+ and TH+/TuJ1+ cell numbers was observed between the cultures with (+) and without (−) LN2 storage.

Fig. S4. High yield of DA neuron differentiation (a-c) and normal karyotype (d) of hES-derived NP cells after 10�12 NP passages. HSF6 hES-derived NP cells were further expanded in the presence of bFGF up to 12 NP passages. (a,b) The hES-NP-P10 and -P12 cells were in vitro differentiated for 8 days before TH/TuJ1 immunostaining. Scale bar, 40 μm. Graph c depicts the percent of cells immunoreactive for TuJ1 of the total (DAPI+) cells or TH of the total TuJ1+ cells. (d) Karyotype of hES-NP with 10 NP passages. Normal karyotype of the hES-NP-P10 cells is in a clear contrast with multiple chromosomal abnormalities of undifferentiated hES cells with 82 ES passages (hES-ES-P82) shown in Fig. 1b. The hES-NP-P10 cells were derived from hES cell with ES-P68. To obtain the hES-NP cells with NP-P10, hES-P68 cells underwent an in vitro period of cultures for 38 days of neural differentiation on MS5 feeder, 7 days of NP expansion in clusters, and 70 days (7 days × 10 passages) of NP passages (total 115 days). The period required from ES-P68 to ES-P82 was 98 days (7 days/passage × 14 passages). It is noteworthy that normal karyotype could be maintained during the period of hES differentiation and 10 NP cell passages, although the in vitro period was longer than that required for undifferentiated hES cell expansion from ES-P68 to ES-P82.

Table S1. Information of RT-PCR primer used in this study.

Please note: Blackwell Publishing are not responsible for the content or functionality of any supplementary materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.

JNC_4898_sm_figS1.jpg696KSupporting info item
JNC_4898_sm_figS2.jpg1551KSupporting info item
JNC_4898_sm_figS3.jpg443KSupporting info item
JNC_4898_sm_figS4.jpg556KSupporting info item
JNC_4898_sm_fig_legends.doc29KSupporting info item
JNC_4898_sm_tableS1.doc39KSupporting info item
JNC_figs1.jpg696KSupporting info item
JNC_figs2.jpg1551KSupporting info item
JNC_figs3.jpg443KSupporting info item
JNC_figs4.jpg556KSupporting info item
JNC_fig_legends.doc29KSupporting info item
JNC_tables1.doc39KSupporting info item

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