Stem Cells 2003;21:41-49
www.StemCells.com
© 2003 AlphaMed Press
A Subset of ES-Cell-Derived Neural Cells Marked by Gene Targeting
Hai-Qing Xian,
Elizabeth McNichols,
Andrew St. Clair,
David I. Gottlieb
Department of Anatomy and Neurobiology, Washington University School of Medicine, St. Louis, Missouri, USA
Key Words. ES cells • Olig2 • Neural differentiation • Gene targeting
David Gottlieb, Ph.D., Department of Anatomy and Neurobiology, Box 8108, Washington University School of Medicine, 660 S. Euclid Avenue, St. Louis, Missouri 63110, USA. Telephone: 314-362-2758; Fax: 314-362-3446; e-mail; gottlied{at}pcg.wustl.edu
 |
ABSTRACT
|
|---|
Embryonic stem cells differentiate efficiently in culture into neural progenitors, neurons, oligodendrocytes, and astrocytes. An embryonic stem (ES) cell line with green fluorescent protein (GFP) inserted into the gene for Olig2, a lineage-specific transcription factor, permits visualization and physical separation of a subset of living ES-cell-derived neural cells. GFP-expressing cells have morphological and antigenic properties of the oligodendrocyte lineage. The differentiation of living GFP-expressing cells can be followed in cultures, and they can be separated by fluorescence-activated cell sorting and cultured as pure populations. This system will allow detailed biochemical and molecular analysis of a neural differentiation pathway at a level not previously feasible. The strategy may have general applicability, since other neural lineages can be marked in an analogous manner.
|
INTRODUCTION
|
|---|
There is currently a great deal of interest in deriving differentiated cells from embryonic stem (ES) cells in tissue culture. Part of this is due to the possibility that ES cells could be a source of cells for transplantation for a variety of diseases. In the case of the nervous system, ES cells might supply neural cells for the treatment of Parkinson’s disease, stroke, spinal cord injury, and other disorders where neural cells are lost. Hope for ES cells as a therapy is justified, because ES cells differentiate efficiently in tissue culture into neural lineage cells including, neurons, oligodendrocytes, and astrocytes [1–4]. Collectively, these cells are referred to as ES-cell-derived neural lineage cells or ESNLCs. In addition to their potential for transplantation, ESNLCs also provide a model system for a wide range of uses, ranging from drug screening to exploring developmental mechanisms.
Two general approaches to differentiate totipotent ES cells into ESNLCs are in widespread use. In the first, aggregates of ES cells are cultured in the presence of retinoic acid (RA) to induce the neural phenotype [1, 2]. In the second, culture in defined medium is used to select neural cells [3, 5]. Both protocols are efficient, producing populations highly enriched in neural cells. Most remarkably, the end-stage differentiated ESNLCs strongly resemble their normal counterparts. For instance, neuronal ESNLCs have axons and dendrites that form physiologically normal synapses [1, 6]. Oligodendrocyte ESNLCs are able to form myelin in culture or when transplanted into injured neural tissue [5, 7]. A third, more recent method for differentiating mouse and primate ES cells by coculture with mesenchymal cells has been described [8, 9]. Taken together, these studies demonstrate that key features of neuronal and oligodendrocyte development are autonomous in the sense that they can take place outside the context of the embryo. Human ES cells are also capable of differentiating into neurons and glia in vitro [10–13], setting the stage for applications involving human ESNLCs.
While the inherent potential of ESNLCs for research and therapy is widely appreciated, major questions about these cells are unanswered. In the normal developing central nervous system (CNS), multipotent neural stem cells give rise to differentiated cells by a series of steps orchestrated by a large set of regulatory genes, many of them transcription factors [14]. It is likely that ESNLCs arise by broadly similar pathways regulated by the same genes. In fact, a few parallels between ESNLC and embryonic development are already known [4, 15]. However, detailed information about cellular and molecular aspects of ESNLC fate choice is limited. One reason is that ESNLCs are heterogeneous, with many cell types developing simultaneously, so it is difficult to identify and study specific cell types individually. New tools for marking and separating subsets of ESNLCs are needed.
Here, we describe an approach utilizing gene targeting to create a green fluorescent protein (GFP) knock-in to a gene that defines a subset of neural cells. Olig2 is a transcription factor in the basic helix-loop-helix family that plays a key role in regulating a CNS differentiation pathway. It is expressed in the ventral portion of the spinal cord and brain [16–19]. In the embryonic spinal cord, it is first expressed in ventral progenitor cells that give rise to oligodendrocytes and motor neurons. In the adult spinal cord, it is expressed in oligodendrocytes but not motor neurons [20, 21]. In the brain, Olig2-expressing progenitors give rise to oligodendrocytes; it is not yet known if they give rise to neurons. Olig2 and a related gene, Olig1, are required for survival of expressing cells [20, 21]. To investigate Olig2-expressing ESNLCs, we genetically engineered ES cells by inserting GFP into the Olig2 gene to create a knock-in ES cell line called G-Olig2. The intent was to produce an ES cell line in which GFP was expressed in Olig2-expressing cells but not in others. Living GFP-expressing cells could then be viewed microscopically and physically isolated via fluorescence-activated cell sorting (FACS). Here, we report that G-Olig2 cells fulfill these expectations, and that GFP-expressing cells have key properties of Olig2-expressing cells. This system provides a means of analyzing one subset of ESNLCs in detail. Because knock-ins can be produced in many regulatory genes, the approach should be applicable to other neural subsets.
|
MATERIALS AND METHODS
|
|---|
Knock-in Cell Line Construction
A BAC clone containing the Olig2 gene was isolated from an arrayed 129/SvJ mouse genomic BAC library (Incyte Genomic, Inc.; Palo Alto, CA) by polymerase chain reaction (PCR) screening. Verification that the clone contained the Olig2 gene rather than a related gene or pseudogene was by Southern mapping of BAC and genomic DNA. A map in which the translational start codon is designated as the +1 position is used throughout to describe the Olig2 gene (Fig. 1
). The targeting vector was constructed in yeast using homologous recombination [22]. To generate the targeting construct, an approximately 21-kb EcoR I fragment (position –2.4 kb to +18 kb) was subcloned into the yeast/ E. coli shuttle vector, pRS316 (GeneBank Accession number GI:417973) [23]. The 5' arm of homology was 2.4 kb, and the 3' arm was 18 kb. This construct was used to transform yeast cells (strain FM393) using uracil (URA) selection. Next, a histidine (HIS) cassette flanked by AscI sites was introduced at position +3 of the Olig2 gene by homologous recombination. In detail, the yeast selection gene, His3, flanked on each side by an AscI site and 39 bp of Olig2 gene sequence was amplified by PCR. This DNA was transfected into yeast carrying pRS316/Olig2 under double URA/HIS selection. Positive colonies were screened for correct recombinants by PCR for the novel junction between HIS and the Olig2 gene. Sequencing showed that the His3 gene with AscI sites was integrated into pRS316/Olig2 by recombination with the Olig2 homologous flanking regions. pRS316/Olig2-His was shuttled back to E. coli. Next, the His3 gene was replaced by an epidermal growth factor protein (EGFP)-Neo cassette flanked by AscI using conventional cloning. The result was pRS316/Olig2-EGFP-Neo. This cassette includes promoterless EGFP and a phosphoglycerol kinase (PGK)-promoter-driven neomycin gene. Ten micrograms of the targeting vector, pRS316/Olig2-EGFP-Neo, linearized by Sal I was electroporated into RW4 ES cells (obtained from Dr. Tim Ley, Washington University). Electroporated cells were cultured in complete medium supplemented with leukemia inhibitory factor (LIF) and beta-mercaptoethanol (see below) plus 250 µg/ml of G418 (neomycin). Colonies resistant to G418 were selected; these included random insertions of the transgene and targeted insertions. Targeted clones were identified by PCR for the novel junction created by homologous recombination in genomic DNA. Four of 96 G418-resistant colonies were positive and, thus, had undergone homologous recombination. Structure of the targeted gene was confirmed by Southern blot analysis. The cell line was designated G-Olig2.
View larger version (27K):
[in this window]
[in a new window]
|
Figure 1. Targeting the Olig2 gene. The strategy for creating the ES cell line, G-Olig2. A) Maps of the Olig2 gene, targeting construct, and targeted gene. Restriction enzyme sites indicated are: B = BamH1; H = HindIII; and EI = EcoR1. ATG was the translational start site. The PGK promoter driving neomycin is indicated by the bent arrow. Internal and external probes for Southern blots are shown. The right arm of the vector is represented with a break ({{) to indicate about 10 kb omitted from the map. B) Southern blots of genomic DNA from G-Olig2 (targeted) and RW4 parental cells (parent). Hybridization to DNA cut with HindIII and probed with internal probe (left panel) and cut with BamH1 and probed with external probe (right panel) are shown. Targeted and normal alleles and their sizes are indicated. Note the predicted upshift in the size of the bands due to insertion of the cassette.
|
|
ES Cell Culture
G-Olig2 cells were cultured essentially as described previously [1]. Undifferentiated ES cells were cultured on irradiated STO cell feeder layers in complete medium (Dulbecco’s modified Eagle’s medium [DMEM]), 10% fetal calf serum, and 10% newborn calf serum) supplemented with LIF and beta-mercaptoethanol. Differentiation into ESNLCs was as previously described [1]. Briefly, embryoid bodies (EBs) were cultured for 4 days in the absence of RA, followed by 4 days in the presence of 5 x 10-7 M RA. Induced EBs (undissociated) were then plated onto glass coverslips coated with gelatin and cultured for 3 days in complete medium followed by the indicated number of additional days in Neurobasal medium +B27 supplement (Life Technologies; Carlsbad, CA; http://www.lifetech.com).
Fluorescence Microscopy
Cells were viewed under a Zeiss Axiovert 100M fluorescent microscope (Carl Zeiss Inc.; Thronwood, NY).
Cell Sorting
Differentiated cultures were dissociated by trypsinization. Cells were washed in DMEM to remove proteins, suspended in trypsin 0.25%-EDTA, 1 mM, and incubated at 37°C for 8 minutes. Next, Hank’s saline with 10% newborn bovine serum was added to quench trypsin. Cells were then triturated to dissociate cells to a suspension consisting predominantly of single cells. Finally, cell suspensions were washed and resuspended in Hank’s saline at 107 cells/ml.
GFP-expressing cells were analyzed and sorted on a Cytomation MoFlo (Cytomation Inc.; Ft. Collins, CO) high-speed cell sorter. An incident beam of 488 nm was used. Forward scatter light (FSC) was used to sense size. Forward and side scatter (SSC) parameters both used 488 nm bandpass filters with a 10-nm bandwidth, and both were plotted on linear scales. A histogram with FSC versus SSC was used to exclude debris, doublets, and clumps. GFP fluorescence was read through filters that pass light in the 510-550 nm range. A second histogram, using FSC versus GFP was used to obtain a two-dimensional bitmap for sort exclusion. GFP was plotted in logarithmic scale. The target population had to satisfy two gates based on its placement in the FSC versus SSC histogram and its GFP expression. Sorted cells were collected in HEPES-buffered DMEM plus 10% fetal bovine serum. Purity and morphology of GFP+ sorted cells was judged by fluorescence microscopy.
Antibody Staining
Cells were fixed with 4% paraformaldehyde for 30 minutes. After washing, the cells were blocked and incubated with primary antibodies at the indicated dilutions against O1 (1:100) or O4 (1:100), NeuN (1:50), TuJ1 (1:100), and GFAP (1:1,000) (Chemicon; Temecula, CA; http://www.chemicon.com). After washing, cells were incubated with species- and isotypespecific Texas-red-conjugated secondary antibodies (Jackson Immunoresearch; West Grove, PA; www.jacksonimmuno.com). Cells were viewed under a Zeiss Axiovert 100M microscope.
|
RESULTS
|
|---|
Construction of G-Olig2
In a knock-in, a promotorless cDNA is inserted into the target gene so that the cDNA is expressed with the same tissue and developmental pattern as the normal target gene. A GFP knock-in to the Olig2 gene was created so that expression of this gene could be visualized in living cells. The targeting strategy used to create the GFP knock-in to Olig2 is shown in Figure 1
. A GFP-neo cassette was inserted 3' to the Olig2 translational start site (ATG) in the proper reading frame. Southern mapping of the normal and targeted Olig2 genes confirmed that the predicted targeted gene was produced.
GFP Expression in G-Olig2 ESNLCs
In the developing CNS, Olig2 is expressed in a highly restricted pattern. This predicts that GFP expression in G-Olig2 ESNLCs should also have a restricted pattern of expression. Expression was analyzed by inducing G-Olig2 cells with retinoic acid by a standard protocol (Fig. 2) and observing GFP fluorescence in living cells via fluorescence microscopy. Undifferentiated ES cells had a low, barely detectable level of GFP expression (data not shown). After retinoic acid induction (4-/4+ stage), a subset of cells, estimated at approximately 30%, had a low level of expression, with the remainder being negative (data not shown).
View larger version (17K):
[in this window]
[in a new window]
|
Figure 2. Schematic of induction. The induction protocol utilized is shown schematically. Undifferentiated ES cells were first cultured as embryoid bodies for 4 days in standard medium without added factors. They were then cultured in the presence of retinoic acid (RA) for four additional days. After induction, they were plated on a gelatin substrate to allow differentiation of neurons, astrocytes, and oligodendrocytes.
|
|
To allow further differentiation after retinoic acid induction, EBs were cultured on a gelatin substrate under conditions that favored terminal differentiation of neurons and glia. The EBs adhered to the substrate and became surrounded by cells that migrated onto the substrate. GFP expression in these cultures was followed for up to 2 weeks. On about day 4 postplating, a few cells showed greater GFP expression. By day 8, there were cells with intense GFP expression in most EBs. Figure 3 shows an example of the expression pattern. GFP-expressing cells were found in restricted zones of the EBs, rather than being scattered at random throughout the EBs. GFP-expressing cells were also found in the cellular outgrowths of the EBs. GFP-expressing cells were relatively small and had branches projecting from the cell body. The GFP-expressing cells consisted of about 2% of the total cell population as determined by FACS (see below).
View larger version (121K):
[in this window]
[in a new window]
|
Figure 3. GFP expression in cultured embryoid bodies. Induced 4-/4+ EBs were plated and cultured for an additional 8 days. Cultures were then viewed for GFP fluorescence (A and C) or with differential interference contrast optics (B and D). A and B: an attached EB at low power. Notice that the EB consists of GFP+ (arrowheads) and GFP- zones; scale bar = 50 µm. C and D: higher magnification along the edge of an attached EB. A subset of cells was intensely stained, while neighboring cells were negative. Arrowheads point to cells with processes; scale bar = 12.5 µm.
|
|
Immunostaining of GFP-Expressing Cells
To determine which lineage GFP-expressing cells belonged to, differentiated cultures were stained with a panel of antibodies specific to neural subtypes. Many of the GFP-expressing cells had complex patterns of branches, consistent with them being oligodendrocyte precursor cells (OPCs). To verify that these cells were indeed OPCs, cultures were stained with O4, an antibody specific for early OPCs [24]. Many of the GFP-expressing cells were O4 positive; examples are shown in Figure 4. These double-labeled cells had the branched morphology of typical OPCs. The O1 antibody stains a cell surface antigen expressed by more advanced oligodendrocyte cells [24]. In a separate experiment, cultures were stained with O1. Many GFP+ cells with oligodendrocyte morphology also stained with O1 (Fig. 5). In the mouse, Olig2 is expressed in the oligodendrocyte lineage. These immunostaining results demonstrate that many of the GFP-expressing ESNLCs in G-Olig2 cultures belonged to this lineage. We also asked if all O4 positive cells were GFP+ and discovered that they were not. Examples of O4 positive cells with typical OPC morphology that did not express GFP were detected in the cultures. This implies that some cells could differentiate along the OPC lineage in the absence of Olig2 expression.
View larger version (41K):
[in this window]
[in a new window]
|
Figure 4. GFP and O4 costaining. Differentiated cultures were stained with the O4 antibody. (A and B) Fields with cells expressing both GFP (green to yellow) and O4 antigen (red). Note double labeling of cells and their oligodendrocyte-like morphology. (C-E) Demonstration of double-label technique. Two cells viewed for GFP (C), O4 staining (D), and the merged images (E). Individual cells clearly expressed both GFP and the O4 antigen. Cells stained with control nonimmune antibodies were negative (data not shown).
|
|
View larger version (53K):
[in this window]
[in a new window]
|
Figure 5. GFP and O1 costaining. Mature plated cultures were stained with O1 antibody to visualize more mature oligodendrocytes. A-C show examples of double-labeled cells. Relative intensity of the two labels varied from cell to cell. In B, one cell (arrowhead) is more intensely labeled with GFP than O1. For another cell (arrowhead), the reverse was true. D-F demonstrate colocalization in a cell. E and F show the single channel images; D is the merged image.
|
|
Differentiated cultures, such as those in Figures 4 and 5, also had many astrocytes and neurons. Based on the expression patterns of Olig2 in the intact animal, it was predicted that these cells would not express GFP. Differentiated cultures were stained with appropriate markers to test this prediction. One set was stained with GFAP to mark astrocytes. Cultures contained many GFAP+ cells with typical astrocytic morphology (Fig. 6A). However, the vast majority of these cells did not express GFP. We did find a few exceptional examples of GFP+GFAP+ cells. Thus, GFP-expressing cells did not differentiate into astrocytes, with a few rare exceptions. It is possible that different culture conditions would cause a larger fraction of GFP+ cells to differentiate into astrocytes. Other cultures were stained with the neuronal marker, NeuN, which stains neuronal nuclei. Cultures contained many positive cells (Fig. 6B). These were often intermingled with GFP-expressing cells. However, GFP and NeuN were never coexpressed. Therefore, we conclude that GFP-expressing cells were not neurons. This conclusion was strengthened by staining cells with TuJ1, another marker for neurons (Fig. 6C). Examples of costained cells could not be found.
View larger version (71K):
[in this window]
[in a new window]
|
Figure 6. Nonoverlap of GFP and neuronal and astrocytic markers in differentiated cultures. A) Field of GFAP stained cells. Positions of cell bodies can be seen as "holes" in the filamentous network (arrowheads are examples). None of the cell bodies or processes expressed GFP. B) A field with GFP+ cells (green; arrowhead on left) and NeuN-positive cells (red; arrowhead on right). Red and green cells were clearly distinct. C) Neuronal marker, TuJ1, stained many cells; none were GFP+. Scale bars = 20 µm.
|
|
In conclusion, many of the intensely GFP+ cells appeared to be oligodendrocyte-lineage cells, as judged by morphology and antigen expression. Neurons were never GFP+, and the vast majority of astrocytes were GFP-. Thus, the pattern of GFP expression was consistent with that predicted from the Olig2 expression pattern in the developing CNS.
Cell Sorting
A second goal of creating the G-Olig2 cell line was to physically separate the subset of ESNLCs that express GFP. We have shown that this goal is feasible. Cultures of differentiated EBs were trypsinized to yield a suspension consisting of predominantly single cells. This suspension was analyzed by FACS (Fig. 7). Consistent with microscopic observation, the great majority of cells were negative. The average fluorescence of negative cells was 5 units on the FL1 scale (Fig. 7A). There was also a small population of positive cells. These ranged from about 20 to almost 1,000 units on the FL1 scale. About 2% of the cells expressed fluorescence above background levels. An experiment was carried out to determine if sorting was feasible. The sorting gate (R2 in Fig. 7A) was set to include about 1% of the cells. When sorted cells were examined immediately after sorting, they consisted mostly of healthy looking, single GFP-expressing cells. A potential problem was that the cells were damaged during sorting and could not develop further. To find out if the cells were capable of further development, they were plated and cultured for 1 day on a laminin substrate. Many of the sorted cells attached to the substrate, spread, and elaborated branches. They had the morphology of OPCs and were costained for O4 antigen (Fig. 7B). Thus, GFP+ cells survived dissociation and sorting and went on to develop oligodendrocyte progenitor cell morphology and antigen expression. Other sorted GFP+ cells did not stain for O4. This was expected because, as noted above, not all GFP+ cells in the unsorted cultures expressed O4.
View larger version (22K):
[in this window]
[in a new window]
|
Figure 7. Sorting cells from differentiated cultures. A) FACS analysis of dissociated cells in terms of forward scatter (FSC) and green fluorescence intensity (FL1). These parameters were used to set sorting gates. Cells falling in the R2 region were sorted and collected. Sorted cells were cultured for 1 day. B) Four examples of sorted GFP-expressing cells that also costained with O4 (red).
|
|
|
DISCUSSION
|
|---|
ES cells can differentiate efficiently in tissue culture into neural lineage cells, termed ESNLCs. ESNLCs are a heterogeneous population of cells in which several lineages, including neurons, astrocytes, and oligodendrocytes, differentiate simultaneously. In order to maximize the utility of ESNLCs as a model system and a source of cells for transplantation, it will be essential to produce more homogeneous cultures. In this report, we explored a selective strategy based on an ES cell line (designated G-Olig2) with a promotorless GFP gene inserted into a cellular gene expressed in a subset of neural cells. A subset of ESNLCs derived from G-Olig2 expressed GFP. The results suggest that this strategy is an exciting way to deal with the problem of heterogeneity.
In the developing mouse, Olig2 is first expressed in a set of ventrally located progenitor cells at about embryonic day 10 [18]. At later stages, expression increases in oligodendrocyte lineage cells and is downregulated in neuronal and astrocytic lineages. Expression of GFP in G-Olig2 ESNLCs resembled this in vivo pattern. Expression was very low in pluripotent ES cells and cells early in the induction procedure. In differentiated cultures, intense expression occurred in a subset of cells. Many of these cells had the morphology and antigenic characteristics of OPCs. Conversely, neuronal ESNLCs did not express GFP, and only rare astrocytic ESNLCs expressed GFP. Thus, the pattern predicted from Olig2 gene expression was obtained. In the culture system used here, only a small percentage of the cells expressed GFP intensely. This could mean that, under our conditions, OPCs were a small percentage of the ESNLCs generated. Another possibility is that not all OPCs expressed the Olig2 gene. The fact that we saw O4-positive cells that were GFP- is consistent with this idea. We are currently exploring ways to increase the numbers of GFP-expressing cells by adding growth factors and utilizing other media.
The technical aspects of our results are worth underscoring. The Olig2 gene is expressed at only moderate levels. An initial concern was that GFP expression would occur at a low level and, thus, not be useful as a marker. We found that, to the contrary, expression of GFP was high enough to visualize cells microscopically and also to sort them with FACS. A second concern was that ESNLCs (many of which are highly branched cells) might not be viable after FACS sorting. Data show that sorted cells remained viable and were able to develop further. Thus, the knock-in to Olig2 is, from a technical point of view, a usable tool. We anticipate that this strategy will be applicable to other genes and, thus, be a general approach for investigating ESNLCs.
In summary, a knock-in ES cell line offered a valuable approach to understanding the developmental pathways of ESNLCs. Since it utilizes ES cells differentiating in vitro, we were able to investigate the full developmental sequence starting with a pluripotent cell type and ending with differentiated OPCs. Olig2-expressing ESNLCs could be visualized and physically separated from other cells. They had the predicted properties of their counterparts in the intact embryo. In future studies, it will be possible to determine the developmental potential of isolated GFP+ cells, analyze their response to growth factors, and profile their gene expression via gene arrays. Such studies will be crucial for a deeper understanding of this subset of ESNLCs and for realizing their potential for transplantation and other applications.
|
CONCLUSION
|
|---|
A mouse ES cell line (G-Olig2) with a GFP knock-in to the Olig2 gene was made and characterized. A subset of ESNLCs derived from G-Olig2 expressed GFP. Many of the expressing cells had key characteristics of OPCs. These cells were readily visible by fluorescence microscopy in living cultures and could be physically separated by FACS. This system will enable new types of analysis of ESNLCs that are not possible with unmarked cells.
|
ACKNOWLEDGMENT
|
|---|
This research was supported by a grant from the Pharmacia/Washington University Research Program and NIH grant P01 NS39577. We thank Amy Boyet and Bill Eades for FACS analysis, Cheryl Rivers for preparing the manuscript, and other members of the Gottlieb lab for helpful discussions.
|
REFERENCES
|
|---|
- Bain G, Kitchens D, Yao M et al. Embryonic stem cells express neuronal properties in vitro. Dev Biol 1995;168:342–357.[CrossRef][Medline]
- Strubing C, Ahnert-Hilger G, Shan J et al. Differentiation of pluripotent embryonic stem cells into the neuronal lineage in vitro gives rise to mature inhibitory and excitatory neurons. Mech Dev 1995;53:275–287.[CrossRef][Medline]
- Okabe S, Forsberg-Nilsson K, Spiro AC et al. Development of neuronal precursor cells and functional postmitotic neurons from embryonic stem cells in vitro. Mech Dev 1996;59:89–102.[CrossRef][Medline]
- Gottlieb DI. Large-scale sources of neural stem cells. Ann Rev Neurosci 2002;25:381–407.[CrossRef][Medline]
- Brustle O, Jones KN, Learish RD et al. Embryonic stem cell-derived glial precursors: a source of myelinating transplants. Science 1999;285:754–756.[Abstract/Free Full Text]
- Finley MF, Kulkarni N, Huettner JE. Synapse formation and establishment of neuronal polarity by P19 embryonic carcinoma cells and embryonic stem cells. J Neurosci 1996;16:1056–1065.[Abstract/Free Full Text]
- Liu S, Qu Y, Stewart T et al. Embryonic stem cells differentiate into oligodendrocytes and myelinate in culture and after spinal cord transplantation. Proc Natl Acad Sci USA 2000;97:6126–6131.[Abstract/Free Full Text]
- Kawasaki H, Mizuseki K, Nishikawa S et al. Induction of midbrain dopaminergic neurons from ES cells by stromal cell-derived inducing activity. Neuron 2000;28:31–40.[CrossRef][Medline]
- Kawasaki H, Suemori H, Mizuseki K et al. Generation of dopaminergic neurons and pigmented epithelia from primate ES cells by stromal cell-derived inducing activity. Proc Natl Acad Sci USA 2002;99:1580–1585.[Abstract/Free Full Text]
- Reubinoff BE, Itsykson P, Turetsky T et al. Neural progenitors from human embryonic stem cells. Nat Biotechnol 2001;19:1134–1140.[CrossRef][Medline]
- Zhang SC, Wernig M, Duncan ID et al. In vitro differentiation of transplantable neural precursors from human embryonic stem cells. Nat Biotechnol 2001;19:1129–1133.[CrossRef][Medline]
- Carpenter MK, Inokuma MS, Denham J et al. Enrichment of neurons and neural precursors from human embryonic stem cells. Exp Neurol 2001;172:383–397.[CrossRef][Medline]
- Schuldiner M, Eiges R, Eden A et al. Induced neuronal differentiation of human embryonic stem cells. Brain Res 2001;913:201–205.[CrossRef][Medline]
- Marquardt T, Pfaff S. Cracking the transcriptional code for cell specification in the neural tube. Cell 2001;106:651–654.[CrossRef][Medline]
- Mujtaba T, Piper DR, Kalyani A et al. Lineage-restricted neural precursors can be isolated from both the mouse neural tube and cultured ES cells. Dev Biol 1999;214:113–127.[CrossRef][Medline]
- Lu QR, Yuk D, Alberta JA et al. Sonic hedgehog—regulated oligodendrocyte lineage genes encoding bHLH proteins in the mammalian central nervous system. Neuron 2000;25:317–329.[CrossRef][Medline]
- Zhou QS, Wang S, Anderson DJ. Identification of a novel family of oligodendrocyte lineage-specific basic helix-loop-helix transcription factors. Neuron 2000;25:331–343.[CrossRef][Medline]
- Takebayashi H, Yoshida S, Sugimori M et al. Dynamic expression of basic helix-loop-helix Olig family members: implication of Olig2 in neuron and oligodendrocyte differentiation and identification of a new member, Olig3. Mech Dev 2000;99:143–148.[CrossRef][Medline]
- Sun TY, Echelard Y, Lu R et al. Olig bHLH proteins interact with homeodomain proteins to regulate cell fate acquisition in progenitors of the ventral neural tube. Curr Biol 2001;11:1413–1420.[CrossRef][Medline]
- Lu QR, Sun T, Zhu Z et al. Common developmental requirement for Olig function indicates a motor neuron/oligodendrocyte connection. Cell 2002;109:75–86.[CrossRef][Medline]
- Zhou Q, Anderson DJ. The bHLH transcription factors OLIG2 and OLIG1 couple neuronal and glial subtype specification. Cell 2002;109:61–73.[CrossRef][Medline]
- Wattler S, Kelly M, Nehls M. Construction of gene targeting vectors from lambda KOS genomic libraries. Biotechniques 1999;26:1150–1160.[Medline]
- Sikorski RS, Hieter P. A system of shuttle vectors and yeast host strains designed for efficient manipulation of DNA in Saccharomyces cerevisiae. Genetics 1989;122:19–27.[Abstract/Free Full Text]
- Sommer I, Schachner M. Monoclonal antibodies (O1 to O4) to oligodendrocyte cell surfaces: an immunocytological study in the central nervous system. Dev Biol 1981;83:311–327.[CrossRef][Medline]
Received June 7, 2002;
accepted for publication August 9, 2002.
This article has been cited by other articles:
|
|
|
|
 
X. Zhang, S. A. Horrell, D. Delaney, and D. I. Gottlieb
Embryonic Stem Cells as a Platform for Analyzing Neural Gene Transcription
Stem Cells,
July 1, 2008;
26(7):
1841 - 1849.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. M. Wobus and K. R. Boheler
Embryonic Stem Cells: Prospects for Developmental Biology and Cell Therapy
Physiol Rev,
April 1, 2005;
85(2):
635 - 678.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. S. O'Shea
Self-renewal vs. Differentiation of Mouse Embryonic Stem Cells
Biol Reprod,
December 1, 2004;
71(6):
1755 - 1765.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
