| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
EMBRYONIC STEM CELLS: CHARACTERIZATION SERIES |
Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
Key Words. Embryonic stem cell totipotency
Correspondence: Saul J. Sharkis, Ph.D., Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, Room 551, 1550 Orleans Street, Baltimore, Maryland 21231, USA. Telephone: 410-955-8508; Fax: 410-502-5742; e-mail: ssharkis{at}jhmi.edu
Received July 5, 2005;
accepted for publication January 26, 2006.
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Two intrinsic transcriptional factors, Oct4 (also known as Pou5f1) and Nanog, are required for the maintenance of the undifferentiated state of embryonic stem (ES) cells [2]. Extrinsic factors can regulate murine embryonic stem cells (mESCs) status. Mouse ESCs remain undifferentiated in the presence of leukemia inhibitory factor (LIF) [3, 4]. However, hESCs can self-renew in the absence of exogenously added LIF [1] and LIF is unable to maintain the pluripotent state of hESCs [5]. All of the extrinsic signals that support hESC self-renewal have yet to be definitively identified. Several reports suggest that fibroblast growth factor (FGF) signaling can play a role in maintaining hESC self-renewal and pluripotency (referred to as hESC properties or hESC stemness) [6, 7]. TGFß-1 has recently been shown to synergize with basic fibroblast growth factor (bFGF) (also called FGF2) to maintain the undifferentiated state of hESCs in feeder-free culture [8]. The Wnt pathway is also involved in the short-term maintenance of pluripotency of both mouse and human embryonic stem cells [9]. BMP synergizes with LIF to maintain self-renewal in mESC [10]. On the other hand, BMP signaling has been shown to promote hESC differentiation into trophoblasts [11]. Thus, several factors play regulatory roles in the ESCs of mouse, human, or both.
Nodal/Activin signals establish the embryonic axes, induce mesoderm and endoderm, pattern the nervous system, and determine left-right asymmetry in vertebrates [12]. Nodal and Activin bind activin receptors and activate Smad2 by phosphorylation. Upon activation by phosphorylation and association with Smad4, the receptor-activated Smad2 translocates to the nucleus and, in concert with other transcription factors, regulates gene expression [12].
Here we report that Activin A functions to maintain hESC self-renewal and pluripotency. We observed that upregulation of Nodal/Activin, FGF, Wnt, and Hedgehog (Hh) signals and downregulation of BMP and Notch signals are correlated with increased expression of Oct4 and Nanog in an aneuploid hESC line (referred to as H1T; T stands for translocation). Activin A, unlike Nodal, was necessary and sufficient for the maintenance of self-renewal and pluripotency of hESCs and was able to support long-term feeder-free growth of hESCs. Importantly, we demonstrated that Activin A was able to induce the expression of other regulators, such as Oct4, Nanog, Nodal, Wnt3, bFGF, and FGF8 and slightly suppressed the expression of BMP7. Therefore, our data revealed a central role of Activin A in maintaining hESCs stemness.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Karyotype Analysis
Karyotype analysis (G-banding) was performed on at least 20 cells from each sample. Multicolor spectral karyotyping (SKY) [13] was performed according to the protocol supplied with the probes (Applied Spectral Imaging, Carlsbad, CA).
Analysis of Cloning Efficiency
The cloning efficiency was determined as described by Amit et al. [6].
Flow Cytometry
Dissociated hESCs were suspended in phosphate-buffered saline (PBS) containing 1% bovine serum albumin (BSA) at a concentration of 1 x 106 cells per 100 µl and add SSEA-4 monoclonal antibody (Developmental Studies Hybridoma Bank, Iowa City, IA, http://www.uiowa.edu/~dshbwww). After 30 minutes at 4°C, the cells were washed once and resuspended in PBS containing 1% BSA. Then fluorescein isothiocyanate-conjugated goat anti-mouse IgG was added. Finally, the cells were washed twice and stained with propidium iodide. Live cells identified by propidium iodide exclusion were analyzed for surface marker expression using FACSCalibur (BD Biosciences, San Diego, http://www.bdbiosciences.com) and Cell Quest software (BD Biosciences).
Immunostaining
Immunostaining was carried out similarly as described [14]. Antibodies used were anti-Oct4 (Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com), anti-Nanog (R&D Systems), anti-SSEA3 (Developmental Studies Hybridoma Bank), anti-SSEA4 (Developmental Studies Hybridoma Bank), anti-tra-1–60 (Chemicon, Temecula, CA, http://www.chemicon.com), anti-tra-1–81 (Chemicon), anti-pan-cytokeratin (Sigma-Aldrich, St. Louis, http://www.sigmaaldrich.com), anti-smooth muscle actin (Sigma-Aldrich), anti-
-fetoprotein (Sigma-Aldrich). For Oct4 and Nanog immunostaining, cells were fixed in 4% paraformaldehyde at room temperature for 20 minutes followed by permeabilization for 20 minutes in 100% ethanol.
Teratoma Formation
Cells were injected intramuscularly into non-obese diabetic/severe combined immune deficient (NOD/SCID) mice (~5 x 106 cells per site). Four mice were injected for each cell line; all of the mice formed teratomas. After 4–6 weeks, tumors were processed for hematoxylineosin staining similarly as described [15]. All animal experiments were conducted in accordance with the Guide for the Care and Use of Animals for Research Purposes and were approved by the Johns Hopkins Animal Care Committee.
Microarray
Total RNA was extracted with TRIZOL reagent (Invitrogen) and further purified using an RNeasy column (Qiagen, Hilden, Germany, http://www1.qiagen.com). The labeling procedure was carried out by using a RNA Fluorescent Linear Amplification Kit (Agilent Technologies, Palo Alto, CA, http://www.agilent.com). Fragmentation was carried out by incubating at 60°C for 30 minutes in fragmentation buffer (Agilent Technologies) and stopped by adding equal volume of 2x hybridization buffer (Agilent Technologies). Fragmented target was applied to a Whole Human Genome Oligo Microarray that contains 41,000 genes (Agilent Technologies). Hybridization proceeded at 60°C for 17 hours in a hybridization oven (Robbins Scientific, Sunnyvale, CA, http://www.robsci.com). The hybridized array was scanned with Agilent G2565BA microarray scanner. The TIFF image generated was loaded into Feature Extraction Software (Agilent Technologies) for feature data extraction.
Immunoblotting
Cells were lysed with 1x lysis buffer: 20 mM Tris (pH 7.5), 150 mM NaCl, 1% Triton X-100, 1 mM Na3VO4, and complete mini protease inhibitor cocktail (Roche Diagnostics, Basel Switzerland, http://www.roche-applied-science.com). Total protein (10 µg) was loaded for each lane. Membranes were blocked in Tris-buffered saline with 0.1% Tween and 5% milk. Antibodies used were anti-phospho-Smad2/3 (Cell Signaling Technology, Beverly, MA, http://www.cellsignal.com), anti-Smad2/3 (Cell Signaling), anti-Oct4 (Santa Cruz Biotechnology), anti-human Nanog (R&D Systems), and ß-Actin (Abcam, Cambridge, MA, http://www.abcam.com). Primary antibodies were incubated overnight, and secondary antibodies were incubated for 2 hours. Proteins were detected by chemiluminescence (Pierce, Rockford, IL, http://www.piercenet.com).
TRAP Assay
Telomerase activity was measured using the telomeric repeat amplification protocol (TRAP) assay as described [16].
RNA Isolation and Real-Time Reverse Transcription-Polymerase Chain Reaction
Total RNA was prepared using the RNAeasy kit (Qiagen) and used as a template for reverse transcription-polymerase chain reaction (RT-PCR). Real-time PCR was performed in MyiQ real-time PCR detection system (Bio-Rad, Hercules, CA, http://www.bio-rad.com) using an Synergy Brand GreenI-based PCR Master mix (Bio-Rad). PCR primers are listed in the supplemental online Table 1
. Each experiment was carried out at least three times. The expression value of each gene was normalized to the amount of glyceraldehyde-3-phosphate dehydrogenase cDNA to calculate a relative amount of RNA present in each sample. The expression level of each gene in non-CM was arbitrarily defined as 1 unit. The normalized expression values for all control and treated samples were averaged, and an average fold change was determined. Analysis of variance (ANOVA) was conducted between the normalized relative expression values for control and treated samples to determine statistical significance.
|
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
; data not shown). We refer to this line as the H1T hESC line.
|
), were able to form teratomas, and demonstrated ability to differentiate into all three germ layers (Fig. 1C), as did the euploid H1 cells. These activities of the H1T line suggest that the maintenance of stem cell properties was not lost with chromosomal translocation. The self-renewal advantage is a desirable feature of hESCs. Therefore, the aneuploid H1T hESCs provide a valuable tool to query the mechanism underlying self-renewal of hESCs. We assumed that if we discovered differences between H1T and H1 cells that conferred a self-renewal advantage on the H1T cells, we would aid in identifying the mechanism(s) that regulated the self-renewal of hESCs. Based on this assumption, we performed a microarray analysis comparing the gene expression profile of H1T hESCs and H1 hESCs.
The microarray analysis was performed according to the manufacturer’s instructions (Agilent Technologies). Fold change and p value were processed with Feature Extraction Software (Agilent Technologies). We observed that the H1T cells expressed 50% more Oct4 than H1 cells did (Table 1). It has been reported that a precise level of Oct4 is required to maintain the undifferentiated state of mESCs. The threshold for inducing differentiation is apparently set at 50% above or below the normal diploid expression level of Oct4 in undifferentiated mESCs [18]. Therefore, a 50% change in Oct4 expression in H1T cells would be biologically significant. We defined the significantly changed genes as the genes that showed a fold change 
1.5 and p value 
.05. According to these criteria, there were 830 significantly upregulated genes and 1,254 down-regulated genes in H1T cells versus H1 cells. Selected genes differentially expressed and associated with the major developmental signal pathways and/or pluripotency are shown in Table 1. We used real-time PCR to confirm the expression change of some key regulators (Table 1).
The expression of ESC hallmarks, Oct4 and Nanog, showed increases of 1.50- ± 0.16-fold and 1.98- ± 0.19-fold in H1T hESCs versus H1 hESCs, respectively (Table 1).
The Wnt signal was upregulated in H1T cells, as evidenced by the increased expression of a ligand, Wnt3, and decreased expression of two inhibitors of Wnt signaling, SFRP1 and FRZB (Table 1). Downregulation of receptors (FZD2 and FZD8) and upregulation of the inhibitor (WIF) might be due to feedback inhibition [19] (Table 1).
The expression of FGF ligands, bFGF and FGF8, increased in H1T hESCs, whereas the expression of FGF receptors 1, 2, and 3 decreased. SPRY4, an inhibitor of the FGF pathway, was upregulated (Table 1). These indicated that FGF signaling was upregulated and that a feedback inhibition mechanism could be involved.
Nodal was upregulated 3.1- ± 0.48-fold. Lefty-A and Lefty-B are the downstream targets and feedback inhibitors of Nodal signaling [12]. They showed increases in the H1T cell line. Activin A increased 1.51- ± 0.01-fold, and a potential inhibitor of Activin A, Follistatin-like 1, was downregulated in H1T hESCs (Table 1). These data indicated that Nodal/Activin signals were upregulated. Activin A receptor type IIB decreased 1.64- ± 0.13-fold, perhaps once again due to feedback inhibition.
BMP signal was downregulated in H1T hESCs, as evidenced by reduced expression of BMP2, BMP7, and BMP11 (Table 1). Notch signaling was also downregulated, as indicated by the reduced expression of ligands JAG1, DLK1, and DLL1 and downstream effect factor HES1 (Table 1).
We did not detect a significant change in expression of the ligands for the Hedgehog pathway (data not shown). Patched is a receptor and an inhibitor of Hh pathway [20], and its down-regulation of 5.67- ± 1.35-fold suggested that Hh signaling might be upregulated. Gli3 is a transcription factor associated with the Hh pathway, and its expression is repressed in response to Hh signals [20]. The downregulation of Gli3 also suggested that Hh signaling might be upregulated (Table 1).
We also observed downregulation of differentiation markers for all three germ layers, including neurofilament (marker of ectoderm), cardiac muscle actin (mesoderm), and -fetoprotein (marker of endoderm) (supplemental online Table 2). There were no trophoblast genes that showed significant change in the H1T cells versus H1 cells. Combined with the evidence that H1T cells expressed hESC markers (Fig. 1B) and maintain pluripotency (Fig. 1C), these observations indicated the absence of differentiation in H1T cells.
In summary, the gene expression data by microarray analysis suggest that the transcription factors associated with pluripotency (Oct4 and Nanog, and Nodal/Activin, Wnt, FGF, and Hedgehog signaling pathways) were upregulated, whereas BMP and Notch pathways were downregulated in H1T hESCs (Table 1). This observation suggested that Nodal/Activin, Wnt, FGF, and Hedgehog pathways positively regulate the expression of Oct4 and Nanog and contribute to maintaining hESCs in an undifferentiated state, whereas BMP and Notch signaling negatively regulate the expression of Oct4 and Nanog and contribute to differentiation.
The Wnt/ß-catenin pathway has been reported to transiently maintain pluripotency of hESCs [9]. bFGF was shown to increase the cloning efficiency of hESCs and currently is used in hESC medium for routine culture [6]. The BMP pathway signal has been shown to induce hESC differentiation into trophoblasts [11]. The function of Nodal/Activin in hESCs has not been definitely determined. Our observations about differential expression of the gene products associated with Wnt, FGF, and BMP pathways are consistent with the existing literature. We hypothesized that the Nodal/Activin pathway might function to maintain the expression of Oct4, Nanog, and the stemness of hESCs.
Activin A Is Necessary and Sufficient for Self-Renewal and Pluripotency of hESCs
Lefty-A blocks Nodal signaling by a dual mechanism, it binds Nodal directly, and it also binds epidermal growth factor-Cripto-FRL-Cryptic domain co-receptors, such as TDGF-1 (also known as Cripto), thus preventing the assembly of an active Nodal/receptor complex [21]. To investigate the function of Nodal in hESCs, we used Lefty-A to block the function of Nodal in H1 hESCs. When Nodal signaling was blocked by adding Lefty-A in the CM, the expression of Oct4 and Nanog was downregulated in a dose-dependent manner (Fig. 2A). This indicates that Nodal does contribute to maintaining the expression of Oct4 and Nanog.
|
).
Follistatin is an inhibitor of Activin A that functions by directly binding with this protein [22]. Follistatin was able to inhibit the expression of Oct4 and Nanog in a dose-dependent manner under the specific conditions used (Fig. 2B, 2G), and the inhibition was reversed by adding Activin A (Fig. 2B). This indicates that Activin A is necessary to maintain the expression of Oct4 and Nanog. Activin A was able to fully maintain the expression of Oct4 and Nanog in a dose-dependent manner (Fig. 2C, 2F). We also used flow cytometry to detect a phenotypic marker of undifferentiated hESC, Tra-1–60, to evaluate the function of Activin A. We found that Activin A was able to significantly increase the number of Tra-1–60 expressing cells in a dose-dependent manner (supplemental online Fig. 1). The effect of Activin A on Oct4 and Nanog was reversed by adding Follistatin (Fig. 2C), indicating the specific action of Activin A on maintenance of Oct4 and Nanog. Activin A also maintained the expression of Oct4 in another hESC line, I-6 (Fig. 2H, 2I; supplemental online Fig. 3A, 3B).
|
). H1 hESCs cultured in 5 ng/ml Activin A in Matrigel-coated flasks showed an appearance similar to that of the cells growing in CM. After cultured in Activin A for 15 passages, H1 cells were found to be still strongly positive for hESC markers Oct4, Nanog, SSEA3, SSEA4, Tra-1–60, and Tra-1–81 (Fig. 3A–3F). Oct4 and Nanog were stably maintained during prolonged culture in Activin A (Fig. 3G). The markers of differentiation were stably inhibited by activin A (supplemental online Fig. 2). H1 cells maintained in Activin A showed high levels of telomerase activity (Fig. 3H). To examine the potential for in vivo differentiation of the hESCs, we injected the H1 hESCs maintained in Activin A for 10 passages into NOD/SCID mice to test their ability to form teratomas. As cells maintained on feeder cells, cells cultured in Activin A generated complex teratomas with various differentiated tissue types comprising all three germ layers (Fig. 4; data not shown). H1 hESCs have been cultured for more than 150 days and >20 passages in feeder-free conditions supplied with 5 ng/ml Activin A in non-CM. Karyotype analysis by G-banding was carried out. The cells cultured with Activin A demonstrated a normal karyotype (supplemental online Fig. 4).
|
) [6, 7, 9, 11]. We analyzed how Activin A might function in this network in order to further address the mechanism that Activin A maintains the stemness of hESCs. H1 (Fig. 5) or I6 (supplemental online Fig. 3) hESCs were cultured in CM and in non-CM plus different concentrations of Activin A (from 0 ng/ml to 100 ng/ml) for 6 days, and the expression of genes was analyzed by real-time PCR. Upon withdrawal of CM and subsequent growth of the cells in non-CM, the expression of Oct4, Nanog, SFRP1, bFGF, FGF8, Nodal, Lefty-A, and Lefty-B decreased significantly (Fig. 5A, 5B, 5D, 5E–5I), whereas the expression of Wnt3 and Activin A increased significantly (Fig. 5C, 5J). These observations indicate that CM contains an activity or activities to induce the expression of Oct4, Nanog, SFRP1, bFGF, FGF8, Nodal, Lefty-A, and Lefty-B and inhibit the expression of Wnt3 and Activin A. Activin A significantly induced the expression of Oct4 and Nanog in a concentration-dependent manner (Fig. 5A p, < .0001, ANOVA; Fig. 5B p, < .0001, ANOVA). Activin A was able to induce the expression of Oct4 and Nanog beyond that of the CM at a concentration of 10 ng/ml. This is consistent with our observations that 5 ng/ml Activin A is sufficient to completely maintain the expression of Oct4 and Nanog and support long-term feeder-free culture of hESCs (Fig. 3). Activin A induced the expression of Wnt3 in a concentration-dependent manner as well (Fig. 5C p, < .0001, ANOVA). The expression of SFRP1 was significantly repressed by Activin A (Fig. 5D p, < .0001, ANOVA). These observations indicate that Activin A regulates Wnt signaling in hESCs by inducing the expression of a ligand, Wnt3, and repressing the expression of an inhibitor, SFRP1. Activin A induced the expression of bFGF (Fig. 5E p, < .00001, ANOVA). The induction reached a peak at 10 ng/ml of Activin A and was reduced at 100 ng/ml. FGF8 expression responded to the induction by Activin A (Fig. 5F p, < .0001, ANOVA), with the maximal induction being 14.74- ± 1.39-fold. Activin A significantly induced the expression of Nodal (Fig. 5G p, < .0001, ANOVA). Lefty-A (Fig. 5H p, < .0001, ANOVA) and Lefty-B (Fig. 5I p, < .0001, ANOVA) were also highly upregulated in response to Activin A. This may explain reports that levels of Lefty-A and Lefty-B correlate with hESC properties [23–26]. Upon withdrawal of CM, the expression of Activin A itself was upregulated fivefold in non-CM, suggesting that inhibitory factors for Activin A expression could exist in CM (Fig. 5J). Interestingly, Activin A suppressed the expression of itself in a concentration-dependent manner (Fig. 5J p, < .0001, ANOVA). This suggests that one of the inhibitory factors on Activin A expression in CM might be Activin A itself. It is also possible that CM may contain other factors that inhibit the expression of Activin. Activin A did not appear to regulate the expression of BMP2 (Fig. 5K p, > .05, ANOVA). The expression of BMP7 was suppressed significantly by Activin A (Fig. 5L p, < .05, ANOVA). Moreover, Activin A was able to inhibit differentiation caused by BMP4 (Fig. 6A).
|
|
). Therefore, our data suggest a complex regulatory network that maintains the properties of hESCs. In this network, Nodal/Activin, FGF, and Wnt positively contribute to the maintenance of the undifferentiated state of hESCs. The BMP pathway is implicated in negatively regulating the maintenance of the hESC properties. Activin A is able to maintain the expression of Oct4 and Nanog in hESCs, at lower concentrations than bFGF.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
It was reported that an increased dosage of chromosome 17q provides a selective growth advantage for the propagation of undifferentiated hESCs [17]. We provide evidence here that the upregulated expression of Oct4 and Nanog might account for the growth advantage in a chromosomal translocated hESC line (Table 1). Our data confirmed a recent report by Mitalipova et al. that the expression of Oct4 was increased in aneuploid hESCs [33]. These observations imply that selection pressure exists within the current hESC culture system, which favors the aneuploid hESC population, with enhanced proliferative ability conferred by increased expression of Oct4 and Nanog.
It has been reported that a precise level of Oct4 is required to maintain the undifferentiated state of mESCs. The threshold for inducing differentiation is apparently set at 50% above or below the normal diploid expression level in undifferentiated mESCs [18]. We observed that H1T cells express 50% more Oct4 and hold a self-renewal advantage without loss of pluripotency (Table 1; Fig. 1). This observation suggests that the expression level of Oct4 in H1T might function to maintain hESCs properties rather than to induce differentiation. Mouse ESC do not spontaneously differentiate when maintained on mouse embryonic fibroblasts (MEFs) and with LIF. H1 hESC on the other hand do spontaneously differentiate at an approximate rate of 10% in standard conditions (on MEFs and with bFGF). Therefore, this suggests that Oct4 levels in human ESCs would be characteristic of the undifferentiated state at a lower amount. We believe the Oct4 expression in H1T cells favors this undifferentiated condition. Thus, we suggest defining the expression level of Oct4 in H1T cells as 100%. Therefore, the expression level of Oct4 in H1 hESCs maintained on MEF feeder layer is 67%, which is sufficient to maintain the undifferentiated state of hESCs. This also suggests that further modification of the culture system to moderately increase the expression of Oct4 should be considered.
Nodal has been shown to play an important role in early vertebrate development [12, 34]. Although Activin A has activity similar to that of Nodal, the function of Activin A remains controversial. Activin A has been identified as the first mesoderm-inducing factor in the amphibian embryo and was proven able to act as a morphogen in embryo patterning [35, 36]. The role of Activin A in the early embryo has, however, remained unclear because Activin A could not been detected until the late gastrula stage in Xenopus [36]; mice with mutations either in Activin A or both Activin A and Activin B did not show defects at early stages [37]. It is generally believed that although Activin A provides an excellent in vitro model for morphogen action and embryo patterning, it may not be important in early vertebrate development in vivo [38].
Nodal is able to support prolonged feeder-free culture (more than 10 passages) when it is constitutively overexpressed in hESCs [39]; however, Nodal failed to fully maintain the expression of Oct4 and Nanog in hESCs even at a concentration of 1,000 ng/ml (Fig. 2A) or support prolonged feeder-free culture when added to medium [39]. These observations suggest that high levels of Nodal only achieved by gene insertion may be able to completely maintain the undifferentiated state of hESCs. These observations also suggest that the endogenous Nodal may not contribute much to the maintenance of hESCs properties because it is negatively regulated by Lefty-A and Lefty-B, which are highly enriched in hESCs [23–26] (Table 1; Fig. 5H, 5I) and thus might block the Nodal function.
Activin A is expressed both in hESCs and in MEF feeder cells (Table 1; Fig. 5J) [25, 40, 41]. James et al. reported that TGFß/Activin/Nodal signaling is necessary for the maintenance of pluripotency in hESCs but failed to show the factor was sufficient for feeder-free culture [42]. We, unlike James et al. [42], provide data that supports the notion that Activin A is both sufficient and necessary for maintaining the undifferentiated condition. Beattie et al. showed that 50 ng/ml Activin A combined with 50 ng/ml KGF (also called FGF7) and 10 mM nicotinamide is able to maintain the long-term feeder-free culture of hESCs on laminin [40]. They did not try to do feeder-free culture with only Activin A on Matrigel. We demonstrate that Activin A is able to maintain the long-term feeder-free culture of hESCs on Matrigel at 5 ng/ml (Fig. 3). Thus, we demonstrated a 10-fold lower concentration, which was sufficient for Activin A acting alone to support undifferentiated hESC growth. Our data indicate that although Activin A may not play an important role in early vertebrate development in vivo, Activin A, when present in the in vitro culture of hESCs, could be used to culture hESCs without either feeder cells or conditioned medium. The importance of this finding is that hESCs may be cultured free of feeder cell contamination or serum requirements, which is an important step in identifying a defined medium for culture. The system described here should be useful for generating the large number of hESCs necessary for therapeutic and other applications.
We report here that a complex signal network contributes to maintain the stemness of hESCs. The signal network includes Nodal/Activin, FGF, Wnt, Hh, BMP, and Notch. Our data indicate that Activin A is both necessary and sufficient to maintain the stemness of hESCs. Our data also suggests that FGF, Wnt, and Hh signaling positively contribute to maintain the self-renewal of hESCs, whereas BMP and Notch signaling negatively regulate the self-renewal of hESCs.
There are three pathways, Wnt, FGF, and Nodal/Activin, that have been shown to maintain the self-renewal and pluripotency of hESCs. Our data indicate that Activin A is central for three reasons. First, Activin A induces Oct4 and Nanog at lower concentrations than other factors studied. Activation of Wnt signals only maintained 70% of the Oct4 expression and less than 50% Nanog expression in 6 days compared with CM [9]. Therefore, Wnt is only able to maintain the self-renewal of human ES cells partially and temporarily. Dravid et al. showed that the Wnt/ß-catenin activation does not suffice to maintain the undifferentiated and pluripotent state of hESCs [43]. bFGF was not sufficient to fully maintain the Oct4 expression and required an unusually high dose of 40 ng/ml ([27, 28, 30]). Compared with Wnt or FGF, Activin A is able to completely maintain the expression of Oct4 and Nanog at 5 ng/ml. Activin A even can induce the expression of Oct4 and Nanog to a level above that of hESCs maintained in CM (Figs. 2C, 5A, 5B). Second, Activin A controls Wnt and FGF signals by inducing the expression of Wnt3, bFGF, and FGF8 and inhibiting the expression of SFRP-1 in two different ESC lines (Figs. 4, 5). Third, BMP signaling has been shown to promote hESCs to differentiate into a trophoblast [11]. Activin A has been shown to reduce phosphorylation of Smad1/5, which is activated by BMP signaling [42], and is able to inhibit the differentiation effect of BMP signals (Fig. 6A). Activation of Wnt signals could not reduce phosphorylation of Smad1/5 [42]. Noggin, a potent BMP antagonist, is required to facilitate the function of bFGF to maintain the properties of hESCs [28, 29]. In conclusion, we believe that the effect of Activin A on hESCs may be equivalent to the effect of LIF on mESCs to maintain the stemness of embryonic stem cells.
ES cells are derived from inner cell mass. During early mouse embryonic development, inner cell mass cells undergo differentiation to a pluripotent cell population termed the epiblast. The epiblast is characterized by expression of a pluripotency marker such as Oct4 and increased expression of fibroblast growth factor 5 [34, 44]. In mice, both Wnt3 and FGF8 are expressed in the epiblast, and their expression is induced by Nodal. Mouse Nodal-null mutants fail to express Wnt3 and FGF8 in the epiblast and display very low levels of Oct-4 expression [45]. It is interesting that the regulation cascade is conserved between species. In hESCs, our observations indicate that Activin A is able to induce the expression of Oct4, Wnt3, and FGF8 (Fig. 5A, 5C, 5F).
There is little information about the functions of Notch and Hh in hESCs so far. Our data suggest that Notch might function to induce the differentiation of hESCs and Hh might function to maintain the undifferentiated state of hESCs. The investigations of the function of Notch and Hh in hESCs in the future might provide more insight into the mechanism underlying hESC self-renewal and pluripotency.
|
DISCLOSURES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  Z. Wu, W. Zhang, G. Chen, L. Cheng, J. Liao, N. Jia, Y. Gao, H. Dai, J. Yuan, L. Cheng, et al. Combinatorial Signals of Activin/Nodal and Bone Morphogenic Protein Regulate the Early Lineage Segregation of Human Embryonic Stem Cells J. Biol. Chem., September 5, 2008; 283(36): 24991 - 25002. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Yu and J. A. Thomson Pluripotent stem cell lines Genes & Dev., August 1, 2008; 22(15): 1987 - 1997. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Saha, L. Ji, J. J. de Pablo, and S. P. Palecek TGF{beta}/Activin/Nodal Pathway in Inhibition of Human Embryonic Stem Cell Differentiation by Mechanical Strain Biophys. J., May 15, 2008; 94(10): 4123 - 4133. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Mayshar, E. Rom, I. Chumakov, A. Kronman, A. Yayon, and N. Benvenisty Fibroblast Growth Factor 4 and Its Novel Splice Isoform Have Opposing Effects on the Maintenance of Human Embryonic Stem Cell Self-Renewal Stem Cells, March 1, 2008; 26(3): 767 - 774. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. C. Robson, D. J. Phillips, T. McAlpine, A. Shin, S. Svobodova, T. Toy, V. Pillay, N. Kirkpatrick, D. Zanker, K. Wilson, et al. Activin-A: a novel dendritic cell-derived cytokine that potently attenuates CD40 ligand-specific cytokine and chemokine production Blood, March 1, 2008; 111(5): 2733 - 2743. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Naujokat and T. Saric Concise Review: Role and Function of the Ubiquitin-Proteasome System in Mammalian Stem and Progenitor Cells Stem Cells, October 1, 2007; 25(10): 2408 - 2418. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Puceat TGF{beta} in the differentiation of embryonic stem cells Cardiovasc Res, May 1, 2007; 74(2): 256 - 261. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Agrawal, W.-K. Hofmann, N. Tidow, M. Ehrich, D. v. d. Boom, S. Koschmieder, W. E. Berdel, H. Serve, and C. Muller-Tidow The C/EBP{delta} tumor suppressor is silenced by hypermethylation in acute myeloid leukemia Blood, May 1, 2007; 109(9): 3895 - 3905. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
< |
