| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
OPEN ACCESS ARTICLE
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
EMBRYONIC STEM CELLS |
cInstitute of Molecular Medicine and Genetics, Medical College of Georgia, Augusta, Georgia, USA;
bBresaGen Inc., Athens, Georgia, USA;
aDepartment of Genetics, University of Georgia, Athens, Georgia, USA
Key Words. Differentiation • 
-Secretase • Human embryonic stem cell • Human development • NOTCH intracellular domain • HES
Correspondence: Brian G. Condie, Ph.D.,Department of Genetics, University of Georgia, Davison Life Sciences Complex, Athens, Georgia 30602, USA. Telephone: 706-542-1431; Fax: 706-583-0691; email: bcondie{at}uga.edu
Received on July 12, 2005;
accepted for publication on March 21, 2006.
First published online in STEM CELLS EXPRESS April 13, 2006.
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
-secretase inhibitors. Our results suggest that NOTCH signaling is not required for the propagation of undifferentiated human ES cells but instead is required for the maintenance of the differentiating cell types that accumulate in human ES cell cultures. Our studies suggest that NOTCH signaling is not required in human embryonic differentiation until the formation of extraembryonic, germ layer, or tissue-specific stem cells and progenitors.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
Another aspect of understanding hESC growth and differentiation is defining the signaling pathways required for the formation of differentiated cell types. As ES cells differentiate to form the various stem cells of the developing embryo, additional pathways supplant or interact with those active in undifferentiated cells. Differences in signaling pathways between undifferentiated ES cells and neural precursors have been exploited to remove undifferentiated cells from ES-derived neural cell cultures, reducing the teratoma-forming ability of the highly heterogeneous mixture of cells in embryoid bodies [8]. In addition, hESC cultures can accumulate a population of differentiated cells over time, particularly in cultures maintained by cell dispersal [9]. Identification of the signaling pathways involved in generating and maintaining the differentiated cell types is a key to controlling differentiation to form specific cell types.
The NOTCH signaling pathway directs a multitude of functions during development [10–12]. Because previous work has shown that the mRNAs encoding NOTCH1–3 and DLL1 were present in human ES cells [13], we tested the role of NOTCH signaling in hESCs and differentiating cells derived from hESCs. Our results indicate that NOTCH can be activated in hESCs by transiently exposing the cells to cation chelators (trypsin-ethylenediaminetetraacetic acid [EDTA]). This result suggests that NOTCH signaling is activated in human ES cells by cell passaging conditions that include cation chelation. Although NOTCH signaling is inducible in hESCs, we find that it is normally inactive in undifferentiated cells. However, inhibition of NOTCH signaling with a -secretase inhibitor markedly reduces the number of differentiating cells in human ES cultures, suggesting that the NOTCH pathway is only required for the maintenance cell types. Our results suggest that NOTCH signaling is not required for the propagation of undifferentiated human ES cells but instead is required for the maintenance of many of the differentiating cell types that accumulate in human ES cell cultures. We also found that NOTCH activity appears to dramatically increase in differentiated cells derived from the hESCs. The level of nuclear NOTCH intracellular domain (NICD) protein is greatly upregulated in the differentiated cells. Our studies suggest that NOTCH signaling is not required in human embryonic differentiation until the formation of extraembryonic, germ layer, or tissue-specific stem cells and progenitors.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
-Secretase Inhibition-secretase experiments, collagenase type IV plus trypsin-EDTA was used to expand the hESCs for no more than 10 passages before stage-specific embryonic antigen (SSEA) 4 selection. Cells were treated with 1 mg/ml collagenase (Gibco, Grand Island, NY, http://www.invitrogen.com) in Dulbecco’s modified Eagle’s medium (DMEM), 20% Knockout Serum Replacement (KSR), 1% minimal essential medium nonessential amino acids, 1 mM L-glutamine, penicillin (0.5 U/ml), streptomycin (0.5 U/ml), and 0.1 mM ß-mercaptoethanol for 3–5 minutes. The collagenase was removed, and 0.05% trypsin-EDTA was added for 45 seconds. Trypsin was neutralized with DMEM containing 10% fetal bovine serum and 10% KSR, and cells were washed in growth medium containing 8 ng/ml basic FGF and plated at 20,000 cells per cm2 on MEFs. Cells were enriched by magnetic cell sorting (MACS) as described previously [14]. Cells were incubated in 5 µM N-[N-(3,5-difluorophenacetyl)-L-alanyl]-S-phenylglycine t-butyl ester (DAPT) in 0.5% dimethyl sulfoxide (DMSO), or 0.5% DMSO alone as a control to test the effect of -secretase inhibition. DAPT was purchased from Calbiochem (San Diego, http://www.emdbiosciences.com).
There is a great deal known about the specificity and molecular mechanism of various -secretase inhibitors, including DAPT. This is due to their potential utility as drugs for the treatment of Alzheimer’s disease. In the case of DAPT, it has been shown that it inhibits -secretase activity by binding to presenilin 1 [15, 16]. DAPT either binds to the active site directly or binds nearby [15–17]. It is a potent NOTCH inhibitor with an IC50 of 10 nM in a cell-free microsome-based assay using a NOTCH reporter as substrate [17]. In mammalian cells, inhibition of NOTCH cleavage is achieved at 10–40 nM [17]. In terms of specificity, DAPT treatment has been shown to produce phenocopies of Drosophila and zebrafish Notch pathway mutations [17–19]. In addition, it has been used as a specific inhibitor of -secretase in numerous studies [20–23]. Although presenilin is also involved in modulating the WNT signaling pathway, this function appears to be independent of its -secretase activity [24, 25].
Immunofluoresence
To detect cell surface antigens, cells were fixed in 4% paraformaldehyde (PFA), 4% sucrose in phosphate-buffered saline (PBS), pH 7.4, and blocked in 3% goat serum, 1% polyvinylpyrollidone in PBS. SSEA4 (1:100; Chemicon, Temecula, CA, http://www.chemicon.com) and NOTCH1 (H-134, 1:10; Santa Cruz Biotechnology Inc., Santa Cruz, CA, http://www.scbt.com) antibodies were added and incubated 2 hours to overnight. Alexa488 anti-mouse antibodies or Alexa594 anti-rabbit secondary antibodies (Molecular Probes Inc., Eugene, OR, http://probes.invitrogen.com) were incubated for 30 minutes to 1 hour at room temperature. Cell nuclei were stained with DAPI. All washes were in PBS. Images were captured with a Nikon TE-2000E microscope (Nikon) with a Retiga 1300 cooled CCD camera and Qcapture software (Qimaging, Burnaby, BC, Canada, http://www.qimaging.com).
To detect the NOTCH1 NICD in differentiated cells, BG01 hESCs were passaged by microdissection to MEFs as above and grown for 5 days under nondifferentiating conditions. The samples were processed for immunocytochemistry detection of NICD using the cleaved NOTCH1 (Val1744) antibody (Cell Signaling Technology, Beverly, MA, http://www.cellsignal.com) according to the manufacturer’s protocol. The cells were fixed in ice-cold methanol, blocked, and exposed to a 1:500 dilution of primary antibody overnight at 4°C. The cells were washed extensively, and primary antibody was detected with a 1:500 dilution of goat anti-rabbit horseradish peroxidase (HRP)-conjugated secondary antibody (Molecular Probes, Inc.) for 2 hours at room temperature. The HRP reaction was developed with the DAB kit (Vector Labs, Inc., Burlingame, CA, http://www.vectorlabs.com) according to the manufacturer’s instructions. Reactions omitting the primary antibody were used as controls.
RNA Analysis
Total RNA was prepared with Trizol (Invitrogen, Carlsbad, CA, http://www.invitrogen.com) and treated with DNAse I. Reverse transcription was performed with oligo(dT) and SuperScript II (Invitrogen). Real-time reverse transcription-polymerase chain reaction (RT-PCR) was used to measure HES1 transcripts. The cDNA was generated from triplicate samples from three independent experiments. TaqMan PCR Master Mix (Applied BioSystems, Foster City, CA, http://www.appliedbiosystems.com) was used in 25-µl reactions with primers at a final concentration of 500 nM and the specific probe at a final concentration of 200 nM. Parallel amplifications of glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA were used for normalization. GAPDH primers were purchased from Applied Biosystems. The primers/probe for HES1 were as follows:
Forward, 5'-CTACCCCAGCCAGTGTCAAC-3'.
Reverse, 3'-TCAGCTGGCTCAGACTTTCA-3'.
Probe, 5'-CGACACCGGATAAACCAAAGACAGC-3'.
Real-time RT-PCRs were run on an ABI7700 Sequence Detection System (Applied Biosystems, Inc.). Data were analyzed with REST-XL (version 2) software to determine significant differences in relative expression.
Flow Cytometry
Cells were washed once in PBS and fixed in 2% PFA for 20 minutes at room temperature. The cells were blocked in 3% goat serum in PBS. SSEA4 (1:100; Chemicon) and NOTCH1 (H-134, 1:10; Santa Cruz Biotechnology) antibodies were added in blocking solution and incubated 2 hours to overnight at 4°C. Alexa488-conjugated anti-mouse antibodies or PE-conjugated anti-rabbit antibodies (Molecular Probes) were used for detection. In some cases, cells were postfixed in 0.5% PFA and washed in PBS, 0.05% azide. Cells were analyzed on a Cytomation CyAn flow cytometer equipped with UV, 488-nm, and 633-nm laser excitation sources. Single stains, isotype controls for directly conjugated primary antibodies, and controls in which primary antibody was omitted in indirect staining protocols were all used for analog and postacquisition compensation. In most cases, 10,000 to 50,000 events were collected. Analysis was performed with Summit analysis or FlowJo software.
Western Blotting
For blotting, equal amounts of protein were resolved on precast gels (Gradipore), transferred to nylon membranes, and blocked in Tris-buffered saline, 0.1% Tween 20, and 5% milk. Antibodies were as follows: 1:10 bTAN20 (NOTCH1, DSHB) and C651.6bHN (NOTCH2, DSHB); 1:100 anti-Oct4 (C-10; Santa Cruz Biotechnology); 1:1,000 anti-presenilin and nicastrin (Chemicon), cleaved NOTCH1 Val1744 antibody (Cell Signaling), and 0.5 µg/ml anti-histone deacetylase (Zymed). Membranes were incubated with primary antibodies overnight and washed in Tris-buffered saline, 0.1% Tween 20. Secondary antibodies were diluted in blocking buffer and incubated with the membranes for 45 minutes at room temperature. Washes were in Tris-buffered saline. Proteins were detected by enhanced chemiluminescence (GE Healthcare Life Sciences, Piscataway, NJ, http://www.amersham.com).
Luciferase Assays
To detect NOTCH activity, reporter plasmids were transfected into undifferentiated hESCs or hESCs differentiated by withdrawal of conditioned medium. The wild-type (4xwtCBF1Luc) and mutant (4xmtCBF1luc) reporters [26] were cotransfected with the control plasmid pRL-CMV (Promega). In some cases, a plasmid expressing NICD (GmN1-ICV-wt, from Dr. R. Kopan, Molecular Biology and Pharmacology, Washington University, St. Louis) was included as a positive control. The hESCs were grown in feeder-free conditions [27], plated on Matrigel-coated 12-well plates, and grown for 2 days prior to transfection. For each transfection, 500 ng of reporter in a total of 1.6 µg of DNA was mixed with 4 µl of Lipofectamine 2000 (Invitrogen) in 200 µl of OptiMEM. This was added to the hESC medium on the cells and incubated overnight. The medium was changed the next morning with MEF-conditioned medium replaced with nonconditioned medium to induce differentiation in some samples. Cells were harvested after 48 or 72 hours for luciferase assay. Transfection efficiency varied with different cell lines. In control transfections, we found that 5%–10% of the cells in culture of H1 hESCs and approximately 15% of the cells in BG01 cultures were transfected.
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
-secretase activity, are also expressed in hESCs and are present in their active forms (Fig. 1C). An antibody that recognizes the loop of the C-terminal fragment of Presenilin-1 shows that the processed and active 20-kDa C-terminal fragment form is much more abundant than the uncleaved and inactive form in hESCs (Fig. 1C). Most of the nicastrin protein is present in its larger glycosylated and active form (Fig. 1C). Our results show that NOTCH receptor proteins, as well as key components required for the cleavage of NOTCH in response to ligand, are expressed in hESCs.
|
-secretase-cleaved NOTCH1 only in cells exposed to EDTA (Fig. 2A). This shows that NOTCH cleavage can be activated in hESCs when passaged by trypsin-EDTA. Treatment of the hESCs with a potent and specific -secretase inhibitor prior to disrupting the cells with trypsin-EDTA blocked the production of the NICD fragment in the cells (Fig. 2B), indicating that the NOTCH activation resulted from intramembrane cleavage of NOTCH1 by -secretase. Thus, an active -secretase complex is present in the human ES cells and can generate NICD upon treatment with EDTA.
|
-secretase inhibitor DAPT (Fig. 2C). EDTA exposure resulted in a fourfold induction of the endogenous HES1 gene in the hESCs (Fig. 2C; p < .001). This induction was reduced by half by the addition of DAPT (Fig. 2C; p < .05). Thus, blocking -secretase blocks the induced upregulation of a NOTCH target gene. We also find that -secretase inhibition does not reduce the expression of HES1 in cells not treated with EDTA (Fig. 2C), suggesting that NOTCH signaling via HES1 is not active in undifferentiated hESCs. The ability of EDTA to activate NOTCH signaling and induce the expression of the HES1 indicates that the NOTCH pathway can be activated by extracellular cation chelation in hESC cultures. To test whether NOTCH signaling is constitutively active in undifferentiated hESCs, we transiently transfected luciferase reporter vectors into BGN1 and H1. To avoid activating NOTCH, we used collagenase to passage the cells without exposure to EDTA. Two reporter vectors were used: one contains four copies of the consensus CBF1 binding site and is upregulated in response to NOTCH signaling, whereas the other is a control vector containing four mutated CBF1 binding sites [26]. We found that the wild-type and mutant CBF1-luciferase reporters were expressed at the same level after 2 days in the presence or absence of conditioned medium; however, transfection of the Notch NICD resulted in robust activation of the reporter (Fig. 2D). Control transfections confirmed that the NICD specifically activates the wild-type CBF reporter (Fig. 2E). The results show that although the pathway can be activated to induce NOTCH-specific gene expression, the canonical NOTCH signaling pathway is not active in hESCs within the limits of detection of our assays. Our results suggest that the NOTCH pathway in hESCs is normally quiescent but can be rapidly activated.
Inhibition of -Secretase Reduces the Proportion of Differentiating Cells in hESC Cultures
Our analysis indicates that NOTCH signaling can be activated in hESCs but normally remains inactive in undifferentiated cells. This suggests that NOTCH may not be involved in the maintenance of undifferentiated hESCs but may be required at some later point in differentiation. We tested this by inhibiting NOTCH signaling while propagating hESCs in conditions known to cause the accumulation of differentiated cells within the cultures. If NOTCH is required for the maintenance of undifferentiated hESCs, its inhibition by DAPT would lead to the rapid accumulation of SSEA4(–) cells and to changes in cell morphology. If it is required for the formation or maintenance of differentiated cell types within the cultures, then NOTCH inhibition would prevent or eliminate the accumulation of the differentiating SSEA4(–) cell population.
To test the role of NOTCH in hESCs, we inhibited NOTCH cleavage by incubating cultures in the -secretase inhibitor DAPT and measured the proportion of cells expressing NOTCH1 and SSEA4. To start the experiment with a relatively uniform cell population, we enriched for cells expressing high levels of SSEA4 by MACS sorting. The retained fraction was plated and subjected to the treatment conditions. Incubation with DAPT led to a statistically significant 27% reduction in total cell number relative to DMSO-treated controls (p = .0078; n = 8; Wilcoxon signed rank test). This reduction is due to a loss of cells expressing low or undetectable levels of SSEA-4 (Fig. 3). To determine the distribution of SSEA4 and NOTCH expression within the -secretase inhibitor-treated cultures and controls, we measured the number of cells expressing NOTCH1 and/or SSEA4 by flow cytometry (Fig. 3A). The majority of cells in the parent culture (Fig. 3A, top left corner plot) express high levels of SSEA4 and NOTCH1 (FrA, 67.8%) indicative of the undifferentiated state of hESCs. A small portion expresses low levels of SSEA4 while expressing moderate levels of NOTCH1 (FrB, 28%). With DAPT treatment (Fig. 3A), the proportion of cells in the culture expressing low levels of SSEA4 is decreased, and the proportion expressing high levels of SSEA4 is increased compared with parent cultures and DMSO-treated or untreated cultures. In addition, compared with the parent culture, both lack of treatment and DMSO treatments decreased the proportion of SSEA4+ cells. Summary data for four independent experiments with two different hESC lines (BGN1 and BGN2) are shown as graphs in Figure 3B. An average of 30% of cells were in SSEA4low/NOTCH1+ fraction in the DMSO-treated cultures, whereas this was reduced by half with inhibitor treatment (p = .0286; n = 4; Mann-Whitney test). Therefore, inhibition of -secretase significantly reduces the number and proportion of differentiating cells in hESC cultures.
|
-secretase-treated cultures suggests that Notch signaling is active in the cells that have differentiated from the hESCs. To test this, we examined the expression of the NICD within spontaneously differentiating hESC cultures. In this case, we used cells that had been maintained exclusively by manual microdissection passaging. In manually passaged cultures, colony fragments sometimes do not attach immediately after being transferred to a new feeder layer. The fragments that are delayed in attachment form heterogeneous colonies containing undifferentiated and differentiated cell types. The differentiating cells within these colonies can be readily recognized by their columnar epithelial morphology. As the cells differentiate, they form tightly packed cell aggregates (Fig. 4B, 4C, black arrows). The tightly packed aggregates of differentiated cells (Fig. 4B, 4C, black arrows) clearly have a different appearance from the epithelial morphology of the undifferentiated cells in these cultures (Fig. 4C). After 5 days of differentiation, we found strikingly high levels of nuclear NICD staining in the cells within the differentiated cell aggregates of the partially differentiated hESC colonies (Fig. 4B, 4C). In undifferentiated cells, we could easily detect Notch receptor expression (Fig. 1A), as well as Oct4 expression (Fig. 4D), but we did not detect NICD (Fig. 4B, 4C). The presence of high levels of nuclear NICD in differentiated cells suggests that Notch signaling becomes active in the differentiating cells derived from hESCs.
|
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
-secretase activity are also present in their processed and active forms. Furthermore, transcription of HES1, a NOTCH target, is induced in hESCs after NOTCH activation. Our analysis shows that NOTCH signaling is normally inactive in undifferentiated hESCs. We found that NOTCH signaling appears to be greatly upregulated as the cells differentiate because the differentiating cells have high levels of NICD present in their nuclei. This observation is consistent with our finding that incubation of hESC cultures in a -secretase inhibitor results in a reduction in the proportion of differentiated cells in the cultures. The results are consistent with a role for NOTCH signaling in the maintenance of differentiating cell types in hESC cultures. The requirement for NOTCH signaling for the survival of differentiated cells in hESC cultures is consistent with the known requirement for NOTCH in the formation and maintenance of differentiating cell types in mouse embryos. Although the mRNAs for Notch pathway components are present in mouse ES cells and in mouse embryos during preimplantation and pregastrula stages [31, 32], Notch signaling is not required in mouse ES cells or during the early stages of mouse embryogenesis [33–35]. A requirement for Notch is first detected at the earliest stages of organogenesis at E8.5–9.5 in Pofut1, CBF1, or presenilin (PS) mutant mice [35, 36]. Detailed analysis of the developing central nervous system in PS1–/– and PS1–/–;PS2+/– mice indicates a greatly reduced self-renewing neural stem cell population, suggesting that Notch is required for the maintenance of neural stem cell populations [36]. The results of genetic analysis in mouse ES cells and mice suggest that Notch signaling components are present but inactive prior to the emergence of tissue-specific stem cell populations during and after gastrulation. Our results suggest that the timing of NOTCH function during human embryonic differentiation is similar to that found in the mouse system.
From a practical standpoint, the accumulation of differentiating cells in hESC cultures interferes with the maintenance of undifferentiated cells and with controlled differentiation. The cell types that can accumulate in hESC cultures are a highly heterogeneous population [9]. The extracellular factors produced by such cells can stimulate the differentiation of the ES cells, leading to increasing heterogeneity within the cultures and the eventual loss of the hESCs. The ability of accumulating differentiated cells to induce the differentiation of the remaining undifferentiated ES cells has been shown previously for mouse embryonic stem cells [37, 38]. Our results indicate that -secretase inhibitors reduce the accumulation of differentiating cells within the hESC and may be extremely useful for the maintenance of undifferentiated hESC lines.
It is important to note that our data also show that NOTCH signaling can be activated in hESCs by reagents used for dissociating cells. We and others have noted that differentiated cells, as well as aneuploid undifferentiated hESCs, accumulate in cultures that are maintained over many cell passages (more than 10) using complete cell dissociation with EDTA-containing solutions [9, 39, 40]. As a result, manual microdissection passaging has been proposed as a method for the maintenance of undifferentiated hESC cultures. We suggest that the repeated activation of NOTCH signaling by cation chelation during cell dissociation may provide a transient survival signal to differentiated cell types. Further studies of NOTCH signaling, as well as the activity of other pathways in hESCs cultured in a variety of conditions, may lead to the development of methods that support routine cell dissociation passaging of most if not all hESC lines. This is a highly desirable goal because manual microdissection passaging is a barrier to the production of the cell numbers required for routine experimentation. A greater understanding of the signals that control the self-renewal, differentiation, and proliferation of hESCs should lead to culture systems that can generate the large numbers of normal and undifferentiated cells required for routine experimentation and cell therapies.
|
DISCLOSURES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  J. Jang, S. Y. Ku, J. E. Kim, K. Choi, Y. Y. Kim, H. S. Kim, S. K. Oh, E. J. Lee, H.-J. Cho, Y. H. Song, et al. Notch Inhibition Promotes Human Embryonic Stem Cell-Derived Cardiac Mesoderm Differentiation Stem Cells, November 1, 2008; 26(11): 2782 - 2790. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Fox, P. J. Gokhale, J. R. Walsh, M. Matin, M. Jones, and P. W. Andrews Cell-Cell Signaling Through NOTCH Regulates Human Embryonic Stem Cell Proliferation Stem Cells, March 1, 2008; 26(3): 715 - 723. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| STEM CELLS | THE ONCOLOGIST | CME | ALPHAMED PRESS JOURNALS |
