| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
EMBRYONIC STEM CELLS |
aGenome Instability and Chromatin-Remodeling Section and
cDevelopmental Genomics and Aging Section, Laboratory of Genetics, National Institute on Aging, National Institutes of Health, Baltimore, Maryland, USA;
bCardiovascular Research Center, Massachusetts General Hospital, Richard B. Simches Research Center, Harvard Medical School, Boston, Massachusetts, USA
Key Words. Chromatin remodeling • Gene targeting • SWI/SNF • Embryonic stem cells • DNA microarray • BAF250B • ARID1B
Correspondence: Correspondence: Weidong Wang, Ph.D., Genome Instability and Chromatin-Remodeling Section, Laboratory of Genetics, National Institute on Aging, National Institutes of Health, Baltimore, Maryland, USA. Telephone: 410-558-8334; Fax: 410-558-8331; e-mail: wangw{at}grc.nia.nih.gov or Minoru S. H. Ko, M.D., Ph.D., Developmental Genomics and Aging Section, Laboratory of Genetics, National Institute on Aging, National Institutes of Health, Baltimore, Maryland, USA. Telephone: 410-558-8359; Fax: 410-558-8331; e-mail: kom{at}mail.nih.gov
Received on October 9, 2007;
accepted for publication on February 23, 2008.
First published online in STEM CELLS EXPRESS March 6, 2008.
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
Disclosure of potential conflicts of interest is found at the end of this article.
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
The SWI/SNF complexes are the prototype of ATP-dependent chromatin remodeling machines, which use the energy of ATP hydrolysis to disrupt the interactions between DNA and histones and then to make the nucleosomal DNA more accessible [4]. In mammalian cells, these complexes contain either BRG1 (SMARCA4) or BRM (SMARCA2), which are ATP-dependent motors and can remodel chromatin by themselves. At least three distinct SWI/SNF-like complexes have been identified in mammalian cells (Fig. 1A): a BAF250A (ARID1A)-associated complex (termed BAF-A; BAF refers to BRG1 or BRM-associated factors), a BAF250B (ARID1B)-associated complex (termed BAF-B), and a BAF180 (PBRM1)-associated complex (termed PBAF) [5–9]. These complexes share eight common subunits, including BRG1, BAF170 (SMARCC2), BAF155 (SMARCC1), BAF60A (SMARCD1), BAF57 (SMARCE1), BAF53A (ACTL6A), ACTIN, and hSNF5/INI1 (SMARCB1). They can be distinguished by the presence of several unique subunit(s): BAF250A and BAF250B are unique to their respective BAF complexes, whereas BAF200 (ARID2) and BAF180 are unique to PBAF [7–11]. The available evidence suggests that these unique subunits play crucial roles in targeting their respective complexes to specific genes [9, 10].
|
In this study, we have purified SWI/SNF complexes from mouse ES cells and shown that the levels of several SWI/SNF components are altered when ES cells are induced to differentiate, implying that complexes containing these components could be important in ES cells. To examine this possibility, we established mouse ES cells deficient in one such component, BAF250B. We found that these cells are viable but fail to maintain the undifferentiated state of ES cells. In particular, they have the altered expression of several genes involved in pluripotency and differentiation. Our data suggest that SWI/SNF complexes play an important role in the pluripotency and differentiation of mouse ES cells.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
Antibodies
Antibodies to SWI/SNF components have been described [7, 8, 10, 11, 21]. Anti-BRM (N-19), anti-BAF155, and anti-POU5F1 (OCT4, OCT3/4) antibodies were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, http://www.scbt.com). Anti-ACTIN antibody was purchased from Bethyl Laboratories, Inc. (Montgomery, TX, http://www.bethyl.com/).
Purification of SWI/SNF Complexes, Superose 6 Gel Filtration, and Immunoblotting
The nuclear extract or cell lysates were prepared from the cells as described [10]. The SWI/SNF complexes were immunoisolated with an anti-BRG1 antibody from the extract of HeLa, mouse ES (MC1), ZHBTc4, or trophoblast-like cells, respectively, by immunoprecipitation as described [10, 20]. The Superose 6 gel filtration analysis (GE Healthcare, Piscataway, NJ, http://www.gelifesciences.com) and immunoblotting were performed as described [10].
Immunofluorescence Staining
Cells were grown on gelatin and tissue culture-coated Thermanox plastic coverslips (Apogent Technologies, Portsmouth, NH, http://www.thermofisher.com), fixed in 4% paraformaldehyde for 10 minutes at room temperature, and then washed in 1x PBT (0.1% Tween 20 in phosphate-buffered saline) and preincubated with serum-free blocking reagent (DakoCytomation protein block; DakoCytomation, Glostrup, Denmark, http://www.dakocytomation.com). Cells were incubated with primary antibodies O/N at 4°C and with secondary antibodies for 1 hour at room temperature, nuclei were counterstained with 4,6-diamidino-2-phenylindole for 5 minutes at room temperature. Coverslips were mounted with Gel/Mount (Biomeda Ltd., Foster City, CA, http://biomeda.com/) and analyzed under a fluorescent microscope. Primary antibodies were anti-mouse POU5F1 (Santa Cruz Biotechnology, Inc.) and anti-mouse ACTN2 (Aviva Systems Biology, San Diego, CA, http://www.avivasysbio.com/). Secondary antibodies were anti-mouse IgG Alexa Fluor 488 (Invitrogen, Carlsbad, CA, http://www.invitrogen.com) and anti-rabbit IgG Alexa Fluor 568.
Real-Time Quantitative Reverse Transcriptase-Polymerase Chain Reaction
Total RNA was extracted using TRIzol (Invitrogen) according to the manufacturer's protocol. Primers for quantitative reverse transcriptase(qRT)-polymerase chain reaction (PCR) were designed using Vector NTI Advance 10 software (Invitrogen) (supplemental online Table 4) and tested for SYBR Green chemistry (Applied BioSystems, Foster City, CA, http://www.appliedbiosystems.com) using an established in-house protocol [39]. Reactions were run on ABI 7900 HT Sequence Detection Systems using the default cycling program, and data were processed using SDS 2.2 software (Applied BioSystems).
Generation of Baf250b–/– ES Cell Lines, Southern Blotting, and RT-PCR
An embryonic stem cell line carrying an insertion in Baf250b (XC389) generated by the BayGenomics group (http://baygenomics.ucsf.edu/; [30]) was cultured and injected into blastocysts to generate Baf250b+/– animals. Blastocyst-stage embryos (3.5 days post coitum [dpc]) were obtained from intercrosses of Baf250b+/– animals, plated into six-well plates covered with a mitotically arrested MEF feeder monolayer, and cultured in Dulbecco's modified Eagle's medium, supplemented with nonessential amino acid, glutamine, β-mercaptoethanol, 2x LIF, and 15% knockout serum replacement (Invitrogen). After 10–12 days in culture, the inner cell mass outgrowth was individually removed from the plates, trypsinized in a 96-well plate, and transferred to a 24-well plate covered with MEF monolayer. The ES cells were then transferred to larger well plates when they reaching near confluence and cultured with medium containing 10% fetal bovine serum instead of the knockout serum replacement.
For Southern analysis, a PCR (forward: TGTGTAACACTGTGTAAGTGCCCC; reverse: CAGATCCTTCCTTGAGAGTTTCCC) amplified intron sequence upstream of the intron inserted by the gene-trap vector was used as the template for probe preparation. SacI digestion produced an 8.9-kilobase (kb) fragment for Baf250b wild-type allele and a 
5.0-kb fragment for mutant allele. Southern blotting was performed using QuickHybridization solution (Stratagene, La Jolla, CA, http://www.stratagene.com) according to the manufacturer's protocol.
RT-PCR was performed as described [40]. Mouse Baf250b primers for RT-PCR were forward, 5'-CAAGCAAGTCTCCCTTCCTG-3 and reverse, 5'-TTGGCTGTGATGCTTCTACG-3'.
Cell Cycle Analysis
Cells were plated on gelatin-coated six-well plates at optimal density (104 cells/cm2) and cultured for 3 days in the standard ES medium. On day 3 cells were harvested, fixed in 70% ethanol, and stained with propidium iodine. Cell cycle analysis was performed on an EasyCyte Mini flow cytometry system (Guava Technologies, Hayward, CA, http://www.guavatechnologies.com).
Time Course Experiments
Cells were started on gelatin-coated six-well plates. When enough cells were available, they were separated into three groups: group 1 was replated every day; group 2 was replated on day 2 for two passages, and group 3 was replated on day 3 for two passages. For proliferation assay and gene expression analyses, cells from each group were plated in six-well plates at a density 105/well and cultured in regular medium for 4 days. On days 1, 2, 3, and 4 of culture in each group, cells from three wells were harvested and counted using a hematocytometer. For gene expression analyses, RNA was extracted from three wells in each group using the TRIzol method on days 1, 2, 3, and 4 of culture. Real-time qRT-PCR was performed for Pou5f1, Nanog, Rex1, and Actn2. Data were normalized to Eef1a1 (Mus musculus eukaryotic translation elongation factor 1
1) as an internal control first and then to the level of expression on day 1 in Baf250b+/+ ES cells.
DNA Microarray
We used Baf250b–/– ES cells, two independently derived clones, at 18 and 72 hours in culture and compared them with the parental cell line (Baf250b+/+) at the same time in culture (two replications each). Total RNAs were extracted using TRIzol according to the manufacturer's protocol. Total RNA samples (2.5 µg) were labeled with Cy3-CTP using a Low RNA Input Fluorescent Linear Amplification Kit (Agilent Technologies, Palo Alto, CA, http://www.agilent.com). A reference target (Cy5-CTP-labeled) was prepared from Universal Mouse Reference RNA (Stratagene). Labeled targets were purified using an RNeasy Mini Kit (Qiagen, Hilden, Germany, http://www1.qiagen.com) according to Agilent's protocol, quantitated by a NanoDrop scanning spectrophotometer (NanoDrop Technologies, Wilmington, DE, http://www.nanodrop.com), and hybridized to the NIA Mouse 44K Microarray v2.2 (whole genome 60-mer oligo, #014117; Agilent Technologies) [41] according to Agilent's protocol (G4140-90030; Agilent 60-mer oligo microarray processing protocol-SSC Wash, v1.0). All hybridizations were carried out in the two-color protocol by combining one Cy3-CTP-labeled experimental target and one Cy5-CTP-labeled reference target. Microarrays were scanned on the Agilent DNA Microarray Scanner using standard settings including automatic photomultiplier tube adjustment.
Statistical Analysis of Microarray Data
The data discussed in this article have been deposited in the Gene Expression Omnibus (GEO) at the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/geo/; [42]) and are accessible through GEO Series accession number GSE8996. The data and analysis software is also available at the National Institute on Aging analysis of variance (ANOVA) tool Web site (http://lgsun.grc.nia.nih.gov/ANOVA/) [43]. Because we were interested in genes substantially affected by the disruption of Baf250b, we compared gene expressions in Baf250b–/– ES cells with those in Baf250b+/+ ES cells as well as those obtained from 12 other ES and EG cell lines reported previously [25]. Log-transformed microarray data were analyzed with ANOVA, considering Baf250b–/– ES cell clones as replications (data from biological replications within the same cell line or clone were averaged before the analyses). To reduce the number of false-positive results we used the maximum of the actual error variances for a gene and the average error variance estimated from 500 genes with similar signal intensity [43]. Difference in expression was considered significant on the basis of false discovery rate (FDR) 
0.05 and fold change 
1.5 between Baf250b–/– ES cells and other ES/EG cells. A minimum 1.5-fold difference in gene expressions between Baf250b–/– ES cells and parental cell lines (Baf250b+/+) was used as an additional criterion. Analysis of over-represented gene ontology (GO) terms was carried out using the hypergeometric distribution (FDR 0.05) and enrichment ratio (>1.5) as significance criteria with the NIA Mouse Gene Index (version mm7) software [44].
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
We next isolated SWI/SNF complexes from mouse ES cells by direct immunoprecipitation (IP) with a BRG1 antibody, following a protocol that has been used previously to isolate the most abundant form of the human SWI/SNF complex (i.e., the BAF250A-associated complex) from HeLa cells [20]. The silver-staining SDS gel and IP-coupled Western blot analyses revealed that the complex from ES cells contains most of the SWI/SNF components of HeLa cells, which include BAF250A, BRG1, BAF155, BAF60A, BAF53A, and SNF5 (Fig. 1C, 1D). These results established the fact that SWI/SNF complexes similar to those present in HeLa cells are also present in ES cells.
One notable difference is that the BAF-A complex in ES cells contains only a trace amount of BAF170, but a high-level of BAF155, whereas that in HeLa cells has nearly equal molar amounts of both proteins. It is known that BAF170 and BAF155 are highly homologous to each other [21], and their yeast ortholog (SWI3) forms a homodimer in the yeast SWI/SNF [22]. Because the SWI/SNF complexes are highly conserved in their subunit compositions and catalytic activities across eukaryotic species [4], the mammalian versions of these complexes will most likely resemble their yeast counterparts in consisting of similar dimeric structures. Therefore, it has been suggested that BAF170 and BAF155 may form a heterodimeric structure within human SWI/SNF complexes isolated from a variety of tissues and cell lines (e.g., HeLa, Jurkat T, and YT cells) [6, 21]. Our results showing that SWI/SNF from ES cells had mostly BAF155 but little BAF170 suggest that, unlike HeLa and other cells examined previously, the SWI/SNF from ES cells may contain mostly BAF155 homodimer. This unique feature of ES cells is most likely related to the distinctive differentiation status of these cells but not related to the species differences, because a previous study showed that the SWI/SNF complex purified from a mouse 3T3-derived cell line is similar to that of HeLa cells in terms of relative amounts of BAF170 and BAF155 [6].
Relative Levels of BAF155 and BAF170 in SWI/SNF Complexes Are Different between Undifferentiated and Differentiated ES Cells
The data above raised a possibility that the relative levels of BAF155 and BAF170 in SWI/SNF are regulated during differentiation of ES cells. We tested this possibility directly by inducing the ES cells to undergo differentiation and comparing the SWI/SNF complexes before and after the differentiation. We used mouse ZHBTc4 ES cells, which can be induced to differentiate into trophoblast-like cells by Tet-controlled repression of Pou5f1 (Oct4, Oct3/4) [23]. We chose this differentiation system, because it induces relatively homogeneous differentiation of ES cells [24] compared with other systems such as those induced by adding retinoic acids or removing LIF from the medium [25, 26]. Immunopurification and silver-staining analyses revealed that the SWI/SNF from the undifferentiated ZHBTc4 ES cells had a high level of BAF155 but a low level of BAF170, whereas that from the differentiated trophoblast-like cells had nearly equal molar amounts of both (Fig. 1E). This difference was also confirmed by IP-Western analyses: the level of BAF155 in both the extract and the BRG1-immunoprecpitated SWI/SNF was higher in undifferentiated cells than that in differentiated cells, whereas the level of BAF170 was higher in those from differentiated cells than in those from undifferentiated cells (Fig. 1F). The data indicate a correlation between the differentiation state of ES cells and the relative levels of BAF155 and BAF170 in SWI/SNF: those from the undifferentiated ES cells contain a higher level of BAF155 and a lower level of BAF170, whereas those from the differentiated cells contain nearly equal amounts of both. This finding is consistent with the fact that SWI/SNF complexes purified from differentiated cells such as a mouse 3T3-derived cell line and HeLa cells also contain nearly equal amounts of both proteins [6]. The composition of SWI/SNF thus seems to be regulated during the transition from undifferentiated cells to differentiated cells.
BAF250B Is a Component of SWI/SNF Complex in Mouse ES Cells
The analyses above are limited to the most abundant BAF250A-associated SWI/SNF (Fig. 1A). The other two complexes (BAF-B and PBAF) could not be detected in the BRG1 immunoprecipitation on the silver-stained SDS gel (Fig. 1C), most likely because of their low abundance as reported previously for SWI/SNF in other cell types [11]. To test whether BAF250B is also a component of SWI/SNF in ES cells, we first attempted to isolate BAF250B-associated SWI/SNF by direct immunoprecipitation. However, such attempts have not been successful, because our antibody against human BAF250B did not react well with the mouse protein in the immunopurification. We then performed gel filtration analysis, which revealed that BAF250B in mouse ES cell nuclear extract fractionated in a single peak, corresponding to a complex larger than 1 MDa (Fig. 2A). This feature was indistinguishable from the BAF250B in human HeLa nuclear extract [11], suggesting that mouse BAF250B is also present in a SWI/SNF complex as its human homolog. Furthermore, the peak of mouse BAF250B coincided with those of SWI/SNF components, such as BRG1 and BAF155 (Fig. 2A). We also carried out IP-Western analysis, which demonstrated that BAF250B in ES cell extract could be coimmunoprecipitated by either BRG1 or BAF155 antibody (Fig. 2B). It also showed that the level of BAF250B obtained by BRG1 antibody from ES cells was comparable to that isolated from HeLa cells (Fig. 2B). Taken together, these results established that BAF250B is a component of SWI/SNF in both HeLa and ES cells.
|
|
To further examine this correlation, we analyzed the differentiation of F9 EC cells, which can be induced to become parietal endoderm-like cells by treatment with RA and dbcAMP (Fig. 3C; supplemental online Fig. 1) [27]. Immunoblots showed that the levels of both BAF250A and BAF250B decreased during the differentiation (Fig. 3D), supporting the correlation between the high levels of BAF250A- and BAF250B-associated complexes and the undifferentiated state of cells. Interestingly, of the two SWI/SNF ATPases, BRM increased its levels during the differentiation of both ES and EC cells, whereas the levels of BRG1 remained relatively constant (Fig. 3B, 3D). These data are in agreement with previous findings [28] and suggest that the lower level of BRM is correlated with the undifferentiated state of ES cells.
We also noticed that there were several differences in SWI/SNF regulation between ES and EC cells. For example, BAF170 and BAF155 levels were altered during the differentiation of ES cells (Fig. 3B) but not EC cells (Fig. 3D). In addition, the level of BAF200 was reduced during the differentiation of EC cells but not ES cells. These differences may be related to the biological differences between ES and EC cells.
Taken together, our data show that the compositions of SWI/SNF complexes are dynamically altered as ES and EC cells differentiate. Such alteration could be important to regulate different subsets of genes between undifferentiated and differentiated ES cells.
Derivation of Baf250b–/– ES Cell Lines
The finding that high levels of BAF250A- and BAF250B-associated complexes correlate with the undifferentiated state of ES cells prompted us to ask whether these specific SWI/SNF complexes are indeed required to maintain the undifferentiated state of ES cells. To address this issue, we developed ES cells deficient in BAF250B (supplemental online Fig. 3A). The ES cells inactivated of BAF250A will be described elsewhere [29].
We obtained a mouse ES cell line carrying an insertion in the Baf250b gene from BayGenomics [30]. We injected the cells into blastocysts to generate Baf250b+/– mice. We then intercrossed Baf250b+/– mice and obtained blastocysts at 3.5 dpc. In vitro cultures of these blastocysts as described in Materials and Methods yielded seven ES cell clones. Southern blotting and RT-PCR analyses showed that these clones consisted of two wild-type (Baf250b+/+: clones 3 and 5), two homozygous mutants (Baf250b–/–: clones 4 and 7), and three heterozygous mutants (Baf250b+/–: clones 1, 2, and 6) (supplemental online Fig. 3B, 3C). These numbers (two Baf250b+/+, three Baf250b+/–, and two Baf250b–/–) were comparable to the predicted Mendelian ratio of 1:2:1, suggesting that Baf250b is not essential for the establishment of mouse ES cells.
Consistent with the Southern blotting and RT-PCR analyses, immunoblotting analysis confirmed that Baf250b–/– ES cells showed no detectable level of BAF250B protein (supplemental online Fig. 3D). Moreover, the same analysis also showed that protein levels of other SWI/SNF components were comparable between Baf250b+/+ and Baf250b–/– ES cells, suggesting that the absence of BAF250B has no significant effect on the stability of other SWI/SNF complex components (supplemental online Fig. 3D). This result should be expected, because most of the SWI/SNF components in ES cells are present in the BAF250A-associated complex as shown above, and thus, the absence of BAF250B should not affect the stability of these proteins.
Cell Proliferation and Cell Cycle Progression Are Disturbed in Baf250b–/– ES Cells
Although the initial establishment of Baf250b–/– ES cell lines (or more accurately ES-like cells) was possible, it was difficult to maintain these cell lines under the standard ES culture conditions, in which ES cells are passaged (replated) every 2–3 days to maintain undifferentiated phenotype and pluripotency. Because of the significant reduction in cell numbers with every passage, we were almost losing the Baf250b–/– ES cell lines. We speculated that more frequent passages of cells may promote cell proliferation by providing cells with fresh growth media and reducing the cell-cell contact that often inhibits cell proliferation. Indeed, we found that when cells were replated every 24 hours (1:2 split), they could be propagated for many passages. Therefore, we maintained the Baf250b–/– ES cell lines by replating them every day, except for the time course experiments described below.
Even in this culture condition (i.e., replating every day), the proliferation rate of Baf250b–/– ES cells was slower than that of Baf250b+/+ ES cells (Fig. 4A). Estimated duplication time was 16.7 ± 0.6 hours for Baf250b–/– ES cells and 11.4 ± 0.4 hours for Baf250b+/+ ES cells. It is known that ES cells have a unique cell cycle structure: the majority of the cells are in S phase [31]. In Baf250b–/– ES cells cultured for 3 days without replating, there was enrichment in G2/M phase of the cell cycle at the expense of S phase (Fig. 4B, supplemental online Fig. 4). The difference in the fraction of cells in S phase and G2/M phase of the cell cycle between Baf250b+/+ ES cells and Baf250b–/– ES cells was statistically significant (Fig. 4C). Interestingly, we noticed that Cdc20 was one of the genes whose expression was down-regulated in Baf250b–/– ES cells compared with that in Baf250b+/+ ES cells on the basis of the microarray analysis described below (supplemental online Table 2 at http://lgsun.grc.nia.nih.gov/data/Publications-Supplemental-download.html) and real-time qRT-PCR data (Fig. 4D). Cdc20 is known to play an important role in cell cycle progression: in particular, it is a substrate of anaphase-promoting complex [32, 33]. Downregulation of Cdc20 in Baf250b–/– ES cells may be the cause of their enrichment in G2/M phase of the cell cycle.
|
|
The morphological differentiation was accompanied by reduced expression of pluripotency marker Pou5f1 (Oct4, Oct3/4) in Baf250b–/– ES cells (Fig. 5B), when Pou5f1 was still highly expressed in Baf250b+/+ ES cells (Fig. 5B). The real-time qRT-PCR analysis also confirmed these observations. Although the expression of these pluripotency markers gradually decreased even in the culture of undifferentiated ES cells (Fig. 6; unpublished results), in Baf250b–/– ES cells at all time points, the expression levels of Pou5f1 and Nanog were lower and declined faster compared with those in Baf250b+/+ ES cells (Fig. 6A, 6C). However, Rex1 expression in Baf250b–/– ES cells was comparable to that in Baf250b+/+ ES cells on day 1 but showed faster downregulation afterward (Fig. 6B).
|
Taken together, the data from time course experiments show that by day 3 of culture, the expression of many pluripotency genes is markedly reduced in Baf250b–/– ES cells. Although Baf250b+/+ ES cells also show the decline in the expression of pluripotency genes, Baf250b–/– ES cells show much faster and more significant decreases. These findings perhaps help to explain why Baf250b–/– ES cells have to be maintained by replating them every day: if the cells are replated every 3 days as regular ES cells, the pluripotency-related genes will be sharply downregulated, and these cells will differentiate. These data suggest that Baf250b–/– ES cells are deficient in maintaining an undifferentiated state, which is further supported by global gene expression profiling described below.
Global Gene Expression Profiling of Baf250b–/– ES Cells
To better understand the molecular mechanism by which Baf250b–/– ES cells fail to self-renew, we compared Baf250b–/– ES cells with their parental line (Baf250b+/+) at 18 and 72 hours after passaging (Fig. 7A, Fig. 7B), as well as with 12 other pluripotent cells (ES and embryonic germ [34] cells from C57BL/6 and 129 mouse strains [25]) using whole-genome mouse microarrays (NIA 44k 60-mer oligo microarrays, V2.2). Only a small number of genes (n = 23) were significantly altered in Baf250b–/– ES cells at 18 hours (supplemental online Table 1 at http://lgsun.grc.nia.nih.gov/data/Publications-Supplemental-download.html). Top overexpressed genes included Actn2 (component of calcium pump), Tmprss11a (serine protease), Adh4 (alcohol dehydrogenase), and Gdf9 (oocyte-specific growth factor). Actn2 is known to couple an SK2 channel with a voltage-gated Ca2+ channel [35]. The differential expression of Actn2 was confirmed by immunohistochemistry and qRT-PCR (see above). The difference in gene expression increased at 72 hours (Fig. 7B): 1,103 genes were overexpressed and 1,380 genes were underexpressed in the Baf250b–/– ES cells (supplemental online Table 2 at http://lgsun.grc.nia.nih.gov/data/Publications-Supplemental-download.html).
|
Our analyses also revealed that Baf250b–/– ES cells have increased expression of several differentiation markers, suggesting that they may have initiated the differentiation toward various lineages. For example, the increased level of Gata2 expression on day 3 (Fig. 7D, supplemental Table 2 at http://lgsun.grc.nia.nih.gov/data/Publications-Supplemental-download.html) suggests mesoderm (endothelial) differentiation [36], whereas increased expression of Esx1 (Fig. 7E, supplemental Table 2 at http://lgsun.grc.nia.nih.gov/data/Publications-Supplemental-download.html) together with decreased expression of Foxd3 (Fig. 7C) suggests trophoblast-giant cell differentiation [37, 38]. Taken together, these data indicate that ablation of Baf250b in ES cells led to the defective self-renewal ability and the accelerated differentiation, suggesting that the BAF250B-associated SWI/SNF complex is required to maintain ES cell pluripotency.
GO analyses showed that many other genes unrelated to the pluripotency of ES cells were also altered in Baf250b–/– ES cells (supplemental online Tables 3a, 3b at http://lgsun.grc.nia.nih.gov/data/Publications-Supplemental-download.html). For example, genes overexpressed in Baf250b–/– ES cells were associated with proteolysis (n = 41), lysosomes (n = 16), glycolysis (n = 13), and lipid biosynthesis (n = 15), whereas genes underexpressed in Baf250b–/– ES cells were associated with the regulation of transcription (n = 146), cell cycle (n = 56), protein biosynthesis (n = 52), nucleic acid unwinding (n = 33), and RNA splicing (n = 39). Baf250b–/– ES cells also showed the increased expression of genes associated with calcium signaling (Actn2, Cab39l, Calca, Calm3, Calr4, Camk2n1, Clca3, Hrc, S100a1, S100a13, and Tram2) and gonad functions (Aym1, Clmn, Cubn, Cyct, Cylc1, Dmrt1, Dppa3, Esx1, Ephx2, Figla, Kit, Mospd1, Mospd2, Obox6, Odf2, Ott, Rhox2, Sat1, Sat2, Sohlh2, Spats2, Strbp, and Sycp3), as well as the decreased expression in genes associated with transcription, mRNA splicing, and protein biosynthesis. It remains to be determined whether BAF250B-associated SWI/SNF directly regulates these genes.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
We found an interesting difference between the BAF-A complexes isolated from ES and HeLa cells: the former have a high level of BAF155 but a low level of BAF170, whereas the latter have near equal amounts of both (Fig. 1C, 1D). This difference appears to correlate with the differentiation status of the cells, because when ES cells are induced to differentiate, complexes containing BAF155 are decreased, whereas those containing both proteins are increased (Fig. 1E, 1F). These results imply that the functions of BAF155 and BAF170 could be different in ES cells, even though these two proteins are highly homologous to each other. Perhaps BAF155 may be more important in undifferentiated ES cells, whereas BAF170 is more important in differentiated cells.
In addition to BAF155 and BAF170, we detected alterations in the protein levels of several other SWI/SNF subunits during differentiation of ES cells. These subunits include BAF250A and BAF250B, whose levels are high in undifferentiated ES cells but low in differentiated cells (Fig. 3). The results predict that the BAF250A and BAF250B-associated SWI/SNF complexes may be one class of the essential remodelers that regulate chromatin structure in ES cells: they may be required for activation of self-renewal-related genes or for repression of differentiation-related genes, or both. To investigate this possibility, we generated mouse ES cells deficient in BAF250B. The data showed that Baf250b–/– ES cells indeed exhibited defective self-renewal capacity and initiation of differentiation, indicating an essential role of the BAF250B-associated SWI/SNF complex in the maintenance of ES cell pluripotency. Similarly, Zhong et al. [29] have demonstrated that the BAF250A-associated SWI/SNF is also required to maintain self-renewal ability of ES cells. Moreover, a recent study showed that a mouse neural-specific SWI/SNF-like complex (npBAF) purified from newborn mouse brains is required for neural stem cell self-renewal [15]. Together, these data suggest that SWI/SNF complexes are essential in governing both embryonic and adult stem cell fates.
How does BAF250B-associated SWI/SNF function to maintain the undifferentiated state of ES cells? Microarray, immunofluorescence staining, and real-time qRT-PCR analyses of Baf250b–/– ES cells showed the reduced expression of several genes related to pluripotency and self-renewal when these cells were cultured for 3 days without replating. These genes include Pou5f1,Bmp4, Klf3, Myc, and Foxd3. Among them, Pou5f1 is one of the major pluripotency genes, and its appropriate level is required to maintain ES cell self-renewal: either repression or overexpression of Pou5f1 can induce ES cell differentiation [23]. It is, therefore, possible that BAF250B-associated SWI/SNF is required for transcriptional activation of these genes. On the other hand, the same analyses revealed increased expression of several differentiation markers in Baf250b–/– ES cells. For example, Gata2 and Esx1 genes were overexpressed, which suggests that Baf250b–/– ES cells may begin to differentiate toward mesoderm (especially, endothelial) [36] and trophoblast giant cells [37, 38]. Together, our data suggest that the BAF250B-associated SWI/SNF complex may play an important role in either activating pluripotency-related genes or repressing differentiation inducers in ES cells or both.
|
DISCLOSURE OF POTENTIAL CONFLICTS OF INTEREST |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosure of Potential...AcknowledgmentsReferences |
|---|
-actinin2. Circ Res 2007;100:112–120.This article has been cited by other articles:
|
|
 |
|
  L. Ho, J. L. Ronan, J. Wu, B. T. Staahl, L. Chen, A. Kuo, J. Lessard, A. I. Nesvizhskii, J. Ranish, and G. R. Crabtree An embryonic stem cell chromatin remodeling complex, esBAF, is essential for embryonic stem cell self-renewal and pluripotency PNAS, March 31, 2009; 106(13): 5181 - 5186. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Ho, R. Jothi, J. L. Ronan, K. Cui, K. Zhao, and G. R. Crabtree An embryonic stem cell chromatin remodeling complex, esBAF, is an essential component of the core pluripotency transcriptional network PNAS, March 31, 2009; 106(13): 5187 - 5191. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Aiba, T. Nedorezov, Y. Piao, A. Nishiyama, R. Matoba, L. V. Sharova, A. A. Sharov, S. Yamanaka, H. Niwa, and M. S. H. Ko Defining Developmental Potency and Cell Lineage Trajectories by Expression Profiling of Differentiating Mouse Embryonic Stem Cells DNA Res, February 1, 2009; 16(1): 73 - 80. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. L. Kidder, S. Palmer, and J. G. Knott SWI/SNF-Brg1 Regulates Self-Renewal and Occupies Core Pluripotency-Related Genes in Embryonic Stem Cells Stem Cells, February 1, 2009; 27(2): 317 - 328. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. S. McKenna, C. G. Sansam, Y.-J. Cho, H. Greulich, J. A. Evans, C. S. Thom, L. A. Moreau, J. A. Biegel, S. L. Pomeroy, and C. W. M. Roberts Loss of the Epigenetic Tumor Suppressor SNF5 Leads to Cancer without Genomic Instability Mol. Cell. Biol., October 15, 2008; 28(20): 6223 - 6233. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Carrera, J. Zavadil, and J. E. Treisman Two Subunits Specific to the PBAP Chromatin Remodeling Complex Have Distinct and Redundant Functions during Drosophila Development Mol. Cell. Biol., September 1, 2008; 28(17): 5238 - 5250. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| STEM CELLS | THE ONCOLOGIST | CME | ALPHAMED PRESS JOURNALS |
