| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Research Article |
Development and Differentiation Laboratory, Developmental Biology Department, Insect and Animal Sciences Division, National Institute of Agrobiological Sciences, Ibaraki, 305-8602, Japan
ABSTRACT
Wepreviously demonstrated that mouse embryonic stem (ES) cells show a wide variation in the expression of platelet endothelial cell adhesion molecule 1 (PECAM1) and that the level of expression is positively correlated with the pluripotency of ES cells. We also found that PECAM1-positive ES cells could be divided into two subpopulations according to the expression of stage-specific embryonic antigen (SSEA)-1. ES cells that showed both PECAM1 and SSEA-1 predominantly differentiated into epiblast after the blastocyst stage. In the present study, we performed pairwise oligo microarray analysis to characterize gene expression profiles in PECAM1-positive and -negative subpopulations of ES cells. The microarray analysis identified 2034 genes with a more than 2-fold difference in expression levels between the PECAM1-positive and -negative cells. Of these genes, 803 were more highly expressed in PECAM1-positive cells and 1231 were more highly expressed in PECAM1-negative cells. As expected, genes known to function in ES cells, such asPou5f1(Oct3/4)andNanog, were found to be upregulated in PECAM1-positive cells. We also isolated 23 previously uncharacterized genes. A comparison of gene expression profiles in PECAM1-positive cells that were either positive or negative for SSEA-1 expression identified only 53 genes that showed a more than 2-fold greater difference in expression levels between these subpopulations. However, many genes that are under epigenetic regulation, such as globins,Igf2,Igf2r, andH19, showed differential expression. Our results suggest that in addition to differences in gene expression profiles, epigenetic status was altered in the three cell subpopulations.
embryo, developmental biology, early development, gene regulation, embryonic stem cells, microarray, epigenetics
INTRODUCTION
Embryonic stem (ES) cells are derived from the inner cell mass of blastocysts. ES cells can be propagated and maintained as pluripotent cells for long periods under suitable culture conditions. Inevitably, however, spontaneous differentiation occurs in some cells in culture, with the consequence that cultures contain cells at different states of differentiation in addition to pluripotent cells. We have attempted to identify cell surface markers that would enable enrichment of the genuine pluripotent cell population. Two cell surface markers, platelet endothelial cell adhesion molecule 1 (PECAM1) and stage-specific embryonic antigen (SSEA)-1, enabled the separation of mouse ES cells into three cell subpopulations: PECAM1–SSEA-1–, PECAM1+SSEA-1–, and PECAM1+SSEA-1+ [1]. Quantitative reverse transcription-polymerase chain reaction (quantitative RT-PCR) revealed a low level of Oct3/4 mRNA expression and an elevation in differentiation marker gene expression in PECAM1-negative cells. To compare the pluripotency of these subpopulations, ES cells carrying the lacZ gene were sorted into three subpopulations, and a single cell from each was injected into eight-cell embryos. Mapping analysis of ES-derived cells by X-gal staining revealed that PECAM1+SSEA-1+/- cells were found to have differentiated into epiblast cells in high numbers at the blastocyst stage. In contrast, PECAM1-negative cell derivatives localized in the primitive endoderm or trophectoderm. At 6.0–7.0 days post coitum (dpc), many PECAM1+SSEA-1+ cells were found in the epiblast, but few ß-gal+ cells were detected in any regions of embryos that were injected with cells from the other two populations. These results showed that the expression level of PECAM1 was positively correlated with pluripotency of ES cells, while that of SSEA-1 was correlated with the capacity to contribute to the epiblast after the blastocyst stage. These results raise the possibility that a comparison of subpopulation cell characteristics would offer important clues to the regulation of ES cell differentiation. To identify the key factors for the maintenance of pluripotency and the acquisition of competency for differentiation into epiblast cells, we examined gene expression profiles of ES cell subpopulations using oligo microarray analysis.
MATERIALS AND METHODS
Cell Culture
TT2 ES cells [2] were maintained on an STO cell feeder layer in Dulbecco's Modified Eagle's media (Invitrogen) containing 17.5% Knockout Serum Replacement (Invitrogen), 10–4 M 2-mercaptoethanol, 1 mM nonessential amino acids (Invitrogen), and human leukemia inhibitory factor (20 ng/ml, prepared in our laboratory). To remove feeder cells, enzymatically dispersed ES and STO cells were suspended in culture medium and plated in a culture dish. After 30 min of incubation at 37°C, nonadherent cells were harvested and plated in a culture dish again. After 30 min, the supernatant cells (pure ES cells) were used for sorting.
Cell Sorting, Blastocyst Collection, and RNA Isolation
Cell staining and sorting were carried out as previously described [1]. Briefly, ES cells were suspended in staining buffer (Hanks balanced salt solution [Invitrogen] containing 0.2% bovine serum albumin and 10 mM HEPES), and were stained with R-phycoerythrin-conjugated anti-mouse CD31/PECAM1 antibody (BD Biosciences, San Jose, CA) and anti-SSEA-1 antibody (Kyowa Medex, Japan) for 60 min on ice. The cells were then washed and resuspended in staining buffer containing fluorescein isothiocyanate-conjugated anti-mouse immunoglobulin M (BD Biosciences). After 60 min of incubation on ice, the cells were washed, resuspended in staining buffer, and kept on ice prior to sorting. The cells were sorted using an Epics ALTRA cell sorter (Beckman Coulter, Fullerton, CA). Total RNA was extracted using the RNeasy Mini Kit (QIAGEN, City, State) with DNase treatment, in accordance with the manufacturer's recommended protocol. Blastocysts (4.0 dpc) were collected from the uteri of superovulated ICR mice, and total RNA was isolated from a pool of 100 blastocysts, as described above.
DNA Microarray Hybridization and Data Analysis
Fluorescence-labeled (Cy3, Cy5) cRNA probes were prepared from 150–300 ng of pooled total RNA from three independent sorted cells using a Low RNA Input Linear Amplification Kit (Agilent Technologies, Palo Alto, CA). Labeled cRNA probes (750 ng each) were hybridized to a Mouse Oligo Microarray or a Mouse Development Oligo Microarray (Agilent Technologies) in hybridization buffer (Gene Expression Hybridization Kit, Agilent Technologies) at 60°C for 17 h. After hybridization, the arrays were washed with 6xSSC, 0.005% Triton X-102 at room temperature for 10 min, followed by 5-min washes in 0.1xSSC, 0.005% Triton X-102 at 4°C. Hybridized arrays were blow dried with N2 gas and scanned using a Microarray scanner (Agilent Technologies) at a pixel resolution size of 10 µm. To filter out unreliable signal intensities, local background subtraction was performed and the locally weighted scatter plot-smoothing algorithm was used for data normalization. The raw data from all of these studies are available online (URL: http://cibex.nig.ac.jp/index.jsp [CIBEX Accession: CBX11].
Quantitative RT-PCR
Total RNA (0.2 µg) was subjected to oligo-dT-primed reverse-transcription with MMLV-reverse transcriptase (Promega, Madison, WI). The single-strand cDNA products were purified using a PCR purification kit (QIAGEN), and 1/200 to 1/50 of the cDNA products were used for each PCR amplification. Quantitative RT-PCR was performed using LightCycler FastStart DNA Master SYBR Green I on a LightCycler instrument (Roche). The sequences of the gene-specific primers used here are shown in supplementary Table 1 (available online at http://www.biolreprod.org). The manufacturer's recommended reaction conditions were used for each amplification. Each gene expression was analyzed in triplicate.
Whole-Mount In Situ Hybridization
To generate probes for use in in situ hybridization experiments, RT-PCR products were subcloned into the EcoRV site of pBluscript II KS (+) plasmids (Stratagene, La Jolla, CA). Template DNAs were amplified by PCR with T7-HindIII (5'-GCGTAATACGACTCACTATAGGGAGATCGACGGTATCGATAAGCTT-3') or T7-EcoRI (5'-GCGTAATACGACTCACTATAGGGAGACCCCCGGGCTGCAGGAATTC-3') primers and cDNA internal sense or antisense primers. Digoxigenin-labeled cRNA antisense and sense probes were synthesized by T7 RNA polymerase from amplified cDNA fragments using a DIG RNA labeling kit (Roche). Blastocysts (4.0 dpc) were collected from the uteri of superovulated ICR mice (Charles River Japan, Kanagawa, Japan) and were fixed with 4% paraformaldehyde-PBS at 4°C overnight. Fixed embryos were washed with PBS containing 0.1% Triton X-100 (PBTX), dehydrated with methanol, and stored at –20°C until use. The embryos were rehydrated with PBTX and incubated for more than 2 h at 65°C with a hybridization buffer (DIG Easy Hyb, Roche). They were then used for hybridization at 65°C overnight in a hybridization buffer with the digoxigenin-labeled probes (
1 µg/ml). The embryos were then washed successively at 65°C with solution 1 (50% formamide, 5xSSC, 0.1% Triton X-100, and 0.5% CHAPS), a 3:1 mixture of solution 1 and 2xSSC, a 1:1 mixture of solution 1 and 2xSSC, and a 1:3 mixture of solution 1 and 2xSSC. This was followed by further washing with 2xSSC containing 0.1% CHAPS and 0.2xSSC containing 0.1% CHAPS at 65°C. After changing the solution to TBTX (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, and 0.1% Triton X-100), samples were blocked with PBS containing 1% skimmed milk powder and 0.2% Triton X-100 for more than 1 h. Samples were then incubated overnight at 4°C with an alkaline phosphatase-conjugated anti-digoxigenin antibody (1:2000, Roche). The samples were washed with TBTX four times for 15 min at room temperature prior to the color reaction. The samples were incubated with 4-nitro-blue-tetrazolium-chloride and 5-bromo-4-chloro-3-indolyl-phosphate in NTMT (100 mM NaCl, 100 mM Tris-HCl, pH 9.5, 50 mM MgCl2, and 0.1% Tween-20). After completion of the color reaction, the samples were washed several times with PBS containing 1% Triton X-100, and were then fixed with 4% PFA in PBTX.
All procedures involving the care and use of animals were approved by the Animal Research Committee, National Institute of Agrobiological Sciences.
RESULTS
Comparison of Gene Expression Profiles in PECAM1-Positive and -Negative ES Cells
To identify genes that might be involved in ES cell pluripotency, we initially compared gene expression profiles in PECAM1–SSEA-1– and PECAM1+SSEA-1– cells (Fig. 1A) using pairwise oligo microarray. We used Agilent's Mouse Oligo Microarray and Development Microarray. Each array contains about 20 000 unique and 3000 overlapping genes; thus, in combination, the two microarrays allow approximately 34 000 genes to be screened. The mouse oligo microarray identified 2034 genes showing a more than 2-fold difference in expression level in the two cell populations: 803 genes were more highly expressed in PECAM1-positive cells and 1231 were more highly expressed in PECAM1-negative cells (Fig. 1B). The mouse development microarray identified 1921 genes with a more than 2-fold difference in expression level: 768 genes were more highly expressed in PECAM1-positive cells and 1153 were more highly expressed in PECAM1-negative cells (data not shown).
|
The 100 genes that showed the greatest upregulation in PECAM1-positive cells in the two microarrays were ranked in order and are listed in Charts 1 and 2 as the "Top 100." Among Top 100 genes in the mouse oligo microarray, 43 were also identified as being upregulated with a more than 2-fold difference in the development microarray (Chart 1), and 52 were not contained in the gene list of the development microarray. Seven other genes (Dazl, Amhr2, Tdh, Galnt6, Rps12, Dpp4, and Zfp40) were only detected in the oligo microarray. As expected, genes known to function in ES cells—such as Pou5f1 (Oct3/4) [3, 4], Ecat genes [5], including Nanog [5, 6], Dppa5 (Esg-1) [7, 8], and Cdh1 (E-cadherin) [9–11]—were upregulated in PECAM1-positive cells. DNA methyltransferase 3l and 3b genes were also assessed as being highly expressed in PECAM1-positive cells. Of the Top 100, the genes that have previously been reported as being highly expressed in ES cells, such as Ung and Fgf4, were also identified as being upregulated with a more than 2-fold difference (Charts 1 and 2).
The Top 100 genes from the PECAM1-negative cells, identified by the mouse oligo microarray and the development microarray, are listed in Charts 3 and 4, respectively. Among the Top 100 genes in the mouse oligo microarray, 46 were also identified as being upregulated with a more than 2-fold difference in the development microarray (Chart 3). Many genes coding for various extracellular matrix (ECM) proteins, such as vitronectin and Efemp2, collagens, and adherent molecules, such as cadherins and activated leukocyte cell adhesion molecule, were detected as enriched genes in PECAM1-negative cells (Charts 3 and 4). Several imprinted genes, such as Igf2, Peg1/Mest, Peg3, and Peg12, were also upregulated in PECAM1-negative cells.
Expression Analysis of Ppet Genes
We focused on 23 uncharacterized genes that were upregulated in PECAM1-positive cells. We have assigned the designation Ppet (PECAM1 Positive ES cell-derived Transcripts) to this group of uncharacterized genes expressed in the PECAM-positive cells temporarily until they are cloned and characterized (Chart 1). Quantitative RT-PCR confirmed the array results (Fig. 2A) and revealed that Ppet genes, except for Ppet 8, were expressed in blastocysts. Using in situ hybridization, we found that Ppet 3, 5, 10, 21, and 22 were specifically expressed at the inner cell mass of blastocysts (Fig. 2B) and that Ppet 23 was strongly expressed in the whole embryo. Other Ppet genes could not detected by in situ hybridization.
|
Comparison of Gene Expression Profiles Between SSEA-1-Positive and -Negative ES Cells
Next, we compared gene expression profiles in SSEA-1-positive and -negative cells of the PECAM1-positive subpopulation in order to identify genes involved in the acquisition of competency for differentiation into epiblast. Using the mouse oligo microarray, only 53 genes were detected with a 2-fold or greater differential expression between the two subpopulations (Fig. 1C). However, we found that many genes under epigenetic regulation showed a more than 1.5-fold higher expression in SSEA-1-negative cells: the globin genes Hbb-bh1 (7.3-fold), Hbb-y (1.9-fold), Hba-a1 (2.1-fold), and Hba-x (1.6-fold); and Igf2 (2.3-fold), Igf2r (1.6-fold), Krt1–18 (Keratin18, 1.6-fold), and H19 (1.5-fold). We performed quantitative RT-PCR on six epigenetically regulated genes and found that expression levels were in the rank order PECAM1–SSEA-1–, PECAM1+SSEA-1–, and PECAM1+SSEA-1+ (Fig. 3).
|
DISCUSSION
Microarray analyses revealed large differences in gene expression profiles between PECAM1-positive and -negative cells. A greater number of genes were identified as being differentially expressed in PECAM1-negative than -positive cells. This might imply that the stringent regulation of gene expression needed to maintain pluripotency has broken down in PECAM1-negative cells. In particular, a remarkable upregulation of genes that encode for cell adhesion molecules and ECM in PECAM1-negative cells, and it may reflect alterations in cell interactions, such as those occurring during the morphological changes in ES cells during differentiation.
Recently, many studies have been carried out to characterize gene expression profiles in ES cells using microarray techniques [8, 12–19]. From these studies, Rao and Stice [20] compiled a list of 83 representative genes that are highly expressed in ES cells. They categorized these genes into six groups: transcription factors/nucleic acid binding, growth factors/receptors, cell cycle regulators, signaling molecules/secreted factors, cell adhesion/membrane proteins, and other miscellaneous genes. Fifteen of the 27 genes classified in the transcription factors/nucleic acid binding category, including Pou5f1 and Nanog, overlapped with those in our microarray analysis (asterisked genes in Charts 1 and 2). In contrast, although genes such as Fgf4, Ccne1, and DNA methyltransferase family genes were observed, the correspondence between the present study and the compiled list was less than expected in the other categories. In addition to differences in methodology and data analysis, differences in the ES cell lines, for example, the mouse strain from which they originated, passage numbers, culture conditions, and the proportion of "undifferentiated cells" in culture, may have contributed to this inconsistency.
We focused on 23 uncharacterized genes, designated Ppet genes, that were enriched in PECAM1-positive cells. Domain search analysis of predicted proteins showed that Ppet 3, 5, 8, 15, 16, 17, and 22 were expected to be transcription factors, and that Ppet 4, 18, 21, and 23 may be involved in signal transduction pathways. Furthermore, we found 22 Ppet genes expressed at the blastocyst stage, and in situ analysis revealed that Ppet 3, 5, 10, 21, and 22 were specifically expressed at the inner cell mass of blastocysts. ES cells are derived from the epiblast of blastocysts, and they preserve the character of epiblast origin. Common mechanisms for cell growth and maintenance of the pluripotency may present in epiblast and ES cells. Therefore, these Ppet genes are expected to encode for essential molecules to maintain pluripotency, not only in ES cells but also in epiblast cells in preimplantation embryos. Studies are under way to define the biological functions of Ppet in ES cells and early embryos.
SSEA-1-positive and -negative cells in the PECAM1-positive subpopulation showed similar gene expression profiles, suggesting that they have almost identical characteristics. In fact, these two subpopulation cells both form typical compact ES colonies, and demonstrate similar proliferation and entirely reciprocal repopulation in culture [1]. Cell fate mapping disclosed that both subpopulation cells can contribute equally to epiblast cells until the blastocyst stage; however, SSEA-1-negative-derived cells disappeared in postimplantation stages [1]. These results suggest that SSEA-1-negative cells lack essential factors and/or important epigenetic competence for organogenesis after the blastocyst stage. Interestingly, many epigenetically regulated genes were identified in the comparison of these two subpopulations, with Hbb-bh1 showing the highest fold change. In mammals, ß-globin family genes are expressed in a tissue-specific and developmental stage-specific manner [21], and are regulated by a locus control region (LCR).
Recently, it was reported that histone acetylation patterns of the LCR and promoter region play important roles in transcriptional regulation of ß-globin [22–24], suggesting that histone modification at the globin locus is different between SSEA-1-negative and -positive cells.
Epigenetic heterogeneity in ES cells may cause the varied contribution of ES cells to chimeric embryos and to phenotype abnormalities that are observed in ES-derived cloned mice [25–27]. It is important to investigate further the patterns of DNA and histone modifications of chromatin domains at epigenetically regulated gene loci. We are attempting to develop assay systems for monitoring epigenetic alterations in living cells.
In this report, we have demonstrated the benefit of purification of ES cell subpopulations to elucidate the mechanism of ES cell pluripotency. A source of homogeneous, undifferentiated cells would also provide an efficient tool for other aspects of stem cell research.
|
|
|
|
FOOTNOTES
1 Supported by a grant from the Program for Promotion of Basic Research Activities for Innovative Biosciences (PROBRAIN). 
2 Correspondence: Tomoyuki Tokunaga, Development and Differentiation Laboratory, Developmental Biology Department, Insect and Animal Sciences Division, National Institute of Agrobiological Sciences, Ikenodai 2, Tsukuba, Ibaraki, 305-8602, Japan. FAX: 81 298 38 7383; tom{at}affrc.go.jp
Received: 24 November 2005.
First decision: 23 December 2005.
Accepted: 5 May 2006.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  A. J. K. Williamson, D. L. Smith, D. Blinco, R. D. Unwin, S. Pearson, C. Wilson, C. Miller, L. Lancashire, G. Lacaud, V. Kouskoff, et al. Quantitative Proteomics Analysis Demonstrates Post-transcriptional Regulation of Embryonic Stem Cell Differentiation to Hematopoiesis Mol. Cell. Proteomics, March 1, 2008; 7(3): 459 - 472. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Singh, T. Hamazaki, K. E. Hankowski, and N. Terada A Heterogeneous Expression Pattern for Nanog in Embryonic Stem Cells Stem Cells, October 1, 2007; 25(10): 2534 - 2542. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z.-X. Wang, C. H.-L. Teh, J. L. L. Kueh, T. Lufkin, P. Robson, and L. W. Stanton Oct4 and Sox2 Directly Regulate Expression of Another Pluripotency Transcription Factor, Zfp206, in Embryonic Stem Cells J. Biol. Chem., April 27, 2007; 282(17): 12822 - 12830. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
