|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Embryo |
Development and Differentiation Laboratory,3 Developmental Biology Department, Insect and Animal Sciences Division, National Institute of Agrobiological Sciences, Ibaraki, 305-8602, Japan
Reproductive Cell Biology Laboratory,4 Department of Animal Breeding and Reproduction, National Institute of Livestock and Grassland Science, Ibaraki, 305-0901, Japan
| ABSTRACT |
|---|
|
|
|---|
early development, embryo, developmental biology
| INTRODUCTION |
|---|
|
|
|---|
| MATERIALS AND METHODS |
|---|
|
|
|---|
Three ES, one embryonic germ (EG), and one embryonic carcinoma (EC) cell line were used in this study. ROSA cells, established from a ROSA26 (C57BL/6 background) x CBA F1 embryo, and TT2 cells [5] were maintained on an STO cell feeder layer in Dulbecco modified Eagle (DME) medium containing 17.5% knockout serum replacement (Invitrogen, Carlsbad, CA), 104 M 2-mercaptoethanol, 1 mM nonessential amino acids (Invitrogen), and human leukemia inhibitory factor (LIF, 20 ng/ml). ZHBTc4 cells [6] were maintained on gelatin-coated dishes in DME medium containing 20% fetal calf serum (FCS) as described. The EG cell line TM1 was established by culturing primordial germ cells obtained from an 8.5-day post coitum C57BL/6 mouse fetus [7] and maintained under the same conditions described above for ROSA and TT2 cells. The EC cell line F9 was cultured on gelatin-coated dishes in DME medium containing 10% FCS and 104 M 2-mercaptoethanol.
Flow Cytometry
Cells were dispersed by Accutase (Innovative Cell Technologies, La Jolla, CA) that was diluted 1:3 with PBS and then suspended in cold staining buffer (PBS containing 0.2% BSA). R-phycoerythrin-conjugated anti-mouse CD31/PECAM-1 antibody (40 ng/106 cells; BD Biosciences, San Jose, CA) and anti-SSEA-1 antibody (200 ng/106 cells, Kyowa Medex, Shizuoka, Japan) were added to the cell suspension; the cell suspension was then incubated for 30 min on ice. Subsequently, the cells were washed with staining buffer and resuspended in staining buffer containing fluorescein isothiocyanate-conjugated anti-mouse IgM (BD Biosciences). After a 30-min incubation on ice, the cells were resuspended in staining buffer and kept on ice prior to fluorescence-activated cell sorting.
Quantitative Reverse Transcription-Polymerase Chain Reaction
Total RNA was extracted using the RNeasy minikit (Qiagen, Hilden, Germany) with DNase treatment, according to the manufacturer's protocol. Total RNA (0.2 µg) was then subjected to oligo-dT-primed reverse transcription (RT) with Muloney murine leukemia virus-reverse transcriptase (Promega, Madison, WI). The single-strand cDNA products were purified using a polymerase chain reaction (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 (Roche, Indianapolis, IN) on a LightCycler instrument (Roche). Gene-specific primers were designed based on published sequences (Table 1). The reaction conditions for each amplification were chosen as recommended by manufacturer.
|
Mapping of ES-Derived Cells and Generation of Chimeric Mice
Sorted ES cells were kept on ice until they were injected into host embryos within 2 h after sorting. A single, sorted ROSA cell was injected into 8-cell embryos [8] that were cultured in M16 media containing 10% FCS. After 1836 h, embryos at Embryonic Day (E) 3.54.0 were fixed with 0.1% glutaraldehyde-PBS containing 0.2% Nonidet P-40 and stained with X-gal. For mapping at later developmental stages, ES cell-injected embryos that were cultured for 1618 h in M16 media containing 10% FCS were transplanted into the uterine horn of pseudopregnant ICR recipient mice. After 4 days, embryos at the E6.07.0 stage were harvested from the uteri and were fixed for X-gal staining. ß-Gal-positive embryos were embedded in paraffin and sectioned for microscopic examinations. To generate chimeric mice, five sorted TT2 cells were injected into 8-cell ICR embryos that were then transplanted into pseudopregnant ICR recipient mice. Surgery was performed under barbiturate anesthesia, and all experiments were carried out under the Guidelines Concerning Animal Experiments at the National Institute of Agrobiological Sciences.
Statistical Analysis
Statistical significance between groups was determined using the chi- square or Fisher test.
| RESULTS |
|---|
|
|
|---|
Flow cytometry analysis revealed a wide degree of variation in the expression of PECAM-1 and SSEA-1 in ROSA ES cells (Fig. 1A). Two additional independent ES cell lines (i.e., TT2 and ZHBTc4 [6]) showed similar variations in their expression of SSEA-1 and PECAM-1 (data not shown), which contrasted with a lesser degree of variation in the expression of these antigens in EG and EC cells. Approximately 80% of the above total cell populations expressed PECAM-1, whereas the expression of SSEA-1 was found only on cells expressing high levels of PECAM-1 (15.2% of the total population).
|
ROSA ES cells were sorted into three subpopulations (i.e., PECAM-1SSEA-1, PECAM-1+SSEA-1, and PECAM-1+SSEA-1+) and repopulation assays were performed. After 4 days, each sorted cell population gave rise to progeny cells representing the other two populations (Fig. 1B). However, cells lacking PECAM-1 expression proliferated very poorly, and almost all of them demonstrated adhesive, differentiated cell morphology. PECAM- 1+SSEA-1+ and PECAM-1+SSEA-1 cells formed compact colonies that were typical of undifferentiated ES cells and showed complete, reciprocal repopulation within 4 days.
Quantitative RT-PCR was performed to determine the level of expression of differentiation marker genes in each cell subpopulation (Fig. 2). Relative to the other subpopulations, PECAM-1 cells showed reduced Oct-3/4 expression. The transcripts of the Oct-3/4 target genes Rex-1 also decreased in this cell population. Nanog [9, 10], a newly reported homeobox transcription factor, was also expressed at low levels in PECAM-1 cells. More significantly, we observed a dramatic activation in primitive endodermal (Gata4 and collagen type IV) and primitive ectodermal markers (activin, Fgf-5) and in early embryonic mesodermal markers (Brachyury) in cells lacking PECAM-1 expression. PECAM-1+SSEA-1 and PECAM-1+SSEA-1+ cells showed nearly identical patterns of expression of these differentiation marker genes. Only the level of mRNA for
1, 3-fucosyltransferase IX (Fuc-TIX) in PECAM- 1+SSEA-1+ cells, which is required for the synthesis of SSEA-1, were higher (2.8-fold) than in PECAM-1+SSEA- 1 cells. High
1, 3-galactosyltransferase (Gal-T) transcript expression was detected in PECAM-1SSEA-1 cells but not in PECAM-1+SSEA-1 cells; it is known that Gal-T conceals the antigenicity of SSEA-1 by masking the epitope with an
-galactose residue.
|
PECAM-1+ SSEA-1+ ES Cells Predominantly Differentiate into Epiblast
To compare the pluripotency of the three ES cell subpopulations, a single cell from each subpopulation was injected into an 8-cell embryo, and the fate of the ES-derived cells was mapped at later developmental stages by X-gal staining. At the blastocyst stage, cells derived from the cell lacking PECAM-1 expression localized to the primitive endoderm or trophectoderm (Table 2 and Fig. 3A). In contrast, cells derived from a PECAM-1-positive cell most often differentiated into epiblast cells (Table 2 and Fig. 3B and C). We did not detect significant differences in the contribution to embryos between the PECAM-1+SSEA-1 and PECAM- 1+SSEA-1+ subpopulations (Table 2). At E6.07.0, ß-gal- positive cells were only detected in three embryos that were injected with both PECAM-1SSEA-1 and PECAM- 1+SSEA-1 cells (5.0% and 4.7%, respectively, Table 3). Only injected PECAM-1SSEA-1 cells were found to differentiate into visceral endoderm (Fig. 4, A, E, B, and F). PECAM-1+SSEA-1 cells differentiated into epiblast cells in three cases, although their contribution was lower than that seen for PECAM-1+SSEA-1+ cell derivatives (Fig. 4C). In contrast, significant numbers of PECAM-1+SSEA- 1+ cell derivatives existed in the epiblast region (15.7% of embryos; Table 3 and Fig. 4, D and G). No ß-galpositive cells derived from any populations were detected in the region of the trophectoderm at this developmental stage.
|
|
|
|
Finally, we compared the frequency of generation of chimeric mice between the nonenriched and enriched cells (PECAM-1+SSEA-1+ cells). Three of thirty mice (10%) derived from nonenriched parental ES cell-injected embryos exhibited chimerism, and one of three chimeras was confirmed as a germ-line chimera by mating (Table 4). In contrast, 8 of 23 pups (35%) from PECAM-1+SSEA-1+ cell- injected embryos exhibited chimerism (Table 4 and Fig. 4H). Unfortunately, two chimeras with high ES coat color contributions from PECAM-1+SSEA-1+ cells died before maturation, but at least two of six viable chimeras (33.3%) were confirmed as germ-line chimeras (Table 4).
|
| DISCUSSION |
|---|
|
|
|---|
In the repopulation assay, PECAM-1-negative cells proliferated poorly, and a minor population of this cell fraction seemed to repopulate the other two subpopulations. In contrast, PECAM-1+SSEA-1+ and PECAM-1+SSEA-1 cells demonstrated entirely reciprocal repopulation. These results suggest that the immunoreactivity of SSEA-1 on PECAM- 1-positive cells may have temporarily changed and is not consistent with irreversible differentiation of ES cells. It has been reported that the change in immunoreactivity of SSEA-1 on F9 cells was due to the activity of Fuc-TIX and Gal-T [26]. We also observed the variation of mRNA expression levels of these carbohydrate enzymes among subpopulations. Gal-T mRNA levels were equivalent in PECAM-1+SSEA-1+ and PECAM-1+SSEA-1 cells, but Fuc- TIX mRNA levels in PECAM-1+SSEA-1+ cells were higher (2.8-fold) than those seen in PECAM-1+SSEA-1 cells, which might result in high SSEA-1 levels in PECAM- 1+SSEA-1+ cells. Although Fuc-TIX mRNA expression in PECAM-1SSEA-1 cells showed the highest levels in three subpopulations, Gal-T mRNA expressed remarkably high levels in this subpopulation, which might lead the reduction of the antigenicity of SSEA-1 in this subpopulation. Taken together, these results suggest that the expression of Gal-T is stringently regulated by the factor(s) involved in the pluripotency of ES cells, but the change in Fuc-TIX is not. Fenderson et al. [27, 28] suggested that the changes in carbohydrate phenotype in the early embryo might play an important role in regulating the cell-cell interactions that are involved in lineage formation and morphogenesis. Similar changes might occur in ES cells, and fluctuations in the amount of SSEA-1 may be responsible for the differences in gene expression that influence their contributions to the epiblast in E6.07.0 embryos. Although it is unclear why PECAM-1+SSEA-1-derived cells that were observed in the blastocyst disappeared in later-stage embryos, we speculate that they might lack essential factor(s) or important epigenetic competence that would allow for organogenesis after the blastocyst stage.
It has been reported that mouse ES cells have the potential to differentiate into the trophectodermal lineage in vitro and in vivo [29]. We observed that the derivatives of ES cells that lacked PECAM-1 expression localized to the trophectoderm at the blastocyst stage but not at E6.07.0. Although PECAM-1-negative cells may have the potential to differentiate into the trophectodermal lineage, their rate of differentiation into functional extraembryonic tissues may be quite low. Interestingly, we have never observed the differentiation of an injected ES cell into multiple lineages in any population. If the injected cell derivatives had still kept multipotency or bipotency just after first cell division, we should observe the derivatives in multiple lineages (i.e., both primitive endoderm and the epiblast lineage). This result raised two possibilities: 1) the injected cell had already committed to a specific lineage or 2) the positional information imparted by the host embryo strongly affected the fate of the injected cell and/or its derivatives, which led them to the particular lineage commitment. To further investigate these possibilities, we are presently attempting to identify new surface markers that can be used to further divide PECAM-1-positive ES cells into more precise and definitive subpopulations and trace their fate in host embryos.
Enrichment of PECAM-1+SSEA-1+ cells was effective in generation of chimeric mice; however, the ES cell contributions in individual mice were equivalent to those in mice-derived nonenriched ES cells. The number of pups from the enriched ES cells was 11.3% less than those from nonenriched ES cells, and offspring with high ES coat color contributions died before maturation, suggesting that mortality of chimeric animals increases with the ES cell proportion. The reason is not clear; however, epigenetic alterations in imprinted genes may occur in some ES cell populations, and it may be associated with high mortality [30, 31].
This article describes the first evidence that the expression levels of PECAM-1 and SSEA-1 in ES cells correlated closely with their pluripotency and/or their ability to incorporate into the epiblast of chimeric embryos. Such studies should eventually enable us to produce homogeneous sources of ES cells that can be used to facilitate the development of effective methods for generating chimeric animals and better protocols for the differentiation of specific cell lineages.
| ACKNOWLEDGMENTS |
|---|
| FOOTNOTES |
|---|
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;caaffrc.go.jp ![]()
Received: 14 October 2003.
First decision: 29 October 2003.
Accepted: 12 January 2004.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
C. L. Bauwens, R. Peerani, S. Niebruegge, K. A. Woodhouse, E. Kumacheva, M. Husain, and P. W. Zandstra Control of Human Embryonic Stem Cell Colony and Aggregate Size Heterogeneity Influences Differentiation Trajectories Stem Cells, September 1, 2008; 26(9): 2300 - 2310. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Toyooka, D. Shimosato, K. Murakami, K. Takahashi, and H. Niwa Identification and characterization of subpopulations in undifferentiated ES cell culture Development, March 1, 2008; 135(5): 909 - 918. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Furusawa, M. Ikeda, F. Inoue, K. Ohkoshi, T. Hamano, and T. Tokunaga Gene Expression Profiling of Mouse Embryonic Stem Cell Subpopulations Biol Reprod, October 1, 2006; 75(4): 555 - 561. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Enver, S. Soneji, C. Joshi, J. Brown, F. Iborra, T. Orntoft, T. Thykjaer, E. Maltby, K. Smith, R. A. Dawud, et al. Cellular differentiation hierarchies in normal and culture-adapted human embryonic stem cells Hum. Mol. Genet., November 1, 2005; 14(21): 3129 - 3140. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Palmqvist, C. H. Glover, L. Hsu, M. Lu, B. Bossen, J. M. Piret, R. K. Humphries, and C. D. Helgason Correlation of Murine Embryonic Stem Cell Gene Expression Profiles with Functional Measures of Pluripotency Stem Cells, May 1, 2005; 23(5): 663 - 680. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |