Abnormal Development of Mouse Embryoid Bodies Lacking p27Kip1 Cell Cycle Regulator
http://www.100md.com
《干细胞学杂志》
a Center for Cell Therapy and Tissue Repair, Charles University, Prague, Czech Republic;
b Department of Molecular Embryology, Institute of Experimental Medicine, Academy of Sciences of the Czech Republic, Prague, Czech Republic;
c Laboratory of Molecular Embryology, Mendel University Brno, Brno, Czech Republic;
d Laboratory of Molecular Neurobiology, Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden;
e Laboratory of Cytokinetics, Institute of Biophysics, Academy of Sciences of the Czech Republic, Brno, Czech Republic
Key Words. Mouse embryonic stem cells ? p27 ? Embryoid bodies ? Lewis-X ? Apoptosis
Correspondence: Ale Hampl, D.V.M., Ph.D., Laboratory of Molecular Embryology, Mendel University Brno, Zemdlská1, 61300 Brno, Czech Republic. Telephone: 420-5-45133297; Fax: 420-5-45133357; e-mail: hampl@mendelu.cz
ABSTRACT
Cell cycle regulatory protein p27Kip1 (from this point referred to as p27) belongs to a Cip/Kip family of cyclin-dependent kinase (CDK) inhibitors. Because its best-studied function is inhibiting the activity of cyclin-CDK complexes that drive the progression of the cell cycle, p27 is mostly understood as a negative regulator of cell proliferation (for review, see ). The phenotype of p27-null animals, which are significantly larger due to a higher number of cells, supports such a view . However, detailed analyses of p27–/– animals have also revealed defects that could not be explained by simple alterations in cell proliferation. Specifically, it was shown that processes such as cellular differentiation and apoptosis were severely affected in p27-deficient mice .
Embryonic stem cells (ESCs) are unique in terms of molecules and mechanisms used to regulate their undifferentiated growth . In regard to this fact, we have recently found that rather than for regulation of proliferation of pluripotent mouse ESCs (mESCs), p27 is essential for the in vitro differentiation of such cells to proceed normally. Specifically, when p27-deficient mESCs grown in monolayer are induced to differentiate into extraembryonal endoderm by withdrawal of leukemia inhibitory factor (LIF) combined with retinoic acid treatment, most of these mESCs die by apoptosis before finishing their differentiation program . Although this phenomenon is well-pronounced in vitro, its relevance to the development of the early embryo is limited due to the used differentiation system. It is well accepted that early developmental processes can be mimicked in vitro by culturing multicellular aggregates of mESCs in the absence of LIF and feeder cells . Under such conditions, mESC aggregates give rise to simple embryoid bodies (EBs) containing an outer layer of endodermal cells and a solid core of undifferentiated ectodermal cells. The inner cells of simple EBs subsequently undergo the wave of programmed cell death to form cystic EBs, a process called cavitation . Because cystic EBs are cultured in vitro, they provide a unique tool for the analysis of the impact of individual genes and proteins on proliferation, differentiation, and apoptosis in early development (for review, see ).
In this study, the phenotypes of EBs produced from normal and p27-deficient mESCs were analyzed in detail to improve our understanding of the significance of p27 for early embryogenesis. The data here show that in the absence of p27, the EBs reach a bigger size than their normal counterparts, most likely due to enhanced cavitation affecting primarily Lewis-X–positive cells.
MATERIALS AND METHODS
p27-Deficient EBs Grow Bigger Than Their Normal Counterparts
Because p27 is known primarily for its proliferation-regulating properties, it was first necessary to determine whether the absence of p27 results in the alteration of growth of EBs. EBs were allowed to form from both normal and p27–/– mESCs, and they were digitally photographed at day 10 of their development. The average diameter of EBs in each wild-type and p27–/– group was then determined using image-analysis software while assuming the EBs to be of round shape. Because data of both normal and p27-deficient cells showed log-normal distribution (Kolmogorov-Smirnov test; p = .02 for p27+/+ data and p = 1.5 x 10–8 for p27–/– data), they were ln-transformed (Kolmogorov-Smirnov test of ln-transformed data; p = .49 for p27+/+ data and p = .07 for p27–/– data) and compared by Student’s t-test. As demonstrated in Figure 1, p27-deficient EBs grew significantly bigger than normal EBs (two-tailed Student’s t-test of ln-transformed data sets; p = 8.8 x 10–16).
Figure 1. Effect of p27 on size of EBs. Normal and p27-deficient EBs were cultured up to day 10, photographed, and analyzed by image analysis software. Typical EBs are shown (A), and statistical analysis (B) of diameter of EBs (box-and-whisker plot; median, 25% to 75% interval, minimal and maximal values) is presented. Abbreviations: EB, embryoid body; K-S, Kolmogorov-Smirnov test of normal distribution of data.
The Activities of CDKs Are Not Deregulated in p27–/– EBs
Increased activities of CDK2 and CDK4 resulting in hyperplasia of various tissues or organs are typical for mice that are deficient in p27 . Therefore, it was important to determine whether the bigger size of p27–/– EBs observed here was correlated with upregulated CDK activities. Undifferentiated mESCs and EBs at days 5, 10, and 20 of their development, both wild-type and p27-deficient, were analyzed to address this question. The following parameters were determined: (a) levels of CDK2, CDK4, cyclin A, cyclin E, all three D-type cyclins, and p27; (b) kinase activities associated with CDK2, CDK4, cyclin A, and cyclin E; and (c) the amounts of p27 associated with the regulators listed under (b). During the development of wild-type EBs, the amount of p27 physically bound to CDK2, CDK4, cyclin A, and cyclin E, respectively, dramatically increases, corresponding to the elevation of the total amount of p27 (Fig. 2). Except for a generally lower amount of cyclin D3 in p27-deficient mESCs and EBs that can be well explained by the molecular mechanism described by us previously , the behavior of no other cyclin was significantly affected by the absence of p27 (Fig. 2). Total amounts of all cyclin A, cyclin E, cyclin D1, and cyclin D3 were downregulated with the progressing development of EBs (Fig. 2). On the other hand, cyclin D2 followed the opposite pattern, being undetectable in mESCs and increased in EBs (Fig. 2). Downregulation of the kinase activities of CDK2, cyclin A, cyclin E, and CDK4 took place in developing EBs irrespectively of their p27 genotype (Fig. 2). These data document that the abnormal growth of EBs produced from p27-deficient mESCs cannot be due to hyperactivation of major CDKs. Furthermore, the results suggest that downregulated cyclins rather than elevated p27 are responsible for the decline of CDK activities that accompanies EB development.
Figure 2. Analysis of cyclin-CDK complexes in normal and p27-deficient EBs. EBs were differentiated from normal and p27-deficient mES cells (D0) and lysed at day 5 (D5), day 10 (D10), and day 20 (D20) of differentiation. Cell lysates were used to immunoprecipitate CDK2, cyclin A, cyclin E, and CDK4. The kinase activities toward histone H1 or GST-Rb were determined by autoradiography. The amount of p27 in immunoprecipitates was determined by Western blotting. The total amount of CDK2, cyclin A, cyclin E, CDK4, D-type cyclins, and p27 as determined by Western blotting is also shown. Densitometry was used to assess the intensity of autoradiographic signal. The average density of p27+/+ (D0) sample was defined as 1.0, and from this value all other values were calculated. Data represent the means, with standard deviations indicated by error bars. Western blots and autoradiograms are representative of at least three independent replicates. Abbreviations: EB, embryoid body; IP, immunoprecipitate; WB, Western blotting.
The Absence of p27 Prevents Normal Expression of Cytokeratin Endo-A (TROMA-I) in Developing EBs
We have previously shown that p27-deficient mESCs suffer from apoptosis when they are induced to differentiate into TROMA-I–positive extraembryonal endoderm in monolayer culture . When wild-type mESCs were allowed to differentiate into EBs, strong positivity for TROMA-I was detected by Western blot as early as at day 5 of differentiation and stayed unchanged until the end of culture at day 20 (Fig. 3A). In contrast, only very low expression of TROMA-I was reached in EBs produced from p27-deficient mESCs, although mESCs of both genotypes progressed in their differentiation, as documented by downregulation of the stem cell marker Oct-4 (Fig. 3A). We next proceeded to determine whether a low proportion of TROMA-I–positive EBs, or decreased numbers of TROMA-I–positive cells in EBs, underlay the lower levels of cytokeratin endo-A determined above. Immunohistochemical visualization of cytokeratin endo-A on sectioned EBs produced by 10-day differentiation of normal and p27–/– mESCs was used to address this question. More than 90% of EBs of wild-type genotype contained TROMA-I–positive cells, with these cells being abundantly distributed in the superficial layer of the EBs (Figs. 3B, 3C). In contrast, only approximately 10% of p27–/– EBs contained TROMA-I–positive cells, which were invariably restricted to only dispersed individual cells (Fig. 3C).
Figure 3. Analysis of expression of TROMA-I in normal and p27-deficient EBs. (A): EBs were differentiated from normal and p27-deficient mouse embryonic stem cells (D0) and harvested at day 5 (D5), day 10 (D10), and day 20 (D20) of differentiation. The expression of Oct-4 and TROMA-I was analyzed by Western blotting. (B): EBs (day 10) were fixed, sectioned, and immunohistochemically stained using antibody against TROMA-I (green) and by PI (red). Typical pattern of TROMA-I staining for p27+/+ and p27–/– EBs is shown. (C): EBs containing at least some TROMA-I–positive cells were considered as TROMA-I–positive, and their proportion within total number of EBs was calculated. n indicates number of analyzed EBs. Abbreviations: EB, embryoid body; PI, propidium iodide; TROMA-I, cytokeratin endo-A.
Lewis-X–Positive Cells Form Cavities in p27-Deficient EBs
The abnormality in development of TROMA-I–positive cells itself could not explain the overgrowth of EBs lacking p27. Therefore, the question remained whether the bigger size of EBs may mirror an abnormal expansion of defined cell types. We have collected a large selection of antibodies against lineage-specific markers and applied them on the sectioned EBs as above to address this question. Interestingly, of all tested markers, only the early neural antigen identified by the antibody FORSE-1 revealed a significant difference between normal and p27-deficient EBs. Specifically, while in normal EBs FORSE-1 positivity was limited to clusters of 10 to 50 cells, under p27–/– conditions, the FORSE-1 positivity was redistributed to both surround and fill the cavities in most EBs (Fig. 4A). Such redistribution was invariably accompanied by a loss of FORSE-1–positive cell clusters in p27–/– EBs. A similar phenotype was observed in all three tested p27-deficient mESC lines.
Figure 4. Expression of FORSE-1 and TEC-01 in normal and p27-deficient EBs. EBs were differentiated from normal and p27-deficient mouse embryonic stem cells, and at day 10 they were fixed, sectioned, and immunohistochemically stained using antibody against FORSE-1 and TEC-01, respectively (green), and PI (red). (A): Typical patterns of FORSE-1 staining are shown. Depending on distribution of FORSE-1–positive cells, two phenotypes of FORSE-1–positive EBs were distinguished: cluster (typical for p27+/+ EBs; EB contains a compact aggregate of 5 to 100 FORSE-1–positive cells without a cell-free space inside the group of positive cells) and cavity (typical for p27–/– EBs; area of FORSE-1–positive cells is characterized by the presence of cell-free space and expression of FORSE-1 is often polarized, having its maximum in lumen of the cavity). Proportion of individual phenotypes of FORSE-1–positive EBs is presented as percentages of EBs typical by clusters and cavities, respectively. n indicates number of analyzed EBs for each genotype. Data represent the means, with standard deviations indicated by error bars. (B): Serial sections of EBs were stained either for FORSE-1 or TEC-1 (green) and compared. Cell nuclei were stained with PI (red). Representative sections are shown. Abbreviations: EB, embryoid body; PI, propidium iodide.
The antigen detected by FORSE-1 antibody involves the oligosaccharide epitope called Lewis-X . However, as the structures that carry the Lewis-X epitope may modify antibody binding, we wanted to confirm the authenticity of the epitope by an independent reagent. We used the monoclonal antibody TEC-01 developed by Draber and Pokorna that is specific for the Lewis-X epitope (identical to the marker of pluripotent cells often referred to as SSEA-1 ). The specificity of FORSE-1 and TEC-01 antibodies was investigated on serial sections made from EBs produced by 10-day differentiation of normal and p27–/– mESCs. As demonstrated in Figure 4B, identical structures, both cells and extracellular matrix, were recognized by FORSE-1 and TEC-01 antibodies. Therefore, in the below text we refer to these structures as Lewis-X–positive.
Lewis-X–Positive Populations in Normal and p27-Deficient EBs Are Different from Each Other
To answer the question of whether increased proliferation within the Lewis-X–positive cell population was involved in establishing the observed phenotype, EBs at days 4, 5, 6, and 10 of their development were trypsinized and the proportion of Lewis-X–positive cells was determined by immunocytochemical staining of coverslip-immobilized cells. As demonstrated in Figure 5A, the number of Lewis-X–positive cells gradually decreased in both normal and p27-deficient EBs down to 3.5% at day 10 of differentiation. Interestingly, although the general trend to decrease was common to EBs of both genotypes, a slightly higher proportion of Lewis-X–positive cells was observable in p27-deficient EBs between days 4 and 6 of their development. This tendency was statistically significant at day 6 and suggests that higher proliferation of Lewis-X–positive cells may contribute to the formation of the cavities in p27-deficient EBs.
Figure 5. Properties of Lewis-X–positive cells in normal and p27-deficient EBs. (A): Normal and p27-deficient EBs were trypsinized at days 4, 5, 6, and 10 (D4, D5, D6, D10), adhered on microscopic cov-erslips, and stained with antibody against Lewis-X. A total of 1,000 cells at two independent regions of the coverslip was counted, and the proportion of Lewis-X–positive cells was determined. The graph represents means and standard deviations obtained from two independent experiments. Statistically significant difference between normal and p27-deficient cells is indicated by an asterisk (Student’s t-test, p < .05). (B): Normal and p27-deficient EBs were cultured until day 10, trypsinized, and stained with antibody against Lewis-X. Cell suspension was subjected to flow cytometric analysis, and Lewis-X–positive cells (region R2 in B) were sorted out. (C): The amounts of p27 in Lewis-X–positive (inside R2 region in B) and Lewis X–negative cells (outside R2 region in B) were analyzed by Western blotting. (D): The levels of p27, Oct-3/4, N-cadherin, E-cadherin, and PARP were determined in sorted Lewis-X–positive cells (region R2 in B) by Western blotting. This experiment was repeated twice, and, except for N-cadherin, the results of only one experiment are presented. Abbreviations: EB, embryoid body; PARP, poly (ADP-ribose) polymerase.
Still, the observed changes in the Lewis-X–positive population also suggest that qualitative differences may exist between the cluster- and cavity-forming Lewis-X–positive cells. To address this issue, Lewis-X–positive cells were sorted out using flow cytometry from 10-day-old EBs (Fig. 5B) and then were analyzed for the expression of several markers by Western blotting. When compared with Lewis-X–negative cells, in normal EBs Lewis-X–positive cells contain much higher amounts of p27 (Fig. 5C) that may make Lewis-X–positive cells more vulnerable to a loss of p27. Within normal EBs, Lewis-X–positive cells, but not Lewis-X–negative cells, express Oct-3/4 (not shown). In the absence of p27, the level of Oct-3/4 is lowered (Fig. 4D). Also, the levels of lineage-specific cadherins are slightly changed, with neural N-cadherin being upregulated and epithelial E-cadherin being downregulated (Fig. 4D). The level of nuclear protein poly(ADP-ribose) polymerase (PARP) that was used as a loading control has remained unchanged in p27-deficient Lewis-X–positive cells. Taken together, these data strongly suggest that the Lewis-X–positive populations in normal and p27-deficient EBs are not identical. The cluster-forming population (most abundant in normal EBs) that has a high level of Oct-3/4 (approximately one fourth of the level typical for undifferentiated mESCs, not shown) most likely represents remnants of undifferentiated ESCs, whereas the cavity-forming population (most abundant in p27-deficient EBs) seems to have a neural phenotype and thus may represent neural progenitor/ stem cells.
Nestin Is Expressed in the Cavities of p27-Deficient EBs
Lewis-X was previously found to be associated with certain regions of the developing nervous system , and it was also identified as a marker of adult neural stem cells . To further address whether Lewis-X–positive cavity-forming cells could represent neural progenitors, the expression of nestin was determined in the set of p27-deficient 10-day-old EBs. Nestin-positive cells were clearly found in the cavities of p27-deficient EBs (Fig. 6A), again suggesting Lewis-X–positive cells that are associated with cavities to be neural progenitors/stem cells.
Figure 6. Neural differentiation of normal and p27-deficient EBs. (A): EBs that were obtained by 10-day-long differentiation of p27-deficient mESCs were fixed, sectioned, and immunohistochemi-cally stained by antibody against nestin (green) and PI (red). (B, C): Mouse E10.5 embryos were fixed and coronal cryosections of embryonal brain were stained with Lewis-X–specific and nestin-specific antibody, respectively (red). Cell nuclei were counterstained with Hoechst (blue). The pattern of Lewis-X (B) and nestin (C) expression in ventral cerebral cortex is shown. Nestin-positive processes coming from ventricular area are indicated by arrows. (D): Normal and p27-deficient mESCs were differentiated according to protocol (for details, see Materials and Methods). Normal and p27-deficient mESCs (D0) and EBs were harvested at day 8 (D8), day 15 (D15), and day 25 (D25) of differentiation, and the expression of neural markers N-CAM, GAP-43, and GFAP was analyzed by Western blotting (E). Abbreviations: EB, embryoid body; mESC, mouse embryonic stem cell; PARP, poly (ADP-ribose) polymerase; PI, propidium iodide.
To identify cells that could represent in vivo analogs of nestin/Lewis-X–positive cells in the cavities of p27-deficient EBs, serial sections were made from the brain of E10.5 mouse embryos and were analyzed for the expression of Lewis-X and nestin. Lewis-X expression was localized to the surface of ventricles in certain parts of the embryonic brain (cerebral cortex, caudal part of dorsal telencephalon, ventral lining of fourth ventricle, and third ventricle lining). Although the expression of nestin was more widespread throughout the brain and the regions of Lewis-X and nestin expression were not completely overlapped, we clearly observed the regions where double-positive cells were residing. As demonstrated on coronal sections of ventral cerebral cortex (Fig. 6B), Lewis-X staining was highly polarized toward the ventricle and resembled the staining of the EB cavities (Fig. 4). In the corresponding region of the brain, we detected nestin-positive processes coming from the ventricular zone (Fig. 6C). These processes most likely belong to the radial glial cells that have their nuclei in the Lewis-X–positive zone and have been shown previously to be positive for nestin . Although similar processes positive for nestin are not developed in the EBs, presumably because the precise spatial organization of the neural tube is lost in the EBs, we hypothesize that the Lewis-X–positive cavities in EBs may represent in vitro analogs of embryonic ventricles.
Neural Differentiation of p27-Deficient ESCs Is Not Affected
This led us to hypothesize that alterations occurring to Lewis-X–positive cells in p27-deficient EBs may be reflected by some abnormalities in their neurogenic potential. To address this issue, mESCs of both genotypes were allowed to differentiate under the conditions that promote neural differentiation (Fig. 6D). Resulting cells were harvested at days 8, 15, and 25 of differentiation and were Western blotted for the expression of the following neural markers: N-CAM (neuron/glia-specific), GAP-43 (neuron/ glia-specific), and GFAP (astrocyte-specific). As demonstrated in Figure 6E, p27-deficient mESCs showed no apparent abnormality in the expression of any of these neural markers compared with their normal counterparts. In other words, despite the molecular characteristics and organization of Lewis-X–positive cells in EBs produced from p27-deficient mESCs, neurogenic potential of p27-deficient mESCs does not seem to be affected.
DISCUSSION
Together, the experiments included in this study recognize p27 as a cell type–specific regulator of both proliferation and differentiation/survival of mESCs differentiating in culture. In some progenitor cell populations, p27 may serve to prevent uncontrolled growth, as for example of neural progenitors positive for Lewis-X epitope, whereas in other cell populations, such as TROMA-I–positive, p27 is necessary for correct differentiation and/or survival. From a general point of view, our study suggests that the emerging concept of a dual role for p27 also applies to ESCs and their progeny.
ACKNOWLEDGMENTS
Sherr CJ, Roberts JM. Inhibitors of mammalian G1 cyclin-dependent kinases. Genes Dev 1995;9:1149–1163.
Fero ML, Rivkin M, Tasch M et al. A syndrome of multiorgan hyperplasia with features of gigantism, tumorigenesis, and female sterility in p27Kip1-deficient mice. Cell 1996;85:733–744.
Kiyokawa H, Kineman RD, Manova-Todorova KO et al. Enhanced growth of mice lacking the cyclin-dependent kinase inhibitor function of p27Kip1. Cell 1996;85:721–732.
Nakayama K, Ishida N, Shirane M et al. Mice lacking p27Kip1 display increased body size, multiple organ hyperplasia, retinal dysplasia and pituitary tumors. Cell 1996;85:707–720.
Tong W, Kiyokawa H, Soos TJ et al. The absence of p27Kip1, an inhibitor of G1 cyclin-dependent kinases, uncouples differentiation and growth arrest during the granulosa->luteal transition. Cell Growth Differ 1998;9:787–794.
de Koning JP, Soede-Bobok AA, Ward AC et al. STAT3-mediated differentiation and survival and of myeloid cells in response to granulocyte colony-stimulating factor: role for the cyclin-dependent kinase inhibitor p27(Kip1). Oncogene 2000;19:3290–3298.
Muraoka RS, Lenferink AEG, Simpson J et al. Cyclin-dependent kinase inhibitor p27Kip1 is required for mouse mammary gland morphogenesis and function. J Cell Biol 2001;153:917–931.
Lin J, Della-Fera A, Li C et al. P27 knockout mice: reduced myostatin in muscle and altered adipogenesis. Biochem Biophys Res Commun 2003;300:938–942.
McAllister SS, Becker-Hapak M, Pintucci G et al. Novel p27kip1 C-terminal scatter domain mediates Rac-dependent cell migration independent of cell cycle arrest functions. Mol Cell Biol 2003;23:216–228.
Bryja V, Pachernik J, Faldíková L et al. A role of p27Kip1 in maintaining the levels of D-type cyclins in vivo. Biochim Biophys Acta 2004;1691:105–116.
Savatier P, Lapillonne H, van Grunsven LA et al. Withdrawal of differentiation inhibitory activity/leukemia inhibitory factor up-regulates D-type cyclins and cyclin-dependent kinase inhibitors in mouse embryonic stem cells. Oncogene 1995;12:309–322.
Stead E, White J, Faast R et al. Pluripotent cell division cycles are driven by ectopic Cdk2, cyclin A/E and E2F activities. Oncogene 2002;21:8320–8333.
Faast R, White J, Cartwright P et al. Cdk6-cyclin D3 activity in murine ES cells is resistant to inhibition by p16INK4a. Oncogene 2004;23:491–502.
Bryja V, Pacherník J, Souek K et al. Increased apoptosis in differentiating p27-deficient embryonic stem cells. Cell Mol Life Sci 2004;61:1384–1400.
Hogan B, Beddington R, Costantini F et al. Manipulating the Mouse Embryo: A Laboratory Manual. New York: Cold Spring Harbor Laboratory Press, 1994.
Coucouvanis E, Martin GR. Signals for death and survival: a two step mechanism for cavitation in the vertebrate embryo. Cell 1995;83:279–287.
Desbaillets I, Ziegler U, Groscurth P et al. Embryoid bodies: an in vitro model of mouse embryogenesis. Exp Physiol 2000;85:645–651.
Bain G, Kitchens D, Yao M et al. Embryonic stem cells express neuronal properties in vitro. Dev Biol 1995;168:342–357.
Tole S, Kaprielian Z, Ou SK et al. FORSE-1: a positionally regulated epitope in the developing rat central nervous system. J Neurosci 1995;15:957–969.
Allendoerfer KL, Magnani JL, Patterson PH. FORSE-1, an antibody that labels regionally restricted subpopulations of progenitor cells in the embryonic central nervous system, recognizes the Le(x) carbohydrate on a proteoglycan and two glycolipid antigens. Mol Cell Neurosci 1995;6:381–395.
Draber P, Pokorna Z. Differentiation antigens of mouse teratocarcinoma stem cells defined by monoclonal antibodies. Cell Differ 1984;15:109–113.
Solter D, Knowles BB. Monoclonal antibody defining a stage-specific mouse embryonic antigen (SSEA-1). Proc Natl Acad Sci U S A 1978;75:5565–5569.
Ashwell KWS, Mai JK. Developmental expression of the CD15 epitope in the hippocampus of the mouse. Cell Tissue Res 1997;289:17–23.
Ashwell KWS, Mai JK. Developmental expression of the CD15-epitope in the brainstem and spinal cord of the mouse. Anat Embryol (Berl) 1997;196:13–25.
Capela A, Temple S. LeX/ssea-1 is expressed by adult mouse CNS stem cells, identifying them as nonependymal. Neuron 2002;35:865–875.
Noctor SC, Flint AC, Weissman TA et al. Dividing precursor cells of the embryonic cortical ventricular zone have morphological and molecular characteristics of radial glia. J Neurosci 2002;22:3161–3173.
Lendahl U, Zimmerman LB, McKay RD. CNS stem cells express a new class of intermediate filament protein. Cell 1990;60:585–595.
Duprey P, Morello D, Vasseur M et al. Expression of the cytokeratin endo A gene during early mouse embryogenesis. Proc Natl Acad Sci U S A 1985;82:8535–8539.
Zhang P, Wong C, DePinho RA et al. Cooperation between the Cdk inhibitors p27KIP1 and p57KIP2 in the control of tissue growth and development. Genes Dev 1998;12:3162–3167.
Cheng T, Rodrigues N, Dombkowski D et al. Stem cell repopulation efficiency but not pool size is governed by p27kip1. Nat Med 2000;6:1235–1240.
Doetsch F, Verdugo JM-G, Caille I et al. Lack of the cell cycle inhibitor p27Kip1 results in selective increase of transit-amplifying cells for adult neurogenesis. J Neurosci 2002;22:2255–2264.
Casaccia-Bonnefil P, Hardy RJ, Teng KK et al. Loss of p27Kip1 function results in increased proliferative capacity of oligodendrocyte progenitors but unaltered timing of differentiation. Development 1999;126:4027–4037.
Casaccia-Bonnefil P, Tikoo R, Kiyokawa H et al. Oligodendrocyte precursor differentiation is perturbed in the absence of the cyclin-dependent kinase inhibitor p27Kip1. Genes Dev 1997;11:2335–2346.
Esner M, Pachernik J, Hampl A et al. Targeted disruption of fibroblast growth factor receptor-1 blocks maturation of visceral endoderm and cavitation in mouse embryoid bodies. Int J Dev Biol 2002;46:817–825.
Fox NW, Damjanov I, Martinez-Hernandez A et al. Immunohistochemical localization of early embryonic antigen (SSEA-1) in postimplantation mouse embryos and fetal and adult tissues. Dev Biol 1981;83:391–398.
Jackson M, Krassowska A, Gilbert N et al. Severe global DNA hypomethylation blocks differentiation and induces histone hyperacetylation in embryonic stem cells. Mol Cell Biol 2004;24:8862–8871.(Vítzslav Bryjaa,b, Luká a)
b Department of Molecular Embryology, Institute of Experimental Medicine, Academy of Sciences of the Czech Republic, Prague, Czech Republic;
c Laboratory of Molecular Embryology, Mendel University Brno, Brno, Czech Republic;
d Laboratory of Molecular Neurobiology, Department of Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden;
e Laboratory of Cytokinetics, Institute of Biophysics, Academy of Sciences of the Czech Republic, Brno, Czech Republic
Key Words. Mouse embryonic stem cells ? p27 ? Embryoid bodies ? Lewis-X ? Apoptosis
Correspondence: Ale Hampl, D.V.M., Ph.D., Laboratory of Molecular Embryology, Mendel University Brno, Zemdlská1, 61300 Brno, Czech Republic. Telephone: 420-5-45133297; Fax: 420-5-45133357; e-mail: hampl@mendelu.cz
ABSTRACT
Cell cycle regulatory protein p27Kip1 (from this point referred to as p27) belongs to a Cip/Kip family of cyclin-dependent kinase (CDK) inhibitors. Because its best-studied function is inhibiting the activity of cyclin-CDK complexes that drive the progression of the cell cycle, p27 is mostly understood as a negative regulator of cell proliferation (for review, see ). The phenotype of p27-null animals, which are significantly larger due to a higher number of cells, supports such a view . However, detailed analyses of p27–/– animals have also revealed defects that could not be explained by simple alterations in cell proliferation. Specifically, it was shown that processes such as cellular differentiation and apoptosis were severely affected in p27-deficient mice .
Embryonic stem cells (ESCs) are unique in terms of molecules and mechanisms used to regulate their undifferentiated growth . In regard to this fact, we have recently found that rather than for regulation of proliferation of pluripotent mouse ESCs (mESCs), p27 is essential for the in vitro differentiation of such cells to proceed normally. Specifically, when p27-deficient mESCs grown in monolayer are induced to differentiate into extraembryonal endoderm by withdrawal of leukemia inhibitory factor (LIF) combined with retinoic acid treatment, most of these mESCs die by apoptosis before finishing their differentiation program . Although this phenomenon is well-pronounced in vitro, its relevance to the development of the early embryo is limited due to the used differentiation system. It is well accepted that early developmental processes can be mimicked in vitro by culturing multicellular aggregates of mESCs in the absence of LIF and feeder cells . Under such conditions, mESC aggregates give rise to simple embryoid bodies (EBs) containing an outer layer of endodermal cells and a solid core of undifferentiated ectodermal cells. The inner cells of simple EBs subsequently undergo the wave of programmed cell death to form cystic EBs, a process called cavitation . Because cystic EBs are cultured in vitro, they provide a unique tool for the analysis of the impact of individual genes and proteins on proliferation, differentiation, and apoptosis in early development (for review, see ).
In this study, the phenotypes of EBs produced from normal and p27-deficient mESCs were analyzed in detail to improve our understanding of the significance of p27 for early embryogenesis. The data here show that in the absence of p27, the EBs reach a bigger size than their normal counterparts, most likely due to enhanced cavitation affecting primarily Lewis-X–positive cells.
MATERIALS AND METHODS
p27-Deficient EBs Grow Bigger Than Their Normal Counterparts
Because p27 is known primarily for its proliferation-regulating properties, it was first necessary to determine whether the absence of p27 results in the alteration of growth of EBs. EBs were allowed to form from both normal and p27–/– mESCs, and they were digitally photographed at day 10 of their development. The average diameter of EBs in each wild-type and p27–/– group was then determined using image-analysis software while assuming the EBs to be of round shape. Because data of both normal and p27-deficient cells showed log-normal distribution (Kolmogorov-Smirnov test; p = .02 for p27+/+ data and p = 1.5 x 10–8 for p27–/– data), they were ln-transformed (Kolmogorov-Smirnov test of ln-transformed data; p = .49 for p27+/+ data and p = .07 for p27–/– data) and compared by Student’s t-test. As demonstrated in Figure 1, p27-deficient EBs grew significantly bigger than normal EBs (two-tailed Student’s t-test of ln-transformed data sets; p = 8.8 x 10–16).
Figure 1. Effect of p27 on size of EBs. Normal and p27-deficient EBs were cultured up to day 10, photographed, and analyzed by image analysis software. Typical EBs are shown (A), and statistical analysis (B) of diameter of EBs (box-and-whisker plot; median, 25% to 75% interval, minimal and maximal values) is presented. Abbreviations: EB, embryoid body; K-S, Kolmogorov-Smirnov test of normal distribution of data.
The Activities of CDKs Are Not Deregulated in p27–/– EBs
Increased activities of CDK2 and CDK4 resulting in hyperplasia of various tissues or organs are typical for mice that are deficient in p27 . Therefore, it was important to determine whether the bigger size of p27–/– EBs observed here was correlated with upregulated CDK activities. Undifferentiated mESCs and EBs at days 5, 10, and 20 of their development, both wild-type and p27-deficient, were analyzed to address this question. The following parameters were determined: (a) levels of CDK2, CDK4, cyclin A, cyclin E, all three D-type cyclins, and p27; (b) kinase activities associated with CDK2, CDK4, cyclin A, and cyclin E; and (c) the amounts of p27 associated with the regulators listed under (b). During the development of wild-type EBs, the amount of p27 physically bound to CDK2, CDK4, cyclin A, and cyclin E, respectively, dramatically increases, corresponding to the elevation of the total amount of p27 (Fig. 2). Except for a generally lower amount of cyclin D3 in p27-deficient mESCs and EBs that can be well explained by the molecular mechanism described by us previously , the behavior of no other cyclin was significantly affected by the absence of p27 (Fig. 2). Total amounts of all cyclin A, cyclin E, cyclin D1, and cyclin D3 were downregulated with the progressing development of EBs (Fig. 2). On the other hand, cyclin D2 followed the opposite pattern, being undetectable in mESCs and increased in EBs (Fig. 2). Downregulation of the kinase activities of CDK2, cyclin A, cyclin E, and CDK4 took place in developing EBs irrespectively of their p27 genotype (Fig. 2). These data document that the abnormal growth of EBs produced from p27-deficient mESCs cannot be due to hyperactivation of major CDKs. Furthermore, the results suggest that downregulated cyclins rather than elevated p27 are responsible for the decline of CDK activities that accompanies EB development.
Figure 2. Analysis of cyclin-CDK complexes in normal and p27-deficient EBs. EBs were differentiated from normal and p27-deficient mES cells (D0) and lysed at day 5 (D5), day 10 (D10), and day 20 (D20) of differentiation. Cell lysates were used to immunoprecipitate CDK2, cyclin A, cyclin E, and CDK4. The kinase activities toward histone H1 or GST-Rb were determined by autoradiography. The amount of p27 in immunoprecipitates was determined by Western blotting. The total amount of CDK2, cyclin A, cyclin E, CDK4, D-type cyclins, and p27 as determined by Western blotting is also shown. Densitometry was used to assess the intensity of autoradiographic signal. The average density of p27+/+ (D0) sample was defined as 1.0, and from this value all other values were calculated. Data represent the means, with standard deviations indicated by error bars. Western blots and autoradiograms are representative of at least three independent replicates. Abbreviations: EB, embryoid body; IP, immunoprecipitate; WB, Western blotting.
The Absence of p27 Prevents Normal Expression of Cytokeratin Endo-A (TROMA-I) in Developing EBs
We have previously shown that p27-deficient mESCs suffer from apoptosis when they are induced to differentiate into TROMA-I–positive extraembryonal endoderm in monolayer culture . When wild-type mESCs were allowed to differentiate into EBs, strong positivity for TROMA-I was detected by Western blot as early as at day 5 of differentiation and stayed unchanged until the end of culture at day 20 (Fig. 3A). In contrast, only very low expression of TROMA-I was reached in EBs produced from p27-deficient mESCs, although mESCs of both genotypes progressed in their differentiation, as documented by downregulation of the stem cell marker Oct-4 (Fig. 3A). We next proceeded to determine whether a low proportion of TROMA-I–positive EBs, or decreased numbers of TROMA-I–positive cells in EBs, underlay the lower levels of cytokeratin endo-A determined above. Immunohistochemical visualization of cytokeratin endo-A on sectioned EBs produced by 10-day differentiation of normal and p27–/– mESCs was used to address this question. More than 90% of EBs of wild-type genotype contained TROMA-I–positive cells, with these cells being abundantly distributed in the superficial layer of the EBs (Figs. 3B, 3C). In contrast, only approximately 10% of p27–/– EBs contained TROMA-I–positive cells, which were invariably restricted to only dispersed individual cells (Fig. 3C).
Figure 3. Analysis of expression of TROMA-I in normal and p27-deficient EBs. (A): EBs were differentiated from normal and p27-deficient mouse embryonic stem cells (D0) and harvested at day 5 (D5), day 10 (D10), and day 20 (D20) of differentiation. The expression of Oct-4 and TROMA-I was analyzed by Western blotting. (B): EBs (day 10) were fixed, sectioned, and immunohistochemically stained using antibody against TROMA-I (green) and by PI (red). Typical pattern of TROMA-I staining for p27+/+ and p27–/– EBs is shown. (C): EBs containing at least some TROMA-I–positive cells were considered as TROMA-I–positive, and their proportion within total number of EBs was calculated. n indicates number of analyzed EBs. Abbreviations: EB, embryoid body; PI, propidium iodide; TROMA-I, cytokeratin endo-A.
Lewis-X–Positive Cells Form Cavities in p27-Deficient EBs
The abnormality in development of TROMA-I–positive cells itself could not explain the overgrowth of EBs lacking p27. Therefore, the question remained whether the bigger size of EBs may mirror an abnormal expansion of defined cell types. We have collected a large selection of antibodies against lineage-specific markers and applied them on the sectioned EBs as above to address this question. Interestingly, of all tested markers, only the early neural antigen identified by the antibody FORSE-1 revealed a significant difference between normal and p27-deficient EBs. Specifically, while in normal EBs FORSE-1 positivity was limited to clusters of 10 to 50 cells, under p27–/– conditions, the FORSE-1 positivity was redistributed to both surround and fill the cavities in most EBs (Fig. 4A). Such redistribution was invariably accompanied by a loss of FORSE-1–positive cell clusters in p27–/– EBs. A similar phenotype was observed in all three tested p27-deficient mESC lines.
Figure 4. Expression of FORSE-1 and TEC-01 in normal and p27-deficient EBs. EBs were differentiated from normal and p27-deficient mouse embryonic stem cells, and at day 10 they were fixed, sectioned, and immunohistochemically stained using antibody against FORSE-1 and TEC-01, respectively (green), and PI (red). (A): Typical patterns of FORSE-1 staining are shown. Depending on distribution of FORSE-1–positive cells, two phenotypes of FORSE-1–positive EBs were distinguished: cluster (typical for p27+/+ EBs; EB contains a compact aggregate of 5 to 100 FORSE-1–positive cells without a cell-free space inside the group of positive cells) and cavity (typical for p27–/– EBs; area of FORSE-1–positive cells is characterized by the presence of cell-free space and expression of FORSE-1 is often polarized, having its maximum in lumen of the cavity). Proportion of individual phenotypes of FORSE-1–positive EBs is presented as percentages of EBs typical by clusters and cavities, respectively. n indicates number of analyzed EBs for each genotype. Data represent the means, with standard deviations indicated by error bars. (B): Serial sections of EBs were stained either for FORSE-1 or TEC-1 (green) and compared. Cell nuclei were stained with PI (red). Representative sections are shown. Abbreviations: EB, embryoid body; PI, propidium iodide.
The antigen detected by FORSE-1 antibody involves the oligosaccharide epitope called Lewis-X . However, as the structures that carry the Lewis-X epitope may modify antibody binding, we wanted to confirm the authenticity of the epitope by an independent reagent. We used the monoclonal antibody TEC-01 developed by Draber and Pokorna that is specific for the Lewis-X epitope (identical to the marker of pluripotent cells often referred to as SSEA-1 ). The specificity of FORSE-1 and TEC-01 antibodies was investigated on serial sections made from EBs produced by 10-day differentiation of normal and p27–/– mESCs. As demonstrated in Figure 4B, identical structures, both cells and extracellular matrix, were recognized by FORSE-1 and TEC-01 antibodies. Therefore, in the below text we refer to these structures as Lewis-X–positive.
Lewis-X–Positive Populations in Normal and p27-Deficient EBs Are Different from Each Other
To answer the question of whether increased proliferation within the Lewis-X–positive cell population was involved in establishing the observed phenotype, EBs at days 4, 5, 6, and 10 of their development were trypsinized and the proportion of Lewis-X–positive cells was determined by immunocytochemical staining of coverslip-immobilized cells. As demonstrated in Figure 5A, the number of Lewis-X–positive cells gradually decreased in both normal and p27-deficient EBs down to 3.5% at day 10 of differentiation. Interestingly, although the general trend to decrease was common to EBs of both genotypes, a slightly higher proportion of Lewis-X–positive cells was observable in p27-deficient EBs between days 4 and 6 of their development. This tendency was statistically significant at day 6 and suggests that higher proliferation of Lewis-X–positive cells may contribute to the formation of the cavities in p27-deficient EBs.
Figure 5. Properties of Lewis-X–positive cells in normal and p27-deficient EBs. (A): Normal and p27-deficient EBs were trypsinized at days 4, 5, 6, and 10 (D4, D5, D6, D10), adhered on microscopic cov-erslips, and stained with antibody against Lewis-X. A total of 1,000 cells at two independent regions of the coverslip was counted, and the proportion of Lewis-X–positive cells was determined. The graph represents means and standard deviations obtained from two independent experiments. Statistically significant difference between normal and p27-deficient cells is indicated by an asterisk (Student’s t-test, p < .05). (B): Normal and p27-deficient EBs were cultured until day 10, trypsinized, and stained with antibody against Lewis-X. Cell suspension was subjected to flow cytometric analysis, and Lewis-X–positive cells (region R2 in B) were sorted out. (C): The amounts of p27 in Lewis-X–positive (inside R2 region in B) and Lewis X–negative cells (outside R2 region in B) were analyzed by Western blotting. (D): The levels of p27, Oct-3/4, N-cadherin, E-cadherin, and PARP were determined in sorted Lewis-X–positive cells (region R2 in B) by Western blotting. This experiment was repeated twice, and, except for N-cadherin, the results of only one experiment are presented. Abbreviations: EB, embryoid body; PARP, poly (ADP-ribose) polymerase.
Still, the observed changes in the Lewis-X–positive population also suggest that qualitative differences may exist between the cluster- and cavity-forming Lewis-X–positive cells. To address this issue, Lewis-X–positive cells were sorted out using flow cytometry from 10-day-old EBs (Fig. 5B) and then were analyzed for the expression of several markers by Western blotting. When compared with Lewis-X–negative cells, in normal EBs Lewis-X–positive cells contain much higher amounts of p27 (Fig. 5C) that may make Lewis-X–positive cells more vulnerable to a loss of p27. Within normal EBs, Lewis-X–positive cells, but not Lewis-X–negative cells, express Oct-3/4 (not shown). In the absence of p27, the level of Oct-3/4 is lowered (Fig. 4D). Also, the levels of lineage-specific cadherins are slightly changed, with neural N-cadherin being upregulated and epithelial E-cadherin being downregulated (Fig. 4D). The level of nuclear protein poly(ADP-ribose) polymerase (PARP) that was used as a loading control has remained unchanged in p27-deficient Lewis-X–positive cells. Taken together, these data strongly suggest that the Lewis-X–positive populations in normal and p27-deficient EBs are not identical. The cluster-forming population (most abundant in normal EBs) that has a high level of Oct-3/4 (approximately one fourth of the level typical for undifferentiated mESCs, not shown) most likely represents remnants of undifferentiated ESCs, whereas the cavity-forming population (most abundant in p27-deficient EBs) seems to have a neural phenotype and thus may represent neural progenitor/ stem cells.
Nestin Is Expressed in the Cavities of p27-Deficient EBs
Lewis-X was previously found to be associated with certain regions of the developing nervous system , and it was also identified as a marker of adult neural stem cells . To further address whether Lewis-X–positive cavity-forming cells could represent neural progenitors, the expression of nestin was determined in the set of p27-deficient 10-day-old EBs. Nestin-positive cells were clearly found in the cavities of p27-deficient EBs (Fig. 6A), again suggesting Lewis-X–positive cells that are associated with cavities to be neural progenitors/stem cells.
Figure 6. Neural differentiation of normal and p27-deficient EBs. (A): EBs that were obtained by 10-day-long differentiation of p27-deficient mESCs were fixed, sectioned, and immunohistochemi-cally stained by antibody against nestin (green) and PI (red). (B, C): Mouse E10.5 embryos were fixed and coronal cryosections of embryonal brain were stained with Lewis-X–specific and nestin-specific antibody, respectively (red). Cell nuclei were counterstained with Hoechst (blue). The pattern of Lewis-X (B) and nestin (C) expression in ventral cerebral cortex is shown. Nestin-positive processes coming from ventricular area are indicated by arrows. (D): Normal and p27-deficient mESCs were differentiated according to protocol (for details, see Materials and Methods). Normal and p27-deficient mESCs (D0) and EBs were harvested at day 8 (D8), day 15 (D15), and day 25 (D25) of differentiation, and the expression of neural markers N-CAM, GAP-43, and GFAP was analyzed by Western blotting (E). Abbreviations: EB, embryoid body; mESC, mouse embryonic stem cell; PARP, poly (ADP-ribose) polymerase; PI, propidium iodide.
To identify cells that could represent in vivo analogs of nestin/Lewis-X–positive cells in the cavities of p27-deficient EBs, serial sections were made from the brain of E10.5 mouse embryos and were analyzed for the expression of Lewis-X and nestin. Lewis-X expression was localized to the surface of ventricles in certain parts of the embryonic brain (cerebral cortex, caudal part of dorsal telencephalon, ventral lining of fourth ventricle, and third ventricle lining). Although the expression of nestin was more widespread throughout the brain and the regions of Lewis-X and nestin expression were not completely overlapped, we clearly observed the regions where double-positive cells were residing. As demonstrated on coronal sections of ventral cerebral cortex (Fig. 6B), Lewis-X staining was highly polarized toward the ventricle and resembled the staining of the EB cavities (Fig. 4). In the corresponding region of the brain, we detected nestin-positive processes coming from the ventricular zone (Fig. 6C). These processes most likely belong to the radial glial cells that have their nuclei in the Lewis-X–positive zone and have been shown previously to be positive for nestin . Although similar processes positive for nestin are not developed in the EBs, presumably because the precise spatial organization of the neural tube is lost in the EBs, we hypothesize that the Lewis-X–positive cavities in EBs may represent in vitro analogs of embryonic ventricles.
Neural Differentiation of p27-Deficient ESCs Is Not Affected
This led us to hypothesize that alterations occurring to Lewis-X–positive cells in p27-deficient EBs may be reflected by some abnormalities in their neurogenic potential. To address this issue, mESCs of both genotypes were allowed to differentiate under the conditions that promote neural differentiation (Fig. 6D). Resulting cells were harvested at days 8, 15, and 25 of differentiation and were Western blotted for the expression of the following neural markers: N-CAM (neuron/glia-specific), GAP-43 (neuron/ glia-specific), and GFAP (astrocyte-specific). As demonstrated in Figure 6E, p27-deficient mESCs showed no apparent abnormality in the expression of any of these neural markers compared with their normal counterparts. In other words, despite the molecular characteristics and organization of Lewis-X–positive cells in EBs produced from p27-deficient mESCs, neurogenic potential of p27-deficient mESCs does not seem to be affected.
DISCUSSION
Together, the experiments included in this study recognize p27 as a cell type–specific regulator of both proliferation and differentiation/survival of mESCs differentiating in culture. In some progenitor cell populations, p27 may serve to prevent uncontrolled growth, as for example of neural progenitors positive for Lewis-X epitope, whereas in other cell populations, such as TROMA-I–positive, p27 is necessary for correct differentiation and/or survival. From a general point of view, our study suggests that the emerging concept of a dual role for p27 also applies to ESCs and their progeny.
ACKNOWLEDGMENTS
Sherr CJ, Roberts JM. Inhibitors of mammalian G1 cyclin-dependent kinases. Genes Dev 1995;9:1149–1163.
Fero ML, Rivkin M, Tasch M et al. A syndrome of multiorgan hyperplasia with features of gigantism, tumorigenesis, and female sterility in p27Kip1-deficient mice. Cell 1996;85:733–744.
Kiyokawa H, Kineman RD, Manova-Todorova KO et al. Enhanced growth of mice lacking the cyclin-dependent kinase inhibitor function of p27Kip1. Cell 1996;85:721–732.
Nakayama K, Ishida N, Shirane M et al. Mice lacking p27Kip1 display increased body size, multiple organ hyperplasia, retinal dysplasia and pituitary tumors. Cell 1996;85:707–720.
Tong W, Kiyokawa H, Soos TJ et al. The absence of p27Kip1, an inhibitor of G1 cyclin-dependent kinases, uncouples differentiation and growth arrest during the granulosa->luteal transition. Cell Growth Differ 1998;9:787–794.
de Koning JP, Soede-Bobok AA, Ward AC et al. STAT3-mediated differentiation and survival and of myeloid cells in response to granulocyte colony-stimulating factor: role for the cyclin-dependent kinase inhibitor p27(Kip1). Oncogene 2000;19:3290–3298.
Muraoka RS, Lenferink AEG, Simpson J et al. Cyclin-dependent kinase inhibitor p27Kip1 is required for mouse mammary gland morphogenesis and function. J Cell Biol 2001;153:917–931.
Lin J, Della-Fera A, Li C et al. P27 knockout mice: reduced myostatin in muscle and altered adipogenesis. Biochem Biophys Res Commun 2003;300:938–942.
McAllister SS, Becker-Hapak M, Pintucci G et al. Novel p27kip1 C-terminal scatter domain mediates Rac-dependent cell migration independent of cell cycle arrest functions. Mol Cell Biol 2003;23:216–228.
Bryja V, Pachernik J, Faldíková L et al. A role of p27Kip1 in maintaining the levels of D-type cyclins in vivo. Biochim Biophys Acta 2004;1691:105–116.
Savatier P, Lapillonne H, van Grunsven LA et al. Withdrawal of differentiation inhibitory activity/leukemia inhibitory factor up-regulates D-type cyclins and cyclin-dependent kinase inhibitors in mouse embryonic stem cells. Oncogene 1995;12:309–322.
Stead E, White J, Faast R et al. Pluripotent cell division cycles are driven by ectopic Cdk2, cyclin A/E and E2F activities. Oncogene 2002;21:8320–8333.
Faast R, White J, Cartwright P et al. Cdk6-cyclin D3 activity in murine ES cells is resistant to inhibition by p16INK4a. Oncogene 2004;23:491–502.
Bryja V, Pacherník J, Souek K et al. Increased apoptosis in differentiating p27-deficient embryonic stem cells. Cell Mol Life Sci 2004;61:1384–1400.
Hogan B, Beddington R, Costantini F et al. Manipulating the Mouse Embryo: A Laboratory Manual. New York: Cold Spring Harbor Laboratory Press, 1994.
Coucouvanis E, Martin GR. Signals for death and survival: a two step mechanism for cavitation in the vertebrate embryo. Cell 1995;83:279–287.
Desbaillets I, Ziegler U, Groscurth P et al. Embryoid bodies: an in vitro model of mouse embryogenesis. Exp Physiol 2000;85:645–651.
Bain G, Kitchens D, Yao M et al. Embryonic stem cells express neuronal properties in vitro. Dev Biol 1995;168:342–357.
Tole S, Kaprielian Z, Ou SK et al. FORSE-1: a positionally regulated epitope in the developing rat central nervous system. J Neurosci 1995;15:957–969.
Allendoerfer KL, Magnani JL, Patterson PH. FORSE-1, an antibody that labels regionally restricted subpopulations of progenitor cells in the embryonic central nervous system, recognizes the Le(x) carbohydrate on a proteoglycan and two glycolipid antigens. Mol Cell Neurosci 1995;6:381–395.
Draber P, Pokorna Z. Differentiation antigens of mouse teratocarcinoma stem cells defined by monoclonal antibodies. Cell Differ 1984;15:109–113.
Solter D, Knowles BB. Monoclonal antibody defining a stage-specific mouse embryonic antigen (SSEA-1). Proc Natl Acad Sci U S A 1978;75:5565–5569.
Ashwell KWS, Mai JK. Developmental expression of the CD15 epitope in the hippocampus of the mouse. Cell Tissue Res 1997;289:17–23.
Ashwell KWS, Mai JK. Developmental expression of the CD15-epitope in the brainstem and spinal cord of the mouse. Anat Embryol (Berl) 1997;196:13–25.
Capela A, Temple S. LeX/ssea-1 is expressed by adult mouse CNS stem cells, identifying them as nonependymal. Neuron 2002;35:865–875.
Noctor SC, Flint AC, Weissman TA et al. Dividing precursor cells of the embryonic cortical ventricular zone have morphological and molecular characteristics of radial glia. J Neurosci 2002;22:3161–3173.
Lendahl U, Zimmerman LB, McKay RD. CNS stem cells express a new class of intermediate filament protein. Cell 1990;60:585–595.
Duprey P, Morello D, Vasseur M et al. Expression of the cytokeratin endo A gene during early mouse embryogenesis. Proc Natl Acad Sci U S A 1985;82:8535–8539.
Zhang P, Wong C, DePinho RA et al. Cooperation between the Cdk inhibitors p27KIP1 and p57KIP2 in the control of tissue growth and development. Genes Dev 1998;12:3162–3167.
Cheng T, Rodrigues N, Dombkowski D et al. Stem cell repopulation efficiency but not pool size is governed by p27kip1. Nat Med 2000;6:1235–1240.
Doetsch F, Verdugo JM-G, Caille I et al. Lack of the cell cycle inhibitor p27Kip1 results in selective increase of transit-amplifying cells for adult neurogenesis. J Neurosci 2002;22:2255–2264.
Casaccia-Bonnefil P, Hardy RJ, Teng KK et al. Loss of p27Kip1 function results in increased proliferative capacity of oligodendrocyte progenitors but unaltered timing of differentiation. Development 1999;126:4027–4037.
Casaccia-Bonnefil P, Tikoo R, Kiyokawa H et al. Oligodendrocyte precursor differentiation is perturbed in the absence of the cyclin-dependent kinase inhibitor p27Kip1. Genes Dev 1997;11:2335–2346.
Esner M, Pachernik J, Hampl A et al. Targeted disruption of fibroblast growth factor receptor-1 blocks maturation of visceral endoderm and cavitation in mouse embryoid bodies. Int J Dev Biol 2002;46:817–825.
Fox NW, Damjanov I, Martinez-Hernandez A et al. Immunohistochemical localization of early embryonic antigen (SSEA-1) in postimplantation mouse embryos and fetal and adult tissues. Dev Biol 1981;83:391–398.
Jackson M, Krassowska A, Gilbert N et al. Severe global DNA hypomethylation blocks differentiation and induces histone hyperacetylation in embryonic stem cells. Mol Cell Biol 2004;24:8862–8871.(Vítzslav Bryjaa,b, Luká a)