- Email: [email protected]
In vitro differentiation of murine embryonic stem cells into keratinocyte-like cells
ARTICLE IN PRESS
European Journal of Cell Biology 86 (2007) 801–805 www.elsevier.de/ejcb
In vitro differentiation of murine embryonic stem cells into keratinocyte-like cells Ingo Haasea,, Renate Knaupa, Maria Wartenbergb, Heinrich Sauerc, Ju¨rgen Heschelerd, Gustav Mahrlea a
Department of Dermatology, University of Cologne and Center for Molecular Medicine, University of Cologne (CMMC), Joseph-Stelzmann-Strasse 9, D-50924 Cologne, Germany b Molekulare Kardiologie und Stammzellforschung, Klinik fu¨r Innere Medizin I, Universita¨tsklinikum Jena, D-07740 Jena, Germany c Physiologisches Institut der Justus-Liebig-Universita¨t, Aulweg 129, D-35392 Giessen, Germany d Center for Physiology, University of Cologne, Robert-Koch-Strasse 39, D-50931 Cologne, Germany Received 5 October 2006; received in revised form 22 June 2007; accepted 2 July 2007
Abstract Embryonic stem (ES) cells are omnipotent; they can differentiate into every cell type of the body. The development of culture conditions that allow their differentiation has made it conceivable to produce large numbers of cells with lineage-specific characteristics in vitro. Here, we describe a method by which murine ES cells can be differentiated into cells with characteristics of epidermal keratinocytes. Keratinocyte-like cells were isolated from embryoid bodies and grown in culture. Potential applications of this method are the in vitro differentiation of cells of interest from ES cells of mice with lethal phenotypes during embryonic development and the production of genetically modified epidermal keratinocytes that could be used as temporary wound dressing or as carriers of genes of interest in gene therapeutic treatments. r 2007 Elsevier GmbH. All rights reserved. Keywords: Keratinocytes; Embryonic stem cells; Embryoid bodies; Differentiation
Introduction Keratinocytes are epithelial cells with the capacity to form a stratified epithelium by undergoing a maturation process that results in the assembly of cornified scales. They express epithelial cadherin (E-cadherin) and, in their immature, proliferative state, cytokeratin 14 (K14). E-cadherin is a classical type I cadherin and a key regulator of epithelial cell–cell adhesion in embryonic Corresponding author. Tel.: +49 221 478 6589; fax: +49 221 478 5949. E-mail address: [email protected] (I. Haase).
0171-9335/$ - see front matter r 2007 Elsevier GmbH. All rights reserved. doi:10.1016/j.ejcb.2007.07.001
and adult tissues. Since it is involved in cell–cell sorting and apoptosis, it is essential for embryonic morphogenesis. E-cadherin is detected in both epidermis and periderm already during the bi-layered epithelial stage (E13) when the ectoderm gives rise to periderm and epidermis (Hardy and Vielkind, 1996). K14, a type I keratin, is constitutively present in basal keratinocytes of the epidermis and stratified epithelia, and in the outer root sheath of the hair follicle. In normal mouse skin, K14 appears as early as the single ectodermal stage (E9.5) and stays in the basal layer after stratification (Byrne et al., 1994). Stratification of epidermal keratinocytes is accompanied by the synthesis
ARTICLE IN PRESS 802
I. Haase et al. / European Journal of Cell Biology 86 (2007) 801–805
of differentiation-specific keratins (e.g. keratins 1, 10) and precursor proteins of the cornified envelope such as involucrin and loricrin (Fuchs, 1990). Embryoid bodies (EBs) formed by aggregation of murine embryonic stem (ES) cells have been widely used for in vitro differentiation into certain lineages, e.g. myocytes and neurons (Guan et al., 1999). Keratinocytes have been differentiated from murine ES cells without EB formation by seeding ES cells onto extracellular matrix and adding bone morphogenetic protein-4 to the culture medium (Coraux et al., 2003). Emigrating cells from human EBs have been shown to give rise to keratinocytes when grown under supportive conditions (Green et al., 2003). We here use EBs derived from cultured murine ES cells to produce keratinocytelike cells in vitro. We show that these keratinocyte-like cells develop on the surface of intact EBs and can be isolated from them and kept in culture.
Materials and methods Maintenance of ES cells ES cells of the line CCE (Robertson et al., 1986), kindly provided by Prof. Max Gassmann, Zurich, were grown on inactivated primary murine embryonic fibroblasts in Iscove’s medium supplemented with 20% heatinactivated (56 1C, 30 min) foetal calf serum (FCS), 2 mM glutamine, 1% minimum essential medium (MEM)-nonessential amino acids, 100 IU/ml penicillin, 100 mg/ml streptomycin (all from PAA, Linz, Austria), 100 mM beta-mercaptoethanol (Sigma, St. Louis, MO), and 10 ng/ml leukemia-inhibiting factor (LIF; ESGROChemicon, Temecula, CA), in a humidified incubator with 5% CO2 at 37 1C, and passaged every 3 days.
Generation of embryoid bodies EBs were generated by two different methods: the spinner flask (Ateghang et al., 2006) and the hanging drop method (Wobus et al., 1991). In brief, for the spinner flask technique, adherent ES cells and feeder cells were enzymatically dissociated 3 days after plating using 0.2% trypsin/0.05% EDTA in phosphate-buffered saline (PBS) (PAA). Cells (1 107/ml) were seeded into a 250-ml siliconized spinner flask (Integra Biosciences, Fernwald, Germany) containing 100 ml Iscove’s medium, with the same supplements as mentioned above, except for LIF. The stirrer was set to 22.5 rpm, using a perpendicular stirring mode. After 24 h, 150 ml medium was added to give a total volume of 250 ml. About half of the culture medium was exchanged every day. After 5–7 days, EBs were harvested and either analyzed directly for the
presence of markers of squamous epithelial cells or plated onto plastic cultures dishes and kept as adherent cultures for different lengths of time and then used for immunostaining or lysed for Western blot analysis. For the generation of EBs in hanging drops we used 600 ES cells/20 ml Iscove’s medium (containing the above-mentioned supplements but without LIF). The droplets were carefully placed on the lid of a 10-cm dish, turned face down in the dish and left there for 3–4 days. One droplet yielded one EB. EBs were plated on cell culture dishes (Falcon, Franklin Lakes, NJ, USA) or chamber slides (Nalge Nunc, Napersville, IL) in FAD medium consisting of 300 ml DMEM, 90 ml HAM’s F12 (both PAA), 1.8 10 4 M adenine (Roche Diagnostics, Mannheim, Germany), 100 IU/ml penicillin, 100 mg/ml streptomycin (PAA), 10% heat-inactivated FCS (PAA), 0.5 mg/ml hydrocortisone (Calbiochem, La Jolla, CA), 5 mg/ml insulin (Sigma), 10 10 M cholera toxin (Calbiochem), 10 ng/ml epidermal growth factor (Progen, Heidelberg, Germany), 2 10 9 M 3-triiodo-thyroxine-Na salt (Serva, Heidelberg, Germany), 5 mg/ml transferrin (Calbiochem) for a minimum of 17 days and a maximum of 24 days. In some experiments, EBs were dissociated with 1 mg/ml collagenase III (Worthington, Lakewood, NJ) and extracted cells were reseeded on inactivated 3T3 murine fibroblasts as feeder cells.
Immunofluorescence staining The EBs grown on chamber slides were rinsed twice with PBS, fixed in 4% formaldehyde for 5 min at room temperature and rinsed again. At this point, we either let the chamber slides air-dry, later to be frozen at 20 1C or we continued immediately. EBs were permeabilized with 0.4% Triton X-100 (Serva) in PBS for 5 min, and then rinsed with PBS. Blocking against unspecific binding was carried out with 10% normal goat serum (Dako, Glostrup, Denmark) for 1 h at room temperature. The primary and secondary antibodies were both applied for 1 h at room temperature. For the primary antibodies, we used polyclonal rabbit anti-mouse K14 at 1 mg/ml (Covance, Richmond, CA) and a monoclonal rat anti-mouse E-cadherin at 10 mg/ml (Zymed, San Francisco, CA). Alexa Fluor-488- or Alexa Fluor-568coupled secondary antibodies were used for detection (Molecular Probes, Eugene, OR). Nuclear staining was performed using 4,6-diamidino-2-phenylindole-dilactate (DAPI) at 1 ng/ml (Sigma) or ToPro III (Molecular Probes). Specimens were analyzed using a Leica TCS upright confocal laser scanning microscope at excitation wavelengths of 488, 543 and 633 nm or with a Nikon Eclipse E800 microscope attached to a Nikon DXM 1200 F digital camera at a power of 200 or 400.
ARTICLE IN PRESS I. Haase et al. / European Journal of Cell Biology 86 (2007) 801–805
Western blotting EBs were lysed as a whole in lysis buffer, the lysates were centrifuged at 14,000g for 10 min, and the supernatant was used for Western blot analysis. Equal amounts of protein were separated by SDS–PAGE, blotted onto nitrocellulose LC 2000 membranes (Invitrogen, Carlsbad, USA) and probed with antibodies against K14 (Covance) and involucrin (kindly provided by F. Watt, London).
Results and discussion EBs were plated onto plastic cell culture dishes, cultured in FAD medium and harvested at different time points. At day 17, a clear separation of EBs into an inner cell mass and a multilayered epithelium that covered its surface could be recognized (Fig. 1(a)). This epithelium expressed E-cadherin as evidenced by immunofluorescence staining (Fig. 1(c–e)) and showed patches of K14 expression (Fig. 1(b), (d–f)), but was negative for keratin 10 or involucrin (data not shown). Morphologically, the multilayered epithelium resembled the developing epidermis that covers murine embryos after embryonic day 11 (M’Boneko and Merker, 1988). In whole mounts of day-18 EBs, K14-positive cells within the epithelial layer were detectable (Fig. 1(d–e)). By day 23, these single cells gave rise to islands of K14expressing epithelial cells which subsequently merged to form aggregates with tongue-shaped branches that seemed to sheathe the surface of the EB (Fig. 1(f)). Occasionally, giant K14-positive epithelial cells were detected that showed widely ramified networks of K14Fig. 1. Development of keratinocyte-like cells in EBs in situ. Haematoxylin/eosin (a) or immunofluorescence (b–f, h, i) staining and Western blot analysis (g) of tissue sections from EBs (a, b) or of total EBs (c–i). Images in (b–f, h, i) were obtained with a confocal microscope. (a) Detail from a section through an EB at day 17 showing the multilayered epithelium that covers its surface. (b) Tissue section through the surface epithelium of an EB stained for keratin 14 (K14; green). Nuclei are shown in red. (c–f) Confocal images of the surfaces of EB whole mounts obtained at day 17 (c), day 20 (d, e) and day 23 (f). Green staining: E-cadherin (c–e), K14 (f). Red staining: nuclei (c, f), K14 (d, e). (g) Western blot analysis of whole EB lysates with antibodies against K14 (upper panel) and involucrin (lower panel). O, P, Q, R, S indicate EBs from different experiments. EBs were analyzed at day 16 (O), 18 (P), 24 (Q), 38 (S), and day 46 (R). +: positive control (lysate from primary murine keratinocytes), : negative control (murine fibroblasts). Molecular weight markers are indicated to the right. (h, i) Confocal images of giant K14-expressing epithelial cells on the surface of an EB at day 20. The image in (i) is a cut out from the image in (h). Green staining: E-cadherin. Red staining: K14. Bars: 50 mm (a, c), 40 mm (f, i), 20 mm (b, d, e, h).
803
positive filaments (Fig. 1(h). These networks contained filaments of varying thickness with the thicker filaments seemingly inserting in membrane domains with clusters of E-cadherin (Fig. 1(i)). To verify expression of K14, we carried out Western blot analysis of whole EBs after different lengths of culture. Most EBs were positive for K14 between culture days 16 and 46 but showed a variable degree of expression (Fig. 1(g)). We also analyzed expression of involucrin, a constituent of the cornified keratinocyte envelope and a terminal differentiation marker (Li et al., 2000). EBs harvested at days 38 and 46 were found to contain involucrin (Fig. 1(g)). These results demonstrate that proteins characteristic of stratified epithelia and
ARTICLE IN PRESS 804
I. Haase et al. / European Journal of Cell Biology 86 (2007) 801–805
epidermis are expressed in EBs cultured under our conditions. Since we had found evidence for the expression of keratinocyte-specific proteins in EBs in situ, we hypothesized that isolation and culture of cells with characteristics of keratinocytes from these EBs should be possible. We therefore allowed EBs to form and differentiate until day 24 and then disaggregated them by enzymatic digestion with collagenase. Enzymatic digestion in addition to repeated aspirating through a narrow pipette tip yielded a cell suspension which was plated onto collagen I-coated cell culture dishes and kept in FAD medium with low calcium concentration. Most of the plated cells died within the first 2–3 days of culture, thus leaving space for rapidly growing colonies of cells that exhibited epithelial morphology (Fig. 2(a)). Immunofluorescence staining revealed that single cells within these colonies and, after several days of culture, entire colonies expressed K14 and E-cadherin (Fig. 2(b–e)). To investigate whether epithelial cells isolated from EBs and grown in culture were able to express markers of more highly differentiated epidermal keratinocytes, we stained cultures of EB-derived cells with an antibody against mouse involucrin (Li et al., 2000). This revealed single positive cells which were partly binucleated (Fig. 2(f–h)), a feature that has been observed before in differentiating epidermal keratinocytes (Gandarillas et al., 2000; Zanet et al., 2005). Although epithelial cells derived from EBs could be kept in culture for up to 3 weeks, we did not succeed in passaging them. The generation of an epidermis-like stratified epithelium from murine ES cells has been reported previously by Coraux et al. (2003). In this study, matrix secreted by dermal fibroblasts in association with the transforming growth factor beta (TGFb)-related bone morphogenetic protein-4 were sufficient to direct differentiation of ES cells towards the keratinocyte lineage without the requirement for EB formation. In our hands, the formation of EBs and their culture in FAD medium were sufficient to induce differentiation into keratinocyte-like cells. Therefore, differentiation towards epidermal precursor cells in vitro is supported by factors autonomously produced by the EB. Recently, EBs have also been used to generate keratinocyte precursors from human ES cells (Green et al., 2003; Ji et al., 2006). Although in this system K14-expressing keratinocyte precursors form independently of exogenously added factors, coating of culture dishes with collagen I or gelatine enhanced the yield of these cells. This suggests that the ability of EBs to give rise to keratinocyte precursor cells is self-contained and that the multiplication of such cells can be propagated by extracellular matrix molecules like gelatine and collagen. In addition to these previous reports with human ES cells, we were able to observe the formation of K14-expressing cells in
Fig. 2. Isolation of keratin 14 (K14)- and involucrin-expressing cells from EBs. Phase-contrast (a) and confocal fluorescence microscopic (b–d) images of cells isolated from EBs after day 18. (b–d) Cells were stained with an antibody against K14 (green); nuclei are shown in red. (e) Double staining for K14 (green) and E-cadherin (red). (f–h) Staining with an antibody against involucrin. Nuclei are shown in blue. Note single involucrin-positive, binucleated cells. Bars: 50 mm (a), 20 mm (b–d), 40 mm (e–h).
whole mounts of EBs in situ. Their formation occurred within a layer of epithelial cells covering the surface of the EBs and did not necessarily require outwards migration from the EB as observed previously (Green et al., 2003; Ji et al., 2006). Ji et al. (2006) reported that a defined serum-free medium (DSFM) was most efficient in generating keratinocyte precursors from human ES cells and that FAD medium did not result in detectable K14 expression within EBs. In our hands, FAD medium did support differentiation of keratinocyte-like cells from EBs derived from murine ES cells which is in line with results obtained by Green et al. (2003). One possible explanation for this difference is that the FAD medium in the study of Ji and co-workers was
ARTICLE IN PRESS I. Haase et al. / European Journal of Cell Biology 86 (2007) 801–805
supplemented with only 2.5% FCS whereas Green and colleagues used 5% FCS and we used 10% FCS. This makes it conceivable that serum may contain factors which promote differentiation of keratinocyte precursors from ES cells. In summary, we show here that EBs derived from ES cells can be used to produce keratinocyte-like cells in vitro. Several applications of this method are conceivable: modification of genes determining antigenic properties would eventually allow to generate epidermal keratinocytes with low antigenic potential from human ES cells that can be kept ‘‘ready to use’’, e.g. as a temporary dressing for large wounds in burn victims. Such keratinocytes could also be used as universal carriers of DNA constructs for gene therapeutic treatments (Gerrard et al., 1993). Finally, our protocol offers the possibility to generate cells of interest from murine ES cells with genetic modifications that would produce lethal phenotypes in mice. Several groups have now succeeded to generate keratinocytes from ES cells. In the future, we will have to learn how to keep these cells in long-term culture and how to make them applicable, e.g. to patients who require skin replacement.
References Ateghang, B., Wartenberg, M., Gassmann, M., Sauer, H., 2006. Regulation of cardiotrophin-1 expression in mouse embryonic stem cells by HIF-1alpha and intracellular reactive oxygen species. J. Cell Sci. 119, 1043–1052. Byrne, C., Tainsky, M., Fuchs, E., 1994. Programming gene expression in developing epidermis. Development 120, 2369–2383. Coraux, C., Hilmi, C., Rouleau, M., Spadafora, A., Hinnrasky, J., Ortonne, J.P., Dani, C., Aberdam, D., 2003. Reconstituted skin from murine embryonic stem cells. Curr. Biol. 13, 849–853. Fuchs, E., 1990. Epidermal differentiation: the bare essentials. J. Cell Biol. 111, 2807–2814.
805
Gandarillas, A., Davies, D., Blanchard, J.M., 2000. Normal and c-Myc-promoted human keratinocyte differentiation both occur via a novel cell cycle involving cellular growth and endoreplication. Oncogene 19, 3278–3289. Gerrard, A.J., Hudson, D.L., Brownlee, G.G., Watt, F.M., 1993. Towards gene therapy for haemophilia B using primary human keratinocytes. Nat. Genet. 3, 180–183. Green, H., Easley, K., Iuchi, S., 2003. Marker succession during the development of keratinocytes from cultured human embryonic stem cells. Proc. Natl. Acad. Sci. USA 100, 15625–15630. Guan, K., Schmidt, M.M., Ding, Q., Chang, H., Wobus, A.M., 1999. Embryonic stem cells in vitro – prospects for cell and developmental biology, embryotoxicology and cell therapy. Altex 16, 135–141. Hardy, M.H., Vielkind, U., 1996. Changing patterns of cell adhesion molecules during mouse pelage hair follicle development. 1. Follicle morphogenesis in wild-type mice. Acta Anat. (Basel) 157, 169–182. Ji, L., Allen-Hoffmann, B.L., de Pablo, J.J., Palecek, S.P., 2006. Generation and differentiation of human embryonic stem cell-derived keratinocyte precursors. Tissue Eng. 12, 665–679. Li, E.R., Owens, D.M., Djian, P., Watt, F.M., 2000. Expression of involucrin in normal, hyperproliferative and neoplastic mouse keratinocytes. Exp. Dermatol. 9, 431–438. M’Boneko, V., Merker, H.J., 1988. Development and morphology of the periderm of mouse embryos (days 9–12 of gestation). Acta Anat. (Basel) 133, 325–336. Robertson, E., Bradley, A., Kuehn, M., Evans, M., 1986. Germ-line transmission of genes introduced into cultured pluripotential cells by retroviral vector. Nature 323, 445–448. Wobus, A.M., Wallukat, G., Hescheler, J., 1991. Pluripotent mouse embryonic stem cells are able to differentiate into cardiomyocytes expressing chronotropic responses to adrenergic and cholinergic agents and Ca2+ channel blockers. Differentiation 48, 173–182. Zanet, J., Pibre, S., Jacquet, C., Ramirez, A., de Alboran, I.M., Gandarillas, A., 2005. Endogenous Myc controls mammalian epidermal cell size, hyperproliferation, endoreplication and stem cell amplification. J. Cell Sci. 118, 1693–1704.
European Journal of Cell Biology 86 (2007) 801–805 www.elsevier.de/ejcb
In vitro differentiation of murine embryonic stem cells into keratinocyte-like cells Ingo Haasea,, Renate Knaupa, Maria Wartenbergb, Heinrich Sauerc, Ju¨rgen Heschelerd, Gustav Mahrlea a
Department of Dermatology, University of Cologne and Center for Molecular Medicine, University of Cologne (CMMC), Joseph-Stelzmann-Strasse 9, D-50924 Cologne, Germany b Molekulare Kardiologie und Stammzellforschung, Klinik fu¨r Innere Medizin I, Universita¨tsklinikum Jena, D-07740 Jena, Germany c Physiologisches Institut der Justus-Liebig-Universita¨t, Aulweg 129, D-35392 Giessen, Germany d Center for Physiology, University of Cologne, Robert-Koch-Strasse 39, D-50931 Cologne, Germany Received 5 October 2006; received in revised form 22 June 2007; accepted 2 July 2007
Abstract Embryonic stem (ES) cells are omnipotent; they can differentiate into every cell type of the body. The development of culture conditions that allow their differentiation has made it conceivable to produce large numbers of cells with lineage-specific characteristics in vitro. Here, we describe a method by which murine ES cells can be differentiated into cells with characteristics of epidermal keratinocytes. Keratinocyte-like cells were isolated from embryoid bodies and grown in culture. Potential applications of this method are the in vitro differentiation of cells of interest from ES cells of mice with lethal phenotypes during embryonic development and the production of genetically modified epidermal keratinocytes that could be used as temporary wound dressing or as carriers of genes of interest in gene therapeutic treatments. r 2007 Elsevier GmbH. All rights reserved. Keywords: Keratinocytes; Embryonic stem cells; Embryoid bodies; Differentiation
Introduction Keratinocytes are epithelial cells with the capacity to form a stratified epithelium by undergoing a maturation process that results in the assembly of cornified scales. They express epithelial cadherin (E-cadherin) and, in their immature, proliferative state, cytokeratin 14 (K14). E-cadherin is a classical type I cadherin and a key regulator of epithelial cell–cell adhesion in embryonic Corresponding author. Tel.: +49 221 478 6589; fax: +49 221 478 5949. E-mail address: [email protected] (I. Haase).
0171-9335/$ - see front matter r 2007 Elsevier GmbH. All rights reserved. doi:10.1016/j.ejcb.2007.07.001
and adult tissues. Since it is involved in cell–cell sorting and apoptosis, it is essential for embryonic morphogenesis. E-cadherin is detected in both epidermis and periderm already during the bi-layered epithelial stage (E13) when the ectoderm gives rise to periderm and epidermis (Hardy and Vielkind, 1996). K14, a type I keratin, is constitutively present in basal keratinocytes of the epidermis and stratified epithelia, and in the outer root sheath of the hair follicle. In normal mouse skin, K14 appears as early as the single ectodermal stage (E9.5) and stays in the basal layer after stratification (Byrne et al., 1994). Stratification of epidermal keratinocytes is accompanied by the synthesis
ARTICLE IN PRESS 802
I. Haase et al. / European Journal of Cell Biology 86 (2007) 801–805
of differentiation-specific keratins (e.g. keratins 1, 10) and precursor proteins of the cornified envelope such as involucrin and loricrin (Fuchs, 1990). Embryoid bodies (EBs) formed by aggregation of murine embryonic stem (ES) cells have been widely used for in vitro differentiation into certain lineages, e.g. myocytes and neurons (Guan et al., 1999). Keratinocytes have been differentiated from murine ES cells without EB formation by seeding ES cells onto extracellular matrix and adding bone morphogenetic protein-4 to the culture medium (Coraux et al., 2003). Emigrating cells from human EBs have been shown to give rise to keratinocytes when grown under supportive conditions (Green et al., 2003). We here use EBs derived from cultured murine ES cells to produce keratinocytelike cells in vitro. We show that these keratinocyte-like cells develop on the surface of intact EBs and can be isolated from them and kept in culture.
Materials and methods Maintenance of ES cells ES cells of the line CCE (Robertson et al., 1986), kindly provided by Prof. Max Gassmann, Zurich, were grown on inactivated primary murine embryonic fibroblasts in Iscove’s medium supplemented with 20% heatinactivated (56 1C, 30 min) foetal calf serum (FCS), 2 mM glutamine, 1% minimum essential medium (MEM)-nonessential amino acids, 100 IU/ml penicillin, 100 mg/ml streptomycin (all from PAA, Linz, Austria), 100 mM beta-mercaptoethanol (Sigma, St. Louis, MO), and 10 ng/ml leukemia-inhibiting factor (LIF; ESGROChemicon, Temecula, CA), in a humidified incubator with 5% CO2 at 37 1C, and passaged every 3 days.
Generation of embryoid bodies EBs were generated by two different methods: the spinner flask (Ateghang et al., 2006) and the hanging drop method (Wobus et al., 1991). In brief, for the spinner flask technique, adherent ES cells and feeder cells were enzymatically dissociated 3 days after plating using 0.2% trypsin/0.05% EDTA in phosphate-buffered saline (PBS) (PAA). Cells (1 107/ml) were seeded into a 250-ml siliconized spinner flask (Integra Biosciences, Fernwald, Germany) containing 100 ml Iscove’s medium, with the same supplements as mentioned above, except for LIF. The stirrer was set to 22.5 rpm, using a perpendicular stirring mode. After 24 h, 150 ml medium was added to give a total volume of 250 ml. About half of the culture medium was exchanged every day. After 5–7 days, EBs were harvested and either analyzed directly for the
presence of markers of squamous epithelial cells or plated onto plastic cultures dishes and kept as adherent cultures for different lengths of time and then used for immunostaining or lysed for Western blot analysis. For the generation of EBs in hanging drops we used 600 ES cells/20 ml Iscove’s medium (containing the above-mentioned supplements but without LIF). The droplets were carefully placed on the lid of a 10-cm dish, turned face down in the dish and left there for 3–4 days. One droplet yielded one EB. EBs were plated on cell culture dishes (Falcon, Franklin Lakes, NJ, USA) or chamber slides (Nalge Nunc, Napersville, IL) in FAD medium consisting of 300 ml DMEM, 90 ml HAM’s F12 (both PAA), 1.8 10 4 M adenine (Roche Diagnostics, Mannheim, Germany), 100 IU/ml penicillin, 100 mg/ml streptomycin (PAA), 10% heat-inactivated FCS (PAA), 0.5 mg/ml hydrocortisone (Calbiochem, La Jolla, CA), 5 mg/ml insulin (Sigma), 10 10 M cholera toxin (Calbiochem), 10 ng/ml epidermal growth factor (Progen, Heidelberg, Germany), 2 10 9 M 3-triiodo-thyroxine-Na salt (Serva, Heidelberg, Germany), 5 mg/ml transferrin (Calbiochem) for a minimum of 17 days and a maximum of 24 days. In some experiments, EBs were dissociated with 1 mg/ml collagenase III (Worthington, Lakewood, NJ) and extracted cells were reseeded on inactivated 3T3 murine fibroblasts as feeder cells.
Immunofluorescence staining The EBs grown on chamber slides were rinsed twice with PBS, fixed in 4% formaldehyde for 5 min at room temperature and rinsed again. At this point, we either let the chamber slides air-dry, later to be frozen at 20 1C or we continued immediately. EBs were permeabilized with 0.4% Triton X-100 (Serva) in PBS for 5 min, and then rinsed with PBS. Blocking against unspecific binding was carried out with 10% normal goat serum (Dako, Glostrup, Denmark) for 1 h at room temperature. The primary and secondary antibodies were both applied for 1 h at room temperature. For the primary antibodies, we used polyclonal rabbit anti-mouse K14 at 1 mg/ml (Covance, Richmond, CA) and a monoclonal rat anti-mouse E-cadherin at 10 mg/ml (Zymed, San Francisco, CA). Alexa Fluor-488- or Alexa Fluor-568coupled secondary antibodies were used for detection (Molecular Probes, Eugene, OR). Nuclear staining was performed using 4,6-diamidino-2-phenylindole-dilactate (DAPI) at 1 ng/ml (Sigma) or ToPro III (Molecular Probes). Specimens were analyzed using a Leica TCS upright confocal laser scanning microscope at excitation wavelengths of 488, 543 and 633 nm or with a Nikon Eclipse E800 microscope attached to a Nikon DXM 1200 F digital camera at a power of 200 or 400.
ARTICLE IN PRESS I. Haase et al. / European Journal of Cell Biology 86 (2007) 801–805
Western blotting EBs were lysed as a whole in lysis buffer, the lysates were centrifuged at 14,000g for 10 min, and the supernatant was used for Western blot analysis. Equal amounts of protein were separated by SDS–PAGE, blotted onto nitrocellulose LC 2000 membranes (Invitrogen, Carlsbad, USA) and probed with antibodies against K14 (Covance) and involucrin (kindly provided by F. Watt, London).
Results and discussion EBs were plated onto plastic cell culture dishes, cultured in FAD medium and harvested at different time points. At day 17, a clear separation of EBs into an inner cell mass and a multilayered epithelium that covered its surface could be recognized (Fig. 1(a)). This epithelium expressed E-cadherin as evidenced by immunofluorescence staining (Fig. 1(c–e)) and showed patches of K14 expression (Fig. 1(b), (d–f)), but was negative for keratin 10 or involucrin (data not shown). Morphologically, the multilayered epithelium resembled the developing epidermis that covers murine embryos after embryonic day 11 (M’Boneko and Merker, 1988). In whole mounts of day-18 EBs, K14-positive cells within the epithelial layer were detectable (Fig. 1(d–e)). By day 23, these single cells gave rise to islands of K14expressing epithelial cells which subsequently merged to form aggregates with tongue-shaped branches that seemed to sheathe the surface of the EB (Fig. 1(f)). Occasionally, giant K14-positive epithelial cells were detected that showed widely ramified networks of K14Fig. 1. Development of keratinocyte-like cells in EBs in situ. Haematoxylin/eosin (a) or immunofluorescence (b–f, h, i) staining and Western blot analysis (g) of tissue sections from EBs (a, b) or of total EBs (c–i). Images in (b–f, h, i) were obtained with a confocal microscope. (a) Detail from a section through an EB at day 17 showing the multilayered epithelium that covers its surface. (b) Tissue section through the surface epithelium of an EB stained for keratin 14 (K14; green). Nuclei are shown in red. (c–f) Confocal images of the surfaces of EB whole mounts obtained at day 17 (c), day 20 (d, e) and day 23 (f). Green staining: E-cadherin (c–e), K14 (f). Red staining: nuclei (c, f), K14 (d, e). (g) Western blot analysis of whole EB lysates with antibodies against K14 (upper panel) and involucrin (lower panel). O, P, Q, R, S indicate EBs from different experiments. EBs were analyzed at day 16 (O), 18 (P), 24 (Q), 38 (S), and day 46 (R). +: positive control (lysate from primary murine keratinocytes), : negative control (murine fibroblasts). Molecular weight markers are indicated to the right. (h, i) Confocal images of giant K14-expressing epithelial cells on the surface of an EB at day 20. The image in (i) is a cut out from the image in (h). Green staining: E-cadherin. Red staining: K14. Bars: 50 mm (a, c), 40 mm (f, i), 20 mm (b, d, e, h).
803
positive filaments (Fig. 1(h). These networks contained filaments of varying thickness with the thicker filaments seemingly inserting in membrane domains with clusters of E-cadherin (Fig. 1(i)). To verify expression of K14, we carried out Western blot analysis of whole EBs after different lengths of culture. Most EBs were positive for K14 between culture days 16 and 46 but showed a variable degree of expression (Fig. 1(g)). We also analyzed expression of involucrin, a constituent of the cornified keratinocyte envelope and a terminal differentiation marker (Li et al., 2000). EBs harvested at days 38 and 46 were found to contain involucrin (Fig. 1(g)). These results demonstrate that proteins characteristic of stratified epithelia and
ARTICLE IN PRESS 804
I. Haase et al. / European Journal of Cell Biology 86 (2007) 801–805
epidermis are expressed in EBs cultured under our conditions. Since we had found evidence for the expression of keratinocyte-specific proteins in EBs in situ, we hypothesized that isolation and culture of cells with characteristics of keratinocytes from these EBs should be possible. We therefore allowed EBs to form and differentiate until day 24 and then disaggregated them by enzymatic digestion with collagenase. Enzymatic digestion in addition to repeated aspirating through a narrow pipette tip yielded a cell suspension which was plated onto collagen I-coated cell culture dishes and kept in FAD medium with low calcium concentration. Most of the plated cells died within the first 2–3 days of culture, thus leaving space for rapidly growing colonies of cells that exhibited epithelial morphology (Fig. 2(a)). Immunofluorescence staining revealed that single cells within these colonies and, after several days of culture, entire colonies expressed K14 and E-cadherin (Fig. 2(b–e)). To investigate whether epithelial cells isolated from EBs and grown in culture were able to express markers of more highly differentiated epidermal keratinocytes, we stained cultures of EB-derived cells with an antibody against mouse involucrin (Li et al., 2000). This revealed single positive cells which were partly binucleated (Fig. 2(f–h)), a feature that has been observed before in differentiating epidermal keratinocytes (Gandarillas et al., 2000; Zanet et al., 2005). Although epithelial cells derived from EBs could be kept in culture for up to 3 weeks, we did not succeed in passaging them. The generation of an epidermis-like stratified epithelium from murine ES cells has been reported previously by Coraux et al. (2003). In this study, matrix secreted by dermal fibroblasts in association with the transforming growth factor beta (TGFb)-related bone morphogenetic protein-4 were sufficient to direct differentiation of ES cells towards the keratinocyte lineage without the requirement for EB formation. In our hands, the formation of EBs and their culture in FAD medium were sufficient to induce differentiation into keratinocyte-like cells. Therefore, differentiation towards epidermal precursor cells in vitro is supported by factors autonomously produced by the EB. Recently, EBs have also been used to generate keratinocyte precursors from human ES cells (Green et al., 2003; Ji et al., 2006). Although in this system K14-expressing keratinocyte precursors form independently of exogenously added factors, coating of culture dishes with collagen I or gelatine enhanced the yield of these cells. This suggests that the ability of EBs to give rise to keratinocyte precursor cells is self-contained and that the multiplication of such cells can be propagated by extracellular matrix molecules like gelatine and collagen. In addition to these previous reports with human ES cells, we were able to observe the formation of K14-expressing cells in
Fig. 2. Isolation of keratin 14 (K14)- and involucrin-expressing cells from EBs. Phase-contrast (a) and confocal fluorescence microscopic (b–d) images of cells isolated from EBs after day 18. (b–d) Cells were stained with an antibody against K14 (green); nuclei are shown in red. (e) Double staining for K14 (green) and E-cadherin (red). (f–h) Staining with an antibody against involucrin. Nuclei are shown in blue. Note single involucrin-positive, binucleated cells. Bars: 50 mm (a), 20 mm (b–d), 40 mm (e–h).
whole mounts of EBs in situ. Their formation occurred within a layer of epithelial cells covering the surface of the EBs and did not necessarily require outwards migration from the EB as observed previously (Green et al., 2003; Ji et al., 2006). Ji et al. (2006) reported that a defined serum-free medium (DSFM) was most efficient in generating keratinocyte precursors from human ES cells and that FAD medium did not result in detectable K14 expression within EBs. In our hands, FAD medium did support differentiation of keratinocyte-like cells from EBs derived from murine ES cells which is in line with results obtained by Green et al. (2003). One possible explanation for this difference is that the FAD medium in the study of Ji and co-workers was
ARTICLE IN PRESS I. Haase et al. / European Journal of Cell Biology 86 (2007) 801–805
supplemented with only 2.5% FCS whereas Green and colleagues used 5% FCS and we used 10% FCS. This makes it conceivable that serum may contain factors which promote differentiation of keratinocyte precursors from ES cells. In summary, we show here that EBs derived from ES cells can be used to produce keratinocyte-like cells in vitro. Several applications of this method are conceivable: modification of genes determining antigenic properties would eventually allow to generate epidermal keratinocytes with low antigenic potential from human ES cells that can be kept ‘‘ready to use’’, e.g. as a temporary dressing for large wounds in burn victims. Such keratinocytes could also be used as universal carriers of DNA constructs for gene therapeutic treatments (Gerrard et al., 1993). Finally, our protocol offers the possibility to generate cells of interest from murine ES cells with genetic modifications that would produce lethal phenotypes in mice. Several groups have now succeeded to generate keratinocytes from ES cells. In the future, we will have to learn how to keep these cells in long-term culture and how to make them applicable, e.g. to patients who require skin replacement.
References Ateghang, B., Wartenberg, M., Gassmann, M., Sauer, H., 2006. Regulation of cardiotrophin-1 expression in mouse embryonic stem cells by HIF-1alpha and intracellular reactive oxygen species. J. Cell Sci. 119, 1043–1052. Byrne, C., Tainsky, M., Fuchs, E., 1994. Programming gene expression in developing epidermis. Development 120, 2369–2383. Coraux, C., Hilmi, C., Rouleau, M., Spadafora, A., Hinnrasky, J., Ortonne, J.P., Dani, C., Aberdam, D., 2003. Reconstituted skin from murine embryonic stem cells. Curr. Biol. 13, 849–853. Fuchs, E., 1990. Epidermal differentiation: the bare essentials. J. Cell Biol. 111, 2807–2814.
805
Gandarillas, A., Davies, D., Blanchard, J.M., 2000. Normal and c-Myc-promoted human keratinocyte differentiation both occur via a novel cell cycle involving cellular growth and endoreplication. Oncogene 19, 3278–3289. Gerrard, A.J., Hudson, D.L., Brownlee, G.G., Watt, F.M., 1993. Towards gene therapy for haemophilia B using primary human keratinocytes. Nat. Genet. 3, 180–183. Green, H., Easley, K., Iuchi, S., 2003. Marker succession during the development of keratinocytes from cultured human embryonic stem cells. Proc. Natl. Acad. Sci. USA 100, 15625–15630. Guan, K., Schmidt, M.M., Ding, Q., Chang, H., Wobus, A.M., 1999. Embryonic stem cells in vitro – prospects for cell and developmental biology, embryotoxicology and cell therapy. Altex 16, 135–141. Hardy, M.H., Vielkind, U., 1996. Changing patterns of cell adhesion molecules during mouse pelage hair follicle development. 1. Follicle morphogenesis in wild-type mice. Acta Anat. (Basel) 157, 169–182. Ji, L., Allen-Hoffmann, B.L., de Pablo, J.J., Palecek, S.P., 2006. Generation and differentiation of human embryonic stem cell-derived keratinocyte precursors. Tissue Eng. 12, 665–679. Li, E.R., Owens, D.M., Djian, P., Watt, F.M., 2000. Expression of involucrin in normal, hyperproliferative and neoplastic mouse keratinocytes. Exp. Dermatol. 9, 431–438. M’Boneko, V., Merker, H.J., 1988. Development and morphology of the periderm of mouse embryos (days 9–12 of gestation). Acta Anat. (Basel) 133, 325–336. Robertson, E., Bradley, A., Kuehn, M., Evans, M., 1986. Germ-line transmission of genes introduced into cultured pluripotential cells by retroviral vector. Nature 323, 445–448. Wobus, A.M., Wallukat, G., Hescheler, J., 1991. Pluripotent mouse embryonic stem cells are able to differentiate into cardiomyocytes expressing chronotropic responses to adrenergic and cholinergic agents and Ca2+ channel blockers. Differentiation 48, 173–182. Zanet, J., Pibre, S., Jacquet, C., Ramirez, A., de Alboran, I.M., Gandarillas, A., 2005. Endogenous Myc controls mammalian epidermal cell size, hyperproliferation, endoreplication and stem cell amplification. J. Cell Sci. 118, 1693–1704.