Stem Cells, Vol. 19, No. 6, 500-504,
November 2001
© 2001 AlphaMed Press
Isolation of Human Embryonal Carcinoma Stem Cells by Immunomagnetic Sorting
Stefan Alexander Przyborski
Department Biological Sciences, University of Durham, Durham, United Kingdom
Key Words. Pluripotent stem cell • Embryonal carcinoma • Isolation • Cloning • Neuron • Glia
Stefan Alexander Przyborski, Ph.D., Department Biological Sciences, University of Durham, South Road, Durham DH1 3LE, United Kingdom. Telephone: 44-0-191-3743341; Fax: 44-0-191-3742417; e-mail: stefan.przyborski{at}durham.ac.uk
 |
ABSTRACT
|
|---|
Embryonal carcinoma cells are pluripotent stem cells derived from germ cell tumors and can be used to study cell differentiation in vitro. This report describes an approach designed to isolate pluripotent stem cells from primary/parent stock cultures of explanted tumor material. Cells expressing the pluripotent stem cell marker, SSEA-3, were isolated from heterogeneous stock cultures of the human teratoma line, TERA2, using immunomagnetic isolation. Single cell selection was performed on isolated SSEA-3+ cells and clonal lines were established. Each line was ultimately grown as a homogeneous monolayer, independent of feeder cells and expressed high levels of markers for pluripotent stem cells. In response to retinoic acid, clone TERA2.cl.SP-12 cells displayed enhanced neural differentiation compared to previously isolated TERA2 sublines and formed both neurons and glia. Deriving human pluripotent stem cell lines that differentiate into a range of cell types will provide useful tools to understand the molecular mechanisms controlling cell differentiation in a manner pertinent to human embryonic development.
|
INTRODUCTION
|
|---|
Pluripotent embryonal carcinoma (EC) stem cells are derived from teratocarcinomas, the tumors of germ cells, and are closely related to embryonic stem cells [1,2]. It has long been recognized that EC clones can be used as models for the investigation of cell differentiation in a manner pertinent to early vertebrate embryogenesis, especially in the mouse [3,4]. Several human EC cell lines have been characterized to extend this approach to human embryonic development. For example, we have recently shown that the molecular program of neuronal development by human NTERA2.cl.D1 EC cells resembles that of differentiating vertebrate neurons in utero [1,5].
There is considerable variation among the number of human EC clones available, such that different lineages have alternate strengths and weaknesses for use in research. This is often reflected in the manner that these cell lines are maintained and their capacity for differentiation. For instance, Pera and coworkers [6] have described several clones isolated from the human teratoma cell line GCT27 that may be termed as either multipotent and give rise to multiple differentiated derivatives or nullipotent with no evidence of differentiation. Multipotent cells derived from GCT27 require feeder cell support for proliferation and are capable of differentiation into all three germ layers. In contrast, human NTERA2.cl.D1 EC cells are independent of feeders [7], but may have a more restricted capacity for differentiation [8]. It is difficult to easily explain such differences especially given their varied karyotype: that such cells were derived from different sources using different techniques and that some clonal lineages derived from a single parent line also appear to have varying abilities for differentiation [6,9,10]. It is therefore possible that the isolation of additional EC cell lines may produce lineages that show further variation in their ability to differentiate and have useful characteristics appropriate for the study of particular pathways of human development. The aim of this study is to isolate new clonal lineages of pluripotent stem cells from the human teratoma line, TERA2, using an approach involving immunomagnetic separation and single cell selection.
|
MATERIALS AND METHODS
|
|---|
Cell Culture
Human EC cell lines, TERA2 and NTERA2.cl.D1, were generously provided by P. Andrews, University of Sheffield, Sheffield, UK. All cells were maintained in Dulbecco's modified Eagle's medium ([DMEM]; Life Technologies Ltd.; Paisley, Scotland; http://www.lifetech.com) as previously described [5,7]. EC cells were induced to differentiate by seeding 1.5 x 106 cells per 75-cm2 tissue culture flask (Nalge Nunc International; Roskilde, Denmark; http://www.nalgenunc.com) in DMEM containing 10 µM retinoic acid (Sigma-Aldrich Company Ltd.; Poole, UK; http://www.sigma-aldrich.com) as previously reported [5,11].
Isolation and Cloning of Human TERA2 Sublines
Confluent TERA2 EC cells (passage 15, earliest available) were briefly treated with 0.25% trypsin (Life Technologies)/2 mM EDTA (Sigma-Aldrich) in phosphate-buffered saline (PBS) for 2-3 minutes to produce a suspension of single cells. Suspended TERA2 cells were diluted to 107 cells/ml and incubated with stage-specific embryonic antigen-3 (SSEA-3) antibody (diluted 1:5), a marker of pluripotent stem cells [1,12], in wash buffer ([WB]; PBS plus 5% volume by volume fetal calf serum [Life Technologies]). Preliminary studies indicate that maximal numbers of cells bind the SSEA-3 antibody after 45 minutes incubation at 4°C (data not shown). Cells immunoreactive for SSEA-3 were isolated using direct positive magnetic separation according to manufacturer's instructions (BioMag; http://www.biomag.helsinki.fi; goat anti-mouse IgM, Polysciences Europe GmbH, Postfach 1130, 69208; Eppelheim, Germany; http://www.polysciences.de). BioMag' magnetic particles are approximately 1µm and because of their non-uniform shape, provide an increased surface area (>100 m2/g) of 20-30 times greater than that of uniform spherical particles allowing for a higher binding capacity while utilizing a lower amount of particle. The magnetic particles detach from the cell membrane automatically as the cell surface is turned over during subsequent culturing for up to 48 hours. Isolated cells were immediately resuspended and washed three times in WB, magnetically separated a second time and finally resuspended in 10 ml WB. Single cells were picked at random with a micropipette under a dissecting microscope and transferred to a drop of DMEM where the presence of a single cell was confirmed. A single cell was added to each well of a tissue culture plate (Nunc) containing irradiated (~10,000 rads) STO-transformed mouse feeder cells [4] (12 wells were seeded in total). Feeder cells were maintained for the first three passages until the newly derived clones formed large enough colonies to grow independently.
Antibodies
Primary monoclonal antibodies SSEA-3, SSEA-4, A2B5, VIN-IS-56, and TRA-1-60 were generously provided by P. Andrews, University of Sheffield, UK. These antibodies recognize specific cell surface antigens associated with globoseries glycolipids and glycoproteins, and show highly regulated expression profiles related to the differentiation of human EC cells [1,9]. For example, SSEA-3, which was originally raised against four-cell-stage mouse embryos [12], is expressed highly in EC stem cells and not their differentiated derivatives [1,9]. Antibodies were pretitered and diluted (1:2 to 1:5) in WB to give maximal binding.
Immunocytochemistry
Cell cultures were fixed in ice-cold methanol (5 minutes), washed three times in PBS and incubated with primary antibody for 60 minutes at 4°C. After three PBS washes, cells were incubated with either fluorescein isothiocyanate-conjugated goat anti-mouse IgG or IgM (ICN Pharmaceuticals, Inc.; Aurora, OH; http://www.icnpharm.com) as appropriate for a further 60 minutes at 4°C, washed three times in PBS, and examined by fluorescence microscopy.
Immunofluorescent Flow Cytometry
Cell surface antigen expression was determined by indirect immunofluorescence using a Coulter (Fullerton, CA; http://www.coulter.com) EPICS XL cytometer in a manner similar to that previously reported [5,9].
Northern Analysis
Poly[A+] RNA was isolated from human EC cells and retinoic acid-induced derivatives, and prepared for Northern blotting as previously described [5]. The blot was hybridized with a 2,127 bp BamHI-XbaI fragment of POU5F1 generously provided by F. Gandolfi, Institute of Anatomy, Milan, Italy [13] and the hybrid signal detected by autoradiography.
Western Analysis
Protein samples were prepared from EC cells and their differentiated derivatives. Samples were separated on SDS-polyacrylamide gels and immunoblotted [14]. Antibodies for neuron-specific enolase (Chemicon; Temecula, CA; http://www.chemicon.com; MAB324; 1:2,000), growth-associated protein 43 ([GAP43]; Sigma-Aldrich, clone 7B10; 1:4,000), glial fibrillary acidic protein ([GFAP]; Sigma-Aldrich, clone GA-5; 1:5,000), and ß-actin (Sigma-Aldrich, clone AC-15; 1:5,000) were localized with IgG-horse radish peroxidase secondary antibody (Amersham; Piscataway, NJ; http://www.apbiotech.com, 1:1,000) in preparation for chemiluminescent detection (Amersham).
|
RESULTS
|
|---|
Close examination of TERA2 cultures enabled the identification of morphologically different cell populations. Distinct colonies of cells displaying typical EC morphology were occasionally observed (Fig. 1
). Such colonies stained positive for markers of human EC cells, including SSEA-3 and SSEA-4, while surrounding cells of non-EC structure did not express these antigens. The frequency of SSEA-3, -4+ cells in TERA2 cultures was determined by flow cytometry and represented only 2%-3% of the total population (data not shown). SSEA-3+ cells were isolated directly by immunomagnetic selection, and single cells were subsequently picked at random to establish clonal sublines that ultimately grew as homogeneous monolayers at high confluency, independent of feeder cells.
View larger version (183K):
[in this window]
[in a new window]
|
Figure 1. Isolation and cloning of EC cells from the human teratoma line, TERA2. Phase image and immunofluorescence localization of SSEA-4 in cultures of TERA2 (p15) cells grown at low seeding density (A and B, respectively). Note a tight colony of EC cells as indicated (ec). Small colony of TERA2.cl.SP-12 EC cells (p2) cocultured with fibroblast feeders (fibro) (C). Corresponding image showing specific SSEA-3 immunoreactivity to TERA2.cl.SP-12 EC cells (D). Expanding colonies of TERA2.cl. SP-12 EC cells (p3) (E) were subsequently grown as confluent homogenous monolayers independent of feeder cells after three passages (F). Scale bars: 100 µm (A, B); 25 µm (C, D, F); 120 µm (E).
|
|
Markers of human pluripotent stem cells were highly expressed in each of the newly derived TERA2 subclones, indicating strong enrichment of a subpopulation of stem cells with a high level of purity. Previous attempts to separate these cells using fluorescence-activated cell sorting have failed to isolate viable EC cells [15]. Biomagnetic sorting is direct and rapid, which is an important factor in conserving the viability of human EC cells. Each of the newly derived TERA2 subclones downregulated their expression of each EC stem cell marker after retinoic acid-induced differentiation. In a reciprocal fashion, antigens expressed during differentiation, notably A2B5 and VIN-IS-56, were upregulated. Clone TERA2.cl.SP-12 was selected for further evaluation on the basis that it consistently displayed the highest levels of EC markers in its undifferentiated state and showed the strongest expression of A2B5 and VIN-IS-56 in response to retinoic acid (Fig. 2).
View larger version (13K):
[in this window]
[in a new window]
|
Figure 2. Expression of cell surface antigens by TERA2.cl.SP-12 EC cells and their differentiated derivatives after a 14-day exposure to retinoic acid. Values represent mean ± SE from three replicates.
|
|
The octamer-binding transcription factor-4 is encoded by the gene POU5F1 and is expressed in human pluripotent stem cells [2,13]. POU5F1 mRNA was detected at greatest concentrations in NTERA2.cl.D1 and TERA2.cl.SP-12 EC cells but at notably lower levels in the TERA2 parent lineage (Fig. 3). TERA2.cl.SP-12 cells showed decreased expression of POU5F1 in response to retinoic acid, indicating that the vast majority of pluripotent stem cells had committed to differentiate after 7 days exposure.
View larger version (37K):
[in this window]
[in a new window]
|
Figure 3. Expression ofPOU5F1 mRNA in TERA2 cells and its sublineages by Northern analysis. Lanes: (1) TERA2 EC cells; (2) NTERA2.cl.D1 EC cells; (3) TERA2.cl.SP-12 EC cells; (4, 5, 6) TERA2.cl.SP-12 cells after 2, 4, and 7 days exposure to retinoic acid, respectively. GAPDH was used as a loading control.
|
|
The reactivity of A2B5 and VIN-IS-56 has previously been associated with neuroectodermal derivatives [9,16], and thus may indicate the ability of TERA2.cl.SP-12 EC cells to form neural derivatives in response to retinoic acid. While TERA2.cl.SP-12 EC cells showed no expression of neural proteins, markers indicative of both neurons and glia were detected after 28 days exposure to retinoic acid (Fig. 4). In an identical experiment, NTERA2.cl.D1 EC cells reacted in response to retinoic acid to produce neurons, but no glial markers were detected (data not shown).
View larger version (16K):
[in this window]
[in a new window]
|
Figure 4. Expression of neural proteins during differentiation of TERA2.cl.SP-12 cells. Lanes: (1) TERA2.cl.SP-12 EC cells; (2) TERA2.cl.SP-12 cells after 28 days exposure to retinoic acid. ß-actin was used as loading control.
|
|
|
DISCUSSION
|
|---|
Earlier studies describing the isolation and cloning of human EC cells from primary explant/stock cultures have recognized that the parent material probably consists of multiple cell types and suggest that pluripotent EC cells represent a fraction of the total parent cell population [7,10,15]. The current data clearly demonstrate that SSEA-3+ and SSEA-4+ EC cells correspond to a minor component of the parent TERA2 line, which correlates favorably with the lower expression of POU5F1 in TERA2 cultures compared to two of its clonal EC derivatives tested in this study. In earlier work, clonal lines established directly from intact stock cultures of the TERA2 parent lineage showed variability, frequently did not express SSEA-3 or SSEA-4, and often displayed limited capacity for differentiation [10,15]. The approach outlined in this study focuses on isolating a population of SSEA-3+ cells from the TERA2 parent lineage prior to deriving clonal lines. This resulted in the majority of cells, especially TERA2.cl.SP-12, expressing and regulating cell surface antigens in a similar manner compared with the well-established human pluripotent EC line, NTERA2.cl.D1 [9]. Human NTERA2.cl.D1 EC cells were originally cloned from a xenograft tumor of the TERA2 parent line [7] whereas TERA2.cl.SP-12 cells have been isolated directly from the earliest available passage of TERA2 [17]. It is well recognized that NTERA2.cl.D1 EC cells produce neurons in vitro [5,11,18], but there is no evidence that glial cells form in response to retinoic acid under the conditions described in this study or by the same conditions used by others [10,11,18]. In contrast, differentiating TERA2.cl.SP-12 cells produced proteins indicative of both neurons and glia. It is not known why TERA2.cl.SP-12 cells appear to have greater ability for neural differentiation than NTERA2.cl.D1, but it may be a consequence of the way in which they were derived or that they simply originate from two different clones.
To summarize, new lineages of human EC stem cells can be derived using an efficient and direct method that has been successfully tested on a recognized human teratoma line. It is proposed that such an approach may also be applied directly to explanted tissue soon after surgical removal to enable the isolation and establishment of additional pluripotent stem cell lines to promote the study of human embryology.
|
ACKNOWLEDGMENT
|
|---|
The author would like to thank Professor Andrews, University of Sheffield, UK, for generously providing cultures of the human EC cell lines, TERA2 and NTERA2.cl.D1, and monoclonal antibodies, and Dr. Gandolfi, Institute of Anatomy, Milan, Italy, for providing the POU5F1 cDNA clone. This work was supported in part by awards to SAP from The Royal Society, London, UK and from The Peel Medical Research Trust, London, UK (Regd. Charity: 214683).
|
REFERENCES
|
|---|
-
Andrews PW, Przyborski SA, Thomson JA. Embryonal carcinoma cells as embryonic stem cells. In: Marshak DR, Gardner R, Gottlieb D, eds. Stem Cell Biology. New York: Cold Spring Harbor Press, Monograph 40. 2001:231-266.
-
Pera MJ, Ruebinoff B, Trounson A. Human embryonic stem cells. J Cell Sci 2000;113:5–10.[Abstract]
-
Martin GR. Teratocarcinomas and mammalian embryogenesis. Science 1980;209:768–776.[Abstract/Free Full Text]
-
Martin GR, Evans MJ. Differentiation of clonal lines of teratocarcinoma cells: formation of embryoid bodies in vitro. Proc Natl Acad Sci USA 1975;72:1441–1445.[Abstract/Free Full Text]
-
Przyborski SA, Morton IE, Wood A et al. Developmental regulation of neurogenesis in the pluripotent human embryonal carcinoma cell line NTERA2. Eur J Neurosci 2000;12:3521–3528.[CrossRef][Medline]
-
Pera MJ, Cooper S, Mills J et al. Isolation and characterisation of a multipotent clone of human embryonal carcinoma cells. Differentiation 1989;42:10–23.[CrossRef][Medline]
-
Andrews PW, Damjanov I, Simon D et al. Pluripotent human embryonal carcinoma clones derived from the human teratocarcinoma cell line Tera-2: differentiation in vivo and in vitro. Lab Invest 1984;50:147–162.[Medline]
-
Pleasure SJ, Lee VM-Y. A human cell line which displays characteristics expected of a human commited neuronal progenitor cell. J Neurosci Res 1993;35:585–602.[CrossRef][Medline]
-
Andrews PW, Nudelman E, Hakomori S-I et al. Different patterns of glycolipid antigens are expressed following differentiation of TERA-2 human embryonal carcinoma cells induced by retinoic acid, hexamethylene bisacetamide (HMBA) or bromodeoxyuridine (BudR). Differentiation 1990;43:131–138.[Medline]
-
Thompson S, Stern PL, Webb M et al. Cloned human teratoma cells differentiate into neuron-like cells and other cell types in retinoic acid. J Cell Sci 1984;72:37–64.[Abstract]
-
Andrews PW. Retinoic acid induces neuronal differentiation of a cloned human embryonal carcinoma cell line in vitro. Dev Biol 1984;103:285–293.[CrossRef][Medline]
-
Shevinsky LH, Knowles BB, Damjanov I et al. Monoclonal antibody to murine embryos defines a stage-specific embryonic antigen expression on mouse embryos and human teratocarcinoma cells. Cell 1982;30:697–705.[CrossRef][Medline]
-
van Eijk MJT, van Rooijen MA, Modina S et al. Molecular cloning, genetic mapping and developmental expression of Bovine POU5F1. Biol Reprod 1999;60:1093–1103.[Abstract/Free Full Text]
-
Przyborski SA, Cambray-Deakin MA. Developmental changes in GAP-43 expression in primary cultures of rat cerebellar granule cells. Mol Brain Res 1994;25:273–285.[Medline]
-
Andrews PW, Damjanov I, Simon D et al. A pluripotent human stem-cell clone isolated from the TERA-2 teratocarcinoma line lacks antigens SSEA-3 and SSEA-4 in vitro, but expresses these antigens when grown as a xenograft tumor. Differentiation 1985;29:127–135.[CrossRef][Medline]
-
Eisenbarth GS, Walsh FS, Nirenberg M. Monoclonal antibody to a plasma membrane antigen of neurons. Proc Natl Acad Sci USA 1979;76:4913–4917.[Abstract/Free Full Text]
-
Fogh J, Trempe G. New human tumor cell lines. In: Fogh J, ed. Human Tumor Cells in Vitro. New York: Plenum Press, 1975:115-159.
-
Andrews PW, Gonczol E, Plotkin SA et al. Differentiation of TERA-2 human embryonal carcinoma cells into neurons and HCMV permissive cells. Differentiation 1986;31:119–126.[CrossRef][Medline]
Received June 20, 2001;
accepted for publication July 27, 2001.
Top
Abstract
Introduction
