Multilineage Differentiation of Rhesus Monkey Embryonic Stem Cells in Three-Dimensional Culture Systems
SILVIA S. CHEN,a ROBERTO P. REVOLTELLA,a,b SANDRA PAPINI,b MONICA MICHELINI,b
WENDY FITZGERALD,a JOSHUA ZIMMERBERG,a LEONID MARGOLISa
aNASA/NIH Center for Three Dimensional Tissue Culture, Laboratory of Cellular and Molecular Biophysics,
National Institute of Child Health and Human Development (NICHD), NIH, Bethesda, Maryland, USA;bCNR-Institute of Biomedical Technologies, Unit of Immunobiology and Cell Differentiation, Pisa, Italy
In the course of normal embryogenesis, embryonicstem (ES) cells differentiate along different lineages inthe context of complex three-dimensional (3D) tissuestructures. In order to study this phenomenon in vitrounder controlled conditions, 3D culture systems are nec-essary. Here, we studied in vitro differentiation of rhesusmonkey ES cells in 3D collagen matrixes (collagen gelsand porous collagen sponges). Differentiation of ES cellsin these 3D systems was different from that in monolay-ers. ES cells differentiated in collagen matrixes intoneural, epithelial, and endothelial lineages. The abilitiesof ES cells to form various structures in two chemicallysimilar but topologically different matrixes were differ-ent. In particular, in collagen gels ES cells formed gland-like circular structures, whereas in collagen sponges ES cells were scattered through the matrix or formed
aggregates. Soluble factors produced by feeder cells oradded to the culture medium facilitated ES cell differen-tiation into particular lineages. Coculture with fibro-blasts in collagen gel facilitated ES cell differentiationinto cells of a neural lineage expressing nestin, neural celladhesion molecule, and class III ββ-tubulin. In collagensponges, keratinocytes facilitated ES cell differentiationinto cells of an endothelial lineage expressing factor VIII.Exogenous granulocyte-macrophage colony-stimulatingfactor further enhanced endothelial differentiation.Thus, both soluble factors and the type of extracellularmatrix seem to be critical in directing differentiation ofES cells and the formation of tissue-like structures.Three-dimensional culture systems are a valuable toolfor studying the mechanisms of these phenomena. StemCells 2003;21:281-295
Embryonic stem (ES) cells are totipotent cells derivedfrom the inner cell mass of preimplantation mammalianembryos. Long-term culture conditions for maintainingundifferentiated ES cells derived from blastocysts of mice,
nonhuman primates, and humans have been established [1-5]. Without these conditions, ES cells differentiate sponta-neously into cells of all three embryonic germ layers [4, 6-8].
Rhesus monkey ES cells injected into muscles of severecombined immunodeficient (SCID) mice formed teratomas
Stem Cells®
Original Article
î�à‰@)Ñ8à‚!f
Chen, Revoltella, Papini et al. 282
that contained elements from all three embryonic germ lay-ers [2, 9, 10]. For example, in this system, ES cells differ-entiated into structures of ectodermal origin resemblingneural tubes, embryonic ganglia, and brain-like gray matter[9], as well as into endoderm-derived tissues includingintestinal and ductal epithelium and pancreas [10].
To understand the mechanisms of ES cell differentiationalong different lineages, in vitro culture models have beendeveloped [11]. In most of these models, ES cells wereinduced to form embryoid bodies (EBs), which were thencocultured with a monolayer of stromal cells [12, 13] orplated on various substrates, in the presence or absence of tis-sue-specific growth factors [7, 14-23]. Alternatively, ES cellswere transfected with appropriate genes [8, 24-26] to identifymolecular mechanisms that control ES cell differentiation. Allof these experiments were performed in two-dimensionalmonolayer cultures in which ES cells were induced to differ-entiate, forming a mixed population of cells of three embry-onic germ lines. However, the multicellular structures, whichare typically formed in normal embryos and in SCID mouseteratomas, require cells to be able to migrate in three dimen-sions and to interact with their microenvironment. To studythis phenomenon in vitro under controlled conditions, three-dimensional (3D) culture systems are necessary. Here, westudied in vitro differentiation of rhesus ES cells and the for-mation of tissue-like structures in 3D collagen matrixes, aswell as the role of intercellular interactions in this process. Weused type I collagen matrixes in two structurally differentforms: as a gel and as a sponge. Nonhuman primate ES cellsgrowing on collagen matrixes were cocultured with humandermal fibroblasts or keratinocytes. We found that in thisenvironment multiple cell-cell interactions and soluble fac-tors led to ES cell differentiation along particular lineages.In particular, in these collagen matrixes, complex tubular orspherical glandular-like structures, similar to those inembryos and teratomas, were formed, and these structuresultimately generated differentiated progeny with character-istics of neural, epithelial, or endothelial lineages. We alsofound that, in 3D collagen matrixes in the presence offeeder cells or exogenous cytokines, ES cell differentiationcould be directed into a particular lineage, accompanied bythe formation of tissue-like structures.
MATERIALS AND METHODS
Culture of Undifferentiated ES CellsUndifferentiated R366.4 rhesus monkey ES cells were
cultured as previously described [4]. ES cells were cocul-tured with mitomycin-C-treated (0.8 mg/ml for 2 hours)(Sigma; St. Louis, MO; http://www.sigmaaldrich.com)murine embryonic fibroblasts (MEFs; Cell Essentials Inc.;
Boston, MA; http://www.cell-essentials.com) in gelatin-coated six-well plates (Nalge Nunc Inc.; Naperville, IL;http://www.nalgenunc.com) to prevent spontaneous differen-tiation. ES cells were maintained in 80% Knock OutDulbecco’s modified Eagle’s medium (DMEM; Invitrogen;Rockville, MD; http://www.invitrogen.com) supplementedwith 20% defined fetal bovine serum (d-FBS; HyClone LabInc.; Logan, UT; http://www.hyclone.com), 1 mM L-gluta-mine, 0.1 mM 2-mercaptoethanol, and 1% nonessentialamino acids (Invitrogen). ES cells were split by releasingthem from the culture plate with 0.8 mg/ml collagenase IV(Invitrogen) and seeded onto a new mitomycin-C-treatedMEF feeder layer in a gelatin-coated six-well plate.
Controlled Differentiation of ES Cells
ES Cells Cultured as Monolayers on Chamber SlidesIn order to verify that the ES cells were still pluripotent,
they were released from the six-well plates with collagenaseIV (0.8 mg/ml) and suspended in DMEM supplemented with10% FBS (Gemini Bioproducts; Woodland, CA). Cells fromeach well of the six-well plates were dissociated by pipettingand seeded on one two-well plastic chamber slide (NalgeNunc Inc.). Some MEFs were also released and were presentamong the ES cells. Three different growth media were used:A) Base medium: DMEM supplemented with 10% FBS con-taining 1% penicillin/streptomycin (Invitrogen). This wasused as a simple base medium that relies on serum to supportcell proliferation; B) A5RT.1-conditioned medium: a mono-layer of high-grade malignant human keratinocyte HaCaT-rasclone A5RT.1 cells [27], which constitutively produce manygrowth factors including GM-CSF and G-CSF [28], was cul-tured with fresh base medium until confluence. Two-day con-ditioned medium was then centrifuged, filtered (0.22-µm poresize filter), diluted 1:3 with fresh base medium, and used; orC) HPI.1-conditioned medium: human dermal HPI.l fibro-blasts, which produce many growth factors including fibro-blast growth factor, epidermal growth factor, stem cell factor,GM-CSF, vascular endothelial growth factor, and insulin-likegrowth factor-1 (S. Papini and R.P. Revoltella, unpublisheddata), were cultured for 2 days with base medium until semi-confluence. The conditioned medium was then centrifuged,filtered, diluted 1:3 with fresh base medium, and used.
All media were changed every 2 days. The morphologiesof the ES cells cultured with different media in the chamberslides were observed with a phase-contrast microscope.
On day 8, chamber slides were rinsed with phosphate-buffered saline (PBS; pH 7.4), fixed in 4% formaldehydefor 30 minutes or in a 1:1 mixture of cold methanol and ace-tone for 10 minutes as recommended by the antibody man-ufacturers, washed in PBS, and kept in PBS at 4°C until
283 ES Cells in 3D Culture System
immunohistochemical analysis. Before immunostaining,cells were permeabilized with 0.1% Triton X-100 in 1×PBS for 10 minutes if required, then blocked in Powerblock(Biogenex; San Ramon, CA; http://www.biogenex.com),5% FBS, or 1%-5% goat serum supplemented with 0.5%bovine serum albumin (BSA; Sigma) for 10-60 minutes, asrecommended by the antibody manufacturers.
3D ES Cultures in Collagen Matrixes
Collagen GelHPI.1 cells (0.1 × 106 cells/well) were seeded in some
wells of six-well plates as a feeder layer 24 hours before EScell seeding. Meanwhile, each of the six-well inserts wascoated with 1 ml of 2.4-mg/ml rat tail type I collagen(Sigma) as described elsewhere [29] with or without 0.1 ×106 HPI.1 cells embedded [30]. After the collagen solutionsolidified at 37°C overnight, ES cells were released fromthe six-well plates, as described above, and seeded on topof the collagen gel (with or without HPI.1 cells embedded).The inserts were then transferred to the six-well plates withor without a 24-hour-old feeder layer at the bottom of thewells (Fig. 1). Culture medium supplemented with 50µg/ml ascorbate (Sigma) and 1% penicillin/streptomycin(Invitrogen) was added to each of the six-well compart-ments outside the insert so that ES cells were cultured at theair-fluid interface. Media were changed every 2 days. After 7and 19 days of culture at 37°C in a 5% CO2 in air atmos-phere, the inserts were rinsed with PBS, fixed in 4%formaldehyde in PBS at 4°C for 3 days, washed with PBS,
and kept in PBS at 4°C until paraffin embedding and sec-tioning. Sections (~5 µm) were stained with hematoxylin andeosin (H&E) or subjected to immunohistochemical analysis.
Collagen SpongeHPI.1 cells (0.3 × 106/well) were seeded as monolayers
in some of the six-well plates as a feeder layer 24 hoursbefore ES cell seeding. Sterile Gelfoam sponges (Pharmacia;Kalamazoo, MI; http://www.pnu.com), from purified porkskin gelatin, were moisturized with culture medium (DMEMcontaining 10% FBS and 1% penicillin/streptomycin) justbefore use. One sponge was cut into four 20 × 15 × 7-mmpieces, and each piece was transferred to a well of a six-wellplate with or without a feeder layer. ES cells, released fromanother six-well plate with 0.8 mg/ml collagenase IV, weresuspended in culture medium and placed on the top of thesponge. On each sponge, ES cells were seeded in six distinctspots (20 µl of the ES cell suspension were applied at eachspot). Then, 3 ml of culture medium were pipetted into eachwell to cover the sponge (Fig. 1). The cultures were thenincubated at 37°C in a 5% CO2 in air atmosphere for 24 days.The medium was changed every 2 days. On day 24, collagensponges were fixed in 4% formaldehyde in PBS at 4°C for 3days, washed with PBS, and kept in PBS at 4°C until paraf-fin embedding and sectioning. To achieve a better view ofcell migration inside and through the sponge, each piece ofthe sponge was cut vertically into halves along the long axisbefore paraffin embedding and sectioning. Sections (~5 µm)were stained with H&E or subjected to immunohistochemi-cal analysis. In order to investigate the effects of different
Figure 1. Experimental procedure. Undifferentiated ES cells cultured on mitomycin-C-treated MEFs were transferred onto two types of colla-gen I matrixes: collagen gel and collagen sponge. In some experiments, ES cells growing on collagen matrixes were cocultured with human dermal fibroblasts or keratinocytes in the presence or absence of GM-CSF.
ESCollagen gel
Culture insert
Embedded cells
Feeder cells
Feeder cellsCollagen sponge
ES
ES
ES
Feeder cells
Chen, Revoltella, Papini et al. 284
feeder cells and growth factors on the differentiation of EScells, we used A5RT.1 (0.3 × 106) cells as an alternativefeeder layer. In addition, we studied the effects of recombi-nant human (rh)GM-CSF (a gift from Dr. F. Bertolero;Upjohn; Milan, Italy) on ES cell differentiation. ExogenousrhGM-CSF (20 ng/ml) was added to each of the cultures. Theresultant cultures were compared with the ones without exogeneous rhGM-CSF.
Histology and Immunocytochemical AnalysisTarget retrieval was performed on all formaldehyde-
fixed paraffin sections by means of 0.01 M citrate buffer(pH 6.0) in a microwave oven. The cells were then incu-bated with primary antibodies. The primary antibodies andtheir dilutions used are listed in Table 1. ES cells grown inchamber slides and unstained sections from ES cells grownon collagen matrixes were incubated with Vector M.O.Mmouse IgG Blocking Reagent (Vector Laboratories;Burlingame, CA; http://vectorlabs.com) to prevent possiblecross-reaction between contaminating MEFs and the primaryor secondary antibodies. All primary antibodies were dilutedin Tris-buffered saline supplemented with 1% BSA. We usedthe immunoperoxidase system (LSAB+ system; DakoCorporation; Capinteria, CA; http://www.dako.com), whichcontained biotinylated anti-rabbit, mouse, and goat Ig as asecondary antibody; streptavidin conjugated to horseradishperoxidase as a link agent; and 3,3′-diaminobenzidine (DAB)as a chromogen. Peroxidase-conjugated rabbit anti-rat IgGs
were used as the secondary antibody, and DAB was used asa chromogen for anti-stage-specific embryonic antigen 3(SSEA-3). After the immunostaining procedures were com-pleted, some sections were lightly counterstained withMayer hematoxylin (Sigma). Positive-stained cells werecounted in nine unconnected fields at 100× magnification,and the percentage of positive staining was calculated onthe basis of the total number of cells in each view. In addi-tion, using the BCIP/NBT substrate system (Dako), wedetected undifferentiated ES cells that expressed alkalinephosphatase.
RESULTS
Two-Dimensional ES Cells Cultured in Chamber SlidesTo evaluate the effect of 3D cultures, we first studied
rhesus monkey ES cell differentiation in conventional cul-ture in chamber slides. ES cells seeded in two-well cham-ber slides were cultured with three different media: basemedium, medium conditioned by A5RT.1 human ker-atinocytes, and medium conditioned by HPI.1 human der-mal fibroblasts. Initially, under all conditions, ES cellsremained undifferentiated, as evidenced by their morpholo-gies and staining with anti-SSEA-3 and anti-SSEA-4 anti-bodies (results not shown) [4, 31, 32], and formed compactcolonies. Later (after 36-48 hours in culture), some cellsbegan to differentiate and migrated out from the colonies,divided, and formed clusters.
Table 1. Primary antibodies used in the study
Antibody Specificity Species Clonality Dilution Source
Anti-SSEA-3 Undifferentiated ES cells Rat Monoclonal 1:5 Developmental Hybridoma Studies Bank (DHSB), (MC-631) University of Iowa, Iowa City, IA
Anti-SSEA-4 Undifferentiated ES cells Mouse Monoclonal 1:10 DHSB(MC-813-70)
Anti-class III Neurons at early Mouse Monoclonal 1:100 Chemicon, Temecula, CAβ-tubulin (TU-20) development stage
After 8 days under all threetested culture conditions, thesecells were predominantly of fourdifferent types (Fig. 2): A) undif-ferentiated ES; B) polygonal,epithelial-like; C) neuronal-like,and D) spread cells with largelamellae. Two types of colonieswere found: A) heterogeneouscolonies consisting of compact,undifferentiated ES cells in themiddle and differentiated cells atthe peripheries (Fig. 2A) and B) homogeneous coloniesformed by polygonal epithelial-like cells (Figs. 2B and4A). The sizes of the epithelial colonies in cultures withA5RT.1-conditioned medium were larger than under otherconditions. The numbers of cells in the epithelial-likecolonies were 14-25, 28-72, and 14-46 cells per cluster inbase medium, A5RT.1-conditioned medium, and HPI.1-conditioned medium, respectively. Neuronal-like cellsgrew on top of MEF cells (transferred to the chamberslides together with ES cells) or adhered directly to theslide; they were unipolar or bipolar and formed networksor aggregates (Fig. 2C). Cells of the fourth type remainedscattered and did not form compact colonies (Fig. 2D).
In order to further characterize the cell types,immunohistochemistry was performed. Cells were stainedfor nestin, neural cell adhesion molecule (NCAM), classIII β-tubulin, and cytokeratins (AE1/AE3). These markersare characteristic of neural precursor cells [33], neuralcells [34], immature (neuronal) cells [35, 36], and epithe-lial cells [37], respectively. Cells at the periphery of EScell colonies appeared to be morphologically differenti-ated and expressed nestin (Figs. 3A-3B). Epithelial-likecells did not express nestin, NCAM, or class III β-tubulin,but they were positive for cytokeratins (Fig. 4B). Cells ofneuronal morphology strongly expressed NCAM (Figs.3D-3E) and class III β-tubulin (Fig. 3F). The large cellswith spread lamellae were also positive for nestin (Fig. 3C).Thus, immunostaining confirmed the morphological evi-dence of cell differentiation into neuronal and epitheliallineages.
3D ES Cell Cultures in Collagen Matrixes
Collagen GelEmbryonic stem cells were seeded on a layer of type I
collagen gel and cultured in inserts in six-well plates. After7 and 19 days of culture, H&E staining of the tangential his-tological sections revealed that some ES cells remained onthe top of the collagen gel, whereas a large proportion ofcells had penetrated into the gel matrix. Unlike cells in thechamber slide cultures described above, ES cells in or oncollagen gel apparently did not form distinguishable clus-ters of epithelial cells. Neither did we observe large spreadcells under these conditions, as observed in chamber slides.Instead, ES cells cultured on 3D collagen gel formed denseclusters as well as tubular or spherical glandular-like struc-tures, which became clearly evident after 19 days in culture.These cells were SSEA-3 and -4 negative. Some structureswere formed by one cell layer (Figs. 5A and 5E), whereasothers were formed by multiple cell layers surrounding alumen (Figs. 5B, 5C, and 5F). We call these structuresmonolayered and multilayered circular structures.
Since HPI.1-conditioned medium stimulated neural dif-ferentiation of ES cells in chamber slides, we cocultured EScells on collagen gels with HPI.1 cells seeded in two dif-ferent ways, as shown in Figure 1: A) HPI.1 cells wereembedded in the collagen gel placed on a porous membraneof the insert, and ES cells were subsequently seeded on topof that or B) HPI.1 cells were seeded at the bottom of theculture plate as a feeder layer, and subsequently ES cellswere plated on top of a collagen gel placed in the insert so
Figure 2. ES cells in a chamber slide,phase-contrast microscopy. Cellswere cultured with HPI.1-conditionedmedium for 8 days. Shown are a com-pact colony of ES-like cells (A), a colonyof epithelial cells (arrow) (B), neuronal-like cells with long processes (C), and acell with a large lamellar cytoplasm(arrow) (D). Scale bars = 50 µm.
Chen, Revoltella, Papini et al. 286
that ES and HPI.1 cells were separated by the porous mem-brane of the insert. As controls, ES cells seeded on collagengel in the absence of HPl.1 cells were used.
The circular structures both on top of and inside the col-lagen gel were observed under all three culture conditions(Fig. 5). Also, under all culture conditions, the numbers ofthese structures on day 19 were greater than those on day 7.At day 7, these circular structures were observed predomi-nantly close to the top surface of the collagen gel, and only afew dispersed cells or small dense clusters had penetratedinto the gel (Figs. 5A-5C). In contrast, on day 19, many clus-ters of cells and a large number of circular structures werefound inside the gel (Figs. 5D-5F). The numbers, sizes, and
morphologies of these structures also varied at different timepoints and culture conditions. At day 7, in controls, the struc-tures were predominantly monolayered (Fig. 5A). In con-trast, when ES cells were cultured with HPI.1 cells as feedersor embedded in collagen gel, multilayered structures pre-dominated (Figs. 5B-5C). At day 19, the majority of the cir-cular structures were multilayered. The numbers of thesestructures in the presence of HPI.1 cells were greater thanthose in the controls. This was especially noted in ES cellscultured with HPI.1 cells embedded in the collagen gel (Fig. 5F). After 7 days and, particularly, after 19 days of cul-ture, many cells within the gel appeared positively stained byanti-Ki-67 antibody (Fig. 5D, inset) and thus were cycling.
Figure 3. ES cells in a chamber slide, immunohistochemical staining. Cells were cultured with HPI.1-conditioned medium for 8 days. Cellswith neuroepithelial morphology (A and B) as well as large spread cells (C) expressed nestin (A-C). Cells with long sprouting out processes onthe top and at the edge of a colony expressed NCAM (D and E) and class III β-tubulin (F). Processes of neuron-like cells aligned in bundles (E)or formed networks (F ). Scale bars = 50 µm.
Figure 4. Epithelial differentiation of ES cells in a chamber slide. ES cells were cultured in A5RT.1-conditioned medium for 8 days. Coloniesof cells with epithelial morphology (A, phase contrast microscopy, scale bar = 50 µm) were positive for cytokeratins (B, scale bar = 25 µm).
287 ES Cells in 3D Culture System
To further characterize the cell types, we stained sectionswith antibodies against cytokeratins and p63 for the epitheliallineage; with antibodies against nestin, class III β-tubulin(TU-20), and NCAM for the neural lineage; with antibodiesagainst chromogranin A for neuroendocrine cells; and withantibodies against factor VIII for endothelial cells. The resultsof these analyses are summarized in Table 2.
Cytokeratins and p63After 7 days, in all culture conditions, almost no cells
positive for cytokeratins were found within or outside of
mono- or multilayered circular structures. However, at day19, the frequency of cytokeratin-positive cells was greater.In particular, in the presence of HPI.1 feeder cells, the num-bers of cytokeratin-positive cells were approximately twotimes higher than controls (Table 2). In monolayered struc-tures, the cytoplasm of the cells facing the lumen was moreintensively stained than the basal site (Fig. 6A). As for mul-tilayered structures, cytokeratin-positive cells were foundonly in the stratified structures and were predominantlylocated around the lumen (Fig. 6B). Again, the part of thecytoplasm that faces the lumen was stained more inten-
Figure 5. ES cells on type I collagen gel, H&E staining. ES cells were cultured for 7 days (A, B, and C) or 19 days (D, E, and F). Shown arecontrol cultures (A and D), cultures with HPI.1 cells as feeder layers (B and E), cultures with HPI.1 cells embedded in a collagen gel (C and F),and cells stained with Ki-67 antibody (D, inset). Scale bars = 50 µm.
ControlNo feeder layer
No embedded cellsHPI.1 as a feeder layer HPI.1 embedded in
collagen gel
Table 2. Percentages of stained cells in collagen gel
Culture conditions Positive cells (%)
Days Feeder Cells in gel Cytokeratin p63 Nestin NCAM Class III ββ-tubulin Chromogranin A Factor VIII
7 No No 0 0 40 15 0 5 07 No HPI.1 0 2 85 55 1 2 27 HPI.1 No 0 0 35 25 2 0 0
19 No No 10 2 20 45 30 20 019 No HPI.1 2 5 40 65 35 45 219 HPI.1 No 20 5 20 35 30 30 0
Percentages of cells stained with various antibodies were estimated in nine unconnected fields at 100×. The numbers were rounded. The typical errorwas about 10%-15% of the average value, except for low average values.
Chen, Revoltella, Papini et al. 288
sively than the basal part (Fig. 6B). The multilayered struc-tures that were negative for cytokeratins seemed to be lessorganized, without apparent stratification (Fig. 6A).
Cells that were positive for cytokeratins were negativefor nestin, class III β-tubulin, and NCAM in serial sections.To further confirm results obtained with anti-cytokeratinantibodies and to expand the range of epithelial markers inthese samples, we stained sections for p63 antigen, which isexpressed by keratinocyte precursor cells [38]. We foundthat after 19 days, about 5% of cells expressed p63 inmonolayered circular structures in cultures with HPI.1 cells
either used as feeders or embedded in collagen gel (Table2). Rarely, p63-positive cells were also found in multilay-ered structures. These p63-positive cells were located onlyin the basal cell layer and never near the lumen (Fig. 6C).In 19-day cultures, p63-positive single cells and smallaggregates without apparent organization (Fig. 6C, inset)were also detected.
NestinThe frequency of nestin-positive cells in control cultures
on day 7 was on average about 40% (Fig. 7A) of the total,
Figure 6. Epithelial differentiation of ES cells on type I collagen gel. ES cells were cultured for 19 days. Controls are shown in panels A-C.Panels A and B show cytokeratin (AE1/AE3) staining. A monolayered circular structure is shown in panel A, and a stratified multilayered cir-cular structure and cell clusters that are cytokeratin positive are shown in panel B. Cells in nonstratified circular multilayered structures areshown in panel A (arrow), and scattered cells that are not stained for cytokeratins are shown in panels A and B. Anti-p63 antibody staining isshown in panel C. The p63-positive cells are located in the basal cell layer of the circular multilayered structures (C) and in aggregates withoutapparent organization (C, inset). Scale bars = 50 µm.
Figure 7. ES cells on type I collagen gel, immunohistochemical staining. ES cells were cultured for 7 days. Shown are control cultures (A andB), cultures with HPI.1 cells as a feeder layer (C and D), and cultures with HPI.1 cells embedded in collagen gel (E and F). Consecutive sec-tions were stained for nestin (A, C, and E) and NCAM (B, D, and F). HPI.1 cells embedded in the collagen gel but not used as a feeder layerfacilitated expression of nestin (E). Scale bars = 50 µm.
ControlNo feeder layer
No embedded cellHPI.1 as a feeder layer
HPI.1 embedded in acollagen gel
289 ES Cells in 3D Culture System
and their frequency decreased to approximately 20% on day19 (Fig. 8A; Table 2). Nestin-positive cells were found pre-dominantly within various layers of the multilayered struc-tures, and, consistently, few were found outside of thesestructures. Nestin-positive cells were also found within multilayered structures in cultures with HPl.1 cells eitherused as feeders (Figs. 7C and 8E) or embedded in collagengels (Figs. 7E and 8I). However, in the latter case, the num-bers of nestin-positive cells was dramatically greater. Onday 7, more than 80% of the cells (Fig. 7E) were nestin
positive, and about 40% were nestin positive on day 19(Fig. 8I; Table 2).
NCAMAbout 15% of cells in control samples were observed to
be NCAM positive on day 7 (Fig. 7B), and their proportionincreased to approximately 40% on day 19 (Fig. 8B). For both7-day (Fig. 7D) and 19-day (Fig. 8F) cultures, coculturing ofES cells with HPI.1 cells as a feeder layer separated by theinsert membrane did not notably affect the proportion or the
ControlNo feeder layer
No embedded cellHPI.1 as a feeder layer
HPI.1 embedded in acollagen gel
Figure 8. ES cells on type I collagen gel, immunohistochemical staining. ES cells were cultured for 19 days. Shown are control cultures (A-D), cultures with HPI.1 cells as a feeder layer (E-H), and cultures with HPI.1 cells embedded in collagen gel (I-L). Consecutive sections werestained for nestin (A, E, and I), NCAM (B, F, and J), class III β-tubulin (C, G, and K), and chromogranin A (D, H, and L). HPI.1 cells embed-ded in the collagen gel (I-L) but not used as a feeder layer (E-H) facilitated expression of neural markers, nestin, class III β-tubulin, and, especially, NCAM. Scale bars = 50 µm.
Chen, Revoltella, Papini et al. 290
distribution of NCAM-positive cells. In contrast, when HPI.1cells were embedded in collagen gel, approximately 50% ofES cells were NCAM positive on day 7 (Fig. 7F) and theirnumbers were greater on day 19 (Fig. 8J; Table 2). Under allthese culture conditions, NCAM-positive cells were confinedto multilayered structures. Cells intensively stained forNCAM were located in the peripheral cell layers.
Class III β-tubulinIn control samples on day 7 of culture, all cells were
negative for class III β-tubulin. Also, only 1%-2% of cellswere observed to be class III β-tubulin positive on day 7when HPI.1 cells were used as a feeder or embedded in col-lagen gel (Table 2). In all culture conditions, the proportionof class III β-tubulin-expressing cells was greater (~30%)on day 19, especially when HPI.1 cells were embedded incollagen gel (Table 2). Similar to results obtained withNCAM staining, cells strongly expressing class III β-tubu-lin were located mainly in multilayered structures, wherethey were found in the cell layers away from the lumen(Figs. 8C, 8G, and 8K).
Chromogranin AChromogranin A was poorly expressed on day 7 both in
control, and in HPI.1 cocultures (Table 2). On day 19, thenumbers of chromogranin A-positive cells in control cul-tures were about 20% (Fig. 8D). When HPI.1 cells wereused as a feeder (Fig. 8H) or embedded in the collagen gel(Fig. 8L), the numbers of cells expressing chromogranin Awere greater (Table 2). These cells were predominantlylocated in multilayered structures.
Factor VIIIFactor VIII was rarely expressed either in controls or in
HPI.1 cocultures at any time (Table 2).In order to prove that the positively stained cells
described above were differentiated ES cells rather thanHPI.1 cells or MEFs that could be released and replatedwhen ES cells were transferred to the collagen matrix, wetested whether these feeder cells expressed nestin, NCAM,class III β-tubulin, chromogranin A, factor VIII, cytoker-atins, or p63. Immunostaining performed on HPI.1 andMEF monolayers showed that these cells were not stainedwith any of the antibodies used.
In conclusion, we demonstrated that ES cells grown in3D collagen gel proliferated and differentiated into neuraland epithelial lineages, forming circular structures resem-bling those evolving during embryogenesis. HPI.1 cellsembedded in the collagen gel but not used as a feeder layerfacilitated expression of neural markers, nestin, class III β-tubulin, and especially, NCAM.
Collagen SpongesTo determine whether the organization of the collagen
matrix affects ES cell differentiation, ES cells were seeded ontop of collagen sponges in six-well plates (Fig. 1). In contrast tothe collagen gel, the collagen sponge was highly porous, withlarge pores that allowed cells to migrate. In some experiments,we plated HPI.1 cells on the bottoms of the wells. As a control,we used ES cells seeded on collagen sponges without feeders.
The H&E staining of tangential histological sections ofthose sponges revealed that after 24 days of culture, only someES cells remained on the top of the sponge, whereas others pen-etrated into the pores and migrated downward (Fig. 9A).Within the sponge, we found many cells that were Ki-67 posi-tive (not shown). The cells within the pores tended to adhere tothe collagen fibers and formed aggregates mostly in the upperand central parts of the sponge. No aggregates were found nearthe bottom of the sponge (Fig. 9B). Besides aggregates, singlecells were found scattered throughout the entire sponge. Theamounts of these cells gradually decreased from the top to thebottom. Similarly to ES cells cultured on a collagen gel, EScells in collagen sponges formed both monolayered and multi-layered circular structures with a central lumen (Fig. 9, inset).However, the numbers of these structures formed within thesponge were consistently lower than those in collagen gel.
To characterize cells in these cultures, we used a set ofantibodies against cytokeratins, nestin, NCAM, chromograninA, factor VIII, and vimentin (Table 3).
CytokeratinsIn control cultures, there were almost no cytokeratin-
positive cells. The presence of HPl.1 cells as a feeder layer
Figure 9. ES cells cultured on a collagen sponge, H&E staining. ES cells werecultured for 24 days (control). Single cells were scattered throughout the entiresponge (A and B). Cell aggregates and circular structures (A, insets) were locatedpredominantly in the upper and central regions (A), whereas no aggregates werefound near the bottom of the sponge (B). Scale bars = 50 µm.
291 ES Cells in 3D Culture System
resulted in slightly greater numbers of such cells (Table 3).Cells immunoreactive for cytokeratins were very rarelyfound among scattered single cells within the pores butrather in mono- and stratified multilayered structures.
NestinA large proportion of nestin-positive cells (more than
50%) was found in cultures without feeder cells. Nestin wasstrongly expressed in single cells and in cells at the periph-eries of multilayered circular structures. Cells in monolayeredstructures were always nestin negative. Qualitatively, a simi-lar distribution of nestin-positive cells was observed when EScells were cocultured with HPI.1. However, the amounts ofnestin-positive cells, relative to the controls, were notablylower when they were cocultured with HPI.1 (Table 3).
NCAMNo cells expressed NCAM in control cultures in colla-
gen sponges. Positively stained cells were observed whenHPI.1 cells were used as a feeder layer (Table 3).
Chromogranin AIn control cultures, more than 30% of cells were chro-
mogranin A positive and were located in multilayeredstructures. The numbers of positive cells were slightlylower when HPI.1 cells were used as feeders (Table 3).
Factor VIIIFactor VIII was expressed in the cytoplasms of scat-
tered single cells but not in cells within circular structures(Fig. 10A). HPI.1 cell feeders did not result in markedlygreater proportions of factor-VIII-positive cells relative tocontrols (Table 3).
VimentinIn control cultures, about 40% of cells were vimentin posi-
tive. The majority of such cells were found among scattered
single cells inside the sponge (Fig. 10B), whereas few werelocated in multilayered structures. The fraction of vimentin-pos-itive cells was dramatically greater (more than 80%) when EScells were cocultured with HPI.1 cells (Table 3).
Thus, HPI.1 feeder layers resulted in dramatically greaterNCAM and vimentin expression and lower nestin expres-sion, whereas their effect on cytokeratins, factor VIII, andchromogranin A expression was less pronounced.
Next, we investigated whether the differentiation of EScells was modified when A5RT.1 cells were used as a feederlayer, as described above for chamber slides. The A5RT.1feeder cells resulted in noticeably greater factor VIII andvimentin expression and lower nestin and chromogranin Aexpression, whereas their effect on cytokeratins and NCAMexpression was less pronounced (Table 3).
Both HPI.1 and A5RT.1 cells produce numerous factors,many of which are as yet unknown. Since A5RT.1 and HPI.1
Table 3. Percentages of stained cells in collagen sponges
Culture conditions Positive cells (%)
Feeder GM-CSF (ng/ml) Cytokeratins Nestin NCAM Chromogranin A Factor VIII Vimentin
No 0 0 50 0 30 15 40
HPI.1 0 5 30 35 20 25 80
A5RT.1 0 5 10 5 5 35 80
No 20 1 30 40 5 20 80
HPI.1 20 15 35 50 20 10 65
A5RT.1 20 5 20 20 1 65 85
Percentages of cells stained with various antibodies were estimated in nine unconnected fields at 100×. The numbers were rounded. The typical errorwas about 10%-15% of the average value, except for low average values.
Figure 10. ES cells cultured on collagen sponges, immunohistochemical stain-ing. ES cells were cultured for 24 days. Shown are staining for factor VIII (A) andvimentin (B). Scale bars = 50 µm.
Chen, Revoltella, Papini et al. 292
cells are known to constitutively produce GM-CSF andexpress GM-CSF receptors [28, 39] (S. Papini and R.P.Revoltella, unpublished data), we added GM-CSF exoge-nously in an attempt to mimic the effect of an HPI.1 orA5RT.1 feeder layer. When GM-CSF (20 ng/ml) was addedto the control cultures of ES cells, the proportions of cellsexpressing NCAM and vimentin were greater, the numbers ofcells positive for nestin and chromogranin A was markedlylower, and cytokeratin and factor VIII expression was mar-ginally affected (Table 3). Thus, by comparing the propor-tions of positively stained cells in each of the different cultureconditions, we found that GM-CSF mimics the effect of anHPI.1 feeder layer on ES cell differentiation in collagensponges, with the exception of chromogranin A expression.Also, similar to the A5RT.1 feeder layer, GM-CSF resulted ingreater vimentin expression of differentiated ES cells and didnot markedly affect expression of cytokeratins and chromo-granin A in collagen sponges. Addition of GM-CSF to EScells cultured with HPI.1 feeders resulted in greater cytoker-atin and NCAM expression, slightly greater nestin expression,lower factor VIII and vimentin expression, and unchangedchromogranin A expression, compared with ES cells grownwith HPI.1 alone. In contrast, when GM-CSF was added toES cultures grown with the A5RT.1 feeder, the expressionof factor VIII was further stimulated relative to that in cul-tures grown with A5RT.1 alone (Table 3). Thus, exogenousGM-CSF, or the keratinocyte cell line producing GM-CSF,facilitated ES cell differentiation into an endothelial lin-eage.
DISCUSSION
In the course of normal embryogenesis, ES cells differen-tiate along different lineages and form complex 3D tissue andorgan structures. This pattern of differentiation is not mimic-ked in monolayer cultures. Here, we cultured rhesus monkeyES cells in 3D collagen matrixes and induced ES cell differ-entiation into various cell types that formed several tissue-likestructures similar to those in a normal embryo. CoculturingES cells with keratinocytes or fibroblasts in 3D collagenmatrixes enabled us to skew ES cell differentiation into par-ticular lineages. Since the use of human ES cells imposes cer-tain restrictions, we used rhesus monkey ES cells, which arecloser to human ES cells both morphologically and in termsof cell surface markers [2, 4] than are mouse ES cells, whichare widely used in in vitro studies [40].
The main types of structure that ES cells formed in 3Dcollagen matrixes were tubular or spherical, each containing acentral lumen. Morphologically similar structures have beendescribed in teratomas formed by monkey ES cells in SCIDmice [4]. Microscopically, these structures were circular andresembled sections of neural tubes, epithelial tubes, or glands.
The physical properties of collagen matrixes seem to affectthe formation of these structures, since, in collagen gel, theywere formed more readily than in collagen sponges, probablybecause of the density and the organization of the collagenfibers. It remains to be elucidated how these multicellularstructures were formed. We think that collagen gels retain cellaggregates that are plated together with single cells, and fromthese aggregates, circular structures are formed in the courseof ES cell differentiation. In contrast, pores in the collagensponge favor cell migration out of aggregates, diminishingtheir sizes and numbers. This speculation is in agreement withthe existence of large migratory zones observed at the periph-eries of circular structures in collagen sponges, whereas, incollagen gels, such zones were narrower. Also, many moresingle scattered cells were found in collagen sponges than incollagen gels. Moreover, in the sponges, the tubular or spher-ical structures were located in the upper part, whereas singlecells preferentially migrated to the bottom of the sponge.Thus, the collagen sponge serves not only as a supportingmatrix but also as a filter that allows migration of single cellsaway from clusters. The high mobility of cells in collagensponges made them a model for testing the invasive ability oftumor cells [41]. Collagen sponges were also used to selecthighly mobile keratinocyte precursor cells [42]. In addition tothe organization of collagen matrixes in which we cultured EScells, soluble factors and extracellular matrixes produced in these cultures also may skew ES cell differentiation to specific lineages, mimicking the process of embryogenesis.
The circular structures formed in 3D collagen matrixeswere of distinct morphological types: a lumen surrounded byone layer or multiple layers of cells, which could be strati-fied or not stratified. In nonstratified multilayered structures, differentiated ES cells were cytokeratin negative butexpressed NCAM, class III β-tubulin, chromogranin A, andnestin. These markers are typically expressed by cells ofneural lineage [33, 34, 43-48]. In contrast, cells in stratifiedmultilayered and in monolayered structures predominantlyexpressed cytokeratins [37] and p63 [38], but not nestin,NCAM, class III β-tubulin, or chromogranin A, and thuswere of epithelial lineage.
The levels of expression of different markers varied. Forexample, in nonstratified multilayered structures resemblingneural tubes, the proportions of cells expressing nestin werelower, whereas the expression of class III β-tubulin and chro-mogranin A were greater with time in culture. Moreover,cells expressing different markers were located differently.Cells expressing class III β-tubulin localized mainly in theperipheries of the multilayered structures. The same patternof temporal and spatial expression of these markers has beenreported for neurogenesis in vivo [34, 35, 46-49]. Thisstrongly suggests ex vivo maturation of these structures.
293 ES Cells in 3D Culture System
To facilitate differentiation of ES cells in 3D matrixesalong the neural lineage, we cocultured them with HPI.1fibroblasts. And indeed, in ES cells cultured with thesefibroblasts, differentiation into the neural lineage was facili-tated (as evidenced by significantly more multilayered, non-stratified circular structures formed by cells positive forNCAM, class III β-tubulin, chromogranin A, or nestin), butonly if HPI.1 fibroblasts were embedded in the collagen gel.In contrast, when HPI.1 fibroblasts were grown as a mono-layer separated from ES cells by a membrane, almost nostimulation of ES cell differentiation was observed. One pos-sible explanation for this difference is that fibroblasts directlyparticipated in the formation of circular structures; however,this is highly unlikely, since they would be distinguishablefrom ES cells both by gross morphology and by the lack ofexpression of cytokeratins, nestin, class III β-tubulin, andNCAM. Another possibility is that HPI.1 fibroblasts producesoluble factors for ES cell differentiation more efficientlywhen they are embedded in collagen gel than when they arecultured as a monolayer. It is more likely, however, thatshort-range ES cell-fibroblast interactions are necessary tostimulate differentiation along the neural lineage. It might bethat fibroblasts produce short-lived soluble factors or thatthey secrete an extracellular matrix or stimulate ES cell differentiation through direct cell-cell contacts.
The importance of cell-cell interactions for ES cell dif-ferentiation is emphasized by the fact that single cells incollagen sponges differentiated substantially differentlyfrom cells in aggregates. The majority of single cells insponges expressed factor VIII and thus were apparently ofthe endothelial type, whereas cells in multicellular struc-tures did not express this marker.
Interactions of cells with each other and with collagenmatrixes in the course of 3D structure formation were asso-ciated with changes in the morphologies of individual cells.For example, in contrast to 3D systems, cells of the neuraltype in two-dimensional culture were more differentiated;they displayed classical neuronal morphology with longdendrite-like processes forming a network, and they oftenassembled in bundles. This network was formed on the topof ES cell colonies and MEFs, and thus cell-cell interactionseems to be an important factor here as well. Also, mouseand monkey ES cells cultured on a stromal cell layer derivedfrom bone marrow differentiated into cells of the neuronaltype, with dendrite processes [13, 14]. Another example ofhow the structures are related to cell morphology is the formation of epithelial-like clusters by cells expressingcytokeratins, but not NCAM, nestin, or chromogranin A. In3D systems, phenotypically similar cells form circular struc-tures. Thus, cells differentiating along a similar lineageacquire different morphologies and assemble in different
structures depending on whether they are restricted to twodimensions or are allowed to interact in 3D matrixes.
Soluble factors significantly modulate ES cell differenti-ation. One such factor is GM-CSF, a pleiotropic cytokineinvolved in the proliferation, maturation, and functional activ-ity of hematopoietic [50] and nonhematopoietic progenitorcells [51-53]. Addition of soluble GM-CSF in the presence ofA5RT.1 cells, which have GM-CSF receptors and also pro-duce GM-CSF, resulted in significantly more factor VIII-pos-itive cells. This observation is consistent with GM-CSFpromotion of angiogenesis in an in vivo model [51]. In addi-tion, GM-CSF modulates sensitivity of progenitor cells of dif-ferent lineages to various cytokines [54, 55]. Recently, it hasbeen shown that embryonic cells express various cytokinereceptors [7, 56]. Therefore, it is possible that GM-CSF mod-ulates ES cell sensitivity to A5RT.1-produced cytokines,which are important for ES cell differentiation along theendothelial lineage. In addition to GM-CSF, both HPI.1 andA5RT.1 cells produce numerous factors, many of which are not yet known and may affect ES cell differentiation. Inthe future, it will be necessary to analyze their effect on EScell differentiation in the context of 3D systems.
In conclusion, we found that ES cells in 3D collagenmatrixes differentiated differently from those in monolay-ers. Moreover, the abilities of ES cells to form various struc-tures and to differentiate along particular lineages in twochemically similar but topologically different matrixes weredifferent. Cells expressing the same differentiation markersacquired different morphological characteristics in differentmicroenvironments. ES cell differentiation and formation oftissue-like structures could be modulated by already differ-entiated cells. Both soluble factors and the type of extracel-lular matrix seemed to be critical in directing differentiationof ES cells and the formation of tissue- and organ-like struc-tures in vitro. Three-dimensional culture systems are a valu-able tool for directing ES cell differentiation and formationof organs and tissue for transplantation.
ACKNOWLEDGMENTS
The work of Silvia S. Chen, Wendy Fitzgerald, JoshuaZimmerberg, and Leonid Margolis was funded by the NIHintramural program and, in part, by the NASA/NIH Centerfor Three-Dimensional Tissue Culture. The work of RobertoP. Revoltella, Sandra Papini, and Monica Michelini wassupported in part by the Italian Ministry of Health (ProjectStem 2001 and FIRB: New Medical Engineering) and inpart by Farmigea SpA. We thank Dr. Daniela Campani,Department of Oncology, University of Pisa, for advice andassistance in immunocytochemical studies. We also thankMs. Emily Graze and Mr. Bennet Porter for their assistancein image analysis and cell counting.
Chen, Revoltella, Papini et al. 294
REFERENCES
1 Evans MJ, Kaufman MH. Establishment in culture of pluripo-tential cells from mouse embryos. Nature 1981;292:154-156.
2 Thomson JA, Kalishman J, Golos TG et al. Isolation of a pri-mate embryonic stem cell line. Proc Natl Acad Sci USA1995;92:7844-7848.
3 Thomson JA, Kalishman J, Golos TG et al. Pluripotent celllines derived from common marmoset (Callithrix jacchus)blastocysts. Biol Reprod 1996;55:254-259.
4 Thomson JA, Marshall VS. Primate embryonic stem cells.Curr Top Dev Biol 1998;38:133-165.
5 Shamblott MJ, Axelman J, Wang S et al. Derivation ofpluripotent stem cells from cultured human primordial germcells. Proc Natl Acad Sci USA 1998;95:13726-13731.
6 Bradley A, Evans M, Kaufman MH et al. Formation of germ-line chimaeras from embryo-derived teratocarcinoma celllines. Nature 1984;309:255-256.
7 Schuldiner M, Yanuka O, Itskovitz-Eldor J et al. Effects ofeight growth factors on the differentiation of cells derivedfrom human embryonic stem cells. Proc Natl Acad Sci USA2000;97:11307-11312.
8 Rathjen PD, Lake J, Whyatt LM et al. Properties and uses ofembryonic stem cells: prospects for application to humanbiology and gene therapy. Reprod Fertil Dev 1998;10:31-47.
10 Jacobson L, Kahan B, Djamali A et al. Differentiation ofendoderm derivatives, pancreas and intestine, from rhesusembryonic stem cells. Transplant Proc 2001;33:674.
11 O’Shea KS. Embryonic stem cell models of development.Anat Rec 1999;257:32-41.
12 Kawasaki H, Mizuseki K, Nishikawa S et al. Induction ofmidbrain dopaminergic neurons from ES cells by stromalcell-derived inducing activity. Neuron 2000;28:31-40.
13 Kawasaki H, Suemori H, Mizuseki K et al. Generation ofdopaminergic neurons and pigmented epithelia from primateES cells by stromal cell-derived inducing activity. Proc NatlAcad Sci USA 2002;99:1580-1585.
14 Li F, Lu S, Vida L et al. Bone morphogenetic protein 4 inducesefficient hematopoietic differentiation of rhesus monkeyembryonic stem cells in vitro. Blood 2001;98:335-342.
15 Carpenter MK, Inokuma MS, Denham J et al. Enrichment ofneurons and neural precursors from human embryonic stemcells. Exp Neurol 2001;172:383-397.
16 Reubinoff BE, Itsykson P, Turetsky T et al. Neural progeni-tors from human embryonic stem cells. Nat Biotechnol2001;19:1134-1140.
17 Bain G, Kitchens D, Yao M et al. Embryonic stem cells expressneuronal properties in vitro. Dev Biol 1995;168:342-357.
18 Okabe S, Forsberg-Nilsson K, Spiro AC et al. Development ofneuronal precursor cells and functional postmitotic neuronsfrom embryonic stem cells in vitro. Mech Dev 1996;59:89-102.
19 Rolletschek A, Chang H, Guan K et al. Differentiation ofembryonic stem cell-derived dopaminergic neurons is enhancedby survival-promoting factors. Mech Dev 2001;105:93-104.
20 Zhang SC, Wernig M, Duncan ID et al. In vitro differentiationof transplantable neural precursors from human embryonic stemcells. Nat Biotechnol 2001;19:1129-1133.
21 Wobus AM, Wallukat G, Hescheler J. Pluripotent mouseembryonic stem cells are able to differentiate into cardiomy-ocytes expressing chronotropic responses to adrenergic andcholinergic agents and Ca2+ channel blockers. Differentiation1991;48:173-182.
22 Kehat I, Kenyagin-Karsenti D, Snir M et al. Human embry-onic stem cells can differentiate into myocytes with structuraland functional properties of cardiomyocytes. J Clin Invest2001;108:407-414.
23 Hamazaki T, Iiboshi Y, Oka M et al. Hepatic maturation indifferentiating embryonic stem cells in vitro. FEBS Lett2001;497:15-19.
24 O’Shea KS. Neuronal differentiation of mouse embryonic stemcells: lineage selection and forced differentiation paradigms.Blood Cells Mol Dis 2001;27:705-712.
25 Lenka N, Lu ZJ, Sasse P et al. Quantitation and functional char-acterization of neural cells derived from ES cells using nestinenhancer-mediated targeting in vitro. J Cell Sci 2002;115:1471-1485.
26 Meyer N, Jaconi M, Landopoulou A et al. A fluorescentreporter gene as a marker for ventricular specification in ES-derived cardiac cells. FEBS Lett 2000;478:151-158.
27 Fusenig NE, Boukamp P. Multiple stages and genetic alter-ations in immortalization, malignant transformation, and tumorprogression of human skin keratinocytes. Mol Carcinog1998;23:144-158.
28 Mueller MM, Fusenig NE. Constitutive expression of G-CSFand GM-CSF in human skin carcinoma cells with functionalconsequence for tumor progression. Int J Cancer 1999;83:780-789.
29 Seftor RE, Seftor EA, Koshikawa N et al. Cooperative interac-tions of laminin 5 gamma2 chain, matrix metalloproteinase-2,and membrane type-1-matrix/metalloproteinase are requiredfor mimicry of embryonic vasculogenesis by aggressivemelanoma. Cancer Res 2001;61:6322-6327.
30 Okamoto E, Kitano Y. Expression of basement membranecomponents in skin equivalents—influence of dermal fibro-blasts. J Dermatol Sci 1993;5:81-88.
31 Kannagi R, Levery SB, Ishigami F et al. New globoseries gly-cosphingolipids in human teratocarcinoma reactive with themonoclonal antibody directed to a developmentally regulatedantigen, stage-specific embryonic antigen 3. J Biol Chem1983;258:8934-8942.
32 Kannagi R, Cochran NA, Ishigami F et al. Stage-specificembryonic antigens (SSEA-3 and -4) are epitopes of a uniqueglobo-series ganglioside isolated from human teratocarci-noma cells. EMBO J 1983;2:2355-2361.
295 ES Cells in 3D Culture System
33 Hockfield S, McKay RD. Identification of major cell classesin the developing mammalian nervous system. J Neurosci1985;5:3310-3328.
34 Rutishauser U, Jessell TM. Cell adhesion molecules in verte-brate neural development. Physiol Rev 1988;68:819-857.
35 Lee MK, Tuttle JB, Rebhun LI et al. The expression and post-translational modification of a neuron-specific beta-tubulinisotype during chick embryogenesis. Cell Motil Cytoskeleton1990;17:118-132.
36 Moody SA, Miller V, Spanos A et al. Developmental expres-sion of a neuron-specific beta-tubulin in frog (Xenopus lae-vis): a marker for growing axons during the embryonicperiod. J Comp Neurol 1996;364:219-230.
37 Woodcock-Mitchell J, Eichner R, Nelson WG et al. Immuno-localization of keratin polypeptides in human epidermis usingmonoclonal antibodies. J Cell Biol 1982;95:580-588.
38 Pellegrini G, Dellambra E, Golisano O et al. p63 identifies keratinocyte stem cells. Proc Natl Acad Sci USA 2001;98:3156-3161.
39 Park KC, Kim DS, Kim HJ et al. Growth related secretion andproduction of GM-CSF by epithelial cell line. J Dermatol Sci2001;25:53-58.
41 Hoffman RM. Three-dimensional sponge-gel matrix histocul-ture: methods and applications. In: Celis JE, ed. Cell Biology:A Laboratory Handbook. San Diego: Academic Press,1998:377-389.
42 Papini S, Cecchetti D, Campani D et al. Isolation and clonalanalysis of human epidermal keratinocyte stem cells in long-term culture. STEM CELLS (in press).
43 Easter Jr SS, Ross LS, Frankfurter A. Initial tract formationin the mouse brain. J Neurosci 1993;13:285-299.
44 Draberova E, Lukas Z, Ivanyi D et al. Expression of class IIIbeta-tubulin in normal and neoplastic human tissues.Histochem Cell Biol 1998;109:231-239.
45 Duittoz AH, Hevor T. Primary culture of neural precursorsfrom the ovine central nervous system (CNS). J NeurosciMethods 2001;107:131-140.
46 Kleinert R. Immunohistochemical characterization of primi-tive neuroectodermal tumors and their possible relationship to
the stepwise ontogenetic development of the central nervoussystem. 1. Ontogenetic studies. Acta Neuropathol (Berl)1991;82:502-507.
47 Zimmerman L, Parr B, Lendahl U et al. Independent regulatoryelements in the nestin gene direct transgene expression to neuralstem cells or muscle precursors. Neuron 1994;12:11-24.
48 Lothian C, Lendahl U. An evolutionarily conserved region inthe second intron of the human nestin gene directs geneexpression to CNS progenitor cells and to early neural crestcells. Eur J Neurosci 1997;9:452-462.
49 Mujtaba T, Piper DR, Kalyani A et al. Lineage-restrictedneural precursors can be isolated from both the mouse neuraltube and cultured ES cells. Dev Biol 1999;214:113-127.
50 Williams ME, Quesenberry PJ. Hematopoietic growth factors.Hematol Pathol 1992;6:105-124.
51 Bussolino F, Ziche M, Wang JM et al. In vitro and in vivoactivation of endothelial cells by colony-stimulating factors. J Clin Invest 1991;87:986-995.
52 Stoof TJ, Boorsma DM, Nickoloff BJ. Keratinocytes andimmunological cytokines. In: Leigh I, Lane EB, Watt FM,eds. The Keratinocyte Handbook. Cambridge: CambridgeUniversity Press, 1994:363-399.
53 Hamilton JA, Piccoli DS, Cebon J et al. Cytokine regulationof colony-stimulating factor (CSF) production in culturedhuman synovial fibroblasts. II. Similarities and differences inthe control of interleukin-1 induction of granulocyte-macrophage CSF and granulocyte-CSF production. Blood1992;79:1413-1419.
54 Ulich TR, del Castillo J, McNiece IK et al. Stem cell factor incombination with granulocyte colony-stimulating factor (CSF)or granulocyte-macrophage CSF synergistically increasesgranulopoiesis in vivo. Blood 1991;78:1954-1962.
55 Bernstein ID, Andrews RG, Zsebo KM. Recombinant humanstem cell factor enhances the formation of colonies by CD34+
and CD34+lin– cells, and the generation of colony-forming cellprogeny from CD34+lin– cells cultured with interleukin-3, gran-ulocyte colony-stimulating factor, or granulocyte-macrophagecolony-stimulating factor. Blood 1991;77:2316-2321.
56 Shamblott MJ, Axelman J, Littlefield JW et al. Humanembryonic germ cell derivatives express a broad range ofdevelopmentally distinct markers and proliferate extensivelyin vitro. Proc Natl Acad Sci USA 2001;98:113-118.
