One of the most frequently used matrices for feeder-free growth of undifferentiated human embryonic stem cells (hESCs) is Matrigel, which supports attachment and growth of undifferentiated hESCs in the presence of mouse embryonic fibroblast–conditioned medium. Unfortunately, application of Matrigel or medium conditioned by mouse embryonic feeder cells is not ideal for potential medical application of hESCs because xenogeneic pathogens can be transmitted through culture conditions. We demonstrate here that human serum as matrix and medium conditioned by differentiated hESCs reduce exposure of hESCs to animal ingredients and provide a safer direction toward completely animal-free conditions for application, handling, and understanding of hESC biology. At the same time, hESCs grown under these conditions maintain all hESC features after prolonged culture, including the developmental potential to differentiate into representative tissues of all three embryonic germ layers, unlimited and undifferentiated proliferative ability, and maintenance of normal karyotype.
Human embryonic stem cells (hESCs) have been derived from the inner cell mass (ICM) of day-5 through -8 blastocysts [1–9]. To date, derivation and propagation of undifferentiated hESCs requires plating of both ICM and hESC colonies on mouse embryonic fibroblast (MEF) cells or human feeder (HFF) cells [4,10–12], and this limits the large-scale culture and genetic manipulation of hESCs . To overcome these obstacles, feeder-free systems have been introduced in which hESCs can be grown on different matrices with addition of MEF-conditioned medium or hESC medium supplemented with serum replacement, different growth factors, or in the presence of 6-bromoindirubin-3′-oxime, a specific pharmacological inhibitor of glycogen synthase kinase-3 [14–22].
One of the most frequently used matrices for feeder-free growth of undifferentiated hESCs is Matrigel, which supports attachment and growth of undifferentiated hESCs in the presence of MEF-conditioned medium [14–16]. Matrigel is ananimal basement membrane preparation extracted from Engelbreth-Holm-Swarm mouse sarcoma, a tumor rich in extracellular matrix (ECM) proteins: laminin, collagen IV, heparan sulfate proteoglycans, entactin, and nidogen 1. Unfortunately, application of Matrigel or MEF-conditioned medium is not ideal for potential medical application of hESCs because xenogeneic pathogens can be transmitted through culture conditions [12, 18].
We previously demonstrated  that hESCs could be successfully grown on Matrigel with addition of medium conditioned by the fibroblasts derived from differentiated hESCs (hES-dF). In this manuscript, we evaluated whether human serum (HS) could be successfully used as a matrix to help the attachment and growth of hESCs with the aim to create feeder-free and more patient-friendly conditions for the long-term growth of undifferentiated hESCs. We demonstrate here that HS and medium conditioned by hES-dF reduce exposure of hESCs to animal ingredients and provide a safer direction toward completely animal-free conditions for application, handling, and understanding of hESC biology. At the same time, hESCs grown under these conditions maintain all hESC features after prolonged culture, including the developmental potential to differentiate into representative tissues of all three embryonic germ layers, unlimited and undifferentiated proliferative ability, and maintenance of normal karyotype.
Materials and Methods
Recovery of hES-dF–Conditioned Medium
We previously described  derivation of fibroblast-like cells from hES-dF, and only cells with fibroblast-like cell morphology were chosen for derivation of hES-dF. These cells were identified, characterized, and successfully used as a feeder or to recover conditioned medium for feeder-free growth of undifferentiated hESCs . To recover conditioned medium, mitotically inactivated hES-dFs were cultured at a density of 56,000 cells per cm2 in T-25 flask with addition of hESC medium (knockout-Dulbecco's modified Eagle's medium [Invitrogen, Carlsbad, CA, http://www.invitrogen.com], 100 μM β-mercaptoethanol [Sigma, St. Louis, http://www.sigmaaldrich.com], 2 mM L-glutamine [Sigma], 100 mM nonessential amino acids, 20% serum replacement (SR) [Invitrogen], 1% penicillin-streptomycin [Sigma], and 4 ng/ml basic fibroblast growth factor [bFGF] [Invitrogen]) for 10 days. hES-dF–conditioned medium was collected every day and then frozen at −80°C before use. After thawing, 8 ng/ml bFGF and 1% of human insulin, transferrin, selenium for adherent cells (Invitrogen) was added.
Growth of hESCs in Feeder-Free System
Two different hESC lines, H1 (WiCell Inc., Madison, WI, http://www.wicell.org) and hES-NCL1, were grown on MEF until passages 43 and 47, respectively, and then transferred on tissue culture plates (Nunc, Roskilde, Denmark, http://www.nuncbrand.com) precoated either with Matrigel (BD, Bedford, MA, http://www.bdbiosciences.com) or with HS (Sigma, Cat. No. H1388). According to the manufacturer, HS had been derived from male clotted blood (all from the U.S.) tested and found negative for hepatitis B surface antigen, anti–hepatitis C virus, and anti-HIV/HIV-2 by FDA-approved tests. To coat plates with HS, the surface of plates was overlaid with HS for 1 hour at room temperature. After that, HS was removed and plates were dried for an additional hour at room temperature. Colonies of hESCs were grown on HS in the presence of hES-dF–conditioned media, and medium was changed every 48 hours. hESC colonies were disaggregated mechanically every 4–6 days and replated onto freshly prepared plates.
Characterization of hESCs Cultured on HS
To investigate whether hESCs grown on HS and in the presence of hES-dF–conditioned medium maintain their undifferentiated and pluripotent state, we performed immunocytochemical live staining of hESC-surface markers as follows: primary antibodies TRA-1-60 (1:100), TRA-1-81 (1:100), and SSEA-4 (1:100) (Chemicon, Temecula, CA, http://www.chemicon.com) were added to hESC for 20 minutes at 37°C. The samples were gently washed three times with embryonic stem cell (ESC) medium before being incubated with the secondary antibodies (Sigma) conjugated to fluorescein isothiocyanate (FITC) at 37°C for 20 minutes. All samples were again washed three times with ESC medium and subjected to fluorescence microscopy. TRA-1-60 and TRA-1-81 samples were additionally stained with 1 μg/ml propidium iodide (PI) for 5 minutes. The bright field and fluorescent images were obtained using a Zeiss microscope and the AxioVision software (Carl Zeiss, Jena, Germany, http://www.zeiss.com). The alkaline phosphatase (AP) staining was carried out using the AP Detection Kit following the manufacturer's instructions (Chemicon). Briefly, cells were fixed in 90% methanol and 10% formamide for 2 minutes and then washed with rinse buffer (20 mM Tris-HCl, pH 7.4, 0.05% Tween-20) once. Staining solution (Naphthol/Fast Red Violet) was added to the wells, and plates were incubated in the dark for 15 minutes. For OCT-4 immunostaining, hESCs were fixed in 3.7% formaldehyde (BDH, Poole, U.K., http://www.bdh.com) for 20 minutes at room temperature, followed by incubation in 3% hydrogen peroxide for 10 minutes. The hESCs were permeabilized with 0.2% Triton X-100 (Sigma) diluted in 4% sheep serum (Sigma) for 30 minutes at 37°C. The hESC colonies were washed with phosphate-buffered saline (PBS) supplemented with 3% H2O2 and then incubated with the primary antibodies (OCT-4 from Santa Cruz Biotechnologies, Heidelberg, Germany http://www.scbt.com) to a final concentration of 10 μg/ml for 30 minutes at room temperature. The hESC colonies were washed twice with PBS for 5 minutes and then incubated with the secondary antibody (biotinylated rat anti-mouse immunoglobulin [Dako Cytomation, Cambridgeshire, U.K., http://www.dakocytomation.com] used at 1:100 dilution) for 30 minutes at room temperature. After that, hESCs were washed again with PBS, incubated with avidinbiotin complex/horseradish peroxidase solution for 25 minutes at room temperature, and washed again with PBS. The detection was carried out by incubation with DAB (Sigma) solution at room temperature for 1 minute. Final washes were done with distilled water. The primary antibody was omitted for the negative control.
For the flow cytometry analysis, the hESC colonies grown on Matrigel or HS were collected using collagenase IV treatment (1 mg/ml for 5 minutes) followed by brief trypsin incubation (1 minute at 0.025% trypsin/0.25 M EDTA). Cell clumps were removed by filtering the cell solution though a nylon mesh (70 μm [BD]). hESCs were suspended in PBS supplemented with 2% fetal calf serum (FCS) at a concentration of 106 cells per ml. Cell suspension (100 μl) was stained with TRA-1-81 (10 μg/ml final concentration [Chemicon]). Three washes in staining buffer were carried out before staining with secondary antibody, goat anti-mouse immunoglobulin-FITC (6-μg/ml final concentration [Sigma]), washed again three times, and resuspended in staining buffer before being analyzed with fluorescence-activated cell sorter Calibur (BD) using the Cell Quest software (BD). Ten thousand events were acquired for each sample, and PI staining (1 μg/ml) was used to distinguish live from dead cells.
Environmental Scanning Electron Microscopy and Scanning Electron Microscopy of hESCs
Environmental scanning electron microscopy (ESEM) samples (noncoated and coated plates) had been predried and examined in a 30XL FEG microscope (FEI Phillips, Acht, Netherlands, http://www.feicompany.com). For scanning electron microscopy (SEM), the images were collected using the Wide Field Gaseous Secondary Electron Detector. For SEM analysis, hESCs grown on Thermanox plastic coverslips (Agar Scientific, Stansted, U.K., http://www.agarscientific.com) coated with Matrigel or HS were treated with 2% gluteraldehyde (TAAB Laboratory Equipment, Aldermaston, U.K., http://www.taab.co.uk) in Sorenson's phosphate buffer (SPB) and then washed again with SPB. After that, the dehydration was undertaken as follows: 25% ethanol for 30 minutes, 50% ethanol for 30 minutes, 75% ethanol for 30 minutes, 100% ethanol for 1 hour, and 100% ethanol for 1 hour. Final dehydration was done with carbon dioxide in a Samdri 780 Critical Point Dryer. The cells were mount on aluminum stub with Achesons Silver Electro Dag (Agar Scientific) and coated with 15 nm of gold using a Polaron SEM Coating Unit. The specimens were examined using a Stereoscan 240 SE microscope.
The reverse transcription (RT) was carried out to investigate the presence of specific hESCs and markers of different germ lineages expressed in undifferentiated or spontaneously differentiated hESCs. RT was done using the cells to cDNA II kit (Ambion, Huntingdon, U.K., http://www.ambion.com) according to the manufacturer's instructions. In brief, hESCs were submerged in 100 μl of ice-cold cell lysis buffer and lysed by incubation at 75°C for 10 minutes. Genomic DNA was degraded by incubation with DNAse I for 15 minutes at 37°C. RNA was reverse transcribed using Moloney murine leukemia virus RT and random hexamers following the manufacturer's instructions. Polymerase chain reaction (PCR) was carried out using the primers as described in Table 1. PCR products were run on 2% agarose gels and stained with ethidium bromide. Results were assessed on the presence or absence of the appropriate size PCR products. RT negative controls were included to monitor genomic contamination.
Table Table 1.. List of primers used for polymerase chain reaction and determination of pluripotency and differentiation genes
5′-GAA GGT ATT CAG CCA AAC -3′
5′-CTT AAT CCA AAA ACC CTG G -3′
Karyotype Analysis of hESCs
To investigate the stability of hES-NCL1 and H1 cells grown on HS and in the presence of hES-dF–conditioned medium, the karyotype of 10 hESCs was determined by standard G-banding procedure.
Tumor Formation in Severe Combined Immunodeficient Mice
Approximately 2,000 to 3,000 hES-NCL1 cells grown on HS and in the presence of hES-dF–conditioned medium were injected beneath the capsule of the testis in adult severe combined immunodeficient (SCID) mice. After 6 weeks, mice were euthanized and tissues were dissected, fixed in Bouin's fluid overnight, processed, and sectioned according to standard procedures and counterstained with either hematoxylin and eosin or Weigerts stain. Sections (5–8 μm) were examined using bright-field light microscopy and photographed as appropriate.
In Vitro Differentiation of hESCs
For this experiment, colonies of hES-NCL1 cells grown on HS and in the presence of hES-dF–conditioned medium (passages 16 and 19) were replated on new HS-coated plates in hES-dF–conditioned medium. After 5–12 days without passaging, spontaneous differentiation was observed and differentiated cells were replated and cultured under the same conditions. Cells were fixed in 4% paraformaldehyde in PBS (Sigma) for 30 minutes and then permeabilized for an additional 10 minutes with 0.1% Triton X (Sigma). The blocking step was 30 minutes with 2% FCS in PBS. Differentiated cells were incubated with antibody against tubulin β III (1:100; Chemicon), α-actinin/sarcomeric (1:800; Sigma), or α-fetoprotein (1:500; Sigma) for an additional 2 hours. Each antibody was detected using corresponding secondary antibodies conjugated to FITC. The nuclei of cells were stained using Hoechst 33342 for 5 minutes. Fat cells were washed twice with PBS, fixed in 4% paraformaldehyde, washed again with PBS, and stained with 0.3% oil red in 60% isopropanol. Then the cells were washed three times with distilled water. The bright field and fluorescent images were obtained using a Zeiss microscope and the AxioVision LE Release 4.2 software (Carl Zeiss).
Surfaces of culture plates coated with Matrigel or HS were investigated using scanning electron microscopy. This analysis revealed that coating with Matrigel or HS provides a substratum with similar shape (Figs. 1A, 1B), which assists cell adhesion and supports undifferentiated growth of both hESC lines, hES-NCL1 and H1 (Figs. 1C, 1D). When transferred on uncoated plates, hESC colonies do not attach; however, they do form embryoid bodies (not shown) or attach and spontaneously differentiate. On the contrary, both lines cultured on HS and in the presence of hES-dF medium kept their pluripotency for over 27 passages. We found that undifferentiated hESCs grown on HS show typical morphology, that is, small cells with prominent nucleoli (Figs. 2A, 2B) expressing typical cell-surface and intracellular hESC markers: TRA-1-60 (Fig. 2C), TRA-1-81 (Fig. 2D), SSEA-4 (Fig. 2E), AP (Fig. 2F), and OCT-4 (Fig. 2G). When hESCs of both cell lines were grown on HS and in the absence of hES-dF medium, spontaneous differentiation was observed 48 hours after passaging (Fig. 2I). RT-PCR analysis of undifferentiated hESCs showed positive expression of OCT-4, REX-1, NANOG, and TERT (Fig. 2J). Comparative flow cytometry analysis revealed that hES-NCL1 cells grown on Matrigel or HS for 21 passages expressed 79.1% and 81.5% of TRA-1-81 antigen, respectively (data not shown).
Karyotyping of the hESCs showed that both lines hES-NCL1 and H1 grown on HS in the presence of hES-dF–conditioned medium have a normal female or male karyotype (Figs. 3A, 3B), keeping their genomic stability even after 21 passages.
When hES-NCL1 or H1 cells grown on HS in the presence of hES-dF–conditioned medium were not replated after 4–6 days (see Material and Methods), spontaneous differentiation into neuronal precursor, fat cells, cardiomyocytes, and endoderm-like cells was observed (Figs. 4A–4D). The presence of the cells of all three germ lineages was confirmed by RT-PCR, in which expression of cytokeratin 3 (CK3), cytokeratin 19 (CK19) PAX6, NESTIN, GATA4, and Indian hedgehog genes was observed (Fig. 4E). This demonstrates that both hESC lines cultured on HS have the potential to spontaneously differentiate into cells of all three germ layers under in vitro conditions. PCR analysis of undifferentiated hESCs using the same markers as for differentiated hESCs showed only expression of CK19 (Fig. 4F).
Under in vivo conditions, hES-NCL1 grown on HS and in the presence of hES-dF–conditioned medium consistently developed into teratomas when grafted into SCID mice. The teratomas were always restricted to the site of transplantation. Gross analysis of excised tumor tissues showed solid teratomas and lesions containing fluid-filled cystic masses accompanied with solid tissues. Histological examination of teratomas revealed advanced differentiation of structures representative of all three embryonic germ layers, including cartilage, muscle, primitive neuroectoderm, neural ganglia, kidney, secretory epithelia, and connective tissues (Figs. 5A–5F). Moreover, such tissues formed complex arrangements, recapitulating the development of complex structures that no doubt require coordinated interactions between different cell types derived from different germ layers.
In this study we demonstrated that HS matrix and medium conditioned by hESC-derived fibroblast-like cells successfully support the attachment and growth of undifferentiated hESCs under feeder-free conditions. These cells expressed all specific cell-surface and intracellular hESC markers, keeping their genomic stability and pluripotency for more than 27 passages, as demonstrated by karyotyping and the ability of hESCs to spontaneously differentiate under in vitro and in vivo conditions.
Several studies described the use of HS as serum supplement to grow hESCs. For instance, Richards et al.  reported the possibility of growing hESCs on human fibroblasts in medium supplemented with HS for at least 10 passages. Under these conditions, hESCs maintained typical features, including morphology, pluripotency, and expression of cell-surface markers. However, the use of HS in culture media was not beneficial for prolonged cultures because increased differentiation rates of hESCs were observed . It is important to point out that HS may be one part of a complex interplay between secreted factors from the feeder cells used in this study. In another study , the human foreskin feeders were cultured using HS continuously for more than 3 months and shown to be similar to the lines derived with bovine serum. In addition, these fibroblast cells were shown to support the growth and pluripotency of hESCs. In this study, we demonstrate that HS can be used as matrix to support attachment and growth of hESCs. Interestingly, we observed that different batches (093K0475, 122K0424, and 052K0983) of commercially available HS or HS recovered from type 1 diabetes patients supported attachment, undifferentiated growth, and spontaneous differentiation of hESCs in a similar manner (data not shown). This demonstrates that different soluble growth factors, adhesion molecules, and ECM components are common and probably consistently present within different batches of HS. HS contains ECM components, including fibronectin, vitronectin, hyaluronic acid, and other factors, which allow attachment and survival of cells . Amit et al.  described growth of hESCs in a feeder-free system, which was based on medium supplemented with SR and a combination of different growth factors, including transforming growth factor β1, leukemia inhibitory factor, bFGF, and fibronectin matrix of different origin. However, when hESCs were grown on human fibronectin, they started to differentiate after several passages . Here, data obtained by flow cytometry demonstrate that both HS and Matrigel in the presence of hES-dF–conditioned medium efficiently supported growth of undifferentiated hESCs. hESCs grown in the presence of ES medium differentiated much faster than those grown in the presence of hES-dF–conditioned medium, which again demonstrates the need for factors present in conditioned medium derived either from MEF  or hES-dF . Significant achievements toward feeder-free conditions for growth of hESCs have been described previously [20, 21]. In these two studies, the authors used a high concentration of bFGF to keep hESCs undifferentiated but used Matrigel to coat the plates. In another study, Klimanskaya et al.  were able for the first time to derive a new hESC line under feeder-free conditions but again used ECM components derived from MEF extracts and in the presence of SR. Therefore, replacement of Matrigel , ECM derived from MEF , or MEF-conditioned medium [14–16] by HS and hES-dF–conditioned medium minimizes source of animal ingredients contained in commercially available serum replacement. Therefore, detailed analysis of HS and secreted factors by human feeders is necessary to identify factors that trigger attachment, survival, and proliferation of hESCs, as previously has been done for MEF . This is of crucial importance because conditioned medium derived from different human foreskin fibroblast cell lines  or conditioned media derived from hESC-derived fibroblasts [20 and this study] successfully support feeder-free growth of hESCs. In addition, to optimize growth conditions and for better understanding of hESC biology, numerous studies are necessary to investigate whether HS or other matrix components affect attachment, proliferation, and survival via CD44, proteinase inhibitors, or focal adhesion kinase pathways as demonstrated for vascular  or airway  smooth muscle cells and whether different matrices and different culture conditions drive specialized hESC differentiation via induction or repression of genes [29, 30].
In conclusion, HS used as matrix and conditioned medium recovered from hES-dF maintain pluripotency and genomic stability of hESCs. This system allowed growth of undifferentiated hESCs, which were able to spontaneously differentiate into cells of all three germ lineages under in vitro and in vivo conditions. This easily accessible feeder-free system could be used to create individual and patient-friendly growth systems, reduce exposure of hESCs to animal ingredients, and offer excellent possibilities to identify human factors that help attachment and proliferation of undifferentiated hESCs. The growth system described here will be very helpful in attempts to develop safe conditions for growth and handling of undifferentiated and differentiated hESCs.
The authors would like to thank Vivian Thompson, Tracey Scott-Davey (Biomedical EM Unit, Newcastle University, Newcastle upon Tyne), and Grant Staines (SAgE Faculty Services, Newcastle University) for SEM and ESEM analysis. This work was supported by Newcastle University Hospitals Special Trustees, One NorthEast Regional Development Agency (Newcastle), Newcastle Health Charity, the Department of Health, and Medical Research Council, London, grant No. G0301182.