Human Cell

, Volume 25, Issue 1, pp 16–23

Efficient derivation of human embryonic stem cell lines from discarded embryos through increases in the concentration of basic fibroblast growth factor


  • Yanwei Wang
    • Shandong Research Center of Stem Cell Engineering Yantai Yuhuangding Hospital
  • Chenming Xu
    • In Vitro Fertilization Center of Women’s Hospital, School of MedicineZhejiang University
  • Haiyan Wang
    • Shandong Research Center of Stem Cell Engineering Yantai Yuhuangding Hospital
  • Juan Liu
    • Shandong Research Center of Stem Cell Engineering Yantai Yuhuangding Hospital
  • Shi Hui
    • Shandong Research Center of Stem Cell Engineering Yantai Yuhuangding Hospital
  • Ning Li
    • Shandong Research Center of Stem Cell Engineering Yantai Yuhuangding Hospital
  • Fujun Liu
    • Shandong Research Center of Stem Cell Engineering Yantai Yuhuangding Hospital
    • Shandong Research Center of Stem Cell Engineering Yantai Yuhuangding Hospital
Research Article

DOI: 10.1007/s13577-011-0039-7

Cite this article as:
Wang, Y., Xu, C., Wang, H. et al. Human Cell (2012) 25: 16. doi:10.1007/s13577-011-0039-7


We describe the derivation and characterization of three novel human embryonic stem (hES) cell lines (YT1, YT2, YT3). One hES line (YT1) was obtained from six discarded blastocysts in a culture medium supplemented with 12 ng/ml basic fibroblast growth factor (bFGF), and two lines (YT2, YT3) were obtained from three discarded blastocysts in the same medium but supplemented with 16 ng/ml bFGF. These cell lines were derived by partial or whole embryo culture followed by further expansion after manual dissection of the passaged cells. These cells were passaged continuously for more than 6 or 8 months and possessed all of the typical features of pluripotent hES cell lines, such as typical morphological characteristics and the expression of hES-specific markers (TRA-1-60, TRA-1-81, SSEA-4, SSEA-3, alkaline phosphatase, Oct4, Nanog) and pluripotency-related genes (Oct4, Nanog, TDGF1, Sox2, EBAF, Thy-1, FGF4, Rex1). The lines maintained normal karyotypes after long-term cultivation. The karyotype of YT1 and YT3 was 46, XX, and that of YT2 was 46, XY. Pluripotency was confirmed by in vitro and in vivo differentiation, and genetic identity was demonstrated by DNA fingerprinting. Our results indicate that higher concentrations of bFGF at the early culture stage support efficient the hES cell derivation.


Basic fibroblast growth factorCharacterizationDerivationHuman embryonic stem cellsPassage


Human embryonic stem (hES) cells are self-renewing cells derived from the inner cell mass (ICM) of human embryos. Since the first hES cell lines were successfully isolated from in vitro-fertilized blastocysts in 1998 [1], a large number of hES cell lines have been derived. As ES cell lines are derived from individual embryos, they are likely to have unique characteristics that may affect their differentiation potential. Consequently, the derivation of new hES cell lines is necessary for both basic research and cell-based therapy. Given the situation of widespread overpopulation and various genetic backgrounds in China, there is an urgent need to establish more hESC lines from the Chinese population for future applications.

Human ES cells are routinely cultured on feeder layers of fibroblasts in medium that is supplemented with basic fibroblast growth factor (bFGF) at a concentration of 4 ng/ml. An obvious interpretation of the effects of bFGF on hESCs is that the former supports the growth of undifferentiated cells and increases cloning efficiency. It is now evident that self-renewal of hES cells requires bFGF but not leukemia inhibitory factor [24]. Although it is widely accepted that exogenous bFGF could support the growth of undifferentiated hES cells, optimal bFGF concentration on the outgrowth of ICMs has not yet been studied.

In this report, we chose to derive hES cell lines in medium supplemented with 12 and 16 ng/ml bFGF to support initial proliferation and then to maintain them in 4 ng/ml. We established three new hES cell lines of the Chinese population. These lines proliferated continuously for 7–8 months and they retained all hES cell features, including typical morphology, the expression of cell-surface markers, and a stable karyotype and pluripotency in vitro and in vivo.

Materials and methods

Preparation and culture of feeder layers

ICR mouse embryos of 13.5 days post-coitum were used to isolate mouse embryonic fibroblasts (MEFs). MEFs were expanded in high-glucose Dulbecco’s modified Eagle’s medium (DMEM; Invitrogen, Carlsbad, CA) containing 10% (v/v) fetal bovine serum (FBS; Hyclone, Logan, UT), 2 mM glutaMAX I (Gibco BRL-Invitrogen, Grand Island, NY), 50 U/ml penicillin, and 50 μg/ml streptomycin (Gibco BRL-Invitrogen). Passages 4–5 of the MEFs were mitotically inactivated with 10 μg/ml mitomycin C (Sigma-Aldrich, St. Louis, MO) for 2.5–3 h and replated onto gelatin-coated dishes at a density of 50,000 cells/cm2.

Isolation and culture of hES cells

Thirty human blastocysts surplus to in vitro fertilization (IVF) treatments were used in this study. All human embryos used were donated by patients undergoing IVF treatment with informed consent. The consent protocol was approved by Yantai Yuhuangding Hospital Research and Ethics Committee. For ICM isolation, the zona pellucida and the outer trophoblast layer were removed manually using a 1 ml-syringe (25 gauge; BD Biosciences, Franklin Lake, NJ) under a stereo microscope. The isolated ICMs were placed on mitomycin C-treated MEF feeder layers in hES cell-derivation culture medium. After 7–13 days, initial colony outgrowths were mechanically dispersed into small clumps and replated on a culture dish with fresh MEF feeders for further propagation. Each hES cell line was routinely passaged mechanically every 4–5 days. Growth medium was changed every day.

The culture medium used for the derivation and culture of the hES cells consisted of 80% (v/v) DMEM/F12 supplemented with 20% (v/v) Knock-out Serum Replacement, 2 mM glutaMAX, 1% (v/v) non-essential amino acids, 50 U/ml penicillin, 50 μg/ml streptomycin, 0.1 mM β-mercaptoethanol, 1% (v/v) insulin-selenium-transferrin, and 4–16 ng/ml bFGF (all from Gibco BRL-Invitrogen). At passages 1–5, cells were cultivated in hES medium supplemented with one of four different concentrations of bFGF (4, 8, 12 and 16 ng/ml). After passage 5, the concentration of bFGF was reduced to 4 ng/ml in all cultures.

Clonal hES cells

Cloning efficiency was determined by peeling away hES cells from the MEF feeder layers and trypsinizing these to single cells with TrypLE Select (Invitrogen). About 1 × 106 cells were plated into the culture dish. Colonies were visible after 8 days. After approximately 10 days of growth, the colonies were counted and the colonizing efficiency was calculated with the following formula: cloning efficiency = (number of colonies/100,000) × 100.

Cryopreservation and thawing

Human embryonic stem cell clumps were suspended in cryoprotective medium consisting of 10% (v/v) dimethylsulfoxide (Sigma-Aldrich), 40% (v/v) FBS, and 50% (v/v) hES cell maintenance media, cooled to −80°C in a Cryo Freezing Container (NALGENE Labware, Rochester, NY) containing isopropyl alcohol overnight, and then plunged into liquid nitrogen for long-term storage.

Embryoid body formation

To analyze the differentiation of hES cells in vitro, we harvested colonies mechanically and cultured these in suspension SR medium without bFGF in Ultra-Low Attachment Dishes (Corning, Corning, NY) for 7 days to form embryoid bodies (EBs). After 7 days, the EBs were plated on gelatin-coated tissue culture dishes for further differentiation. The expression of markers associated with differentiation of the three germ layers was assessed by first extracting total cellular RNA from EB cultures and then evaluating lineage markers representing the ectoderm, mesoderm, and endoderm, respectively, by reverse transcriptase (RT)-PCR. The negative control contained sterilized water instead of cDNA template.

The PCR primer sequences used were:


  • bone morphogenetic protein (BMP4): 5′-TTTGTTCAAGATTGGCTGTC and 5′-AGATCCCGCATGTAGTCC;




Teratoma formation

To estimate the differentiation capacity in vivo, manually dissected hES cell clumps or TrypLE-digested single cells (1 × 106 cells) after about 20 passages were injected into either the leg muscle or beneath the skin of 4- to 6-week-old SCID-beige mice. After 8–10 weeks, the mice were killed and the resulting teratomas excised and fixed in 10% (v/v) neutral buffered formalin overnight. Hematoxylin and eosin-stained paraffin sections of representative tissues of all three embryonic germ layers were evaluated histologically.

Immunofluorescence staining for specific markers of hES cells

For demonstration of stem cell surface markers, hES cell colonies were fixed in 4% (w/v) paraformaldehyde for 15 min and permeabilized in 0.2% (v/v) Triton-X-100 for 30 min at room temperature. After blocking with phosphate buffered saline (PBS) containing 3% (w/v) bovine serum albumin (BSA) for 30 min, the colonies were incubated in primary antibodies overnight at 4°C and in secondary antibodies for 1 h at room temperature with washes in PBS between incubations. The primary antibodies were SSEA-4, SSEA-3, TRA-1-60, TRA-1-81, and Oct4 (all from Chemicon, Temecula, CA; 1:40 dilution), and the secondary antibodies were FITC-labeled goat anti-mouse immunoglobulin G (IgG)/IgM and goat anti-rat-IgG (Jackson ImmunoResearch Laboratories, West Grove, PA; 1:100 dilution). All fluorescent images were captured using a Zeiss confocal microscope (Zeiss, Jena, Germany). Alkaline phosphatase (AKP) activity was detected with the AKP kit according to the manufacturer’s instructions (Millipore, Temecula, CA).

RT-PCR detection of pluripotent markers

Total RNA from hES cell colonies was extracted with Trizol reagent (Sigma, St. Louis, MO) following the manufacturer’s instructions. RNA was treated with DNase, and cDNA was synthesized according to standard protocols. Primer sequences were listed in Table 1.
Table 1

Primer sequences and conditions for PCR analysis


Primer sequences (forward, upper/reverse, lower)

Annealing temperature (°C)

Products size (bp)














































PCR products were run on 1% (w/v) agarose gels and stained with ethidium bromide.

Karyotyping analysis

Karyotyping was performed by incubation of hES cell colonies with 10 μg/ml colcemid solution (Karyomax; Invitrogen) for 3 h at 37°C. After trypsinization and resuspension in hypotonic KCl (0.075 M) solution for 30 min at 37°C, the cells were repeatedly fixed (3 times) with 3:1 methanol:glacial acetic acid, and dropped onto pre-cleaned glass slides. Chromosomes were visualized from G-band staining. At least 20 metaphases were captured for modal number determination.

DNA fingerprinting

Total genomic DNA was extracted for fingerprinting that involved using short tandem repeat (STR) loci amplified with 15 STR loci and Amelogenin, followed by analysis on an ABI310 genetic analyzer with Genotype software (Applied Biosystems, Foster City, CA). The 15 STR loci analyzed were D8S1179, D21S11, D7S820, CSF1PO, D3S1358, TH01, D13S317, D16S539, D2S1338, D19S433, vWA, TPOX, D18S51, D5S818, and FGA.


Isolation and maintenance of hES cells

A total of 28 blastocysts were used to isolate hES cell lines, resulting in six initial outgrowths and three hES cell lines, as shown in Table 2. Among these, four blastocysts showed relatively large ICMs and others contained small ICMs or none at all. ICMs isolated mechanically from blastocysts were plated on MEF layers from which six ICMs grew successfully and displayed typical hES cell morphology within 7–12 days (Fig. 1). After prolonged culture, three stable hES cell lines, termed YT1, YT2, and YT3, were established (Fig. 1), and at the time of the study these were in continuous culture for 52, 48, and 46 passages, respectively. The derivation efficiency of the hES cells varied with the concentration of bFGF during passages 1–5. None of cell lines were derived in culture medium supplemented with 4 or 8 ng/ml bFGF. One hES cell line was generated in culture medium containing 12 ng/ml bFGF and two lines were isolated in the presence of 16 ng/ml bFGF. As found for hES cells reported elsewhere, the cells existed as flat multicellular colonies with well-defined borders and prominent nucleoli; they had a high nuclear/cytoplasm ratio. The cloning efficiency of the three cell lines was 0.4, 0.6 and 0.5%, respectively. A hES cell colony derived from a single hES cell is depicted in Fig. 1e. All the lines have been frozen and thawed, and all lines grew well after thawing.
Table 2

Different basic fibroblast growth factor concentration on the derivation of human embryonic stem cell lines


Number of blastocysts used for derivation

bFGF concentration (ng/ml)

Number of ICM outgrowths

Number of cell lines

Derivation efficiency (%)

19 May 2010






30 May 2010






3 June 2010






16 November 2010






bFGF, Basic fibroblast growth factor; ICM, inner cell mass
Fig. 1

Derivation of human embryonic stem (hES) cells and morphology of three hES cell lines cultured on mouse embryonic fibroblasts (MEFs). a, f, k Day-6 blastocysts, b outgrowth of inner cell masses on day 5 after plating, c primary hES cell colonies grown on inactivated MEFs (YT1), d, j, o colonies of hES cell line YT1, YT2, and YT3 growing on the feeder layer, e a hES cell colony derived from a single hES cell (YT1), g, h, I primary outgrowths and hES cell-like colonies observed after mechanical dissection at 4 (g), 6 (h), and 10 (I) days after plating (YT2), kn ICM outgrowth on days 4 (l), 6 (m) and 10 (n) after initial plating (YT3). Scale bars: 200 μm (a, f, k); 100 μm (be, gj, lo)

Characterization of hES cell lines

In addition to being studied for their representative morphology characteristics, hES cells were also examined for their expression of pluripotency markers. The three hES cell lines tested positive for the expression of SSEA-3, SSEA-4, Oct4, TRA-1-60, TRA-1-81, and AKP activity (Fig. 2), but not for the expression of SSEA-1 (Fig. 2). The expression of pluripotency genes Nanog, Oct4, FGF4, TDGF1, Rex1, Thy-1, Sox2, and EBAF the three hES cell lines were also identified by RT-PCR (Fig. 3d).
Fig. 2

Photomicrographs showing alkaline phosphatase (AKP) activity and the expression of Oct4, Nanog, SSEA-3, SSEA-4, TRA-1-60, TRA-1-81 and the lack of expression of SSEA-1 in the three hES cell lines. Scale bars: 100 μm
Fig. 3

In vitro differentiation of hES cell lines and analysis of hES marker gene expression by reverse transcription-PCR. a 10-day-old embryoid bodies (EBs), b spontaneous differentiation to different cell lineages from EBs. Scale bars: 100 μm. c Expression of tissue-specific marker genes in differentiated EBs from line YT1, including nestin, neurofilament (ectoderm), myoglobin, BMP4 (bone morphogenetic protein; mesoderm), AFP, and albumin (endoderm). N Negative control (water as template). d hES cells of YT1, YT2 and YT3 express hES cell-specific marker genes: Oct4, Nanog, Sox2, Rex1, EBAF, TDGF1, FGF4, Thy-1. M Marker

Karyotypic analyses revealed all three cell lines to be karyotypically normal, with one line being male and two being female (Fig. 4d–f).
Fig. 4

DNA fingerprinting and karyotypic analysis of all cell lines. a Short tandem repeat (STR) loci analysis of YT1-hES, b STR loci analysis of YT2-hES, c STR loci analysis of YT1-hES, d YT1-hES showing normal 46, XX karyotype, e normal 46, XY karyotype of YT2-hES, f normal karyotype 46, XX of YT3-hES

To evaluate the spontaneous differentiation potency in vitro, the three hES cell lines were induced to form EBs in suspension culture on non-adhesive petri dishes for 7 days followed by plating on gelatin-coated dishes for further differentiation (Fig. 3b). Differentiated cells grew out from the EBs to form monolayers. After 21 days, the differentiated cells that had grown out of the EBs were examined for their gene expression by using RT-PCR. These results demonstrated that the EBs from all cell lines expressed albumin, myoglobin, α-AFP, neurofilament, nestin, and BMP4, all of which are marker genes for the three embryonic germ layers (Fig. 3c).

In this study we investigated the potency of teratoma formation at two different anatomical locations (muscle and the subcutaneous space) in SCID mice with either hES cell lumps or single cells. Cell clumps or single cells from each line were implanted at six intramuscular and six subcutaneous sites. No tumor mass was generated when single-cell hES cells digested by TrypLE Select were injected at two sites. After 8–10 weeks, seven solid tumors and two cystic tumors were dissected and analyzed, as described previously [5, 6].The solid tumors contained a mixture of mostly mature tissues derived from all three germ layers. Paraffin sections of each tumor showed that they consisted of the three embryonic germ layers: ectoderm (neural epithelium); mesoderm (cartilage, smooth muscle, adipose tissue; endoderm (gut epithelium, glandular epithelium, connective tissue). Figure 5 depicts a representative histological section and tissue components derived from the three lines. Based on the histological analysis, the tumors were considered to be teratomas (Fig. 5). The hES cell lines were distinguishable by DNA fingerprinting analysis (Fig. 4a–c).
Fig. 5

Histological analysis of teratomas. ad YT1, eh YT2, Il YT3. The teratomas contained three embryonic germ cell layers. Ectoderm (b, f, j): neural epithelium. Mesoderm (c, g, k): adipose tissue (YT1), cartilage (YT2), muscle (YT3). Endoderm (d, h, l): glandular epithelium. Scale bars: 100 μm


At the early stage of the derivation of stem cell lines, the goal is to increase ‘putative human ES cell’ number and not be too concerned with removing differentiated cells from the outgrowths. Published data indicate that bFGF may enhance the efficiency of the formation of the initial colonies by promoting the expression of extracellular matrix molecules through the activation of the PI3K/Akt/PKB pathway [7]. The amount of bFGF required in medium to maintain self-renewal is a function of various culture conditions, including unknown factors in the serum and serum replacement and factors secreted by MEFs [8, 9]. Current data suggest that MEFs respond optimally to 20 ng/ml of bFGF, although many researchers have reported the successful use of hES cell culture medium supplemented with 4 ng/ml bFGF [10, 11], and two groups have successfully derived hES cell lines using relatively high concentrations of bFGF (10–20 ng/ml) [12, 13]. In our study such a high bFGF concentration would have been suboptimal for the derivation of hES cell lines since none of the outgrowths in hES cell media containing 4 ng/ml bFGF produced a cell line displaying hES cell morphology that continued to proliferate. We chose to derive hES cell lines in 12 or 16 ng/ml bFGF and maintain them in the presence of 4 ng/ml bFGF. Higher concentrations of bFGF would encourage the proliferation of ICMs. These cells had morphological characteristics typical of hES cells. The lines were propagated for many passages and continued to maintain a normal karyotype. These results demonstrate that these three cell lines can be cultured for prolonged periods without compromising chromosomal stability.

The proper maintenance of stem cells is important in the study of hES cells. Various methods are used by different laboratories for passaging hES cells, such as enzymatic or mechanical dissociation methods. Both differentiated and undifferentiated cells are transferred by the use of enzyme digestion. It has been reported [1417] that the passage of hES cells with enzymatic dissociation (collagenase or trypsin) causes chromosomal aberrations, whereas mechanical methods maintain a stable karyotype. Enzymatic passage may reduce the efficiency of isolating putative embryonic stem cell lines [18]. Therefore, the most reliable method for passaging undifferentiated hES cells is manual mechanical dissection of the colonies. Because of this concern, we chose to use mechanical dissection for the maintenance of the line with a low number of differentiated cells.

The ability to form teratomas is the standard assay for evaluating the pluripotency of hES cells. There are various methods for producing teratomas in mice, such as grafting under the kidney capsule and intramuscular, intratesticular, and subcutaneous injection. Injection under the testicular and renal capsule is a very effective method to produce teratomas and has the advantage of requiring fewer cells; however, these methods require specialized training of personnel. In contrast, implants are easily performed, and teratomas are readily seen at intramuscular and subcutaneous locations. A series of studies has shown that teratoma formation is influenced by the implantation site [19, 20]. As we found that single-cell hES did not produce any teratomas at all, it is possible that line YT1-hES could not adjust to enzymatic dissociation; alternatively, intramuscular injection may limit the migration of the implanted cells away from the transplantation site. Analysis of all the teratomas isolated revealed only one fluid-filled cyst. Therefore, our study results show that intramuscular hES cell implantation can result in highly efficient teratoma formation.

In conclusion, we obtained three stable fully characterized hES cell lines from embryos discarded from IVF clinics. In early passage culture, 16 ng/ml bFGF was able to promote the outgrowth of ICMs from blastocysts and the early hES-like cells. Follow-up experiments are needed to develop new clinical-grade hES cell lines and pancreatic differentiation.


We thank Prof. Yao Weidong for technical assistance in the histological analysis of teratoma sections, Hou Nanying for karyotype analysis, and Dr. TG Cooper for language editing. This research was supported by grants from the National Basic Research Program of China (973 Program, 2005CB522705-2), National High Technology Research and Development Program of China (863 program, 2003AA205008), Shandong Province Key Project of China (021100104), and Yantai Science and Technology Project (2009107).

Copyright information

© Japan Human Cell Society and Springer 2012