Evidence of a Pluripotent Human Embryonic Stem Cell Line Derived from a Cloned Blastocyst

See allHide authors and affiliations

Science  12 Mar 2004:
Vol. 303, Issue 5664, pp. 1669-1674
DOI: 10.1126/science.1094515

This article has a correction. Please see:


Somatic cell nuclear transfer (SCNT) technology has recently been used to generate animals with a common genetic composition. In this study, we report the derivation of a pluripotent embryonic stem (ES) cell line (SCNT-hES-1) from a cloned human blastocyst. The SCNT-hES-1 cells displayed typical ES cell morphology and cell surface markers and were capable of differentiating into embryoid bodies in vitro and of forming teratomas in vivo containing cell derivatives from all three embryonic germ layers in severe combined immunodeficient mice. After continuous proliferation for more than 70 passages, SCNT-hES-1 cells maintained normal karyotypes and were genetically identical to the somatic nuclear donor cells. Although we cannot completely exclude the possibility that the cells had a parthenogenetic origin, imprinting analyses support a SCNT origin of the derived human ES cells.

The isolation of pluripotent human embryonic stem (ES) cells (1) and breakthroughs in somatic cell nuclear transfer (SCNT) in mammals (2) have raised the possibility of performing human SCNT to generate potentially unlimited sources of undifferentiated cells for use in research, with potential applications in tissue repair and transplantation medicine. This concept, known as “therapeutic cloning,” refers to the transfer of the nucleus of a somatic cell into an enucleated donor oocyte (3). In theory, the oocyte's cytoplasm would reprogram the transferred nucleus by silencing all the somatic cell genes and activating the embryonic ones. ES cells would be isolated from the inner cell mass (ICM) of the cloned preimplantation embryo. When applied in a therapeutic setting, these cells would carry the nuclear genome of the patient; therefore, it is proposed that after directed cell differentiation, the cells could be transplanted without immune rejection to treat degenerative disorders such as diabetes, osteoarthritis, and Parkinson's disease (among others). Previous reports have described the generation of bovine ES-like cells (4) and mouse ES cells from the ICMs of cloned blastocysts (57) and the development of cloned human embryos to the 8- to 10-cell stage (8, 9). Here we describe evidence of the derivation of human ES cells after SCNT (10).

Fresh oocytes and cumulus cells were donated by healthy women for the express purpose of SCNT stem cell derivation for therapeutic cloning research and its applications. Before beginning any experiments, we obtained approval for this study from the Institutional Review Board on Human Subjects Research and Ethics Committees (Hanyang University Hospital, Seoul, Korea). Donors were fully aware of the scope of our study and signed an informed consent form (a summary of the informed consent form is available in the supporting online text); donors voluntarily donated oocytes and cumulus cells (including DNA) for therapeutic cloning research and its applications only, not for reproductive cloning; and there was no financial payment. A total of 242 oocytes were obtained from 16 volunteers (there were one or two donors for each trial) after ovarian stimulation: 176 metaphase II (MII) oocytes were used directly for SCNT, whereas the remaining 66 oocytes were allowed to mature to the MII stage before use in SCNT. Autologous SCNT was performed; that is, the donor's own cumulus cell, isolated from the cumulus-oocyte complex (COC), was transferred back into the donor's own enucleated oocyte. Before enucleation, the oocytes were matured in vitro in G1.2 medium (Vitro Life, Goteborg, Sweden) for 1 to 2 hours. Enucleation, SCNT, and electrical fusion were performed as described (11). To directly confirm that the oocyte's DNA was removed during enucleation, we imaged the extruded DNA MII spindle complex from every oocyte with Hoechst 33342 fluorescent DNA dye (Fig. 1, A and B; arrows).

Fig. 1.

Confirmation of enucleation, photographs of human SCNT ES cells and their differentiated progeny, and karyotype analysis. (A and B) Images (×200) of extruded DNA MII spindle complexes (arrows) from an oocyte before (A) and after (B) enucleation. (C to E) Bright-field [(C), ×100] and phase contrast [(D), ×100] micrographs and higher magnification image [(E), ×200] of a colony of SCNT-hES-1 cells. Immunofluorescence staining for nestin [(F), ×200] and G-banded kayotyping (G) inSCNT-hES-1 cells are shown. Scale bars, 20 μm in (A) and (B) and 100 μm in(C) to (F).

Without any report of an efficient protocol for human SCNT, several critical steps had to be optimized (2), including reprogramming time, activation method, and in vitro culture conditions. Reprogramming time, or the lapse of time between cell fusion and egg activation, returns the gene expression of the somatic cell to that needed for appropriate embryonic development. Initially, we investigated the effect of simultaneous fusion and activation, as used for porcine SCNT (12, 13), but observed low fusion and cleavage rates, with no blastocyst development. Instead, we adapted the bovine SCNT procedure of waiting a few hours between fusion and activation. By allowing 2 hours for reprogramming, we were able to develop ∼25% of the embryos to the blastocyst stage.

Because sperm-mediated activation is absent in SCNT, an artificial stimulus is needed to initiate development. Various chemical, physical, and mechanical agents induce parthenogenetic development in mice (14), but human data are limited. Oocyte activation using the calcium ionophore A23187 (calcimycin) or ionomycin and the protein synthesis inhibitor puromycin induces parthenogenetic development of human oocytes at different efficiencies (15). We found that incubation in 10 μM A23187 for 5 min, followed by incubation with 2.0 mM 6-dimethylaminopurine (DMAP) for 4 hours, gave efficient chemical activation for human SCNT eggs. Other investigators have reported encouraging results in overcoming inefficiencies in embryo culture by supplementing the culture with different energy substrates (16). Furthermore, the recent development of serum-free sequential media has led to considerable improvement in the rate of clinical pregnancies produced by in vitro fertilization (IVF) (17). In this study, human modified synthetic oviductal fluid with amino acids (hmSOFaa) was prepared by supplementing mSOFaa (18) with human serum albumin (10 mg/ml) and fructose (1.5 mM) instead of bovine serum albumin (8 mg/ml) and glucose (1.5 mM). The replacement of glucose with fructose improves the developmental competence of bovine SCNT embryos (11, 19). Culture of human SCNT-derived embryos in G1.2 medium for the first 48 hours followed by hmSOFaa medium produced more blastocysts, as compared to culture in G1.2 medium for the first 48 hours followed by culture in G1.2 medium or in continuous hmSOFaa medium (Table 1). Cibelli et al. (8) reported that the treatment of human oocytes with 5 μM calcium ionomycin followed by 2 mM DMAP in G1.2 culture medium triggered pronucleus formation, embryonic cleavage, and the formation of a blastocoelic cavity in human parthenotes. However, they did not obtain human SCNT blastocysts when their protocol was applied to SCNT embryos. Limitations in oocyte supply precluded full optimization of all the parameters for human SCNT; nonetheless, the protocol described here produced cloned blastocysts at rates of 19 to 29% (as a percentage of oocytes used) and was comparable to those produced by established SCNT methods in cattle (∼25%) (11) and pigs (∼26%) (12, 13).

Table 1.

Conditions for human SCNT.

Experiment Activation conditionView inline Reprogramming time (hours) 1st step mediumView inline 2nd step medium No. of oocytes No. (%) of cloned embryos developed to
Two-cell Compacted morula Blastocyst
1st set 10 μM ionophore 6-DMAP 2 G 1.2 hmSOFaa 16 16 (100) 4 (25) 4 (25)
10 μM ionophore 6-DMAP 4 G 1.2 hmSOFaa 16 15 (94) 1 (6) 0
10 μM ionophore 6-DMAP 6 G 1.2 hmSOFaa 16 15 (94) 1 (6) 1 (6)
10 μM ionophore 6-DMAP 20 G 1.2 hmSOFaa 16 9 (56) 1 (6) 0
2nd set 10 μM ionophore 6-DMAP 2 G 1.2 hmSOFaa 16 16 (100) 5 (31) 3 (19)
5 μM ionophore 6-DMAP 2 G 1.2 hmSOFaa 16 11 (69) 0 0
10 μM ionomycin 6-DMAP 2 G 1.2 hmSOFaa 16 12 (75) 0 0
5 μM ionomycin 6-DMAP 2 G 1.2 hmSOFaa 16 9 (56) 0 0
3rd set 10 μM ionophore 6-DMAP 2 G 1.2 hmSOFaa 16 16 (100) 4 (25) 3 (19)
10 μM ionophore 6-DMAP 2 G 1.2 G 2.2 16 16 (100) 0 0
10 μM ionophore 6-DMAP 2 Continuous hmSOFaa 16 16 (100) 0 0
4th set 10 μM ionophore 6-DMAP 2 G 1.2 hmSOFaa 66 62 (93) 24 (36) 19 (29)
  • View inline* Fused donor oocytes and somatic cells were activated in either calcium ionophore A23187 (5 or 10 μM) or ionomycin (5 or 10 μM) for 5 min, followed by 2 mM 6-DMAP treatment for 4 hours.

  • View inline Oocytes were incubated in the first medium for 48 hours.

  • A total of 30 SCNT-derived blastocysts were cultured, 20 ICMs were isolated by immunosurgical removal of the trophoblast, and one ES cell line (SCNT-hES-1) was derived. The resulting SCNT-hES-1 cells had a high nucleus-to-cytoplasm ratio and prominent nucleoli. The cell colonies displayed similar morphology to that reported previously for hES cells derived after IVF (Fig. 1, C to E). When cultured in defined medium conditioned for neural cell differentiation (20), SCNT-hES-1 cells differentiated into nestin-positive cells, an indication of primitive neuroectoderm differentiation (Fig. 1F). The SCNT-hES-1 cell line was mechanically passaged by dissociation every 5 to 7 days and successfully maintained its undifferentiated morphology after continuous proliferation for >70 passages, while still maintaining a normal female (XX) karyotype (Fig. 1G) (21). Furthermore, the SCNT-hES-1 cells expressed ES cell markers such as alkaline phosphatase, SSEA-3, SSEA-4, TRA-1-60, TRA-1-81, and Oct-4, but not SSEA-1 (Fig. 2). As previously described in monkey (22) and human ES cells (1, 23, 24) and in mouse SCNT-ES cells (6), SCNT-hES-1 cells did not respond to exogenous leukaemia inhibitory factor, suggesting that a pluripotent state is maintained by a gp130-independent pathway. The pluripotency of SCNT-hES-1 cells was investigated in vitro (fig. S1) and in vivo (Fig. 3). Clumps of the cells were cultured in vitro in suspension to form embryoid bodies. The resulting embryoid bodies were stained with three dermal markers and were found to differentiate into a variety of cell types, including derivatives of endoderm, mesoderm, and ectoderm (fig. S1). When undifferentiated SCNT-hES-1 cells were injected into the testes of severe combined immunodeficient (SCID) mice, teratomas were obtained 6 to 7 weeks after injection. The resulting teratomas contained tissue representative of all three germ layers. Differentiated tissues seen in Fig. 3 include neuroepithelial rosset, pigmented retinal epithelium, smooth muscle, bone, cartilage, connective tissues, and glandular epithelium. The DNA fingerprinting analysis with human short tandem repeat (STR) markers indicates that the cell line originated from the cloned blastocysts reconstructed from the donor cells, not from parthenogenetic activation (Fig. 4, A to C). The statistical probability that the cells may have derived from an unrelated donor is 8.8 × 10–16. Reverse transcription polymerase chain reaction (RT-PCR) amplification for paternally expressed (hSNRPN and ARH1) and maternally expressed (UBE3A and H19) genes further confirmed that the cell line originated from the donor cells (Fig. 4D).

    Fig. 2.

    Expression of characteristic cell surface markers inhuman SCNT ES cells. SCNT-hES-1 cells expressed cell surface markers, including alkaline phosphatase (B), SSEA-3 (H), SSEA-4 (K), TRA-1-60 (N), TRA-1-81 (Q), and Oct-4 (T), but not SSEA-1 (E). The differentiated SCNT-hES-1 cells were not stained with alkaline phosphatase (A). The IVF-derived human ES cells (Miz-hES) were used for comparison and also expressed alkaline phosphatase (C), SSEA-3 (I), SSEA-4 (L), TRA-1-60 (O), TRA-1-81 (R), and Oct-4 (U), but not SSEA-1 (F). Negative controls not treated with first antibodies are shown (D, G, J, M, P, and S). Magnification in (A) to (U), ×40. Scale bars, 100 μm.

    Fig. 3.

    Teratomas formed by human SCNT ES cells in the testes of SCID mice at 12 weeks after injection. Neuroepithelial rosset (A), pigmented retinal epithelium (B), ostoid island showing bony differentiation (C), cartilage (D), and glandular epithelium with smooth muscle and connective tissues (E). Magnification in (A) to (D), ×200; in (E), ×100. Scale bar, 100 μm.

    Fig. 4.

    DNA fingerprinting analysis and expression of imprinted genes. (A) Isogenic analysis inloci D3S1358 (chromosome location 3p), vWA (chromosome location 12p 12-pter), and FGA (chromosome location 4q28). (B) Isogenic analysis in loci amelogenin (chromosome location X:p22.1-22.3 and Y:p11.2), THO1 (chromosome location 11p 15.5), TPOX (chromosome location 2p23-2per), and CSF1PO (chromosome location 5q33.3-34). (C) Isogenic analysis in loci D5S818 (chromosome location 5p22-31), D13S317 (chromosome location 13q22-31), and D7S820 (chromosome location 7q11.21-22). The boxed numbers and corresponding peaks represent locations of polymorphisms for each short tandem repeat marker. (D) RT-PCR amplification of paternally expressed (hSNRPN and ARH1) and maternally expressed (UBE3A and H19) genes. Cyno-1, maternally derived monkey parthenogenetic stem cell line (25); mFBLST, monkey fibroblasts; hFBLST, human fibroblasts; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; Tm(-), without template for PCR amplification.

    Simerly et al. (26) recently reported defective mitotic spindles after SCNT in nonhuman primate embryos, perhaps resulting from the depletion of microtubule motor and centrosome proteins lost to the meiotic spindle after enucleation. In this study, Fig. 1G demonstrates that SCNT-hES-1 cells have the normal karyotype. We speculate that SCNT blastocysts from which ES cell lines were not derived might have been aneuploid. However, it is important to note that our investigations differ from those of Simerly et al. in a few ways: Media and protocols for in vitro development were optimized for human oocytes and embryos, whereas the protocols for nonhuman primate studies are extrapolated from clinical procedures; the enucleation method here differs, because we squeeze the MII oocyte so that the DNA spindle complex is extruded through a small hole in the zona pellucida, instead of aspirating the DNA spindle complex with a glass pipette as others have described (27); and the DNA spindle complex is extruded shortly after the appearance of the first polar body, so that it may even be at the prometaphase II stage.

    In this report, we provide three lines of evidence supporting the nuclear transfer origins of the SCNT-hES-1 cell line: (i) DNA extraction was verified for each of the 242 enucleated oocytes (Fig. 1, A and B; arrows); (ii) DNA fingerprinting showed heterozygous, not homozygous, chromosomes (Fig. 4, A to C); and (iii) RT-PCR showed biparental, and not unimaternal, expression of imprinted genes (Fig. 4D). Although the Cyno-1 parthenogenetic cells retained their strictly maternal imprints, that evidence came from a single monkey cell line. Given the aberrant expression of imprinted genes after murine SCNT (28), perhaps the SCNT-hES-1 cells' biparental expression of imprinted genes might have been influenced by SCNT or subsequent culture. Heterologous along with autologous SCNT will provide more definitive molecular evidence. Although overwhelming ethical constraints preclude any reproductive cloning attempts, complementary investigations in nonhuman primates might provide additional and confirmatory information. Consequently, although we cannot exclude the possibility of a parthenogenetic origin, the studies reported here support the conclusion that the SCNT-hES-1 cell line originated from the donor's diploid somatic cumulus cell after SCNT.

    In order to successfully derive immunocompatible human ES cells from a living donor, a reliable and efficient method for producing cloned embryos and ES isolation must be developed. Thomson et al. (1), Reubinoff et al. (23), and Lanzendorf et al. (29) produced human ES cell lines at high efficiency. Briefly, five ES cell lines were derived from a total of 14 ICMs, two ES cell lines were derived from four ICMs, and three ES cell lines were derived from 18 ICMS, respectively. In our study, one SCNT-hES cell line was derived from 20 ICMs. It remains to be determined whether this low efficiency is due to faulty reprogramming of the somatic cells or to subtle variations in our experimental procedures. We cannot rule out the possibility that the genetic background of the cell donor had an impact on the overall efficiency of the procedure. Further improvements in SCNT protocols and in vitro culture systems are needed before contemplating the use of this technique for cell therapy. In addition, the mechanisms governing the differentiation of human tissues must be elucidated in order to produce tissue-specific cell populations from undifferentiated ES cells. This study shows the feasibility of generating human ES cells from a somatic cell isolated from a living person.

    Supporting Online Material

    Materials and Methods

    SOM Text

    Fig. S1

    References and Notes

    View Abstract

    Stay Connected to Science


    Navigate This Article