Differentiation of Embryonic Stem Cell Lines Generated from Adult Somatic Cells by Nuclear Transfer

See allHide authors and affiliations

Science  27 Apr 2001:
Vol. 292, Issue 5517, pp. 740-743
DOI: 10.1126/science.1059399


Embryonic stem (ES) cells are fully pluripotent in that they can differentiate into all cell types, including gametes. We have derived 35 ES cell lines via nuclear transfer (ntES cell lines) from adult mouse somatic cells of inbred, hybrid, and mutant strains. ntES cells contributed to an extensive variety of cell types, including dopaminergic and serotonergic neurons in vitro and germ cells in vivo. Cloning by transfer of ntES cell nuclei could result in normal development of fertile adults. These studies demonstrate the full pluripotency of ntES cells.

Stem cells are able to differentiate into multiple cell types, representatives of which might be harnessed for tissue repair in degenerative disorders such as diabetes and Parkinson's disease (1). One obstacle to therapeutic applications is obtaining stem cells for a given patient. A solution would be to derive stem cells from embryos generated by cloning from the nuclei of the individual's somatic cells. We have previously cloned mice by microinjection using a variety of cell types as nucleus donors, including embryonic stem (ES) cells (2–4). We sought to perform the converse experiment by deriving ES cell lines in vitro from the inner cell mass (ICM) of blastocysts clonally produced by nuclear transfer.

To this end, nuclei from adult-derived somatic donor cells of five strains, including inbred (e.g., 129/Sv and C57BL/6nu/nu, nude) and F1 hybrid (e.g., C57BL/6 × DBA/2) representatives were transferred by microinjection (5) to produce cloned blastocysts (Table 1). When plated on fibroblast feeder layers in culture medium (6), cloned blastocysts from all five strains tested yielded at least one nuclear transfer ES (ntES) cell line (Table 1) (8). Cultures were established from XX embryos derived via cumulus cell nuclear transfer (14.2% of blastocysts) and both XX and XY embryos derived from tail-tip cells (6.5%; Table 1). In total, 35 successfully cryopreserved stable ntES cell lines were produced.

Table 1

Establishment of ntES cell lines after nuclear transfer from adult-derived cumulus or tail-tip cells and examination of pluripotency after injection into fertilization-derived blastocysts.

View this table:

Clonal origin of ntES cell lines was confirmed by polymerase chain reaction (PCR) analysis of polymorphic markers (8, 9). The ntES cell morphology of most lines was similar to that of widely disseminated lines such as E14 (11). We found no evidence for a pronounced difference in the efficiency of ntES cell line establishment between inbred and hybrid backgrounds (Table 1). All ntES cell lines tested expressed markers diagnostic for undifferentiated ES cells (12), including alkaline phosphatase (8) and Oct3/4 (13).

Embryonic stem cells have been induced to differentiate in vitro to produce cardiomyocytes (14), neurons (15), astrocytes and oligodendrocytes (16), and hematopoietic lineages (17). To assess the pluripotency of ntES cells, we therefore sought (i) to differentiate them in vitro to a wide variety of ectodermal, mesodermal, and endodermal lineages, and (ii) to induce a highly differentiated cell type. We chose a particularly specialized example with therapeutic potential: dopaminergic neurons.

Differentiation of embryoid bodies (8, 18) derived from three different ntES cell lines resulted in a mixed population of ectodermal, endodermal, and mesodermal derivatives (19). Efficient neural differentiation of ntES cells could be readily induced in each of the seven lines tested. Generation of specific midbrain dopaminergic neurons from ntES cells was achieved with a range of efficiencies by using a multistep differentiation protocol described previously (15, 20) (Fig. 1, A and B). One ntES cell line yielded dopaminergic neurons in excess of 50% of the total cell number. The functional nature of these neurons was confirmed by reversed-phase HPLC (RP-HPLC) determination of dopamine release (21) (Fig. 1C). Serotonergic neurons were also detected histochemically, although in smaller numbers, and serotonin release was confirmed by RP-HPLC (Fig. 1, D and E).

Figure 1

Dopaminergic and serotonergic differentiation of ntES cells in vitro. Embryoid bodies were plated under conditions favoring CNS selection followed by dopaminergic induction. Images shown are of C15. (A) Colocalization of tyrosine-hydroxylase (TH, green) and β-III tubulin (red). The inset shows the morphology of neuritic TH+ cells at higher magnification. (B) The presence of serotonergic (Ser, green) and TH (red) neurons. Scale bar = 100 μm. (C) Yield of TH+ neurons varied among the cell lines tested, with >50% of total cell number in C15 cells. Other commonly used ES lines (E14, AB2.2) generated a percentage of TH+ cells falling within the range shown by our ntES cells. C4, C15, C16, CN1, CN2, CT1, CT2 represent ntES, AB2.2, and E14 ES cell lines. (D) Representative chromatogram showing elution and electrochemical detection of dopamine (DA) and serotonin (Ser) from conditioned medium by RP-HPLC. (E) Quantification of dopamine and serotonin release. Neurotransmitter concentration was determined in conditioned medium (CM; 24 hours after last medium change), basal condition (15 min in buffer solution) and upon evoked release (KCl; 15 min in 56 mM KCl buffer). Serotonin release was low under basal and evoked conditions, probably reflecting a lower number of serotonergic neurons.

Two recent reports (22, 23) describe a total of five mouse ES cell-like lines derived from the ICMs of cloned blastocysts, although none contributed to the germ line. We characterized the contribution of 19 ntES cell lines to chimeric offspring after ntES cell injection into fertilization-derived blastocysts from the ICR strain (24). The contribution of ntES cells to 105 chimeric offspring after 348 blastocyst injections is summarized inTable 1. The contribution can be readily approximated by coat color, because all ntES cell lines are derived from black-eyed strains with dark coat color, whereas ICR is albino (Fig. 2, A and B). ntES cell lines generally contributed strongly to the coats of chimeric offspring (Table 1). This was corroborated for ntES cells derived from a hybrid strain ubiquitously expressing high levels of the reporter transgene,EGFP (25). All internal organs examined from two EGFP Tg chimeras contained an extensive contribution from the EGFP-expressing ntES cells (13).

Figure 2

Totipotency of ntES cells demonstrated in vivo. (A) Contribution of C57BL/6nu/nu-nude ntES cells (line CN1) to chimeric offspring after injection into ICR × ICR fertilization-derived blastocysts is demonstrated in offspring 14 days after birth in which dark coat color derives from the ntES cell contribution. The adult mouse in the cage is the tail-tip donor used to generate the CN1 ntES line. (B) The male indicated with an asterisk in (A) was crossed at 8 weeks with a white (ICR) female, producing a litter containing three dark offspring, confirming the contribution of C57BL/6nu to the germ line. Asterisks in panels (A) and (B) indicate the same male. (C) Cloning using ntES cells as nucleus donors, exemplified by a B6D2F1 clone (line C4) shown at 12 weeks with her litter. (D) PCR analysis of microsatellite markers in genomic DNA from ntES cell lines (CN1, CN2, CN3, CN4) and cloned offspring (cCN1) confirms the clonal origin of the C57BL/6nu/nupup derived from line CN1. Polymorphic markers D8Mit248,D9Mit191, and D4Mit204 are conserved between genomic DNA from the ntES cell lines and the cloned pup, but differ from those of the ICR surrogate mother (CD1) or ooplast recipient strain B6D2F1 (F1).

As a comprehensive measure of pluripotency, the ability to contribute to the germ line is considered a defining characteristic of ES cells. Chimeric offspring were crossed with the albino strain, ICR. In ongoing experiments, 29 pups have been derived after chimera × ICR crosses as judged by eye and coat color and, where appropriate, EGFP expression (Table 1). Germ line transmission was demonstrated for seven ntES cell lines derived from male and female representatives of all mouse progenitor strains. These data confirm that ntES cells contribute to both male and female gametogenesis when derived from either inbred, hybrid, or mutant strains (Table 1), consistent with the universality of the phenomenon among diverse genetic backgrounds.

To determine whether the reprogramming that produced fully pluripotent ntES cells could be reversed, we attempted to re-derive the original nucleus donor cell types in offspring cloned by nuclear transfer from ntES cells (2). Nuclei from all ntES cell lines supported development in vitro to the blastocyst stage after microinjection into enucleated oocytes (Table 2). When transferred to pseudopregnant surrogate mothers, blastocysts derived from six of the ntES cell lines developed to term, resulting in a total of 20 live-born pups. Of these, one was derived from the nucleus of a C57BL/6nu/nu (nude, inbred) background, and the remaining 19 from the nuclei of hybrid strains (Fig. 2C; Table 2). Hybrid genomes thus preferentially supported cloning in these experiments. Moreover, 11 (all cumulus-derived females; see Fig. 2C) of the 19 were derived from B6D2F1 ntES cell lines, of which 10 survived to adulthood and are healthy, exhibiting normal fertility (Fig. 2C). The remaining nine, which died perinatally of unknown cause(s), also contained genomic contribution from the hybrid, B6D2F1 (129/Sv Tac × B6D2F1; Table 2), albeit diluted. This possibly reflects a subtle, yet critical contribution made by the hybrid genetic background of B6D2F1. We corroborated the clonal origin of ntES cells and cloned offspring by PCR analysis of polymorphic markers (Fig. 2D).

Table 2

Cloning using ntES cells as nucleus donors.

View this table:

We have demonstrated that adult-derived somatic cell nuclei can efficiently be used to generate ES cell lines that exhibit full pluripotency; they can be caused to differentiate along prescribed pathways in vitro, contribute to the germ line after injection into blastocysts, and support full development following nuclear transfer. Because ES cells support homologous recombination at a relatively high efficiency, genetic lesions in ntES cells might be repaired by gene targeting or transgenic complementation before they are used to establish germ line chimeras or in cloning. This should facilitate the establishment of germ cells, individuals, and cell lines containing targeted alleles.

Reports of human ES cell-like cell lines (26,27), coupled to the success of mammalian cloning by somatic cell nuclear transfer, have raised the possibility that ntES cells could provide a source of differentiated cells for human autologous transplant therapy: therapeutic cloning (28). In this context, demonstration of the full pluripotency of ntES cells is particularly relevant; for example, adult-derived stem cells are apparently restricted in their range of potential cell fates and may be unable to contribute to all tissues, including hematopoietic lineages (29). Indeed, the efficient generation of midbrain dopaminergic neurons in vitro has been achieved to date only with mesencephalic precursors (30) and ES cells (15), but not from adult-derived cells. In combining ES and nuclear transfer technologies, we have here addressed this limitation and demonstrated the feasibility of the first steps required for the application of cloning to transplant therapy.

  • * Present address: Advanced Cell Technology, One Innovation Drive, Worcester, MA 01605, USA.

  • To whom correspondence should be addressed. E-mail: teru{at}; studerl{at}


View Abstract

Navigate This Article