Stem cells

Recycling the abnormal

Using human eggs in the quest to make donor-specific embryonic stem cells is controversial. A method developed in mice, if applicable to humans, could eliminate the need to obtain eggs for this purpose.

On page 679 of this issue, Egli et al.1 describe a promising method for generating embryonic stem-cell (ESC) lineages using the technique of somatic-cell nuclear transfer (SCNT). Conventional SCNT involves replacement of the nuclear genetic material of an unfertilized egg (oocyte), with that of a somatic (non-germ) cell. After 'fertilization', which is induced by chemical or electrical triggers, the embryo undergoes several rounds of cell division and, after implantation into a foster mother, may develop to term. So far, this technique has been used successfully to clone 12 species. It also has been used in mice to generate ESCs from a 3.5-day-old mouse embryo2 — a blastocyst.

Since Dolly the Sheep was cloned by SCNT more than ten years ago3, it has been hoped that this technique would serve to create patient-matched ESCs for therapy, and human-disease-specific ESC lines for use in basic research and drug development. However, in contrast to SCNT in mice, the use of this technique in humans has been thwarted by technical difficulties, as well as logistical and ethical concerns about obtaining oocytes. Now, Egli and colleagues1 describe a different approach to produce donor/disease-specific ESC lines that may well revolutionize the field of human stem-cell research, and that removes one of the main ethical objections to such work. The crux of their contribution is the use of fertilized eggs, instead of oocytes, as SCNT recipients.

Historically, fertilized mouse eggs at the one-cell stage — the zygote — have been successfully used as recipients of nuclear genetic material4, but only when the donor cells were also zygotes and not from later developmental stages5. Possible reasons for this limitation include loss of essential non-DNA factors with the removed genetic material6, and inadequate time for the reprogramming of the donor's genetic material in its new environment5.

Egli et al. reasoned that the loss of the crucial factors could be minimized or eliminated if nuclear transfer is conducted when both the recipient zygote and the donor cell are temporarily arrested at the mitotic cell division. To test this, they used the drug nocodazole to arrest mitosis in mouse zygotes at the stage when chromosomes condense. Replacing nocodazole with another inhibitor allowed chromosome alignment along the mitotic spindle, but prevented further cell-cycle progression. The spindle could then be seen using optical devices, and removed mechanically.

Donor zygotes and two- and eight-celled embryos were also arrested with nocodazole. The condensed chromosomes were then identified, removed from individual cells, and injected into the cytoplasm of treated recipient zygotes. Removal of the inhibitors allowed development to resume, and the resulting blastocysts were returned to foster mothers. Donors from all three stages of development led to some live births.

Next, the authors used mouse ESCs as donors. The resultant blastocysts were either returned to foster mothers — leading to nine live births — or were used to make new ESC lines. By injecting these SCNT-derived ESCs into normal host blastocysts they showed that these cells had the full range of developmental potencies expected from bona fide mouse ESCs.

Finally, adult tail-tip cells were used as recipients to make donor-specific SCNT-derived ESCs. Previously, mitotic, embryonic7 and somatic cells8 have all been used as donors in nuclear transfer experiments. But Egli et al. are the first to use a mitotic cytoplasm as a recipient. Using cells at this stage of the cell cycle as recipients may expedite reprogramming of donor chromosomes, because at other stages reprogramming factors are probably sequestered within cells' nuclei9.

In terms of efficiency, the method reported by Egli et al.1 is not better than previous ones. So why all the excitement? After all, the new method seems to be ethically inferior, as in generating SCNT-derived ESCs, two, rather than one, developing embryos are disrupted — the original zygote and the SCNT-derived embryo. The answer can be found in the results of their last experiment.

The researchers generated an embryo containing three sets of chromosomes — in which two sperm cells fertilized a single oocyte. Such embryos never develop normally. Nevertheless, replacement of these three sets of chromosomes with one set from an ESC led to a normally developing embryo, which could potentially be used to generate a new ESC line (Fig. 1). This finding could have a profound effect on developing a viable and tractable method of SCNT in humans.

Figure 1: Somatic-cell nuclear transfer using abnormal embryos.

Egli et al.1 generated abnormal mouse zygotes, in which the egg was fertilized with two sperm cells (n indicate pronuclei). Such zygotes are often useless by-products of human in vitro fertilization procedures. However, using inhibitors, the authors allowed the abnormal mouse zygote to progress through the cell cycle up to the point during mitosis at which its chromosomes aligned on the mitotic spindle. They then mechanically removed the spindle and replaced it with the condensed chromosomal content of a donor embryonic stem cell (ESC), also arrested at mitosis. Removal of the inhibitors allowed development to resume, and a blastocyst formed. This is a promising technical feat, as the authors also found that blastocysts formed in this way but using ESCs or adult tail-tip cells as donors and normal zygotes as recipients led to live offspring or, alternatively, new ESCs.

The failure of SCNT in humans and monkeys has been attributed by some10 to fundamental differences between primate and non-primate unfertilized eggs in the way their spindles form during cell division. However, even if this difficulty could be surmounted, obtaining freshly ovulated human oocytes would remain of logistical and ethical concern; unfortunately, in contrast to recent success in mice11, aged, unfertilized oocytes — a by-product of normal in vitro fertilization (IVF) procedures — have been inadequate for SCNT in humans12. However, if the technique developed by Egli and colleagues could be used successfully in humans, all of these problems would be circumvented.

It is estimated that 3–5% of fertilized human zygotes contain supernumerary sets of chromosomes13. Such zygotes are always excluded from clinical use in IVF centres because they cannot develop, and are therefore disposed of. The possibility of recycling non-viable zygotes to produce ESC lines obviates the need for oocyte donation. So those who have been troubled by this ethical aspect of human SCNT stem-cell research will be very encouraged by the results of Egli and his colleagues1.


  1. 1

    Egli, D., Rosains, J., Birkhoff, G. & Eggan, K. Nature 447, 679–685 (2007).

    ADS  CAS  Article  Google Scholar 

  2. 2

    Munsie, M. et al. Curr. Biol. 10, 989–992 (2000).

    CAS  Article  Google Scholar 

  3. 3

    Wilmut, I., Schnieke, A. E., McWhir, J., Kind, A. J. & Campbell, K. H. Nature 385, 810–813 (1997).

    ADS  CAS  Article  Google Scholar 

  4. 4

    McGrath, J. & Solter, D. Science 220, 1300–1302 (1983).

    ADS  CAS  Article  Google Scholar 

  5. 5

    McGrath, J. & Solter, D. Science 226, 1317–1319 (1994).

    ADS  Article  Google Scholar 

  6. 6

    Polejaeva, I. A. et al. Nature 407, 86–90 (2000).

    ADS  CAS  Article  Google Scholar 

  7. 7

    Kwon, O. Y. & Kono, T. Proc. Natl Acad. Sci. USA 93, 13010–13013 (1996).

    ADS  CAS  Article  Google Scholar 

  8. 8

    Ono, Y., Shimozawa, N., Ito, M. & Kono, T. Biol. Reprod. 64, 44–50 (2001).

    CAS  Article  Google Scholar 

  9. 9

    Do, J. T. & Scholer, H. R. Stem Cells 22, 941–949 (2004).

    CAS  Article  Google Scholar 

  10. 10

    Simerly, C. et al. Science 300, 297 (2003).

    Article  Google Scholar 

  11. 11

    Wakayama, S. et al. Curr. Biol. 17, R120–R121 (2007).

    CAS  Article  Google Scholar 

  12. 12

    Hall, V. J. et al. Hum. Reprod. 22, 52–62 (2007).

    CAS  Article  Google Scholar 

  13. 13

    Aoki, V. W. et al. J. Exp. Clin. Assist. Reprod. 2, 3 (2005).

    Article  Google Scholar 

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Colman, A., Burley, J. Recycling the abnormal. Nature 447, 649–650 (2007).

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing