Developmental biology

Asexual healing

The development of healthy monkeys from embryos in which the egg contains nuclear DNA from one donor and mitochondrial DNA from another suggests a method to prevent inheritance of certain human diseases.

Preventing the transmission of harmful genetic mutations from parent to baby is a double challenge for reproductive biologists as there are two genomes in our cells that can carry mutations — one in the nucleus and one in the mitochondria. The mitochondrial genome is tiny, encoding only a handful of proteins, all essential for cellular energy production. However, despite its small size, disorders associated with mutations in mitochondrial DNA are among the most common genetically inherited metabolic diseases, and there are currently no effective treatments1. On page 367 of this issue, Tachibana et al.2 report the development of a technique that could prevent transmission of defective mitochondrial DNA from mother to baby.

About 1:5,000 people have, or are likely to develop, a mitochondrial disease3, and as many as 1:200 newborns carry one of the common mitochondrial DNA mutations4. All mitochondria in the developing embryo come from the egg, or oocyte, and none from the sperm. So transmission of mutated mitochondrial DNA from mother to offspring could theoretically be prevented by correcting the defect in the egg. This is exactly what Tachibana et al. have done.

Working with rhesus macaque monkeys (Macaca mulatta), the authors2 reconstructed mature oocytes containing the nuclear genome from one oocyte and the mitochondrial genome from another. To do this they removed the nuclear genetic material from one oocyte, leaving behind all of the mitochondrial DNA, then transferred it to another oocyte whose nucleus had been removed (a cytoplast), but which contained a full complement of mitochondrial DNA (Fig. 1).

Figure 1: Swapping mitochondrial DNA in mammalian oocytes.

Working with rhesus macaque monkeys, Tachibana et al.2 removed the nuclear material plus a cellular membrane (a karyoplast) from a mature oocyte, leaving behind its mitochondria (a). They transferred the karyoplast to an oocyte whose nucleus had been removed (a cytoplast; b). The nuclear material in the karyoplast consists of condensed chromosomes attached to thread-like spindle fibres (the spindle–chromosomal complex). The authors fused the karyoplast with the cytoplast and then fertilized the reconstructed oocyte (c). The developing blastocyst (d) was implanted in a surrogate mother, which gave birth to a healthy baby (e). This technique has the potential to prevent transmission of human mutated mitochondrial DNA from mother to baby.

Although this sounds simple, it is, in fact, a technical tour de force. Mature oocytes are arrested in the second meiotic division, a stage of the cell cycle in which the nuclear genetic material is present as a spindle–chromosomal complex, consisting of condensed chromosomes attached to the thread-like spindle fibres that distribute chromosomes to daughter cells as meiosis progresses. The nuclear membrane has broken down by the second meiotic division and so the spindle–chromosomal complex is hard to visualize except with toxic dyes. It is also incredibly difficult to remove this complex intact. To get around these problems, Tachibana et al.2 used polarized microscopy — a technique that they had developed5 to create spindle-free oocytes for reprogramming nuclei from non-gametes (somatic cells). Using this technique, the authors could remove the spindle–chromosomal complex from the donor oocyte with essentially 100% efficiency, taking with it only a small amount of cytoplasm (and so avoiding transferring mitochondria) and a cell membrane (a structure called a karyoplast). This manoeuvre was crucial to the success of the experiments.

Next, Tachibana and colleagues had to find a way to fuse the karyoplast with the cytoplast containing the mitochondrial DNA. They first tried fusion using an electric current, but this induced the oocyte to resume meiosis prematurely. The authors therefore resorted to using an extract from a virus that is known to promote cellular fusion.

The reconstructed oocytes could be fertilized and went on to develop normally as pre-implantation embryos in vitro. Embryonic stem cells derived from these embryos had a normal set of chromosomes and could mature into different cell types in culture, indicating that they developed normally. The real test, however, was to implant the embryos into female macaques to see whether they could produce normal offspring. Of the nine female macaques that were implanted, three became pregnant, one with twins. So far, three apparently healthy baby macaques have been born. Tests to determine the genetic make-up of these animals showed that the nuclear genome was inherited exclusively from the karyoplast donor, and that the mitochondrial DNA was derived from the cytoplast donor.

It is encouraging that the technique used by Tachibana and colleagues2 seems to have worked so efficiently in a primate model, but there are many hurdles, some practical, some ethical, that need to be overcome before this method could be transferred to the clinic. The practical issues relate to the safety of the viral agent used to fuse the spindle–chromosomal complex with the enucleated oocyte, as well as the efficiency with which spindle–chromosomal complexes can be safely removed and transferred from human oocytes. Also, it's not known whether the presence of foreign mitochondrial DNA in cells will have any biological consequences for the offspring, and this will have to be carefully investigated.

As discussed in a recent Editorial in Nature6, the ethical debates that surround human reproductive research will probably be revived by this work. The procedure used by Tachibana and colleagues2 requires the use of donor eggs with normal mitochondrial DNA, and certainly the research necessary to test the efficiency and safety of the procedure will require the destruction of embryos. However, unlike therapeutic cloning procedures, in which somatic-cell nuclei are transferred to enucleated eggs with the goal of isolating embryonic stem cells, here, the donor egg is not destroyed, but rather allows the birth of a healthy child.

But the mixing of nuclear and mitochondrial genomes brings other ethical issues to the fore, not least the production of offspring with genetic contributions from three parents — a combination forbidden by most jurisdictions. Thus, the laws regulating human germline DNA manipulation would have to be rewritten. It is worthwhile pointing out that if donor eggs could be obtained from a maternal relative who has not inherited the mutation, the reconstructed embryo would be genetically identical to one conceived naturally because of the exclusive transmission of mitochondrial DNA through the female germline.

Currently, the only way to prevent transmission of mutated mitochondrial DNA is pre-implantation genetic diagnosis, in which cells of the early embryo are tested for the presence of mitochondrial-DNA mutations, and only genetically normal embryos are implanted7. There is, however, no guarantee that embryos without the mutation will be identified, and it is also not always clear whether embryos with low proportions of mutated mitochondrial DNA will be free of disease. The technique reported in the present paper, if proven safe and effective, could provide a universal solution to the problem.


  1. 1

    Taylor, R. W. & Turnbull, D. M. Nature Rev. Genet. 6, 389–402 (2005).

  2. 2

    Tachibana, M. et al. Nature 461, 367–372 (2009).

  3. 3

    Schaefer, A. M. et al. Ann. Neurol. 63, 35–39 (2008).

  4. 4

    Elliott, H. R., Samuels, D. C., Eden, J. A., Relton, C. L. & Chinnery, P. F. Am. J. Hum. Genet. 83, 254–260 (2008).

  5. 5

    Byrne, J. A. et al. Nature 450, 497–502 (2007).

  6. 6

    Nature 460, 1057 (2009).

  7. 7

    Steffann, J. et al. J. Med. Genet. 43, 244–247 (2006).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Shoubridge, E. Asexual healing. Nature 461, 354–355 (2009) doi:10.1038/461354a

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.