Making new eggs in old mice

A textbook-challenging finding revives debates about extending female fertility.

Adult mice might be able to generate new eggs. Credit: David Spears Ltd/ Science Photo Library

Researchers in China today announced a discovery that challenges a canonical belief in reproductive biology: that women are born with a set number of immature egg cells, called oocytes, which become depleted with age. In a paper in Nature Cell Biology,1 the Chinese team reports that it has found precursors to oocytes in adult mice. When the researchers transplanted those cells into sterilized mice, they produced offspring — a finding that feeds into an ongoing debate about the limits of mammalian fertility.

"It provides a smoking gun," says Jonathan Tilly of Massachusetts General Hospital in Boston, who was not involved in the work but has previously published evidence that new eggs can be formed in adult mice in a series of argument-sparking papers.2 Although the current work hasn't settled the question, Tilly says, it represents an addition to the "critical mass of data" suggesting that old ovaries can produce new eggs, a finding that some say could have implications for fertility medicine.

“It provides a smoking gun. Jonathan Tilly , Massachusetts General Hospital”

Roger Gosden of Cornell University in New York has been critical of such research in the past. "It's a dramatic paper," he says, but as such, must pass very high technical standards and be replicated elsewhere. "I'm wanting more before I swallow the story whole," he says. David Albertini of the University of Kansas Medical Center in Kansas City says he does not think the data support the claims in the paper because of outstanding questions about the offspring and methods they used to derive the cell line. "It's really hard to know what's going on here," he says.

Ji Wu of Shanghai Jiao Tong University in Shanghai and her colleagues report that they've identified and isolated a small population of cells from the surface of mouse ovaries that have the ability to self-replenish for many generations in a Petri dish. Moreover, they share some of the protein markers that scientists look for when trying to identify primordial germ cells, a population of cells that during foetal development are believed to produce all of the oocytes a female will have for the rest of her life. The group's stem cell–like cells, capable of replenishment and differentiation, were extracted and cultured from the ovaries both of neonatal mice, just five days old, and adult mice.

Once the cell lines were established, the researchers infected the cells with a virus that would cause them to produce green fluorescent protein (GFP), allowing them to be tracked by microscopy. Wu's group injected the cells into the ovaries of mice that had been treated with chemicals commonly used for chemotherapy, which severely damage existing egg cells and leave the mice essentially sterile. Roughly 80% of the mice that received cell transplants gave birth, regardless of whether the transplants came from young or old mice. About a third of these offspring were positive for the GFP gene, supporting the claim that the mice came directly from donor cells.

The results seem to go beyond those of a 2007 study by Tilly and colleagues, which showed that a bone marrow transplant could restore fertility in chemically sterilized mice. In that earlier study, however, the offspring were biologically related to their mothers and not to the bone marrow cells.3

Questions remain

Demonstrating that offspring came from the donor cells is a critical leap for the current paper, but not all offspring have the GFP gene. One potential explanation, say the authors, is that not all transplanted cells were infected by the virus carrying the GFP gene. Another, says Albertini, is that the donor cells did not actually turn into oocytes. Instead the chemotherapy may not have completely sterilized all of the original oocytes in the adult mice, or the cell-transplant procedure could have re-established or somehow protected the poisoned oocytes in the host ovary. In either of those cases, some of those original oocytes may have picked up the GFP gene from donor cells.

Wu's group provides some evidence against that kind of gene transfer into the original oocytes, showing that the virus used to infect the germline stem cells is incapable of infecting cells in healthy ovarian tissue.

But the paper does not show that the same holds true after chemotherapy or that the virus cannot infect other stem-cell populations. Moreover, the article does not mention what percentage of developing oocytes in the treated mouse ovaries glow green. In response to a list of e-mailed questions, Wu wrote that there were GFP-negative oocytes, but she did not provide a ratio of positive to negative. She wrote that it is not possible that offspring came from host oocytes in the ovary.

Joshua Johnson, an assistant professor at Yale School of Medicine in New Haven, Connecticut, who worked with Tilly, agrees. Incomplete sterilization, he adds, shouldn't matter, given the high rate of fertility in the treated mice and the high proportion of offspring with the GFP gene. Gosden says that although he has his doubts, "the fertility pattern in these animals is remarkable."

Still, the paper leaves a key question open: the relevance of these cells physiologically. The authors say they appear similar to spermatogonial stem cells, the cells that in males replenish sperm reserves throughout life. But whether these female germline stem cells are adding to the ovarian reserve of eggs is unclear. "There's the possibility that they're present in adult tissue but don't do anything," says Tilly. "If that's true, it's disappointing, but I don't think it is the case." Wu is similarly optimistic: she says that she believes under normal conditions "they can renew themselves and differentiate into oocytes".

Albertini says he's unconvinced that the newfound cells represent anything more than an odd, laboratory-created transgenic stem-cell line. If that's true, he says, researchers could capitalize on this procedure as an efficient means for creating transgenic offspring. As for fertility treatments, however, he says there's "no clinical relevance".

“If it's replicated, we do have to rewrite the textbooks. Roger Gosden , Cornell University”

Gosden says there's much more work to be done to prove a physiological role for the cells that the Chinese group found. "I'm not bowled over by there being cells in the ovary after birth that have germ cell–like properties," he says. Still, the cells themselves in culture could help in understanding the complexities of egg development. Johnson says he's excited to work with them, and hopes that others will try to replicate the work.

Though debate lines hold firm for now, both sides agree that the work, from a group that few in the field had heard of, will require independent verification. Gosden says, "If it's replicated, we do have to rewrite the textbooks."


  1. 1

    Zou, K. et al. Nat. Cell Biol., doi:10.1038/ncb1869 (2009).

  2. 2

    Tilly, J.L., Niikura, Y. & Rueda, B.R. Biol. Reprod. 80, 2–12 (2009).

  3. 3

    Lee, H.J. et al. J. Clin. Oncol. 25, 3198-3204 (2007).

Download references

Related links

Related links

Related links in Nature Research

Nature Outlook: Fertility

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Maher, B. Making new eggs in old mice. Nature (2009).

Download citation


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.