THE mammalian egg, the cell that holds the secret to fertility, cloning and cell rejuvenation, has been created outside the body for the first time. And, 快猫短视频 has learned, researchers are very close to creating sperm in a similar way.
The feat was accomplished with cells originally derived from mouse embryos, but most experts see no reason why the technique would not work with human embryonic stem cells too. If human eggs and sperm created this way are healthy 鈥� and it is a big if 鈥� the implications for reproductive technology and regenerative medicine would be immense.
Most immediately, a cheap, limitless supply of human eggs would greatly accelerate research in keys fields such as infertility and therapeutic cloning. 鈥淚t鈥檚 terrific,鈥� says Jose Cibelli of Michigan State University, the first scientist to publish details of attempts to clone human cells. 鈥淓ggs were one of those cell types we never thought we could produce.鈥� In the longer term, the applications might even include allowing infertile women or a pair of men to have their own children.
Advertisement
The key step in the formation of eggs and sperm is a form of cell division called meiosis, which produces cells with one set of chromosomes instead of two. Fertilisation then creates an embryo with the full two sets. There have been attempts to create artificial eggs or sperm by persuading normal cells to spit out one set of chromosomes (快猫短视频, 22/29 December 2001, p 24), but embryos created from such cells never survive long.
This failure involves imprinting, the process whereby chemical changes to DNA turn some genes on or off. For an embryo to develop normally, its genome must have the correct imprinting pattern, which forms as eggs and sperm cells mature. You cannot create an embryo merely by adding the set of chromosomes from one egg to those of another, because paternal chromosomes have a different pattern of imprinting from maternal ones.
The method used to create eggs from embryonic stem cells (ESCs) was astonishingly simple. Instead of searching for chemicals that coax ESCs into becoming eggs, as many have attempted, Hans Sch枚ler鈥檚 team at the University of Pennsylvania just let mouse ESCs grow at a high density. In these conditions, some of the cells form floating aggregates most scientists would discard as useless debris. But team member Karin H眉bner instead placed the clumps in new dishes.
鈥淚n four days they proliferated like crazy,鈥� says Sch枚ler. The aggregates seem to behave like miniature ovarian follicles in which small cells nurture a bigger cell that forms an egg.
Further studies by the team revealed that these egg-like cells form by meiosis, and switch on the same key genes as normal eggs as they develop. The follicle-like structures make hormones such as oestradiol in amounts that rise and fall on the same timescale as the menstrual cycle. Adding a hormone called gonadotrophin triggers the expulsion of the egg cell into the culture dish, mimicking ovulation (Science, DOI: 10.1126/science.1083452).
Intriguingly, eggs form from both female (XX) and male (XY) ESCs. That is because mammalian germ cells will go down the egg route unless signals produced by the testes tell the cells to become sperm.
This is also why getting ESCs to turn into sperm is more complex. However, a team led by Toshiaki Noce at the Mitsubishi Kasei Institute of Life Sciences in Tokyo may already have succeeded. According to a document found by 快猫短视频, the team allowed male mouse ESCs to develop spontaneously into various different types of cell, and picked out those that had begun turning into germ cells.
These cells do not develop far in culture, but when Noce鈥檚 team transplanted them into testicular tissue he found after three months that they had undergone meiosis and formed what appeared to be normal sperm.
Both teams have yet to perform the next, crucial step: fertilising the artificial eggs with normal sperm, or using the artificial sperm to fertilise normal eggs. Since Sch枚ler鈥檚 egg have a thinner outer layer and are more fragile than normal eggs, they could be harder to fertilise.
The big question is whether the resulting embryos will have normal imprinting and develop into healthy baby mice. 鈥淭here is always a possibility that the imprints would not be normal,鈥� says Azim Surani of Cambridge University.
He points out that imprinting can be affected simply by growing cells in the lab. Even children born through normal IVF may have a higher risk of certain defects. 鈥淵ou could create lots of abnormalities,鈥� warns Surani.
But if the animals are normal, the race will be on to create artificial human egg and sperm cells this way. It is likely to be much harder with human cells. Attempts to obtain human eggs simply by growing slices of ovarian tissue have failed, and the only mouse created from an egg obtained this way, dubbed Eggbert, was sickly and died young. It will also be slow: it takes up to six months for a human egg to mature.
But if it can be done, the possibilities are astounding. In cloning, for instance, there are massive problems with imprinting. When a nucleus is transferred to an empty egg, chemicals in the egg seem to alter the imprinting, 鈥渞eprogramming鈥� the genome to ensure the right genes are expressed in the growing embryo. But the process is far from perfect, and many cloned embryos die or give rise to animals with major defects.
The offspring of cloned animals, however, seem to be normal, prompting some cloning experts to suggest that the formation of eggs and sperm corrects any lingering imprinting defects. It might be possible to take an individual鈥檚 cell, create ESCs from it by therapeutic cloning, and then derive healthy eggs or sperm from them for use in IVF (see Graphic).
The most obvious application would be to treat infertile women who cannot produce any eggs suitable for IVF, or men who cannot produce sperm. And because male ESCs can be turned into eggs as well as sperm, two men could both be biological parents of a child, with the help of a surrogate mother; two out of three of such children would be male. The technique would not allow two women to have children together, though, as female ECSs lack the Y chromosome, without which sperm cells appear unable to form.
While it sounds far-fetched, the method would even allow 鈥渟elf-fertilisation鈥� by a man. Such a child would not be an identical clone of its parent, as it would have lost some of his genetic diversity. Even more controversially, it might be possible to genetically engineer children by modifying ESCs in culture and then deriving and selecting sperm and eggs carrying the modifications.
For now, the potential risks of the technology rule out such uses, Surani says. And Sch枚ler agrees. 鈥淭he one thing I don鈥檛 want is that the oocytes be used to make humans,鈥� he says.
Yet while some countries have regulations that would prohibit the use of the technology, in the US and other countries there is far less control of IVF clinics. Now may be the time to consider the potential benefits and concerns, before experiments even begin with human cells. 鈥淚t鈥檚 always good to have these kinds of debates of potential ethical issues,鈥� says Surani.