It's official. Primate genomes can be reset to an early embryonic state. Years of unsuccessful attempts to create embryonic stem (ES) cells by combining monkey eggs and nuclei from adult cells had led to speculation that the genomes of differentiated primate cells could not be turned back to the embryonic state of pluripotency — the ability to produce any cell type in the body. But this week, an article in Nature reports success at last in using nuclei from adult rhesus monkey cells to make ES cell lines1.

Semos: the donor of the cells used to create cloned embryos Credit: Oregon Health & Science University

If the same can be done with human cells, ES cell lines could be created that are genetically identical to a patient with a particular disease. These cells could then be used to study the disease or, perhaps, grown to produce replacement tissues that would not be rejected.

A team led by Shoukhrat Mitalipov at Oregon Health & Science University in Portland took nuclei from skin cells of an adult monkey, injected them into enucleated eggs collected from fertile monkeys, coaxed the eggs into forming hollow ball-shaped embryos (blastocysts) in laboratory dishes, and then scooped out the innermost cells to create two ES cell lines1. It wasn't easy: it took 304 eggs to create the two cell lines, one of which is genetically abnormal. Mitalipov reports that 35 of the 304 oocytes developed into blastocysts, 20 of which were of high enough quality to attempt nuclear transfer.

Now that Shoukhrat has gone that much further we can raise our expectations a bit. Mary Herbert, Newcastle University

Therapeutic cloning (also called oocyte-assisted reprogramming) and reproductive cloning both start by making a cloned embryo, and so Mitalipov's success indicates that cloning monkeys (and perhaps even people) might be possible. But reproductive cloning tends to be much more difficult than making ES cells from an embryo, partly because cloned embryos often don't survive long when transplanted to a surrogate mother and make abnormal placentas when they do. The Oregon team has not yet produced a live cloned monkey, despite many attempts. Dolly the sheep, the first cloned mammal, was born only after attempts with 277 oocytes. And when it comes to human reproductive cloning, scientists in the field, as well as the general public, object on many ethical grounds.

Mitalipov's work is highly significant for showing what is possible, says Alan Colman, an animal cloning and human stem-cell expert and head of the Singapore Stem Cell Consortium, but he thinks the number of eggs used per cell line surprisingly high. “It didn't give me much encouragement that the technique would be worth pursuing in humans because the numbers are so poor.”

Still, Mary Herbert at Newcastle University in the UK is collaborating with Mitalipov and hopes to start applying his techniques to human eggs before the month is out. She expects that her team will be able to use more than 500 freshly harvested human oocytes from women who opt to donate eggs collected during fertility treatments. Other researchers across the world are also attempting nuclear transfer with human oocytes. Still, few or none will have access to so many high-quality oocytes as the Newcastle team.

What can be learned?

One could question how much effort should be devoted to adapting the monkey work to humans, given the low efficiency, the scarcity of human eggs, and the success in reprogramming differentiated mouse skin cells without even needing to use eggs2. In fact, Shinya Yamanaka of Kyoto University is publishing an article in Cell that human cells as well as mouse cells can be reprogrammed by inserting a series of genes associated with pluripotency.

But the cells produced through each method are likely to have different applications, with oocyte-derived cell lines being more suitable for clinical applications and genetically reprogrammed cell lines more suitable for use in screens for new drugs, says Colman.

More importantly, figuring out how an oocyte reprogrammes a donor nucleus could be key to unlocking any cell's potential. The 'epigenetic' modifications to DNA that control gene activation differ between cells derived from fertilized embryos, embryos produced through nuclear transfer, adult animals and directly reprogrammed cells, says Jose Cibelli of Michigan State University in East Lansing, who helped to produce the first cloned human blastocyst (which didn't get past the six-cell stage). Figuring out which techniques improve success won't just provide additional research material — it could suggest new reprogramming techniques and refine existing ones.

Keys to success

Mitalipov attributes much of his success to a less invasive technique for removing the oocyte's own genetic material. Rather than using a fluorescent dye (known as bisbenzimide or Hoeschst 33342) and ultraviolet light to find the egg's chromosomes, he used an imaging system called Oosight, which doesn't require such harsh pretreatment.

At first, Mitalipov says, he didn't think the DNA stain would matter much because the stained chromosomes are all removed. But closer inspection revealed that the stain remained behind in the egg, binding the egg's mitochondrial DNA as well as the donor DNA — and mitochondria provide essential energy for reprogramming and cell division.

Although he does not know the exact mechanism, Mitalipov speculates that the bisbenzimide method could also activate the oocyte too early or interfere with a protein complex called maturation-promoting factor (MPF), which helps break down the nuclear envelope. After a donor nucleus is injected into an egg, its nuclear membrane must, according to conventional wisdom, be stripped away for reprogramming to occur. Mitalipov, along with researchers from the Chinese Academy of Sciences in Beijing, found that the nuclear membrane seemed to remain intact if the DNA stain was used but broke down if it was not3.

However, many researchers think that factors other than eliminating the stain might have led to success. “People get better at this procedure; that can make as much of a difference as changing a single variable,” says Kevin Eggan at Harvard University. Eggan is working on alternative techniques for therapeutic cloning that consist of cloning from chromosomes instead of whole nuclei. But Mitalipov has also described studies conducted under identical conditions that used nuclei from fetal fibroblasts3. In that work, 10 blastocysts developed from 67 embryos created through nuclear transfer using Oosight; only 3 out of 235 developed that far when conventional staining and ultraviolet light were used.

Primate eggs, which contain less fat than cow and pig eggs, seem particularly sensitive to ultraviolet light, and many researchers attempting somatic cell nuclear transfer with human oocytes have already abandoned the conventional staining technique. Not surprisingly, they think more than an advanced imaging system is key to success. Scientists at the Oregon National Primate Research Center have been trying to clone monkey embryos for more than ten years, and many credit Mitalipov's accomplishment to prowess under a microscope — the speed and precision with which his team can remove chromosomes from an egg and inject donor nuclei into it. “It's not an either-or or a black-and-white answer. The stain could be part of it, but not all of it,” says Cibelli.

Mitalipov and others not involved in the work believe that the procedure will be improved to yield higher success rates. “Once you get one embryo [to develop], the efficiency improves quite quickly,” says Andrew French, chief scientific officer at stem-cell company Stemagen in La Jolla, California, who has worked on nuclear transfer with both cows and humans.

On the learning curve

Mouse cells were the first mammalian ES cells cloned, in 2000 (ref. 4). Megan Munsie, now at the Australian Stem Cell Centre in Adelaide, was the first author on that paper. She credits a gentler way of injecting the donor nucleus into the egg as particularly important to success, but says she made tweaks to each step of the protocol. “There are bottlenecks in each technique, so whether you're doing stem-cell derivation or enucleation, there are technical hurdles to be overcome.”

Getting the mouse ES cells through nuclear transfer was a process of trial and error and “many teary moments,” recalls Alan Trounson, a member of that team and president-designate of the California Institute of Regenerative Medicine. “I think we changed nearly every factor we worked with,” he says. “We were obviously so far out of the optimum space that it wouldn't work for quite some time.”

Trounson, who helped verify that Mitalipov's monkey ES cells were made by nuclear transfer5, believes that the morale boost will be more important than changes in technique. “It will encourage people to do the work. That's what the impact will be.”

For both mouse and monkey, a major hurdle seems to be getting embryos to keep dividing after the embryonic genome becomes active (two cells for mice; eight cells for primates). Almost all techniques for making ES cells gather cells from the inner cell mass of the blastocyst, at which point the embryo contains dozens of cells. This procedure destroys the embryo and is one reason why some people oppose human ES cell research. Although the overall efficiency is low, the rate at which Mitalipov coaxed early embryos to develop to the blastocyst stage is much improved over previous work.

Quality counts

Human eggs may even have some advantages over monkey eggs. Mitalipov believes that one reason so few of his attempts yielded ES cell lines was because the monkey oocytes may have been of low quality. These monkeys' ovaries were stimulated with hormones so that they released as many as 40 eggs at a time, which may have compromised egg quality. Procedures worked out for women seeking fertility treatment are more refined and produce fewer eggs. Mitalipov speculates that they could yield higher-quality oocytes.

In Newcastle, Herbert and her colleague Allison Murdoch will obtain eggs through an egg-sharing programme that allows researchers to pay some of the cost of fertility treatment in exchange for some of the eggs collected. She says that they already have a waiting list of donors, so her team should be able to select high-quality oocytes. This would mean that her team could have more eggs than Mitalipov used. Herbert is also looking to improve the efficiency at which eggs with donor nuclei grow into blastocysts by changing the protocol, in particular how eggs are activated to begin dividing. The goal is to “try to mimic more truthfully the signals that a sperm induces,” she says.

Originally, Herbert had written a grant requesting funding primarily to develop reliable techniques for growing human blastocysts produced through nuclear transfer. Now she hopes there will be enough blastocysts to derive stem-cell lines she says. “Now that Shoukhrat has gone that much further we can raise our expectations a bit.”