Bad programming. These hybrid embryos, in which human DNA has been placed in animal eggs, don't turn on the same genes as human embryos.

ACT/Cloning and Stem Cells

Research Questions Hybrid Approach to Stem Cells

Can a person be cloned? And can human-animal hybrid embryos produce stem cells that shed light on human diseases? "Probably" and "no" are the respective answers to these provocative questions, according to a study out today.

Ever since researchers cloned Dolly the sheep in 1996 by transferring the nucleus of one of her cells into the nucleus-free egg of another sheep, scientists, ethicists, politicians, and the public have wondered whether a person could be cloned in the same way. Most have rejected the idea of human cloning as unethical. But if this technique--called somatic cell nuclear transfer (SCNT)--can produce healthy human embryos, it might also provide medically useful stem cells, ones that would have the same genetic profile as a patient and would thus avoid immune-rejection issues.

A South Korean team famously claimed a few years ago to have made human embryonic stem cells using SCNT, but that cloning work was found to have been fraudulent (ScienceNOW, 12 May 2006), and no one has reported a similar success since (Science, 22 December 2006, p. 1853). One roadblock to understanding SCNT has been a lack of human eggs, so some scientists are experimenting with a hybrid approach in which they place human DNA into animal eggs. No cells from such hybrid embryos would be transplanted into people, but if the DNA came from a person with a genetic disease, scientists might derive human stem cell lines from the embryos to better study the condition. Aside from a 2003 report of stem cells resulting from human DNA in rabbit eggs, however, there have been no successes using this approach. Advanced Cell Technology in Worcester, Massachusetts, for example, has gone through thousands of animal eggs trying to produce human stem cells with SCNT, says the company's chief scientific officer, Robert Lanza.

To understand its lack of success, ACT researchers, along with their partners, used SCNT on human eggs to create cloned human embryos, as well as hybrid embryos, using mouse, cow, and rabbit eggs. They then took a close look at what genes were on and off in those embryos. For the transferred nuclear DNA to generate a healthy embryo, molecular factors within the egg must essentially turn back time, reprogramming the DNA to turn on the genes essential for embryo development.

Lanza and his colleagues report online today in Cloning and Stem Cells that the human eggs effectively reprogrammed the human DNA so that gene activity mirrored that in normal human embryos over the first few days of development. Not so with the hybrid embryos; key developmental genes, including several needed for stem cell formation, were silenced or remained off. The researchers conclude that the hybrid approach will not generate human embryonic stem cells, presumably because eggs have species-specific factors that reprogram DNA.

Stem cell scientists, particularly those pursuing hybrid embryo research, vehemently challenge the study's conclusions. They criticize ACT, for example, for not letting the hybrid embryos develop longer, in case the reprogramming happens on a delayed basis when mixing species. They also note that ACT didn't try the hybrid approach with monkey eggs--which are closer, genetically, to human eggs--or exhaust all the tricks cloning researchers have developed. "I don't think they've definitively shown anything," says Stephen Minger of King's College London, who has one of several U.K. licenses to conduct hybrid stem cell research.

Notwithstanding this debate, Lanza takes encouragement from his team's results with human eggs. "For the first time, this paper shows that SCNT activates all the core genes necessary for cellular reprogramming," he says. That suggests scientists should eventually be able to produce patient-matched stem cells from a cloned human embryo. And, Lanza acknowledges, the work also indicates that cloning an actual person is indeed possible.