The technique for deriving embryonic stem cells in mice described on page 212 of this issue offers a rare example of science being driven by biomedical ethics.

Rudolph Jaenisch, a professor of biology at the Massachusetts Institute of Technology, had become frustrated at the US government's policy on human embryonic stem-cell research. The idea of taking DNA from a patient and deriving stem cells from this material in an embryo — often described as therapeutic cloning — could offer medical benefits for conditions such as Parkinson's disease. Although such work is usually carried out on ‘excess’ embryos generated during in vitro fertilization procedures, extracting the stem cells necessarily means the destruction of a human embryo. The US government has put a moratorium on such work, saying that it destroys human life as the embryos could theoretically have been implanted and brought to term.

Jaenisch wondered whether the idea of ‘accelerated nuclear transfer’, or ANT, put forward by bioethicist William Hurlbut of Stanford University, might offer a way out of this dilemma. In essence, Hurlbut suggested that if research in human embryonic stem cells used a biological entity that could never have developed into a fetus, the argument that life was being destroyed would no longer stand.

Jaenisch accepted the challenge and set about turning this theory into practice. Working with mice, he reasoned that the criteria for the concept would be met if he could create an embryo that was unable to implant in the uterus. The target for this work became the gene Cdx2, which in mice is responsible for placental implantation.

His student, Alexander Meissner, began the project using mouse skin cells. With a little work, he was able to silence the Cdx2 gene in the cells and then cloned them to create an embryo that could not be implanted successfully. Meissner completed the work in about four months. “That is really fast for such a complicated experiment,” Jaenisch says.

The work, which Nature published online on 16 October 2005, has had a mixed response from politicians and ethicists. “Some people will still say it is a particularly devious way of murder,” Jaenisch says. “But many very prominent ethicists think this is a very serious way to proceed.”

The next step, Jaenisch says, is to try this approach in humans. Theoretically this should be straightforward, as humans have a gene that is equivalent to Cdx2. Jaenisch remains optimistic that the concept could prove acceptable to the US government, and so allow stem-cell research to proceed.

“I think that the ANT approach gives people in Congress who want to support this, but don't dare, some confidence that the ethical problems may not be insurmountable,” Jaenisch says. But he acknowledges that any work on human embryonic stem cells — no matter how they are derived — will still attract some criticism. “It has been controversial before and it will be controversial after this paper,” he says.