TORONTO--One of the biggest questions in stem cell biology is how the cloning process manages to turn back the clock of mature cells, resetting them to their embryonic potential. Ideally, researchers would like to find a way to convert adult cells directly into to embryonic stem (ES) cells--without having to create an embryo at all. At a meeting of the International Society for Stem Cell Research here, Shinya Yamanaka of Kyoto University in Japan reported that boosting the activity of just four genes can apparently turn mouse skin cells into cells that closely resemble ES cells.
Yamanaka and his colleagues wondered whether the factors that give ES cells their unique properties might also be able to reprogram adult cells to behave like ES cells. They identified 24 genes that are specifically expressed in mouse ES cells and used viral vectors to introduce extra copies of the genes into skin cells taken from mouse tail tips. When they inserted extra copies of all 24 genes, they found that a small percentage of cells that took up the genes did indeed seem to take on characteristics of ES cells. But no single gene introduced alone was able to manage the transformation.
Through a process of elimination, the team whittled down the candidates to a suite of just four genes that, when introduced together into the tail-tip cells, could produce colonies of ES-like cells. As Yamanaka described, three of the four factors are old friends: Oct4, Sox2, and c-Myc are all key genes in both early embryos and ES cells. Yamanaka did not name the fourth gene, but he said it is a transcription factor that until now has not been recognized as playing a major role in ES cells.
The ES-like cells the group produced with the four introduced genes seemed to have almost all the key properties of ES cells derived from embryos. They formed several kinds of tissue in the culture dish and produced tumors called teratomas when they were injected under the skin of immune-compromised mice--both classic characteristics of ES cells.
Yamanaka says his group has not yet tried the technique with human cells. Because of differences in human and mouse embryo development, he says, it's possible that a different set of genes would be required to reprogram human cells.
Other researchers at the meeting were impressed. "It's huge," says Kevin Eggan of Harvard University, who also works on cell reprogramming. Still, he notes that the process is not yet very efficient; the four introduced genes managed to reprogram just 1 out of 1000 cells that received them. That suggests that the four genes are perhaps not the whole story, and that another factor could improve the efficiency of the process. "But this is the litmus test" for finding the genes that are essential for reprogramming, he says.