On Friday 24 March, researchers at the Georg August University of Göttingen in Germany announced they had found a source of reprogrammable cells in the testes of adult mice. By Sunday morning, more than ten companies had e-mailed senior team member Gerd Hasenfuss about collaborating on future work.

Hasenfuss, a cardiologist, admits that much work must be done before the discovery could lead to any applications in humans. But the finding drew attention immediately: it promises, at least for men, a simple and uncontroversial method for harvesting therapeutic stem cells. Two other research groups have made similar claims in the past, and a patent showdown among the three is looming.

As Hasenfuss's team reports in Nature, the reprogrammable cells come from sperm-producing stem cells in mice (K. Guan et al. Nature 200610.1038/nature04697). It is not clear whether the cells in the testes have the ability to produce all other kinds of bodily cells, as embryonic stem cells can. But after the team extracted, cultured and multiplied the cells 700-fold, they converted to an embryo-like state in 4 out of 15 cases. When injected into early embryos, the converted cells helped build various organs of the resulting mouse. And, in vitro, the converted cells turned into several different types of cell, including heart, brain and skin cells.

Hopes and fears: sperm-producing cells can stand in for embryonic stem cells, in mice at least.

Hoping to move the work into humans, the team has already received tissue samples from testicular-cancer patients. Hasenfuss thinks that, eventually, a simple biopsy could extract the necessary cells. Such cells would be a genetic match with the patient, which makes them less likely to be rejected if used in therapy.

But Hasenfuss's work has its sceptics, including Takashi Shinohara, a mouse germ-cell expert at Japan's Kyoto University. Shinohara's group was the first to isolate reprogrammable cells from mouse testes, but it used two-day-old pups (M. Kanatsu-Shinohara et al. Cell 119, 1001–1012; 2004). His experience differs from that of the Göttingen researchers in some ways — for example, they report that stem cells injected into embryos are able to integrate into various tissues. “Perhaps they have some different kind,” he says, “but I don't think that type of cell exists.”

Shinohara also points out that many researchers have tried doing similar experiments, with more sophisticated culture media, without the same results. “It's too good to be true,” he says. Kaomei Guan, of the Göttingen team, counters that simplicity may be the key. Applying a complicated mix of certain growth factors in a lab dish for a long period of time, he says, can cause other cells to take over and the sperm stem cells to die.

A third, US-based group has already claimed some success in related experiments with human cells. Francisco Silva is vice-president of research at PrimeGen Biotech, based in Irvine, California. He says that, working with samples from 22 testes, the company “has successfully therapeutically reprogrammed human germ cells and differentiated them to multiple cell types in vitro”. Silva has presented the data at scientific meetings, and says he plans to submit results for publication soon. But Shinohara is again sceptical.

Hasenfuss, Shinohara and PrimeGen all say they have filed for international patents on their technologies. Some legal squabbles are likely. Hasenfuss, for example, claims work in adult mice is different from that in neonatal mice. Shinohara says the two experiments are essentially the same.

Whether any of the mouse-based claims would cover humans is not clear. Even if a patent office grants a patent, it may be invalidated by the courts, says Lynn Pasahow, an intellectual-property attorney with Fenwick & West in Mountain View, California (see ‘US to rule on research patent’). “It may be difficult to know the scope of the valid claims,” he says.

In any case, researchers still have to show that the cells can be derived, grown and manipulated just like those from the alternative source, cloned embryos. In theory, the reprogrammable testes cells could circumvent the ethical difficulties of stem-cell work that involves destroying human embryos. “This could put the embryonic stem-cell people out of business,” says Peter Donovan, a stem-cell expert at the University of California, Irvine. “But it remains to be seen whether they work in humans.”