PrimeGen Biotech of Irvine, Calif., says its embryonic-like stem cells are created without human eggs or embryos - as were the cells reported by researchers in Wisconsin and Japan last fall.

The difference is that PrimeGen has shortcut the process of turning a cell's developmental clock backward, using what it calls "purified proteins." That not only hastens the reprogramming - a week or two instead of a month - but also avoids the theoretical danger of triggering cancer, company president John Sundsmo told a stem-cell industry conference Tuesday.

Sundsmo was vague about this shortcut during his 10-minute presentation at the New York City conference, attended by potential investors. In an interview yesterday, he said he could not go into detail until a new corporate partner signs off.

Stem-cell experts were intrigued but skeptical of PrimeGen's announcement, reiterated in a news release yesterday, given that its new research has not been published or even submitted to outside experts for peer review.

"Whether they have accomplished what the press release says, we have to wait for peer review," said Jonathan Epstein, a stem-cell researcher at the University of Pennsylvania. "But if they have, I think it's very exciting."

Konrad Hochedlinger, a molecular biologist at the Harvard Stem Cell Institute, doubted that PrimeGen's methods could produce a large-enough colony of embryonic-like stem cells in only a few weeks.

Last week, Hochedlinger published a paper that defined, for the first time, the step-by-step molecular events required to turn a mature cell back into its earliest, or "pluripotent," state. Embryonic stem cells are pluripotent, meaning they can produce copies indefinitely or, with the right chemical cues, morph into any specialized cell type in the body - a versatility that makes these cells potentially useful in treating disease.

Hochedlinger's experiments showed that cells need about two weeks to become pluripotent, and a few weeks more to grow into a sizable colony.

"Technically and biologically, it would be possible to use proteins to reprogram the cells," Hochedlinger said. "But it would take at least two weeks to 20 days, so I'm rather skeptical about five to seven days."

While he would not give details, Sundsmo said PrimeGen's shortcut involves "delivery molecules" that "essentially inject" reprogramming proteins into the cell. The technique is similar, he said, to one in which cells take in molecules or viruses that lock on to a receptor on the cell wall.

In contrast, the approach reported by scientists in Wisconsin and Japan - and, most recently, at the University of California at Los Angeles - is more manipulative. It requires using an inactivated virus to carry three or four powerful genes into the cell. These genes - which end up exerting control over all the other genes - issue instructions for producing the proteins that actually carry out the molecular march backward to pluripotency.

The theoretical problem, experts agree, is that both the virus and the extra genes could interfere with the cell's normal DNA in a way that would trigger abnormal, uncontrolled growth - in a word, cancer.

PrimeGen, a privately held company set up five years ago, has been underwritten primarily by John Tu, founder of flash-drive-memory maker Kingston Technology, and Thomas C.K. Yuen, president of audio-equipment maker SRS Labs.

PrimeGen plans to present its latest data in June in Philadelphia at an international stem-cell research convention, Sundsmo said.

The data include evidence, such as key molecular markers, that verifies the embryonic-like stem cells are truly pluripotent.

Sundsmo acknowledged, however, that his researchers were still waiting to see if the cells induce nonmalignant tumors, called teratomas, in rodents. Teratomas are the ultimate proof of pluripotency because they contain tissues derived from all three basic embryonic layers, such as bone, muscle, cartilage, nerve, hair, even teeth.

"Making teratomas in animals is the acid test," Sundsmo said. "We've got everything but the acid test."

Even if PrimeGen's reprogramming technology works well, the next big challenge will be figuring out how to reliably control the stem cell's transformation into cells suitable for treating Parkinson's, diabetes, spinal injury and other diseases, Hochedlinger said.

Nonetheless, Sundsmo said PrimeGen hopes to complete animal testing by the end of next year, and begin human clinical trials of its regenerative cells by 2010.

"This field is moving incredibly fast," he said.