All of the human embryonic stem cells available to federally funded scientists under President Bush's three-year-old research policy share a previously unrecognized trait that fosters rejection by the immune systems, diminishing their potential as medical treatments, new research indicates.

A second study has concluded that at least a quarter of the Bush-approved cell colonies are so difficult to keep alive they have little potential even as research tools.

The two studies -- the second still incomplete and the first one provisionally accepted for publication in a top-tier scientific journal but not yet published -- add new elements to the escalating debate over U.S. stem cell policy.

Embryonic stem cell research has become an unexpected wedge issue in the neck-and-neck race for the White House, with Bush insisting that it would be immoral to expand the research to include new cell colonies and Democratic challenger John F. Kerry promising to loosen the restrictions that today limit federal funding to 22 of the more than 150 known cell colonies.

The first study, led by Fred Gage of the Salk Institute in La Jolla, Calif., and Ajit Varki of the University of California at San Diego, focused on a peculiar aspect of the federally approved cell lines: Unlike colonies being derived using newer techniques, all the Bush-approved colonies were initially cultivated in laboratory dishes that also contained mouse cells.

Scientists and the Food and Drug Administration have already expressed concern that animal viruses lurking in those mouse cells might infect the human cells and cause trouble when they are transplanted into patients, as doctors hope to do.

With adequate testing, those cells may yet gain approval for use in patients, the FDA has said. But the new work suggests that the mouse-exposed cells have an additional drawback.

At the heart of the problem is that all mammalian cells -- with the exception of human cells -- bear certain molecules on their surface, known as N-glycoylneuraminic acid. (Human cell surfaces bear a different but related molecule, N-acetyl neuraminic acid.)

Varki had previously demonstrated that the vast majority of people have antibodies against this molecule, perhaps as a result of eating mammalian meat such as beef. The new work shows that human embryonic stem cells grown on mouse cells "consume" the mouse molecules and then display them on their own surfaces.

When human blood serum was added to the mouse-cultivated human stem cells in lab dishes, antibodies attacked the stem cells and killed them. In the eyes of the immune system, "these human cells look like animal cells . . . which leads to [their] death," Gage said at a recent scientific meeting.

Details of the experiment are embargoed until the research report is published, but Gage described the work on Oct. 12 to a panel of experts at the National Academies of Science, which is drawing up policy recommendations on stem cell research.

Several teams around the world have lately had success growing human embryonic stem cells without mouse cells, and proponents of stem cell research said yesterday that the findings strengthen the case for letting federally funded researchers work on newer stem cell colonies.

"This study appears to point out yet another flaw with the president's policy," said Sean Tipton of the Coalition for the Advancement of Medical Research, a Washington-based consortium of patient advocacy groups, scientific societies and research institutions. "It means these cells are unlikely to be useful for medical purposes."

James Battey, who heads the stem cell program at the National Institutes of Health, said, "No question, this raises important safety questions." But he said he could envision techniques that might remove the problem molecules.

"As with so many safety issues, there are a variety of clever solutions that could potentially be brought into play that could mitigate against this kind of problem," he said.

Former FDA commissioner Mark McClellan added that some patients have already been treated with tissues grown on mouse cells without rejection problems. But he conceded that anti-rejection drugs -- which doctors hope to avoid with new stem cell therapies -- may have been needed.

The second study, also summarized at the academy meeting, is comparing the characteristics of 14 of the 22 Bush-approved colonies. At least five of those colonies "will never be useful for the clinic" because they are so difficult to grow, said Carol Ware of the University of Washington, who led the study.

Moreover, she said in an interview, she has found that each colony has its own quirky propensity to turn into one kind of body cell or another, suggesting many more than the 22 colonies available will be needed if the field is to reach its full potential to treat a wide variety of failing organs.

Richard Doerflinger of the U.S. Conference of Catholic Bishops said the new findings could not justify expanding the research arena.

"It's throwing good money after bad," he said, "but here the cost is not in money but in nascent lives."