Scientist of landmark paper on stem cells urges retraction

The lead scientist of a landmark paper on cloning human stem cells has asked the journal Science to retract the paper, the editors said yesterday. South Korean scientist Hwang Woo Suk, who became a national hero on the strength of his research, made the request on Thursday night after two colleagues publicly charged that the paper included misleading data. In a press conference in Seoul yesterday, Hwang said that he would soon provide proof that his team had done the work they claimed, but that the team had made errors in the tests used in the paper to verify that they had cloned cells.

''After analyzing the data, our team concluded that the results . . . could not be trusted," Hwang wrote in a note delivered to Science. The note was read to journalists at a press conference yesterday.

After a week of mounting questions, the retraction was expected, but it nonetheless adds to the cloud of suspicion surrounding the South Korean group and its work. The paper Hwang intends to retract claims that the group found a way to efficiently create cloned human embryonic stem cells. It builds on another paper, published in 2004, which is considered the first time that scientists had ever created cloned human embryonic stem cells.

Once Science receives an explanation of the errors in the 2005 paper, its editors plan on scrutinizing the 2004 one. ''We are going to be giving that paper careful scrutiny, guided by what we can learn in the course of time about the 2005 one and what went wrong there," said Donald Kennedy, editor in chief of Science.

The retraction represents a setback for the field of embryonic stem cell research, though how much of one depends on how much of the South Korean group's claims turn out to be incorrect. Creating cloned embryonic stem cells will give scientists a unique way to study how some diseases develop. The South Korean team is the only one to have claimed to accomplish this feat, and other teams, including two at Harvard University, were hoping to build on the South Koreans' work. Now it is unclear how much the South Koreans really accomplished.

Kennedy said that he would urge Hwang and his coauthors to be as specific as possible in explaining the errors that led them to retract the paper.

Seoul National University, where Hwang works, is conducting an investigation of the case, as is the University of Pittsburgh, home to Hwang's American collaborator, Gerald Schatten.

Earlier in the week, Schatten delivered a letter to Science, requesting that his name be removed from the paper, and saying he suspected some of the paper's claims were misleading. Science responded that it does not remove a scientist from the list of authors. Schatten then joined Hwang in asking for a retraction, the editors said yesterday. Hwang is contacting all the paper's authors, at Science's request, to write a retraction.

The episode could also have serious implications for Schatten. For reasons that remain unclear, Schatten was listed as the senior author of the paper. This puzzled scientists, because all the work was done in South Korea, and the senior author is usually the person who oversees the research -- and vouches for it. Now Schatten will have to explain to university investigators when he learned of the problems with the paper and why, as the person credited with senior authorship, he was not aware of them earlier.

In addition to the 2005 paper on cloning, Schatten also worked with the team to create what they claimed was the first cloned dog.

The editors of Science said yesterday that they are not planning substantial changes in the way that they review manuscripts, and that they believe that the scientists who reviewed the manuscript did their job well. Scientific reviewers do not typically look for evidence of fraud unless there is some reason to be suspicious, they explained.

Hwang said at Seoul National University that five batches of frozen stem cells were in the process of being thawed for analysis, to verify that they were cloned stem cells. He expected results in 10 days.

Material from the Associated Press was used in this report. Gareth Cook can be reached at cook@globe.com.