A Japanese research center’s investigation of two controversial stem cell papers has found “significant discrepancies” that its president on Friday termed “extremely regrettable,” and three coauthors have agreed to consider retracting the work based on interim results of the ongoing review.
Earlier this week, a fourth Japanese researcher, a prominent scientist and senior author of one of the two papers, called for them to be withdrawn from the journal Nature. But the Boston scientist who was the senior author of the other paper said he continues to stand by the core findings.
The research, coauthored by Boston and Japanese scientists, reported a shockingly simple way to make stem cells from mature mouse blood cells, by bathing them in a weak acid. If the work is verified by other scientists and reproduced in humans, it could make it easier for scientists to produce stem cells, which have the capacity to become any of the numerous cell types in the body and are seen as providing potential therapies for a range of conditions from diabetes to heart failure to paralysis.
Since the papers were published in January, however, numerous allegations have emerged in online forums about possible problems with images and alleged plagiarism.
Officials of the RIKEN institute disclosed during a four-hour press conference in Tokyo that they had found two cases of inappropriate handling of data so far and that four other issues are still being examined. No evidence of outright fraud has been discovered.
“It is extremely regrettable that significant discrepancies have been found to have been generated in the process of preparing the Nature articles for publication,” Ryoji Noyori, a Nobel laureate who is the president of RIKEN, said in a statement. “We are investigating these discrepancies, with the understanding that it may become necessary to demand the withdrawal of the articles.”
Separately, Dr. Rudolf Jaenisch, a stem cell scientist at the Whitehead Institute, an MIT-affiliated biomedical research center in Cambridge, said that his own lab had tried and failed to create STAP cells, the name given the type of stem cells reportedly made in the challenged experiments, which the authors said behave similarly to embryonic stem cells.
Jaenisch said it was very concerning that Teruhiko Wakayama , a coauthor and professor at the University of Yamanashi in Japan, called for the papers to be withdrawn earlier this week.
“He is one of the key authors on both papers, and he must feel very uncomfortable about this,” Jaenisch said. “The investigation is still going on, and I assume RIKEN considers it very serious to do. I think we can’t make a final conclusion, but it doesn’t look very good.”
A spokeswoman for RIKEN said by e-mail that three scientists there have agreed to consider a retraction: Yoshiki Sasai, Hitoshi Niwa, and Haruko Obokata, who was the lead researcher. But, the spokeswoman added, the papers will not be retracted unless all the numerous coauthors agree.
Dr. Charles Vacanti of Brigham and Women’s Hospital, senior author of one paper, said in a prepared statement Friday that while he awaits the outcome of the complete RIKEN investigation, he still believes the results are sound.
“I continue to feel that the findings presented in these papers are too significant to disregard based on relatively minor errors or external pressures,” Vacanti said. “In the absence of compelling evidence that the data presented is incorrect, I do not believe that the manuscripts should be retracted.” He said he would discuss the issue with his coauthors before making a final decision.
“I firmly believe that the most appropriate course of action at this time is to clarify, in a very specific manner, all of the subtleties associated with the creation of STAP cells by posting specific details of our most effective protocol on our laboratory website.”
In an e-mail, Vacanti clarified his role in the research. He said the initial idea for the stem cell technique emerged from work he and his brother, Martin, had been developing for years. He extended that research in one of the two recent papers, working at the Brigham with Dr. Koji Kojima and Obokata, who worked in Vacanti’s laboratory.
“In regards to these most recent publications, Dr. Kojima and I developed the initial experiments with Dr. Obokata and then met regularly with her to develop, design, review, and discuss the results of many of the experiments reported in the first paper,” Vacanti wrote. “Later experiments and revisions to the manuscripts were, to a large degree, designed and overseen by our collaborators at the RIKEN, with intermittent updates given to us from Japan.”
RIKEN scientists said at the press conference that they did not know of any independent scientists who had successfully replicated the experiments, according to the Wall Street Journal. A Hong Kong scientist posted a report Thursday of his failed efforts to repeat the experiment on a website called ResearchGate. Such failures can be difficult to interpret because initial difficulty repeating a new technique is not rare.
RIKEN officials gave a detailed list of issues that have been brought up with the papers. The institution has concluded its investigation of two items: an image of colored cell parts that had an “unnatural appearance” and two images that appeared strikingly similar.
The colored cell image was not found to be research misconduct. The possibly duplicated figure was not referred to in the captions or in the text, “but there was nothing to contradict the explanation that one of the figures had inadvertently been left undeleted during the process of manuscript creation,” the RIKEN report said.
The four remaining issues include possible image manipulation, possible plagiarism of part of the methods section from a different paper, a description of how something was done that is different from the procedure actually used, and use of images that are similar to ones in Obokata’s doctoral thesis.
RIKEN said in its statement that it would issue a final report when the investigation is completed.Carolyn Y. Johnson can be reached at email@example.com. Follow her on Twitter @carolynyjohnson.