* Stem cell studies were initially hailed as breakthrough
* But independent teams could not replicate findings
* Nature says "multiple errors" impair work's credibility (Adds statement from Harvard researcher Charles Vacanti)
By Kate Kelland, Health and Science Correspondent
LONDON, July 2 A stem cell paper published by a team of Japanese and American scientists in the influential journal Nature has been retracted due to "several critical errors", the journal said on Wednesday.
The research, which when published in January was described as game-changing by many experts in the field, was subsequently investigated by Japan's RIKEN scientific institute, which "categorised some of the errors as misconduct", Nature said.
The paper, led at RIKEN by Japanese researcher Haruko Obokata, detailed simple ways to reprogram mature animal cells back to an embryonic-like state, allowing them to generate many different types of cells.
The results appeared to offer a promise that human cells might in future be simply and cheaply reprogrammed into embryonic-like cells - in this case cells dubbed Stimulus-Triggered Acquisition of Pluripotency, or STAP, cells - suggesting a simple way to replace damaged cells or grow new organs for sick and injured people.
Obokata initially staunchly defended her work in the face of serious doubt and criticisms, but last month agreed to retract the papers.
Nature confirmed the retraction on Wednesday, saying "multiple errors impair the credibility of the study as a whole.
"Ongoing studies are investigating this phenomenon afresh, but given the extensive nature of the errors currently found, we consider it appropriate to retract both papers," it said.
Charles Vacanti of Boston's Brigham and Women's Hospital and Harvard University in the United States, who was a co-author on the paper, conceded in a statement that "multiple errors" had been found, and agreed the work should be withdrawn.
"In science, the integrity of data is the foundation for credible findings," he said. "I am deeply saddened by all that has transpired, and after thoughtful consideration of the errors presented in the RIKEN report and other concerns that have been raised, I have agreed to retract the papers."
Stem cell experts, who at the time expressed excitement about Obokata's findings, said they were disappointed at the outcome.
"The STAP technology, indeed, sounded too good to be true," said Dusko Ilic, senior lecturer in stem cell science at King's College London. "I hoped that Haruko Obokata would prove at the end all those naysayers wrong. Unfortunately, she did not."
According to the original Nature papers and briefings given by Obokata in January, she and her team took skin and blood cells, let them multiply, then subjected them to stress "almost to the point of death" by exposing them to trauma, low oxygen levels and acidic environments.
The Japanese-U.S. team said that within days they found the cells had not only survived but had also recovered by naturally reverting to a state similar to that of an embryonic stem cell.
These stem cells were then able to differentiate and mature into different types of cells and tissues, they said.
But other research teams around the world were unable to replicate the research, and within weeks questions were raised about the validity of the findings.
RIKEN, a semi-governmental research institute, launched an investigation and in April said it had concluded "this was an act of research misconduct involving fabrication".
Chris Mason, an expert in regenerative medicine bioprocessing at University College London, said the incident exposed flaws in the peer-review process whereby fellow scientists are asked to review and comment on scientific studies before journals accept them for publication.
"This ... highlights that the peer review process does not end at the recommendation to publish a paper, but continues with even greater rigour by a wide range of experts in their laboratories and increasingly across social media," he said.
"Final validation is the reproduction of the data by independent scientists. This final step is the most important step in the entire peer review process." (Editing by Janet Lawrence and Sonya Hepinstall)