The head of a Japanese committee investigating claims that stem cells could be made using mechanical stress or acid resigned from the committee today over anonymous allegations that at least one of his own papers contained problematic data. He says he resigned out of concern that the incident could complicate the current investigation.
Shunsuke Ishii, a molecular biologist at the RIKEN Advanced Science Institute in Tsukuba, was leading a team looking into two papers that reported a new kind of pluripotent stem cells — called STAP cells — in Nature on 30 January. The papers, whose first author was Haruko Obokata of the RIKEN Center for Developmental Biology in Kobe, were found to contain problems, including gel lanes that had been spliced together such that different lanes looked as though they were part of the same experimental procedure.
In a report released on 1 April, the committee deemed that two of the problems — the spliced gel lanes and the use of an image that had appeared in Obokata’s dissertation to represent a different experimental finding — constituted misconduct. Obokata appealed the judgment, and she has delivered supplementary materials in response to a request by the committee. Now the committee is deciding whether to reopen the case.
On 24 April, an eight-page document alleging problems in two papers Ishii co-authored in 2004 and 2008 was posted anonymously and widely circulated. The alleged problems included gel lanes that were spliced and set next to each other, without clear indication that splicing had taken place.
Ishii responded on the same day with a post on his laboratory’s homepage. The first paper in question is a 2008 study of a tumour suppressor and published in Oncogene (Oncogene is published by the Nature Publishing Group, which hosts this website; Nature’s news and comment team is editorially independent of Nature Publishing Group journals’ research editorial teams.) Ishii confirmed that the order of the gel lanes was changed to align with the explanation in the text and that errors were made when making that alignment. But he insists that the data were all part of the same experiment. In his post he includes copies from the notebooks showing the original data for the experiments. He also includes a correction that, he says, the editor of the journal has already accepted. “These changes do not affect any of the results or conclusions of our study,” reads the correction.
Today, Ishii supplemented the post with explanatory materials to address the problems raised about the second paper, a 2004 article in Journal of Biological Chemistry. He says that different gel lanes in that paper were juxtaposed for “space saving”, and he provides clearly marked original data. He says this presentation was “not a problem, according to the rules ten years ago”. He acknowledges, however, that “stricter rules now” require a clear line indicating when gel lanes are not in their original setting. He is not seeking a correction for this paper.
Some researchers have criticized Ishii, saying that if splicing of gel lanes counted as misconduct for Obokata, it should also count as misconduct in his case. Obokata’s lawyer cites the Ishii case to argue for a reinvestigation of her own.
One difference in Obokata’s case, however, is that the gel lanes were not only spliced, they were also stretched and rotated, showing a greater degree of manipulation. Also, according to the committee, Obokata’s laboratory notes lacked a clear indication of the path from original data to published data.
RIKEN says that it is now gathering information to decide whether to launch an investigation into Ishii’s research. This preliminary investigation should take about a week, according to a RIKEN spokesperson.
Today Ishii announced his resignation from the committee investigating Obokata’s papers. In a written statement, he says the “copies of laboratory notes and original experimental data [posted on the laboratory homepage] clearly show that no misconduct was involved” in the two papers.