A Japanese researcher falsified figures in three published papers while working as a visiting postdoc at the NIH's National Institute of Dental and Craniofacial Research (NIDCR), the Office of Research Integrity (ORI) reported last week.

Kazuhiro Tanaka, a cancer researcher formerly at Kyushu University in Japan, fidgeted with Western blots, Northern blots, and gel shift assay images by duplicating bands in the results of three papers published from 2000 to 2002.

The dodgy studies stem from work done from 1996 to 1998 when Tanaka was a visiting postdoc in Yoshihiko Yamada's lab at the NIDCR investigating transcriptional regulation of type II and type XI collagen genes in mouse and rat cells.

The fudged figures were included in one paper in the Journal of Biological Chemistry (JBC) in 2000 (with Tanaka as a middle author), which identified a cartilage-specific enhancer in the first intron of the collagen gene, Col11a2, and has been cited 43 times according to ISI, and two papers in Molecular and Cellular Biology (MCB) in 2000 and 2002 (with Tanaka as the first author), which characterized zinc finger factors that negatively regulate cartilage-specific expression of Col11a2 and another collagen gene, Col2a1, and have been cited 39 and 18 times, respectively.

In 2006, the papers' authors issued a correction for the problematic figure in the JBC study and retracted figures in both MCB papers, although the authors maintain that they stand by the basic results of the papers. "Some figures needed to be corrected, but overall the conclusions I still feel are correct," Yamada told The Scientist.

Yamada learned of the image manipulations after someone in his lab noticed that some of the figures looked "very strange," he said. When Yamada questioned Tanaka about the figures, he was told that a "personal friend" in Japan carried out the experiments. Although Yamada was given the friend's name, neither he nor the NIH investigators could locate her.

The NIH's Office of Intramural Research conducted an inquiry from January to June 2005 in which a committee interviewed Tanaka in person in the US with the help of a translator. "[Tanaka] claimed that somebody else had done all the fabrication," Joan Schwartz, the NIH's intramural research integrity officer, told The Scientist. "To be honest, we don't know that there ever was such a person." The inquiry committee tentatively concluded that Tanaka was guilty of misconduct, although they couldn't prove it, Schwartz said. "We ended the case at the point thinking we couldn't go any further."

The committee sent its findings to the ORI, which probed the datasets and concluded that the image manipulations were carried out during Tanaka's time at the NIDCR, not in Japan. The NIH then established a formal investigation from January to August 2007, at which time Tanaka returned to the US with a lawyer for another interview. "We finally concluded that, yes, it had been he who had committed most of the misconduct," said Schwartz.

Although certain figures have already been corrected or retracted, Schwartz said she is now working with Yamada to completely retract the two MCB papers, and to correct one more figure in the JBC paper. "Now that we have the [ORI's] final findings, I'm working with [Yamada] to sort out the final wording to send to the journals to say that the [MCB] papers should be retracted," she said.

Tanaka, who holds both a PhD and an MD, was working at Kyushu University together with Yukihide Iwamoto, and published papers as recently as January, but according to Yamada he has since moved to a private clinic in the Kumamoto Prefecture of Kyushu Island and is no longer conducting research. Tanaka and Iwamoto did not respond to email requests for interviews from The Scientist.

According to the ORI report, Tanaka acknowledged that original data relating to the falsified figures were missing, though he did not admit misconduct. As part of the settlement agreement, Tanaka is barred from performing research funded by US taxpayers until 2012.


Related stories:

Images faked by UCSF postdoc
[16th February 2009]

UCLA prof falsified cancer data
[9th February 2009]