Retraction Watch

Tracking retractions as a window into the scientific process

More on Hattori case from co-author: Did grudge lead to scientist’s fall?

without comments

We have an update on the case of Yoshiyuki Hattori, the Japanese endocrinologist who has had a half-dozen papers retracted because of issues involving reused data. We’ve reported on some of those retractions, and report on three new ones here.

As a trainee, Hattori spent some time in England, where he met Steven Gross, a prominent pharmacology researcher at Cornell. Gross was impressed with the young physician-scientist, and invited him back to his New York City laboratory to do a postdoc.

Gross’ name appears on one of the retracted articles, “NO suppresses while peroxynitrite sustains NF-κB: a paradigm to rationalize cytoprotective and cytotoxic actions attributed to NO,” which appeared in 2004 in the journal Cardiovascular Research and has been cited 42 times, according to Thomson Scientific’s Web of Knowledge.

According to the notice:

This article has been retracted by the journal due to the fact that it contains data that are identical to those shown in a previously published article:

In Fig. 2A, the photomicrographs of cells in (a) (“control”) and (b) (“LPS”) are the same as those shown in Fig. 1C (a) (“control”) and (b) (“TNFα”) of a previous article by this group[1] (except that the image in (a) is turned 180°). In Fig. 2B, the leftmost three bands (for “Cont”, “LPS”, and “+ NOR3”) are the same as those shown Fig. 1B (for “Cont”, “TNFα”, and “TNFα + competitor”) of the previous article[1] (except that the latter lanes are underexposed).

That’s more or less what another retraction notice in Cardiovascular Research says about a second Hattori paper, “Cilostazol inhibits cytokine-induced nuclear factor-kB activation via AMP-activated protein kinase activation in vascular endothelial cells,” published in 2009 and cited 16 times:

This article has been retracted by the journal due to the fact that it contains data that are identical to those shown in a previously published article:

In Fig. 4A, the bands for phospho-IkBa (“Time after TNFa”) are the same as those shown for phospho-IkBa (“Time after LPS”) in Fig. 6A of a previous article by this group.1 In Fig. 4B, the bands for IkBa (“Cilostazol”) are the same as the leftmost four bands shown for IkBa in Fig. 6A

That referenced article: the 2004 paper by Hattori and Gross (update: the article was written long after Hattori had left Gross’ lab, in the mid-1990s).

Hattori also retracted “Statin enhances cytokine-mediated induction of nitric oxide synthesis in vascular smooth muscle cells,” a 2002 Cardiovascular Research paper that has been cited 53 times. The notice — we should note that the original study abstracts all say a “correction” has been published, not a retraction:

This article has been retracted by the journal due to it being highly redundant in both data and text with another publication.1

The citation is for a 2002 Hattori-co-authored paper in Pteridines, “3-Hydroxy-3-methylglutaryl-CoA reductase inhibitors enhance cytokine-mediated induction of nitric oxide synthesis in vascular smooth muscle cells.”

Although it’s hard to tell exactly what was going on here — we’ve attempted to contact the editor of the journal to find out more — Hattori pretty clearly 1) reused data, in the form of images and 2) also seems to have manipulated them.

But according to Gross, who helped Hattori write their manuscript but did not control the data, his co-author was not exercising fraudulent intent.

It was a mistake on his part, it was an error in judgment. As far as I understand, the science is all perfect. The issue was that there was sharing of data between the papers, and that’s a no-no. The right way would have been to make a reference to it but he chose to show it. I thought that was rather ridiculous [to issue a retraction] for the shared data.

Gross had more to say about the case, some of it quite disturbing. Hattori appears to have been up for a major promotion at his institution, Dokkyo University School of Medicine. According to Gross, a rival found the prospect of his advancement so upsetting that he or she scoured Hattori’s publications for mistakes and other weaknesses, then sent anonymous letters to journal editors (including Cardiovascular Research, we presume) about the findings.

While we haven’t been able to confirm Gross’ account, the part about an anonymous antagonist does jibe circumstantially with what we were told by Dennis Vance, editor of Biochimica et Biophysica Acta, which earlier this year retracted one of Hattori’s papers. Vance said “a reader” had alerted him and another editor to the dodgy figures.

Hattori, Gross added, has been devastated by the retractions and will be leaving academic life to practice medicine in the countryside.

 He’s leaving his institution in shame. It hurts me to see him go down in flames.

If Gross is wrong about his former post-doc, and Hattori was guilty of outright fraud, it’s hard to feel all that sorry for him. However, if Gross is right, and Hattori made honest mistakes — repeatedly, it seems — then he deserves some pity. The question is whether retraction is the best outcome in cases like these in which the science is solid but some data are recycled.

Comments
  • QStel September 26, 2011 at 3:44 pm

    Clearly this is not a honest mistake. See the retraction notice, which reads: “In Fig. 2A, the photomicrographs of cells in (a) (“control”) and (b) (“LPS”) are the same as those shown in Fig. 1C (a) (“control”) and (b) (“TNFα”) of a previous article by this group”. If data would be reused in this case, then also the labelling of the photomicrographs should have been identical.

  • Neuroskeptic September 27, 2011 at 10:18 am

    But in the one WHY links to, it fits with Gross’s account. They’re both pictures of the same two conditions CONT and LPS/IFN, and are labelled as such in both cases.

    So it fits with it being correct data, that was reused in two papers.

    However the one QStel points to is more worrying since it appears that the same data was labelled as “LPS” in one paper and “TNFalpha” in another.

    And again in another notice

    “In Fig. 4A, the bands for phospho-IkBa (“Time after TNFa”) are the same as those shown for phospho-IkBa (“Time after LPS”) in Fig. 6A of a previous article by this group.”

  • conradseitz September 27, 2011 at 1:46 pm

    So Hattori has gone to practice medicine in the countryside? I hope he has the appropriate “training” for that! heheheh. (From one who was there but fortunately didn’t have to have his papers retracted to convince him to leave for the country.)

  • canoinve December 20, 2011 at 4:27 am

    New retraction in diabetologia.

    Diabetologia. 2011 Dec 15. [Epub ahead of print]
    Retraction Note: A glucagon-like peptide-1 (GLP-1) analogue, liraglutide, upregulates nitric oxide production and exerts anti-inflammatory action in endothelial cells.
    Hattori Y, Jojima T, Tomizawa A, Satoh H, Hattori S, Kasai K, Hayashi T.

    http://www.ncbi.nlm.nih.gov/pubmed/22170464

  • Post a comment

    Threaded commenting powered by interconnect/it code.