JBC cancer paper felled by duplication is one author’s second retraction this month

A 2002 paper in the Journal of Biological Chemistry on how lung cancer cells resist death has been retracted for duplicating figures from a 2001 paper.

The retracted paper, “Fibroblast growth factor-2 induces translational regulation of Bcl-XL and Bcl-2 via a MEK-dependent pathway: correlation with resistance to etoposide-induced apoptosis,” shares the first and last authors with the 2001 paper, in Oncogene, as well as two other co-authors.

Here’s JBC’s entire retraction note, a sub-genre with which we’ve become intimately familiar by now:

This article has been withdrawn by the authors.

But that’s apparently not the note the authors wanted to provide.

First, let’s take a look at a correction note to the 2001 Oncogene paper, published last year. In full:

We have discovered and would like to unreservedly apologise for some errors in a paper we submitted to J Biol Chem in 2001 (published in April 2002, 277(14): 12040–6) that relate to our Oncogene paper that was published in Sep 2001 (20(52): 7658–67). Most of these errors concern duplication of small amounts of negative control data that appeared in the Oncogene manuscript and that were reused in the J Biol Chem paper without proper attribution to the original publication in Oncogene. Thus, Figures 6A, B and 3A (right panel) from our Oncogene paper were duplicated as Figures 2B, C and 4A (upper panel) in the J Biol Chem paper, respectively. These data show that FGF-2 does not activate PI3K/PKB signalling in H-510 cells or Erk1/2 in H-69 cells. To rectify this, we have retrospectively asked permission from Oncogene to reproduce Figures 6A and B in J Biol Chem and have provided a fresh replacement experiment to J Biol Chem for Figure 4A (upper panel). In addition, we noted that the wrong lamin B loading controls were used in Figure 3A and the lower panel of Figure 4A of the J Biol Chem paper and appeared to have been incorrectly duplicated from separate experiments presented in the Oncogene paper. We have therefore repeated these experiments and presented the replacement figures to J Biol Chem. Importantly, the results confirm our original findings. Moreover, none of the errors have any impact on the interpretation of the data or conclusions presented in our original Oncogene manuscript.

So, interestingly, the correction note in Oncogene talks a lot about the issues in the JBC paper, and implies they can be resolved by a correction, not retraction.

Last author Michael Seckl, at Imperial College London, explained more in an email to Retraction Watch:

JBC were originally going to consider a correction note but on reflection decided that they wanted a retraction.

So he and his co-authors tried to assemble a note, to explain what happened:

We noticed a problem with duplication of negative control data between our Oncogene 2001 and JBC 2002 paper in addition to loading control errors in the JBC paper. We reported this to both journals. Both considered the matter. Whilst Oncogene accepted a correction note, in the end, JBC decided that they wanted a retraction which we complied with providing the following attached statement:

‘We would be most grateful if JBC would retract our paper (Pardo OE, Arcaro A, Salerno G, Raguz S, Downward J, Seckl MJ. (2002) Fibroblast growth factor-2 induces translational regulation of Bcl-XL and Bcl-2 via a MEK-dependent pathway: correlation with resistance to etoposide-induced apoptosis. J Biol Chem. 277, 12040-6.). We feel this is required because we have discovered duplication of control blots used in a prior paper that appeared in Oncogene (Pardo OE, Arcaro A, Salerno G, Tetley TD, Valovka T, Gout I, Seckl MJ. (2001) Novel cross talk between MEK and S6K2 in FGF-2 induced proliferation of SCLC cells. Oncogene 20(52):7658-67) as well as additional loading control errors. Although these do not invalidate the novel data presented in the JBC paper, we feel that the safest course of action is to withdraw the paper.’

However, we were informed that it is standard practice for JBC not to print any explanation and so the above text did not appear with the retraction note.

The statement to us was signed: “Michael Seckl on behalf of all the JBC co-authors.”

We contacted Kaoru Sakabe, Manager of Publishing Issues at JBC, to get more information, who told us:

JBC policy is to maintain confidentiality in such matters.

The Oncogene paper has four authors in common with the retracted JBC paper: Pardo is the first author on both, and Arcaro, Salerno and Seckl are co-authors on both.

The JBC paper — which has been cited 117 times, according to Thomson Scientific’s Web of Knowledge — sparked a discussion on PubPeer about figures and the correction note in Oncogene.

There’s more that’s interesting here: the second-to-last author on the JBC paper, Julian Downward, had to retract a Nature paper nine days prior due to misconduct from the first author, who was not involved in the JBC paper.

Downward told us he was not “closely involved in this:”

…I was a co-author on the paper only for providing reagents and no members of my lab were on the author list – but I do know a bit about it. You should probably get in touch with Michael Seckl, the senior and corresponding author on the article. However, I copy below the original text that the authors wanted JBC to print, but apparently it is not journal policy to explain retractions. As you will see, the problem is that a control blot panel appeared in the JBC paper that was previously published in the Oncogene paper. I do not know the detail of how this happened or who was responsible. I believe that the authors originally felt that JBC would accept a correction to the paper, but the editor decided it should be retracted. For further detail I would recommend you get in touch with Dr Seckl.

We also heard back from first author Pardo:

The reason why JBC decided to proceed to retraction are due to the duplication of data between our Oncogene and JBC papers (control experiments that only aimed at pointing out the lack of ERK signalling in H69 cells and the absence of PI3K signalling downstream of FGF-2 stimulation). Indeed, the conclusions of the JBC manuscript are still fully valid and have been not only further investigated by ourselves (see our own publications and those of collaborating teams) but by others. Hence , we regret this turn of event but accept the decision made by JBC.

Like Retraction Watch? Consider supporting our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, and sign up on our homepage for an email every time there’s a new post.