Data highjinx forces retraction of tumor paper in JBC

jbccover414The Journal of Biological Chemistry has an illuminating retraction notice — we’re happy to be able to say — about a 2001 article from a group of researchers at the National University of Singapore.

The paper, “Intracellular acidification triggered by mitochondrial-derived hydrogen peroxide is an effector mechanism for drug-induced apoptosis in tumor cells,” was written by Jayshree L. Hirpara, Marie-Véronique Clément and Shazib Pervaiz.

It has has been cited 110 times, according to Thomson Scientific’s Web of Knowledge. According to the retraction notice:

Figs. 1 and 9C contain images that were taken from previous publications where they were described as representing the results of different experiments. Some of the data were labeled incorrectly, and permission for re-use of the data was not obtained from the copyright holder, although the articles in which the data in Figs. 1 and 9C were originally published were referenced in the text. Therefore, in accordance with accepted scientific practice, the authors have withdrawn the paper.

The paper had come to the attention of Abnormal Science back in 2011, when the blog reported that three articles by Pervaiz and Hirpara were under scrutiny for questionable findings. Pervaiz was a co-author on a paper that Nature Immunology retracted in 2011 — again for image issues — from a group led by Arilio Melendez, who once worked at the National University of Singapore.

9 thoughts on “Data highjinx forces retraction of tumor paper in JBC”

  1. 2 other Pervaiz papers were questioned on

    Cell Death Differ 2007;14;1617-27; PMID17510660
    Cell Death Differ 2010;17;408-20; PMID19834492

    Figure 5b of the 2007 paper is reproduced in Figure 1b of the 2010 paper. The publisher was notified of the problems in October 2012.

    In addition, the 2007 paper was subject to an erratum for use of an incorrect b-actin loading control. The erratum appears on the PubMed listing for the paper, but the journal home page for it does not contain a link to the erratum. This is a common problem in NPG journals – they don’t make retraction or correction notices easy to find. In this case you have to arrive at the site via PubMed to get any idea that there’s a problem.

    1. Incidentally, another similar case was observed in plant science, a few days ago.

      Pejman Azadi, Dong Poh Chin, Kiyo Kuroda, Raham Sher Khan, Masahiro Mii (2010) Macro elements in inoculation and co-cultivation medium strongly affect the efficiency of Agrobacterium-mediated transformation in Lilium. Plant Cell, Tissue and Organ Culture (PCTOC) Volume 101, Issue 2, pp 201-209

      Pejman Azadi, Ntui Valentaine Otang, Hasthanasombut Supaporn, Raham Sher Khan, Dong Poh Chin, Ikuo Nakamura, Masahiro Mii (2011) Increased resistance to cucumber mosaic virus (CMV) in Lilium transformed with a defective CMV replicase gene. Biotechnology Letters Volume 33, Issue 6, pp 1249-1255

      Fig. 2a of the 2011 paper appears to be the same as Fig. 1a of the 2010 paper, just cropped.

    2. Oh to be reminded of that darkly mysterious, anti-scientific time long, long ago before PPPR was officially sanctioned in our newly green and pleasant land where we are at last living happily ever after.

      It is said that the internet never forgets and so a personal thank you Retraction Watch for this reminder. For, as it happens, I saved one of the Pervaiz entries from the Abnormal Science blog because of the strange and not very scientific nature of some of the comments. Some soundbite-sized key fact chunks…

      Title of the AS blog entry:

      “Singapore: SMS alert #4 at NUS part 2
      Veröffentlicht am 14. November 2011”

      A 2007 paper being called out on the blog entry (same journal, NPG group, no doubt purely coincidental):

      “Example 5
Kumar AP, Chang MK, Fliegel L, Pervaiz S, Clément MV.
Oxidative repression of NHE1 gene expression involves iron-mediated caspase activity.
Cell Death Differ. 2007 Oct;14(10):1733-46.
National University Medical Institutes, Yong Loo Lin School of Medicine, National University of Singapore”

      My favourite, though grammatically unorthodox, comment:

      “17. November 2011 um 05:34
      hello Joerg Zwirner
dont fear about extradition. but things do go wrong on freeways in germany too!
 be careful. you are challenging an entire govt (which has german collaborations in science and technology)”

      From time to time, things are reported to go wrong on German autobahns. Even so, it is still not publicly known whether “things going wrong on German freeways” is official Singaporean government policy when faced with issues of scientific credibility. Conversely, and in the interests of fair and balanced reporting, it is also not known how much weight to place on the rumours that Dr. Joerg Zwirner laughed so much on receiving this threat that his health was seriously endangered.

    1. I have seen many nostalgic posts about the science fraud web-site. Surely there is a way of resurrecting it by substituting the title, ensuring that the word fraud is never used, to avoid libelous claims, and having hawkish moderators at the helm?

    2. I miss Science-fraud too.

      Still, one can be happy that PubPeer is widely known in the scientific community now. It will be interesting to see if PubPeer is also seven-times more effective than the “other channels”, as we now know was the case for Science-fraud.

      Although now I think about it, in the case of PubPeer, I am not sure that this can be meaningfully analysed as there is no comparable dataset for the “other channels”. Might a ROC curve be able to track the signal to noise if there is a boost in corrections/retractions in the next 1-2 years?

  2. Here is the methodology for Figure 1B

    Detection of Cytosolic Cytochrome c

    In a separate experiment, M14 cells (2 × 103) were grown on coverslips, exposed to C2 or C5, fixed with methanol:acetone (1:1, v/v), and incubated for 2 h at 37 °C with 1:150 dilution (in 3% bovine serum albumin) of anti-cytochrome c antibody (clone 7H8.2C12; PharMingen) as described previously (Ref 6). Following three washes with 1× phosphate-buffered saline + 1% fetal bovine serum, cells were exposed to 1:20 dilution of anti-mouse fluorescein isothiocyanate-conjugated IgG (Pharmingen, San Diego, CA) for 1 h, washed twice, and analyzed by confocal microscopy.

    The cited reference (Ref 6) is this:-

    Shazib Pervaiz, Mohamed A. Seyed, Jayshreekumari L. Hirpara, Marie-Véronique Clément, and Kok W. Loh
    Purified Photoproducts of Merocyanine 540 Trigger Cytochrome C Release and Caspase 8-Dependent Apoptosis in Human Leukemia and Melanoma Cells
    Blood 1999 93:4096-4108

    and here they show images that demonstrate that cytochrome c has exited the intermembrane space and entered the cytosol

    It appears that fixation of cells with methanol/acetone, at room temperature, allows antibodies to cross the cell membrane at will, but not cross the outer mitochondrial membrane.

    This is odd as one can measure mitochondrial creatine kinase, which sits in the inter membrane space in methanol/acetone fixed cells

    So how does methanol/acetone know that it must permeabilize the plasma membrane and not the outer mitochondrial membrane for these workers?

  3. Correction: LY303511 Enhances TRAIL Sensitivity of SHEP-1 Neuroblastoma
    Cells via Hydrogen Peroxide–Mediated Mitogen-Activated Protein Kinase Activation
    and Up-regulation of Death Receptors

    In this article (Cancer Res 2009;69:1941–50), which appeared in the March 1, 2009,
    issue of Cancer Research (1), there was an error in the histograms appearing in
    Fig. 6B and C. The authors have provided the correct histograms, along with an
    annotated version of Fig. 6, below.

Leave a Reply

Your email address will not be published. Required fields are marked *