“Mixed up” images earn biologists four retractions

Four different journals have pulled papers from the same authors due to alleged duplication or manipulation of images.

All four papers have two authors in common — Jianting Miao and Wei Zhang, both based at The Fourth Military Medical University in Xi’an City, Shaanxi, China. Many of the other co-authors are also listed in two or three of the retracted papers.

Miao claims that the photographs got “mixed up” due to the researchers’ “great carelessness” and “insufficient knowledge.” He told us:                                                                                                                                                                      

When we were informed by the journal editors that there are some inappropriate duplications of photographs in our publications, we carefully re-checked all the courses of the different experiments. We have several research groups respectively focused on the underlying mechanisms involved in the cognitive deficits of Alzheimer’s disease. During the above research work, a large number of similar photographs from different research groups were obtained, and all the photographs (i.e., GFAP-stained astrocytes, ThS-stained fluorescent plaques, and Nissl-stained neurons) from different research groups were saved in the same computer connected with microscope camera.

He added:

Unfortunately, these images obtained from different research groups were not clearly labeled by the researchers. Due to the author’s great carelessness and insufficient knowledge, when selecting representative images for different research papers, the authors inadvertently mixed up some photographs obtained from differently treated animals. Thus, the incorrect images with overlaps were inadvertently used in different research papers by mistake.

Miao noted that it was “unfortunate” that the corresponding authors of the papers didn’t notice these issues before submitting the papers:

We sincerely apologize to the editors and readers of the journal for the trouble we caused and express our regret for any inconvenience this has caused. We have learned much more lessons from these events. We will carefully supervise and inspect everything of the experiment to prevent the mistakes in the future.

Let’s take a look at the retracted papers and their retraction notices.

First, a 2012 paper published in Neurobiology of Aging, “Multiple inflammatory pathways are involved in the development and progression of cognitive deficits in APPswe/PS1dE9 mice,” which has been cited 20 times, according to Thomson Reuters Web of Science.

Here’s its retraction note:

This article has been retracted at the request of the Editor-in-Chief and the authors due to inappropriate duplication of photomicrographs and errors in the description of the material. The authors’ account is that the identity of material was confused during preparation of the report and that the errors were inadvertent. Specifically, Figure 2F duplicates Figure 2B in Zhang et al. (2012, Pharmacol. Biochem. Behav., 100:361e69; http://dx.doi.org/10.1016/j.pbb.2011.09.012). Figures 4A and 4B are the same histological preparation, but are reported to be from different animals and experimental conditions, and both are duplicates of Figure 5A in Zhang et al., 2012 (Pharmacol. Biochem. Behav., 100:361e69; http://dx.doi.org/10.1016/j.pbb.2011.09.012). Figure 4C is a duplicate of Figure 5B in Zhang et al., 2012 (Pharmacol. Biochem. Behav., 100:361e69; http://dx.doi.org/10.1016/j.pbb.2011.09.012). The associated quantitative results reported in the paper cannot be considered reliable.

The Editors of Neurobiology of Aging and the scientific community take a very strong view on this matter and apologies are offered to readers of the journal that these issues were not detected earlier.

Peter Rapp, editor-in-chief of Neurobiology of Aging, who is based at the Biomedical Research Center in Baltimore, Maryland, told us:

…we were altered to problems in the paper by a reader

Rapp added:

…the reader outlined the issues detailed in the retraction notice, the authors were notified and they confirmed the accuracy of the concerns

He noted that the journal did not determine whether the problems with the article were “deliberate misrepresentations:”

…the authors maintain the errors were inadvertent.

Next, the British Journal of Pharmacology has pulled a 2014 paper, “Rutin protects against cognitive deficits and brain damage in rats with chronic cerebral hypoperfusion,” which has garnered seven citations so far.

Its retraction notice reads:

The above article, published online on 17th July 2014 in Wiley Online Library (http://onlinelibrary.wiley.com/doi/10.1111/bph.12725/full), has been retracted by agreement between the authors, the journal Editor in Chief, Professor J. McGrath and John Wiley & Sons Limited. The retraction has been agreed owing to duplication of some images. In this paper, a series of adjacent sections from each group were stained using immunohistochemical and histological methods, and the authors obtained a large number of photographs from each group. When selecting representative images from each group, some photographs obtained from differently treated groups were inadvertently mixed up. Incorrect photographs were used in this paper as follows: (1) the photographs shown in Figure 7A and B are overlapping images from the same section; (2) the photographs shown in Figure 10a and c, and in Figure 10b and d are overlapping images from the same tissue slices. In addition, because the authors failed to identify correctly the damaged neurons with Nissl staining, the same neurons that are intact in the sham-operated group are reported to be damaged in the BCCAO group with treatment. Thus, this may have caused the incorrect number of the intact neurons obtained from each group.

We discovered this retraction a few months ago, after it was released in August, 2015; at the time, we contacted the journal’s then-editor, Ian McGrath, who told us an alert scientist wrote the British Journal of Pharmacology with concerns about the images in the paper. He noted in 2015:

Our policy is to investigate thoroughly — according to the Publisher’s Ethics policy, which are based on COPE guidelines — all ethical concerns brought to our attention. We also believe that we should publish a fully transparent and straightforward account of the events leading to any retraction, and where possible to agree the retraction statement with the authors in question.  I believe, and I hope you will agree, that we have done so here.

More recently, we contacted current editor-in-chief Amrita Alhuwalia, from the Queen Mary University of London, who concurred with McGrath’s account:

The final opinion was that there had been a number of mistakes that compromised data analysis.

Next, a 2008 paper published in Neuropeptides, “S14G-Humanin ameliorates Aβ25-35-induced behavioral deficits by reducing neuroinflammatory responses and apoptosis in mice,” which has been cited 37 times so far.

Here’s its retraction notice:

This article has been retracted: please see Elsevier Policy on Article Withdrawal (http://www.elsevier.com/locate/withdrawalpolicy).This article has been retracted at the request of the Editor and the Authors. There are some inappropriate manipulations or alterations of photographs in this publication. Specifically, in Figure 4, Panels A and B show identical brain sections, 90 degrees rotated. According to the legend, panel A is from a non-treated animal, whereas panel B is from an HNG treated animal. In Figure 5, Panels A and B show nearly identical brain sections, mirrored. A structure present in panel A is absent from panel B. Both panels contain repetitions that could result from deliberate manipulation. According the legend, panel A is from a non-treated animal, whereas panel B is from an HNG treated animal. In Figure 7, panels A, B, and D show nearly identical brain sections. All three panels contain repetitions that could result from deliberate manipulation. According the legend, panel A is from a non-treated animal, whereas panel B is from an HNG treated animal, and panel D is from an Aβ/HNG treated animal.The Editors of Neuropeptides and the scientific community take a very strong view on this matter and apologies are offered to readers of the journal. Unfortunately, issues of deliberate manipulation of this nature, despite our rigorous review process, are sometimes difficult to detect.

We contacted Elsevier about this one since the notice alleges “deliberate manipulation” on the part of the authors, but they referred us back to the retraction notice. We also followed up with Miao, who added:

We did not see the retraction text about this paper from the journal editor before it was published. After carefully re-checking all the courses of the experiments, we confirmed that the incorrect images used in this paper were inadvertently made by mistake.

Finally, a 2014 paper published in Neurobiology of Learning and Memory, “Neuronal damage, central cholinergic dysfunction and oxidative damage correlate with cognitive deficits in rats with chronic cerebral hypoperfusion,” which so far has 22 citations to its name.

Here is its retraction notice:

This article has been retracted at the request of the Authors. There are some inappropriate duplications of photographs in this publication. When preparing images for different research papers, the authors maintain that they inadvertently mixed up some photographs obtained from differently treated animals. Thus, incorrect images with overlaps with other published research papers were used in this article by mistake, according to the authors. Specifically, Figure 3 panel A shows overlap with Figure 10 panels B and D in a previous paper publisher by the authors (Qu et al., Br J Pharmacol. 2014 Aug;171(15):3702–3715; http://dx.doi.org/10.1111/bph.12725); Figure 4 panel B shows overlap with Figure 5 panel D in their previous paper (Pharmacol Biochem Behav. 2012 Jan;100(3):361–369; http://dx.doi.org/10.1016/j.pbb.2011.09.012); Figure 4 panel C shows overlap with Figure 5 panel B in their previous paper (Pharmacol Biochem Behav. 2012 Jan;100(3):361–369; http://dx.doi.org/10.1016/j.pbb.2011.09.012); Figure 4 panel C shows overlap with Figure 4 panel C in their previous paper (Zhang et al., Neurobiol Aging. 2012 Nov;33(11):2661–2677; http://dx.doi.org/10.1016/j.neurobiolaging.2011.12.023); Figure 4 panel C shows overlap with Figure 7 panels A, B, and D in their previous paper (Qu et al., Br J Pharmacol. 2014 Aug;171(15):3702–3715; http://dx.doi.org/10.1111/bph.12725).

We’ve reached out to the editors of Neuropeptides and Neurobiology of Learning and Memory, and will update the post with anything else we learn.

With reporting from Emily Willingham

Hat tip: Rolf Degen

Like Retraction Watch? Consider making a tax-deductible contribution to support our growth. You can also follow us on Twitter, like us on Facebook, add us to your RSS reader, sign up on our homepage for an email every time there’s a new post, or subscribe to our new daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.