A journal has retracted six papers by a cancer researcher at the University of Maryland in Baltimore, bringing his total to 10.
The retractions cite an investigation by the university, and detail problems ranging from duplicated images, to tweaking an image to conceal particular bands, to including unreliable data.
Three of the papers had already been flagged by the journal with expressions of concern. The last author on all the papers is Anil Jaiswal, a professor in the pharmacology department. He has issued four previous retractions.
Bruce Jarrell, the Chief Academic and Research Officer and Senior Vice President at the University of Maryland, told us at least two more retractions are forthcoming:
The outcome of [the investigation] is confidential. But there was a serious investigation.
Jarrell added the investigation was prompted by a complaint (he wouldn’t say from who). Jaiswal is still on the faculty at the university, he noted.
Kaoru Sakabe, the data integrity manager at the American Society for Biochemistry and Molecular Biology, which publishes the journal, told us:
In some instances, we were alerted to problems by readers. The University did contact us and provide us details regarding their investigation as stated in the retraction notices.
She added that all the papers were retracted by the publisher:
At the JBC, articles are retracted by the publisher when the authors do not agree to withdraw their article. I do, however, want to point out that one author did agree with the retraction.
We contacted Jaiswal, but haven’t heard anything back.
Here’s the first notice:
This article has been retracted by the publisher. An investigation at the University of Maryland, Baltimore determined that the data shown in Fig. 2A are unreliable and do not support the hypothesis of this work.
“NAD(P) H:quinone Oxidoreductase 1 (NQO1) Competes with 20S Proteasome for Binding with C/EBP alpha Leading to Its Stabilization and Protection against Radiation-induced Myeloproliferative Disease” has been cited five times since it was published in 2012, according to Clarivate Analytics’ Web of Science, formerly part of Thomson Reuters.
Here’s another notice (a doozy), which states the second-to-last author agrees with the retraction:
This article has been retracted by the publisher. Phillip M. Shelton agrees with the retraction. An investigation at the University of Maryland, Baltimore determined that bands depicted in several figures were digitally altered as follows.
In Fig. 2, the LaminB immunoblot from “Src siRNA” was digitally mirrored and duplicated as LDH in the “YES siRNA” panel. Bands 4, 5, and 6 in the LDH panels of the Fyn and Lyn panels were duplicated.
In Fig. 4A, in the Nrf2-V5 panel, bands 1 and 2 and bands 4 and 5 are identical. Similarly, in the IP:pTyr lane, bands 5 and 6 are identical. In Fig 4B, there is evidence of splicing in the Nrf2-V5 panel. In addition, in the FLAG panel, bands 1 and 2 and bands 4 and 7 are identical.
In Fig. 5A, there is evidence of splicing in the Nrf2 panel. In addition, bands 1 and 2 in the Src panel are identical. In Fig. 5C, in the Nrf2 panel, lanes 1 and 2 and lanes 5 and 8are identical. Similarly, in the Fyn panel, lanes 1 and 2 are identical. In addition, lanes 4and 5 were duplicated and presented as mirror images of each other. In Fig. 5E, lanes 4and 5 from the IP:V5 WB:pTyr panel were duplicated in Fig. 7A, right panel LDH, lanes 4and 5.
In Fig. 7A, the left Lyn panel and the right Yes panel were duplicated after resizing. In Fig. 7B, the left and right GSKβ panels were duplicated after resizing.
In Fig. 8A, in the NQO1 panel, bands 2 and 6 are identical. In the HO-1 panel, bands 2 and 6 were resized and duplicated. In the GCLC panel, bands 1 and 6 were duplicated.
In addition, the journal determined that lanes 2–4 of the HO-1 immunoblot from the “Yes siRNA” panel from Fig. 1A was flipped horizontally and reused in lanes 2–4 of the Lyn immunoblot from the “Lyn siRNA” panel of the same figure. In Fig. 2, lanes 3 and 6of the Src immunoblot from the “Src siRNA” panel were duplicated. In Fig. 5D, lanes 2and 5 of the IP:V5 WB:pSer immunoblot from the right panels were duplicated. In Fig. 6D, lanes 1–4 of the Lamin B immunoblot from SYF+/+ MEFs were duplicated in the LDH immunoblot from Fig. 7B, right panel. The LDH immunoblot from Fig. 7A, left panel, was duplicated as LDH in Fig. 7B, left panel. The right Fyn panel and the right Src panel were duplicated in Fig. 7A. In Fig. 7B, lanes 3 and 5 of the GSKβpY216 immunoblot, right panel, were duplicated.
“Src subfamily kinases regulate nuclear export and degradation of transcription factor Nrf2 to switch off Nrf2-mediated antioxidant activation of cytoprotective gene expression,” published in 2011, had already been flagged with an expression of concern. It’s been cited 51 times.
Another notice details multiple duplications in a paper that had also been under an EOC:
This article has been retracted by the publisher. An investigation at the University of Maryland, Baltimore determined that there is evidence of duplication in Fig. 2B. The investigation also determined that in Fig. 3, data in Cul3, INrf2, and Rbx1 show clear evidence of manipulations in which individual blots were spliced and rearranged. In Fig. 5A, duplication of bands is apparent. In addition, the journal determined that in Fig. 1B, left panel, lanes 6 and 9 were duplicated. The Lamin B immunoblot from Fig. 5D was reused as Lamin B in Fig. 5E. Also in Fig. 5E, the INrf2 immunoblot from 25 nM and 50 nM PTMα siRNA were duplicated. In Fig. 9B, lanes 5 and 6 of the PTMα immunoblot were duplicated.
“Prothymosin-alpha mediates nuclear import of the INrf2/Cul3Rbx1 complex to degrade nuclear Nrf2” has been cited 44 times since it was published in 2009.
A third paper that had received an EOC, “Phosphorylation and dephosphorylation of tyrosine 141 regulate stability and degradation of INrf2. A novel mechanism in Nrf2 activation,” has also been retracted, with the journal again citing multiple duplications:
This article has been retracted by the publisher. In Fig. 3C, cytosol panels, lanes 5 and 6 were duplicated in the INrf2-V5 immunoblot. Additionally, lanes 5 and 6 were duplicated in the INrf2-Y141A-V5 immunoblot. In Fig. 3C, nucleus panels, lanes 1 and 2of the INrf2-V5 immunoblot were duplicated. Additionally, the Lamin B immunoblot was duplicated in the top and bottom panels.
The 2008 paper has been cited 38 times.
The fifth notice details more image problems:
This article has been retracted by the publisher. An investigation at the University of Maryland, Baltimore determined in both the submitted but unpublished and in the published versions of Fig. 2, empty lanes were purported to depict absence of labeling in pcDNA. Examining the unedited version shows that the image was obtained from a region outside of the gel, where no labeling can occur. The figure does not support that there is no labeling after transfection with the empty pcDNA vector and the claim that transfection with the active constructs had a specific effect. The investigation also determined that in Fig. 6, in the versions submitted to EMBO, JBC (first submission), and to Mol. Cell. Biol., images of transfected Hepa-1 cells were obtained from a single sample, captured from a single microscope field, and then presented as if they originated from different samples. Whereas in the originally submitted versions, the treatment is described as “LMB,” in the published version, the same data are purported to show effects of PP2.
“Phosphorylation of tyrosine 568 controls nuclear export of Nrf2” has been cited 102 times since it was published in 2006.
Finally, this notice says the authors had tweaked an image to conceal particular bands:
This article has been retracted by the publisher. An investigation at the University of Maryland, Baltimore determined that contrast and brightness were enhanced to selectively conceal LDH and Lamin B bands in Fig. 4. This drastically reduced the apparent presence of Lamin B in the cytosol fraction and of LDH in the nuclear fraction, thereby artificially increasing the difference in the proteins compared to that in the other fraction examined.
“Nuclear import and export signals in control of Nrf2” has been cited 140 times since it was published in 2005.
The journal also issued a correction for one of Jaiswal’s papers, “NQO1 and NQO2 regulation of humoral immunity and autoimmunity,” cited 43 times since it was published in 2006. According to the notice:
The wrong graphs were inadvertently used in Fig. 3B. This error has now been corrected and does not affect the results or conclusions of this work.
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 daily digest. Click here to review our Comments Policy. For a sneak peek at what we’re working on, click here.