The Journal of Virology has retracted a 2010 article on avian viruses marred by signs of bogus data.
The paper, “Avian Reovirus Nonstructural Protein p17-Induced G2/M Cell Cycle Arrest and Host Cellular Protein Translation Shutoff Involve Activation of p53-Dependent Pathways,” came from a group at National Pingtung University of Science and Technology, in Pingtung, China. It purported to find that:
The effects of avian reovirus (ARV) p17 protein on cell cycle progression and host cellular protein translation were studied. ARV infection and ARV p17 transfection resulted in the accumulation of infected and/or transfected cells in the G2/M phase of the cell cycle. The accumulation of cells in the G2/M phase was accompanied by upregulation and phosphorylation of the G2/M-phase proteins ATM, p53, p21cip1/waf1, Cdc2, cyclin B1, Chk1, Chk2, and Cdc25C, suggesting that p17 induces a G2/M cell cycle arrest through activation of the ATM/p53/p21cip1/waf1/Cdc2/cyclin B1 and ATM/Chk1/Chk2/Cdc25C pathways. The G2/M cell cycle arrest resulted in increased virus replication. In the present study, we also provide evidence demonstrating that p17 protein is responsible for ARV-induced host cellular protein translation shutoff. Increased phosphorylation levels of the eukaryotic translation elongation factor 2 (eEF2) and initiation factor eIF2α and reduced phosphorylation levels of the eukaryotic translation initiation factors eIF4E, eIF4B, and eIF4G, as well as 4E-BP1 and Mnk-1 in p17-transfected cells, demonstrated that ARV p17 suppresses translation initiation factors and translation elongation factors to induce host cellular protein translation shutoff. Inhibition of mTOR by rapamycin resulted in a decrease in the levels of phosphorylated 4E-BP1, eIF4B, and eIF4G and an increase in the levels eEF2 but did not affect ARV replication, suggesting that ARV replication was not hindered by inhibition of cap-dependent translation. Taken together, our data indicate that ARV p17-induced G2/M arrest and host cellular translation shutoff resulted in increased ARV replication.
The paper has been cited 13 times, according to Thomson Scientific’s Web of Knowledge. But according to the retraction notice:
After publication, this article was found to have evidence of data duplication of the following images. Figure 3B, lanes labeled p-Akt(S473) and p-p21(S145) from the pcDNA3.1 transfection samples; Fig. 4A, lane labeled Cdc25C from the ARV infection samples and Fig. 4B, lane labeled Cdc2 from the pcDNA3.1 transfection samples; Fig. 4A, lanes Cdc2 and actin from the mock infection samples; Fig. 6A, lane labeled actin from the p17 samples and Fig. 8B, lane labeled actin from p17+ caffeine samples.
Since the integrity of the data as presented was compromised, the authors retract this publication. We apologize to the editors and readers of the Journal of Virology and wish to express our regret for any inconvenience this has caused.
Hat tips: Rolf Degen, Simon Rayner
Kudos to ASM and JV for doing the right thing with this manuscript.
First, it is important that publishers retract compromised papers – one bad figure and a paper should be retracted.
Second, the issue of inappropriate image manipulation is so pervasive now that authors should be required to include in supplemental material all the original non-manipulated images that go into compiling figures.
Access to the original images is esp needed for those figures that are composites of multiple gels. For example, the figures in this JV paper must have come from dozens if not hundreds of individual images.
Which publishing house will lead on this or is it something that COPE might champion?