Another retraction and a correction that retracts two figures — ie, a partial retraction — have been posted for Duke University lung researchers, Erin Potts-Kant and Michael Foster.
These latest notices move the count up to 8.5 retractions for Potts-Kant and 7.5 for Foster (counting the partial retraction as 1/2), along with the correction for both. In both cases and in a familiar note from previous retractions, authors found “potential discrepancies” between two sets of data (partial retraction) and study figures that weren’t “reliable” (retraction).
The retraction comes after the authors discovered problems with three of the study figures. In the corrected paper, the authors were able to validate some of their findings after repeating the experiments, but retracted two of the study figures that they were “unable to verify.”
The American Journal of Physiology – Lung Cellular and Molecular Physiology posted the retraction notice for “MARCKS-related peptide modulates in vivo the secretion of airway Muc5ac” online in August. It’s the second paper authored by Foster and Potts-Kant that the journal has retracted.
The paper was published in 2010 and detailed “the effects of intratracheal instillation of the MANS peptide” on the airways of mice, according to the abstract. It has been cited 17 times, according to Thomson Scientific’s Web of Knowledge.
The notice reads:
The authors have discovered that Figs. 1, 2, and 6 are not reliable. We offer our sincere apologies for the errors and for any inconvenience associated with the publication of the article. The paper is therefore being retracted by the American Physiological Society at the request of Dr. Voynow, and with the approval of the coauthors.
Christina Bennet, Publications Ethics Manager for the journal’s publisher, the American Physiological Society, said that the final notice will be published in October:
The final version of the retraction notice will be published in the October 1 issue of AJP-Lung.
The corrected paper was published in 2009 by The Journal of Immunology. The journal issued a lengthy correction notice — which is behind a paywall but has been appended to the pdf of the original paper.
The study, “SP-A Preserves Airway Homeostasis during Mycoplasma pneumoniae Infection in Mice” examined how the protein SP-A works during pulmonary infections in mice with different levels of the protein. It has been cited 19 times.
Here’s the full notice, which explains the correction in detail (and may qualify as a “mega-correction“):
We have recently become aware of potential discrepancies between the machine-generated raw data and the data provided by our collaborating pulmonary function laboratory that were used to calculate the average resistance and compliance results on the Flexivent. We therefore repeated the original experiments, and in contrast to our published work, we now find that SP-A KO mice do not have greater airway hyperresponsiveness as compared with wild type (WT) mice. We verified that WT mice have significantly greater airway resistance when infected with Mycoplasma pneumonia(Mp) as compared with WT saline controls. These newly generated data are provided here as a replacement for our original Fig. 1. [See full note for updated figure]
In addition, we would like to retract Fig. 5A and 5B from the published article, as they build upon the data in Fig. 1, which we have been unable to verify. In light of the changes toFig. 1 and the retraction of Figs. 5A and 5B, the following corrections are needed to the text of the published article.
In the Abstract, the sentence “Likewise, physiologic responses (airway hyperresponsiveness and lung compliance) to Mp infection were more severely affected in SP-A−/− mice,” is no longer valid and needs to be omitted.
In the last paragraph of the Introduction, the phrase “…and significantly elevated airway hyperresponsiveness (AHR)…” is no longer valid and needs to be omitted.
In the Results section, in the second paragraph under the subheading AHR in Mp-infected mice, the sentence “Shown in Fig. 1A, all animals had minimal AHR to the methacholine challenge when instilled with saline,” and the sentence “In contrast to WT mice, the infected SP-A null mice had significantly elevated AHR, which was evident even at the 25 mg/ml dose of methacholine (Fig. 1A),” are no longer valid and need to be omitted. The entire third paragraph, “In addition to increased airway resistance in the Mp-infected SP-A null mice, they also showed significantly reduced dynamic compliance as shown in Fig. 1B. Dynamic compliance measures the ease with which the lungs can be extended and a drop in compliance values indicates increased stiffness in the lungs. The compliance measurements in the WT infected mice were comparable to saline-treated mice. However, the compliance of the infected SP-A null mice was significantly lower at the baseline pre-challenge measurements well as throughout the methacholine challenges (Fig. 1B),” is no longer valid and needs to be omitted.
In the Results section, in the first paragraph under the subheading Inhibition of TNF-α attenuates AHR, the text “To determine whether the increased AHR observed in Mp-infected SP-A null mice may be due to overproduction of TNF-α, and therefore a myriad of other effects due to TNF signaling…” needs to be omitted, as does the following text at the end of that paragraph: “…in which they continued to have significantly higher AHR compared to WT infected mice. However, this heightened AHR was significantly attenuated if the SP-A null mice were pretreated with the TNF inhibitor, LMP-420, before infection (Fig. 5A). This suggests that SP-A can modulate factors related to physiologic airway function during the acute phase of an infection, and in the absence of SP-A, Mp enhances TNF signaling and secretion, leading to an enhancement in AHR.” In the second paragraph, the text “As discussed above, dynamic compliance was also significantly elevated in Mp infected SP-A null mice before methacholine challenge. However, compliance was attenuated at baseline as well as with methacholine challenge in the Mp infected SP-A null mice in which TNF-α activity had been inhibited with LMP-420 treatment (Fig. 5B),” is no longer valid and needs to be omitted.
The data shown in Fig. 5A and 5B and part of the text in the figure legend, “AHR” and “A, Airway responsiveness and B, airway compliance to methacholine challenge were analyzed 72 h post infection by Flexivent technology. n=8–10/group and *, p < 0.05; **, p < 0.01 and is SP-A−/− Mp infected/vehicle vs all other groups,” are no longer valid.
In the first paragraph of the Discussion, the following text needs to be omitted: “…and elucidate a new role for the protein in mediating airway hyperreactivity…” and “Coincident with up-regulation of the biologic response, physiologic changes inclusive of AHR and decreased lung compliance were also more severe in the absence of SP-A.” In addition, “AHR” in the sentence “Inhibition of TNF-α production during Mp infection by systemic intervention resulted in striking reductions in AHR, cellular inflammation, and mucus hypersecretion,” needs to be omitted. In the last sentence of the second paragraph of theDiscussion, the text “…and further exacerbate AHR…” should be disregarded. In addition, in the last sentence of the third paragraph of the Discussion, the last sentence “The heightened increase in TNF-α activity in those mice lacking SP-A could account for the increase in AHR observed in the SP-A null mice by several mechanisms,” needs to be omitted.
In the first sentence of the fourth paragraph of the Discussion, the text “…not only significantly decreased AHR in SP-A null mice but also…” and the last sentence “Thus, the decrease in T lymphocytes via reduction in TNF-α production may partially explain the apparent decrease in AHR of the infected SP-A null mice,” are no longer valid and need to be omitted. Finally, in the last paragraph of the Discussion, the sentence “Although discriminating between these factors cannot clearly define one as the causative agent of enhanced AHR in the infected mice lacking SP-A, these multiple factors should all be taken into consideration as contributors to the heightened response in these mice,” is no longer valid and needs to be omitted.
We are confident that the other data we reported are valid, as they were collected and analyzed independently of the pulmonary mechanics data. All authors agree to this Correction and to the retraction of Figs. 5A and 5B. We apologize to our colleagues and the scientific community for any inconvenience this might have caused.
Pamela Fink, the journal’s editor-in-chief, cited JI policy when asked for a statement on the correction:
As per our policy outlined at http://www.jimmunol.org/site/misc/editorialpolicies.xhtml#scientificfraud, we communicate only with the Corresponding Author regarding all issues involving corrections and retractions.
Potts-Kant was arrested in 2013 for embezzling almost $15,000 from Duke University.
We’ve contacted Foster for a statement on the two notices and we’ll update with any response.
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, and sign up on our homepage for an email every time there’s a new post. Click here to review our Comments Policy.
How does one retract two figures and leave the paper intact? In my opinion, there should be no such thing as a partial retraction. It simply makes no sense. Did the publisher put a big red cross or something over each “retracted” figure, and then also a red strikethrough through the relevant text?