On Dec. 2, 2013, Alison Lakin, the research integrity officer at the University of Colorado Denver, received a concerning email.
The emailer was alleging several problems in a 2012 paper in the Journal of Clinical Investigation, co-authored by one of its high-profile faculty members. Lakin discussed the allegations with some administrators and agreed they had merit; Lakin sequestered an author’s laptop and other materials. Over the next few months, the university learned of additional allegations affecting other papers — and discovered even more serious problems in the JCI paper. Namely, the first author had inserted changes to 21 figures in the paper after submitting it, without alerting the other authors, journal, or reviewers.
That journal retracted the paper this month, citing numerous problems:
An investigative committee at the University of Colorado Denver recently reported multiple findings of data falsification and fabrication or lack of underlying supporting data regarding Figures 1, 2, 3, 5, 6, 7, 8, and 9, and Supplemental Figures 6 and 8 in this publication. Due to the numerous manipulations and lack of data to support the published findings, the JCI is retracting this article.
The article, “Equilibrative nucleoside transporter 1 (ENT1) regulates postischemic blood flow during acute kidney injury in mice,” received an expression of concern in 2014 over duplicated images. It has been cited 77 times since 2012 (including 40 since 2014), according to Clarivate Analytics’ Web of Science.
According to the University of Colorado Denver’s investigation report, which we obtained through a public records request, former professor Almut Grenz committed research misconduct. Additional papers by the last author on the JCI paper, Holger Eltzschig, were also found to contain issues that were the result of “error,” not misconduct. Nevertheless, the committee recommended an additional paper (on which Grenz is the last author) be retracted, a correction to a PNAS paper on which she is the second author, and a correction or retraction for a Journal of Immunology paper, on which Grenz is the first author.
A spokesperson from the University of Colorado Denver told us that Grenz has left the university:
As of 2016, Dr. Grenz is no longer an employee of the university. I cannot say the terms under which she left the university.
Dr. Grenz was the only person confirmed to have engaged in research misconduct.
We were unable to locate Grenz, nor her contact information.
All papers were co-authored by Eltzschig, an anesthesiology researcher who has published more than 160 articles, which have collectively been cited nearly 8,000 times. In 2016, Eltzschig left Denver for the McGovern Medical School (part of The University of Texas Health Science Center at Houston), where he became the first to hold a McGovern chair position. A press release announcing his appointment calls him:
The “Number One” physician scientist in the field of perioperative medicine
We contacted Eltzschig; a representative responded and referred us to the spokesperson at the University of Colorado Denver.
Although the first allegations about the JCI paper appeared in December 2013, the following month another complainant contacted Lakin, alleging additional errors in four papers Eltzschig helped write, including three by Grenz. In February 2014, an inquiry committee determined a full investigation was warranted.
The investigation committee, along with Lakin, met 18 times, and interviewed 10 people. They read interview transcripts, laboratory notebooks, computer files and other aspects relating to the five papers.
The committee concluded that the five papers contained 36 duplicated histology images. But during the investigations, they discovered additional problems, including with the now-retracted JCI paper. Specifically, they found:
The respondents revised many figure panels with altered data that changed mean values and error bars, in 21 graphs presented in the JCI 2012 paper without informing the scientific reviewers, journal editors, and coauthors after the review process had commenced on June 28, 2011, following the submission of the manuscript by [name redacted]. These changes were not asked for by the reviewers after their initial critique of the manuscript. In addition, a pair of western blots for 2 equilibrative nucleoside transporters (ENT1 and 2) were switched prior to the original submission of the JCI paper and differ from their public presentation at several poster sessions. The switched protein bands were substituted after the final manuscript was sent to the coauthors for comments but prior to submitting the original manuscript to the JCI.
The report notes:
There does not seem to be a clear rationale for making the changes between versions but a number of the changes do serve to better illustrate the hypothesis of the paper.
Although many names in the report — including those of the complainants, respondents, and committee members — are redacted, in a handful of cases, the report makes its conclusions clear:
The committee finds that Dr. Grenz committed research misconduct in relation to the 21 figures detailed in section 3A above.
The report also notes:
…the committee heard testimony from a number of former members of the laboratory describing a strained work environment with pressure to produce data that conformed to the Principal Investigator’s [redacted] general hypothesis.
The hypothesis of the JCI paper, according to the abstract, is:
These studies identify ENT1 and adenosine receptors as key to the process of reestablishing renal perfusion following ischemic AKI. If translatable from mice to humans, these data have important therapeutic implications.
The university spokesperson told us she was “unaware” if Eltzschig had any commercial interest in the development of ENT1-based therapies for ischemia.
THE AFFECTED PAPERS
Another paper that was found to contain additional problems was published in Hepatology in 2013; Eltzschig and Grenz are co-corresponding authors. In that paper, the committee found the mean value error bars were changed in 36 conditions between the original and final versions of the paper, and 23 changes to p-values. In this case, the committee found the changes stemmed from “a case of careless misuse of error bar calculations.”
It recommended the Hepatology paper be retracted, as well. Chris Graf, the Director of Research Integrity and Publishing Ethics at Wiley, which publishes the journal, told us it had not yet received a request from the institution or author to publish a retraction.
“Equilibrative nucleoside transporter (ENT)-1-dependent elevation of extracellular adenosine protects the liver during ischemia and reperfusion” has been cited 15 times since it was published in 2013.
The other affected papers are:
“Signaling through hepatocellular A2B adenosine receptors dampens ischemia and reperfusion injury of the liver,” PNAS 2013, cited 17 times. Recommended for correction.
“Adora2b Adenosine Receptor Signaling Protects during Acute Kidney Injury via Inhibition of Neutrophil-Dependent TNF-α Release,” Journal of Immunology 2012, cited 22 times. Recommended for correction or retraction.
“Hypoxia-inducible factor–dependent induction of netrin-1 dampens inflammation caused by hypoxia,” Nature Immunology 2009. Cited 208 times. Grenz is not a co-author; Eltzschig is the last author. The report notes that the paper “was not under the jurisdiction of this committee as all the data was collected external to the University of Colorado Denver.” What’s more, the paper was corrected in 2015, noting:
In the version of this article initially published, the contrast in lane 4 of Figure 2c had been altered, and the top right image in Figure 5a and bottom right image Figure 6e were incorrect. The correct gel and images are now presented. The error has been corrected in the HTML and PDF versions of the article.
When the spokesperson sent us the redacted report, she included this statement:
The University of Colorado Anschutz Medical Campus is committed to ensuring the integrity of research and takes scientific misconduct allegations seriously. After concerns about research misconduct were raised, the university conducted an investigation in accordance with federal mandates. At the conclusion of that investigation, it was determined that Almut Grenz falsified data while employed as a researcher in the Organ Protection Laboratory in the Department of Anesthesiology. Affected publications, co-authors and funding agencies have been notified. The university remains vigilant in fostering an environment that promotes the responsible conduct of research and delivers an exceptional educational experience to our students.
Update, 20:37 UTC time, June 19, 2017: The PNAS and Journal of Immunology papers have been retracted. Here is the PNAS notice:
The authors wish to note: “We were informed by the Office of Research Integrity of the University of Colorado of the inappropriate duplication of several histological images in Figs. 2 A and D, 3 A–C, S2B, and S3B of our recent PNAS paper. There were no findings of scientific misconduct made against any of the authors. The problems with the histologic slides appear to be due to honest errors. We apologize to our colleagues and the scientific community for any inconvenience this might have caused. Based on these concerns, we request to retract our paper.”
Here is a link to the notice from the Journal of Immunology retraction of “Adora2b Adenosine Receptor Signaling Protects during Acute Kidney Injury via Inhibition of Neutrophil-Dependent TNF-α Release:”
A request was received from the Research Integrity Officer of the University of Colorado Denver, Anschutz Medical Campus, and by Dr. Clambey, to retract this article: “Adora2b Adenosine Receptor Signaling Protects during Acute Kidney Injury via Inhibition of Neutrophil-Dependent TNF-α Release” by Almut Grenz, Jae-Hwan Kim, Jessica D. Bauerle, Eunyoung Tak, Holger K. Eltzschig, and Eric T. Clambey, The Journal of Immunology, 2012, 189: 4566–4573.
The Editor-in-Chief of The Journal of Immunology was informed that the University of Colorado Denver had conducted a review of the work and concluded that data used for the upper right hand panel of Figure 3B (−Infliximab +Ischemia in a homozygous Adora2b−/− mouse) was duplicated with the upper right hand panel of Figure 3E (−Infliximab +Ischemia in a WT mouse). The authors were not able to provide the original data for these figures, and thus they cannot be validated. There were no findings of scientific misconduct made against any of the coauthors relating to the article. The identified errors appear to be the result of honest error, but due to the lack of original data, the article is hereby retracted.
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.