Two papers (one highly cited) on the mechanism of estrogen signaling have been retracted after an investigation by Wyeth found that the research data of its former employee Boris Cheskis were "unreliable."

Image: Wikimedia commons

The retractions do "clear up an area of uncertainty," said molecular endocrinologist David Ray of the University of Manchester School of Medicine, whose studies on a related topic conflicted with the now-retracted findings. "You don't want [to spend] your whole career refuting other people's work; you want to be figuring out entirely new stuff for yourself."

The primary action of steroid receptors, such as the estrogen, progesterone, and glucocorticoid receptors, is to regulate transcription when bound by a natural or drug ligand. When the ligands bind, however, there are many cellular effects that occur more rapidly than could be mediated through gene expression, Ray explained, suggesting these receptors may have other ways of effecting changes in the cell. Then researchers discovered that the progesterone receptor could bind directly to c-Src -- a signaling kinase that can activate a variety of pathways in the cell.

In 2002, Cheskis and colleagues published that they had identified a way that the estrogen receptor could also affect c-Src activity -- through interactions with a protein they called MNAR (modulator of nongenomic action of estrogen receptor). Their report, published in Proceedings of the National Academy of Sciences and cited 190 times, according to ISI, suggested that the estrogen receptor (ER) interacted with MNAR, which in turn mediated the activity of various signaling kinases and downstream signaling molecules, including c-Src. At the time, MNAR was the first molecule identified that linked a nuclear hormone receptor to the intracellular communication system. It was a "novel interacting molecular mechanism," Ray said.

Last July, the paper was retracted by the authors, who claimed that the data demonstrating the activation of the cellular signaling cascade by MNAR were "unreliable," although the physical interactions between the ER, MNAR, and Src were "valid." Daniel Salsbury, the managing editor of PNAS, told The Scientist that the journal retracted the paper on the request of the authors, and declined to go into any further detail.

The retraction helped validate the work of Ray and his colleagues on the structurally similar glucocorticoid receptors (GR), which had found that MNAR was not required for the activation of Src. At the time, Ray and his colleagues just chalked up the difference to the fact that they were studying a different receptor system. The GR and the ER "are similar, not identical," he said. "So we weren't unduly alarmed when it was different." But with the mechanism reported by Cheskis being retracted, "it now appears that they're rather more similar than we'd given them credit for."

"In general, the retraction of this specific mechanism does not have a major impact in the scientific field," molecular gynecologist Tommaso Simoncini of the International Society of Gynecological Endocrinology and the University of Pisa, Italy, who recently cited the PNAS paper, told The Scientist in an email. But it does negate "one potentially important advancement in the molecular understanding of the interaction."

A Molecular and Cellular Biology report, cited 27 times since 2007, which further detailed the mechanism of MNAR's interactions with the ER and Src, was also retracted due to "unreliable" data. MCB did not respond to requests for comment.

Wyeth (now Pfizer) declined to comment further on the details of their internal investigation, but did confirm that Cheskis is no longer employed by Wyeth or Pfizer.

Cheskis was also found guilty of research misconduct by the Office of Research Integrity, having falsified four figures in two National Institutes of Health grant applications, according to a report in the Federal Register earlier this month.

"It's unfortunate," said cell and molecular biologist John White of McGill University in Québec, who cited one of the retracted papers in a recent publication. "None of this knocks any pillars out from under our understanding of how the estrogen receptor functions," he added, "but the notion that there was misconduct [is] itself upsetting."

As a result of the ORI's finding of misconduct with respect to the two NIH grants filed, Cheskis signed a voluntary settlement agreement which states that he is responsible for the figure falsifications. For the next two years, he may receive funding from the US Public Health service, but only under supervision "to ensure the scientific integrity of his research contribution," according to the Federal Register.

Cheskis could not be reached for comment.


Related stories:

PhD student admits misconduct
[15th April 2010]

Science paper pulled
[30th November 2009]

Misconduct from cancer researcher
[25th November 2009]

Misconduct from NIH postdoc
[17th February 2009]

Rumblings over Science retractions
[26th March 2008]