Nature to retract plant study

By Edyta Zielinska | December 9, 2008

Comments

A highly cited __Nature__ paper that identified a long-sought receptor critical for mediating plant response to stress is being retracted after researchers were unable to reproduce the results. Corresponding author on the paper, linkurl:Robert Hill;http://home.cc.umanitoba.ca/~rhill/ from the University of Manitoba, first discovered a problem with the results over the summer when one of his students failed to reproduce the findings. "The binding assay procedures, at least in our hands, did not give the correct results," said Hill. Hill and his coauthors were the first to identify a receptor for abscicic acid (ABA), which regulates plant stress response. Researchers have been trying to identify an ABA receptor for decades -- a finding that would be of major interest to the agriculture industry. In the linkurl:study,;http://www.nature.com/nature/journal/v439/n7074/abs/nature04373.html published in 2006 and cited more 120 times, first author Fawzi Razem, who conducted the work as a postdoc in Hill's lab, found that FCA, a nuclear protein that binds RNA and is involved in plant flowering, also acted as a receptor for ABA. Two more ABA receptors were found in linkurl:2006;http://www.nature.com/nature/journal/v443/n7113/abs/nature05176.html and linkurl:2007;http://www.sciencemag.org/cgi/content/full/315/5819/1712 by other groups who used the same method. Researchers in the field were surprised when the finding was published in early 2006. "That paper was perplexing," said linkurl:Sean Cutler;http://cutlerlab.blogspot.com/ a plant biologist at the University of California, Riverside. "Every paper that's come out claiming to be an ABA receptor has been perplexing." The reason for the hesitation, said linkurl:Peter McCourt,;http://www.ratemyprofessors.com/ShowRatings.jsp?tid=486497 a plant geneticist from the University of Toronto, was that researchers have long known that concentrations of the ABA hormone "goes through the roof" -- increasing about 100-fold -- when plants are stressed. According to Razem's paper, however, FCA binds abscisic acid with a high affinity, meaning that fairly low concentrations of the hormone would trigger a stress response. "Corn plants are always stressed, so by definition the receptor will always be on," said McCourt, "and that doesn't make sense." "I [had] always been very skeptical of binding assays in whole plants," said McCourt. "I read the [Razem] paper and I thought, 'maybe I was wrong.'" In a study to be published in __Nature__, the proofs of which were obtained by __The Scientist__, corresponding author linkurl:Catherine Day;http://biochem.otago.ac.nz/staff/day/cday.html and colleagues from the University of Otago in New Zealand report that they were unable to reproduce the Razem results. They suggest that the sensitivity of the binding assay used by Razem, as well as the authors of the subsequent two ABA papers, was too high, bringing all three into question. A note in that paper states that the Razem paper has been retracted, but since it has not yet been published, the study contains no dates. In an email to __The Scientist__, Day said she had nothing more to add about the Razem study. Peter McVetty,head of the university's plant science department, told the linkurl:Winnipeg Free Press;http://www.canada.com/topics/news/national/story.html?id=1028551 that a calculation error may have "crept in" to the data. The retraction letter for the Razem paper, which Hill coauthored, will be published in the December 11th issue of __Nature__, according to Manitoba University spokesperson John Danakas. __Nature__ declined to confirm the retraction or say when it would be published. Hill said he corresponded with Razem in regards to the retraction, but declined to say what they discussed. Razem could not be reached for comment by phone or email. According to Danakas, Razem, who became an assistant professor at the University of Manitoba shortly after the publication of this study (September of 2006), has resigned from his position, effective by the end of this year. Danakas said Razem is leaving for another appointment, but where is "not known to us." Danakas also said he could not comment whether an investigation was currently underway, but did say that in cases where scientific findings are not reproducible the university usually performs a "thorough and rigorous investigation" into possible fraud allegations. "It obviously put a big dent in what we're doing," said Hill. "It's meant that we've had to go back and reinterpret data." One graduate student has had to "chop a publication" which was based on the assumption that the ABA receptor was real. "I've come to reconcile with the problem," said Hill, adding that he is working to correct it "without hurting too many people." "All one can do it be upfront about it, once one gets in a situation" like this, said Hill.
**__Related stories:__*** linkurl:The drought receptor;http://www.the-scientist.com/article/display/54050/
[Jan 2008]*linkurl:Science retracts major Arabidopsis paper;http://www.the-scientist.com/news/display/53081/
[20th April 2007]*linkurl:US postdoc fabricates DNA data;http://www.the-scientist.com/blog/display/54637/
[9th May 2008]

Comments

The Scientist

Comments

anonymous poster

Posts: 1

December 9, 2008

What they care is not the truth but " not hurting too many people".............

Adam Smith

Posts: 20

December 9, 2008

Clearly they do care about the truth, or they would not be the ones who figured out the problem and retracted the paper.

anonymous poster

Posts: 1

December 9, 2008

I want to retract one paper from ACS (I am a co-author of that paper, and don't want to publish that in present version). But they ignore me and don't reply my email. Is that only corresponding author can do that?

Ruth Rosin

Posts: 117

December 10, 2008

It's nice to see that an erring paper is due to be retracted so quickly.\n\nIn flagrant contrast, there is an old, extremely serious scientific error, touted as a great achievement in hundred, if not thousands of serious scientific publications, as well as in the popular scientific media, that no one seems to even consider retracting. \n\nI refer to v. Frisch's Nobel Prize winning "discovery" of the very famous, but non-existent honeybee "dance language" (DL). \n\nThe still ongoing honeybee DL hypothesis was first discredited by Wenner & his team in 1967, after the hypothesis had already become a revered ruling paradigm. Thus, instead of being rewarded they were quickly turned into paraiahs. The DL controversy has, since then, been going on for over 40 years, even tough, (because DL opponents are generally denied the right to be heard, many serious scientists "know" that the controversy had been resolved long ago in favor of DL supporters.\n\nAfter this very lengthy controversy it has finally become possible to discredit the v. Frisch's whole DL hypothesis in a very short shrift: The DL hypothesis turns out to have been a stillborn hypothesis, rooted in outright scientific fraud, disguised by a deliberate cover-up.\n\nThe DL hypothesis, first published by v. Frisch in a scientific journal, in 1946, as presumably already fully adequately experimentally confirmed, was preceded by v. Frisch's own initial conclusion, based on his own first study on honeybee-recruitment, (published in a very extensive summary in 1923), that honeybee-recruits use only odor, and no information about the location of any food. Moreover, the results obtained in tat study already grossly contradicted his later, sensational DL hypothesis, long before its inception, in terms from the expectations from round dances. \n\nAccording to the DL hypothesis round dances cause recruits to find food with the foragers' food-odor, only near the hive, within not more than 100 m. from the hive for any honeybee species and strain. V. Frisch's first study on honeybee-recruitment showed that this was not at all the case, in all 3 tests which included such sources further from the hive. The honeybee DL hypothesis was, therefore stillborn.\n\nThe techniques used in that first study were extremely simple, and fully adequate, to the point that neither v. Frisch, nor anyone else ever saw any fault with the study.\n\nHowever, after the inception of his sensational DL hypothesis v. Frisch "eliminated" the results of that first study, which refuted the hypothesis. The act fully qualifies as an act of outright scientific fraud. Eventually, in his 1967 definitive book on the honeybee DL (translated from the German edition of 1965), he substituted for the "eliminated" results, the results of new tests with round dances, actually done in 1962, using a drastically different experimental design than that used in his first study on honeybee-recruitment. This time the results fit the expectations from his DL hypothesis. (None of the sources that were more than 100 m. from the hive were found by the recruits.) The act of the substitution fully qualifies as an act of deliberate cover-up.\n\nV. Frisch, however, truly, and innocently believed he had discovered the DL he had attributed to honeybees, and that he acted correctly, in reviving the stillborn DL hypothesis, "eliminating the early results that refuted the hypothesis, and covering up the "elimination", because, in his own mind he only valiantly saved The Theory of Evolution from an erroneously perceived crisis it would presumably face in the absence of the honeybee DL. He, as well as many others, became convinced that the honeybee DL simply had to exist to explain to provide an otherwise inconceivable adaptive value for honeybee-dances and dance-attendance.\n\nHowever, irrespective of his noble purpose, his honeybee DL remains none other than a stillborn hypothesis, rooted in outright scientific fraud, disguised by a deliberate cover-up.\n\nThe reason he, as well as many others, became convinced that the honeybee DL simply had to exist is that even though on the face of it the DL controversy seems to be a controversy over some idiosyncratic behavior of honeybees, the controversy actually constitutes the most important reflection of a much more basic controversy over the very foundations of the whole field of Behavioral science, and Biology in general, i.e. the controversy over whether genetically predetermined individual traits, (known as "instincts" in behavior) exist in ontogeny. They don't. But I shall not go into that general controversy here.\n\nAT any rate, we are all owed retractions by very prestigious, refereed journals like Nature, and Science, and by very many others, that repeatedly touted the honeybee DL, as well as by the 1973 Nobel Committee. The retractions are going to severely damage the scientific reputation of very many scientists, both dead, and alive. \n\nWould anyone who should publish a retraction ever master the courage, and decency required to do just that? I can't say.