Doing the right thing: Researchers retract quorum sensing paper after public process

We’ll say it again: We like being able to point out when researchers stand up and do the right thing, even at personal cost.

In December 2011, Pamela C. Ronald, of the University of California, Davis, and colleagues published a paper in PLOS ONE,”Small Protein-Mediated Quorum Sensing in a Gram-Negative Bacterium.” Such quorum sensing research is a “hot topic” right now, so not surprisingly the paper caught the attention of other scientists, and the media, including the Western Farm Press. The study has been cited eight times, according to Thomson Scientific’s Web of Knowledge.

One of those scientists who took notice was Ronald’s UC Davis colleague Jonathan Eisen, who posted about the paper on his blog. That was on January 9, 2012. But if you go to that post today, you’ll see that Eisen struck through most of it, and added this comment:

Pam Ronald asked me to retract this post because she discovered that one of the strains used in the reported experiments was compromised.  She notes that he laboratory is now in the process of repeating each experiment with newly validated strains and that much of the work has been independently validated in other laboratories  (McCarthy et al., 2011, J Bacteriology 193:6375-6378; Shuguo et al. 2012, Appl Biochem Biotechnol.  166:1368-79).

Please contact her if you have questions.  I am leaving the text of the post below based on the notion that one should not completely delete anything from the record but rather post corrections and retractions.  More detail will be coming from Pam soon.  Kudos to Pam for trying to make sure the scientific record is accurate and for contacting me about this.

That’s because on January 29, 2013, Ronald and her co-authors posted this comment on their paper:

As a result of recent experiments in my laboratory, I wish to comment on the report of Han et al., 2011. Specifically, members of my laboratory have recently discovered critical errors in strain PXO99Δax21 used in the reported work. Because key parts of the work depend on the integrity of strain PXO99Δax21, we are now repeating all of the experiments using newly generated and validated strains. We will report the outcome of each of these experiments on this page. We note that much of the work has been independently validated in other laboratories (see McCarthy et al., 2011, J Bacteriology 193:6375-6378).

Sang-Wook Han, Malinee Sriariyanun, Sang-Won Lee, Manoj Sharma, Ofir Bahar, Zachary Bower and Pamela C. Ronald

Those experiments are apparently now complete, and Ronald et. al. have retracted the paper. Here’s the notice, which went live on September 9:

The authors retract this publication due to concerns about the integrity of strain PXO99Dax21 employed in the study.

During additional experiments carried out at our laboratory, we discovered that the strain PXO99Dax21 employed in the study was mixed up with another strain in our collection. This compromises the validity of Figures 1, 2 and 3 in the article as we are unsure as to whether the correct strain was employed in the studies.

We have taken steps to repeat the experiments but we have been unable to validate the knockout mutants generated so far. We have also attempted to replicate the work using strains obtained from another group, but in our hands, those strains did not infect rice plants.

Kudos to Ronald and her colleagues.