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.
Very very excited for this “Story behind the paper” post. For those who do not know, I have been hosting posts here on my blog written by authors of Open Access papers telling the story behind the paper (I have also been writing some of my own). This one comes from my friend and UC Davis colleague Pamela Ronald. Pam is a fascinating person – great scientist – fun in person – blogger – book author – prize winner for work for the developing world – and much more. She was in the office across the hall from me for a while but now has moved to new diggs on campus and I miss the regular interactions with her. Anyway – without further ado – here is her post describing a new paper of hers from PLoS One, among many other things (note – she wrote the post – I added a few headers and pictures and links). Finally I note – if anyone else wants to tell the Story behind one of their papers – please let me know – I would love to host it here.
The Deep Background
The windowless room, dank an dark, was not an obvious place for inspiration. I took notes, wondering if I would be able to glean anything meaningful from Professor Helen Stafford’s (1922-2011) meandering lecture. I was skeptical. After all, this was the same teacher who, annoyed with our choice of vegetarianism, had told us that “plants have feelings, too”.
But what I learned that day, 33 years ago, would trigger a grand curiosity about the natural world and draw me into the greatest scientific puzzle of my career.
Helen informed us that human language is not the only way that species communicate. Plants form intimate associations with fungi and bacteria, which allow them to thrive in stressful environments. Establishment and maintenance of the relationship depends on the passing and receiving of coded information between partners. She also told us that plants can only defend themselves against microbes that they can sense.
All these interactions dramatically affect human health and farm productivity.
I was hooked.
We showed that single-celled bacteria communicate with each other using coded messages to coordinate attacks on their targets. Until now, the diversity of codes employed by these invading bacteria was thought to be extremely limited. We showed that one group of bacteria communicate by a novel, previously undescribed signal – and, as if to prove Helen Stafford correct- some plants have evolved a complementary cypher-breaking detection system that intercepts this bacterial code and then uses the information to trigger a robust immune response, preventing disease.
The Recent Background
The author Carol Shields has said that “The scientific life is the most complex of all to write about. In the case of scientists, impulse becomes compulsion”. And so it was with me. A year after college, I moved to Sweden to study with Professor Nils Fries who had identified a compound in pine trees that could coax some fungi to establish a lasting, mutually beneficial relationship. But pine trees and fungi and are difficult to study because they grow slowly. For this reason, when I returned to the US, I shifted my research to bacterial-plant interactions and their impact on plant health.
|Rowland Biffen In 1905, British geneticist and plant breeder Rowland Biffen demonstrated that it as possible to generate wheat varieties with resistance to a devastating diseases by moving genes around. He cross-pollinated a resistant wheat variety with a susceptible wheat variety and showed that the resulting seed carried the resistance of the parent.
Today, more than 100 years after Biffen’s discovery, plant breeders have introduced “resistance genes” into virtually every crop plant that we consume. But what are these genes? How do they function to detect the disease-causing microbe? For many years, the molecular basis of resistance remained elusive.
|Brian Staskwicz These were central questions addressed by the laboratory of Brian Staskwicz, where I carried out my PhD work. He had recently identified a bacterial protein that triggered an immune response in infected plants. He later identified one of the first plant resistance genes. After graduating, I moved to the laboratory of plant breeder Steven Tanksley at Cornell who was pioneering gene mapping in plants using molecular methods. Focusing on Xa21
My focus was a genetic locus, called Xa21, described in 1977 by Drs. S. Devadath, Gurdev Khush and collaborators. Rice plants carrying Xa21 have an unusual property: they are resistant to all known races of the bacterial pathogen Xanthomonas oryzae pv. oryzae (Xoo), which normally caused a devastating disease of rice in Asia and Africa. In 1992, I mapped this locus to a specific region in the rice genome. I hypothesized that it must consist of a cluster of tightly linked genes each recognizing a single race of the pathogen or a single gene that encoded a receptor recognizing a conserved microbial signature present in all races. When I moved to UC Davis I began a map-based cloning approach to isolate Xa21.
In 1995, two talented postdoctoral fellows in my lab, Guoliang Wang and Wenyuan Song, successfully isolated Xa21 and showed that it encoded a single gene: a receptor kinase. The structure immediately suggested that the XA21 protein could detect a microbial molecule present outside the plant cell and that this perception would activate an immune response inside the cell.
Toll and Toll-Like Receptors – Similar to XA21 – get the Nobel Prize in 2011
A few years after the discovery of the first plant resistance genes, the fly Toll and mouse Toll-like receptor (Tlr4) genes were isolated and shown to have striking structural similarities to XA21. Like XA21, TLR4 is membrane bound extracellular receptor that was predicted to bind a conserved microbial signature. TLR4 also carry the Toll /IL-1 Receptor (TIR) domain found in fly TOLL, the tobacco N resistance gene and the flax L6 resistance gene. Thus, the discovery of a role for Toll and TLR4 in immunity provided a structural link between sensors utilized by plants and animals to detect infection. Professors Bruce Beutler and Jules Hoffman were awarded the 2011 Nobel Prize in Physiology or Medicine for their important work.
[Bruce and I share more than an interest in science; my father (Robert Rosenthal) and Bruce’s father (Ernst Beutler) were young cousins in Berlin in the 1920 and early 1930s. Their families fled the Nazi’s and reunited in the US after the war. I was honored to hear Bruce discuss XA21/Ax21 and our shared family history during his Nobel lecture last month (starts at 40:45)]
The next challenge was to isolate the conserved microbial signature produced by Xoo that was recognized by XA21.
To isolate this molecule, which we named Ax21 (Activator of Xa21-mediated immunity), we screened for bacterial mutants altered in their ability to produce active Ax21. In this way, we isolated and characterized eight genes required for Ax21 activity (rax genes). raxA, raxB and raxC encode components of a predicted type I secretion system. Ax21 requires this RaxABC system for activity and secretion. The RaxB protein carries two highly conserved domains characteristic of proteins in Gram-positive bacteria that cleave N-terminal peptides prior to secretion of small proteins. Another set of rax genes encoded proteins important for sulfation.
These data suggested that Ax21 was a small sulfated protein with an N-terminal leader peptide, which was cleaved by the RaxB transporter prior to secretion outside the bacterial cell.
Much to our delight, that is exactly what we found.
Ax21 is a small protein with a tyrosine sulfation site and a predicted N-terminal leader sequence. Ax21 is not secreted in the absence of the RaxABC type 1 system. XA21/Ax21 binding triggers XA21-mediated innate immunity.
The Ax21 sequence is conserved in all Xanthomonas spp., in Xylella fastidiosa and the human pathogen Stenotrophomonas maltophilia. This conservation suggested that Ax21 serves a key biological function.
In our latest paper, we demonstrated, using two different methods, that the predicted Ax21 leader is cleaved as predicted. We further showed that only mature Ax21, missing the leader, is found outside the cell. When we add the purified protein back to a mutant lacking the wild-type gene, we can restore all Ax21 functions. These experiments demonstrate that the Ax21 protein acts outside the cell. We then used a genetic approach to reveal that the likely bacterial receptor for Ax21, is a histidine kinase called RaxH. Finally, we show that the Ax21 mature protein has some pretty impressive capabilities: bacteria use it “talk” to each other.
Over the last 20 years, researchers have shown that bacteria employ specific signals to communicate. These signaling molecules, called “bacterial Esperanto” by Professor Bonnie Bassler, an early pioneer in studies of bacterial communication, accumulate in the external environment as the cells grow. When the concentration reaches a certain threshold level, the bacteria mobilize together to carry out concerted, group actions. This process is called quorum sensing.
Until now, it was thought that the two major groups of bacteria (Gram-positive and Gram-negative) use distinctly different types of communication codes. However, Ax21 doesn’t fall into either class. While the previously characterized signals in the bacterial coding repertoire were all relatively small molecules, Ax21 is a small protein, which makes it much larger.
Perception of Ax21 by the RaxH receptor triggers a massive change in the bacterial genetic program, altering the expression of nearly 500 genes, or approximately 10% of the bacteria’s genome. These changes allow the bacteria to assemble into elaborate protective bunkers, called biofilms, which render the bacteria resistant to dessication and antibiotic treatment. Thus, by virtue of communication and communal living, bacteria increase their chances of survival and proliferation. Ax21 perception also regulates the production of a virulent arsenal, including “effectors” that are shot directly into the host to disrupt its defenses and that initiate motility, allowing the bacteria to colonize new sites for infection.
Most rice plants are virtually defenseless against this Ax21-mediated bacterial attack – except for those plants that carry the XA21 immune receptor. This early detection gives the plant time to mobilize its defenses and mount an early and potent immune response.
The discovery that a small protein from a Gram-negative bacterium has a dual role in bacterial communication and in activation of the host innate immune response has not previously been demonstrated. We do not, however, believe this is an anomaly or that the biological importance of Ax21 is restricted to plant pathogens. We previously reported that Ax21 is also conserved in the nosocomial pathogen S. maltophilia and proposed a similar role for Ax21 in this species. Consistent with our hypothesis, a synthetic Ax21 protein has now been shown to regulate gene expression, motility, and biofilm formation in S. maltophilia, extending our findings to an animal pathogen.
Exploration of other bacterial genomes reveals the presence of an abundance of small secreted proteins similar to Ax21, suggesting the intriguing possibility that other species of bacteria also use small proteins to communicate and coordinate infection.
Control of Gram-negative bacterial infections in plants and animals remains a major challenge for the medical profession and for farmers, because conventional approaches are often not sufficient to eradicate these infections. One major reason for their persistence seems to be the capability of most bacteria to grow within biofilms that protect them from adverse environmental factors and antibiotics. The knowledge that bacteria use Ax21 to communicate can be used to develop reagents to immunize hosts against infection or antagonists to disrupt Ax21-mediated virulence activities and biofilm formation, a process thought to be involved in 65-80% of bacterial infections of plants and animals.
A Note on the Publication Process
Convincing the scientific community that a Gram-negative bacterium uses a small, N-terminal processed protein for bacterial communication has not been easy.
We initially submitted the manuscript to Science in 2009. One year, two manuscript versions, seven reviews, 16 figures and 11 tables later, we were still revising.
All but one of the six reviews were enthusiastic: they found the work “highly interesting”, the experimental work of “high quality”, carried out “to an exceptionally high standard” and the “evidence compelling”. The initial set of experiments the reviewers asked us to carry out were challenging but reasonable. We completed them over 7 months of intensive experimentation. Happily, the reviewer said we had “addressed all the fundamental issues” requested in the first round of reviews and that “the revisions were satisfactory”. Most of the additional suggestions were easily addressed. None of the reviewers challenged our central conclusion: that the Ax21 protein processed and secreted outside the cell where it serves as a quorum sensing factor. Indeed it is almost impossible to envision a different interpretation of these results.
However there was a problem.
One of the new reviewers pointed out that extracellular Ax21 must traverse the outer membrane to bind to the candidate RaxH receptor, which was predicted to localize to the inner membrane. The reviewer therefore asked that we isolate the genes encoding the proteins that make up the protein complex that imports Ax21. Such experiments, however, would require months, if not years, of additional work. Although we agreed that the requested set of experiments were interesting (and were already underway), successful completion would clearly not alter the fundamental conclusions of the paper. In addition, although the editor was kindly willing to consider our response to reviews, the editor cautioned us that the next version of the paper would be sent out to a THIRD set of reviewers.
(With my academic compulsion for accuracy, I initially wanted to include the second set of reviews, verbatim, as part of this post. Conversations with colleagues and the editor convinced me otherwise. They felt it would rupture an important agreement: that the conservations between editor, reviewers and author would remain confidential).
Because of the work described in the paper had already been talked about in public forums and included in grant applications, and because publication was important for moving forward with our grant applications, job applications and other papers, we felt we could not spend another year in the review process. The very essence of the scientific process is to challenge paradigms and share the experimental details with other scientists who can then reproduce or refute the findings. Publication is key for this process. We needed to publish.
After much discussion in the lab and several more discussions with the editor at Science, we decided to take advantage of the explosion of open access journals as an alternative route to publication.
I asked the editors at PLoS One if they would consider an expedited review. After reading the 4 science reviews, they generously agreed. They sent the paper out for an eighth review, which also came in positive. Finally, the paper was accepted. It was a long road but we are very glad the work is now available to other scientists to test and tweak. We hope it will be widely read.
As one distinguished colleague pointed out, ultimately it may not matter where the results are published; if it has legs, it will stand.
Han, S., Sriariyanun, M., Lee, S., Sharma, M., Bahar, O., Bower, Z., & Ronald, P. (2011). Small Protein-Mediated Quorum Sensing in a Gram-Negative Bacterium PLoS ONE, 6 (12) DOI: 10.1371/journal.pone.0029192