(105f) Listening to New Signals with Old Ears: Sensing and Recognition of the Quorum Sensing Signal AI-2 by the Tsr Chemoreceptor

Bacteria constantly monitor the concentration of specific molecules in their external environment using cell-surface receptors. By comparing the current concentration to the concentration detected a few seconds earlier, bacteria direct their movement to approach sources of attractant chemicals or to avoid sources of repellent chemicals. This movement, termed chemotaxis, is potentially an important determinant of infections, because the migration of pathogens to epithelial cell surfaces is an essential first step in gastrointestinal tract infections. Since the gastrointestinal tract is typically heterogeneous and contains multiple chemoeffectors with potentially opposing effects, the integrated chemotactic response in such environments is likely to be an important factor in bacterial infection. Understanding the mechanisms underlying resolution of different chemotaxis signals is the long-term objective of this work.

Here we present results on the mechanism utilized by E. coli for sensing the inter-species communication signal autoinducer-2 (AI-2). AI-2 is known to be produced by nearly 80 bacterial species, and our prior work has shown it to be a chemoattractant for E. coli (Appl. Environ. Microbiol. 75: 4557-64, 2009; Appl. Microbiol. Biotechnol. 78: 811-9, 2008). In this study, we used E. coli RP437 with different single and multiple chemoreceptor knockouts of its four primary chemoreceptors (Tsr, Tar, Tap, Trg) to show that chemotaxis towards AI-2 is mediated through the Tsr receptor. A novel microfluidic chemotaxis model and fluorescence microscopy were used to quantify the extent of migration of E. coli in stable gradients of AI-2. Our data demonstrate that Tsr is both necessary and sufficient for sensing AI-2. Since E. coli responded to a shallow (i.e., linear) gradient of AI-2 but required a much steeper (non-linear) gradient of serine (which is the canonical chemoeffector sensed by E. coli), we conclude that the sensitivity of Tsr to AI-2 is significantly higher than that to serine.

Using isogenic mutant strains, we also show that AI-2 does not need to be internalized into the cytoplasm as a lsrC mutant defective in AI-2 uptake exhibited normal chemotaxis. This result suggests that the interaction between AI-2 and Tsr occurs in the periplasmic space. We further investigated whether AI-2 binds directly to Tsr or whether interactions with the receptor require other binding proteins. Our data show that the periplasmic AI-2 binding protein LsrB, that is part of the AI-2 uptake system, is required for chemotaxis towards AI-2. We conclude that LsrB, when bound to AI-2, interacts directly with the periplasmic domain of Tsr, making LsrB the first known periplasmic protein partner for Tsr, and points to a chemosensing model similar to that for maltose, galactose, ribose, and dipeptides (i.e., in which a binding protein is involved in chemotaxis.