Variable sizes of Escherichia coli chemoreceptor signaling teams
Robert G Endres1,2,a, Olga Oleksiuk3,a, Clinton H Hansen4, Yigal Meir5, Victor Sourjik3 & Ned S Wingreen1
- Department of Molecular Biology, Princeton University, Princeton, NJ, USA
- Division of Molecular Biosciences and Centre for Integrated Systems Biology at Imperial College, Imperial College London, London, UK
- Zentrum für Molekulare Biologie der Universität Heidelberg (ZMBH), Heidelberg, Germany
- Department of Physics, Princeton University, Princeton, NJ, USA
- Department of Physics, Ben Gurion University, Beer Sheva, Israel
Correspondence to: Victor Sourjik3 Zentrum für Molekulare Biologie der Universität Heidelberg (ZMBH), Im Neuenheimer Feld 282, Heidelberg, Germany. Tel.: +49 622154 6858; Fax: +49 622154 5894; Email: v.sourjik@zmbh.uni-heidelberg.de
Correspondence to: Ned S Wingreen1 Department of Molecular Biology, Princeton University, Washington Road, Princeton, NJ 08544-1014, USA. Tel.: +1 609 258 1894; Fax: +1 609 258 8616; E-mail: Email: wingreen@princeton.edu
Received 3 April 2008; Accepted 21 June 2008; Published online 5 August 2008
aThese authors contributed equally to this work
Top of pageArticle highlights
- In vivo FRET data and a model for cooperative signaling by bacterial chemoreceptor-signaling teams revealed that signaling-team size increases from about 6 receptor dimers to about 20 receptor dimers with receptor modification, such as occurs during adaptation to an attractant.
- Sizes of receptor-signaling teams and other parameters are obtained with high confidence from noisy, but correlated data by Principal Component Analysis.
- The variable sizes of receptor-signaling teams are shown, in theory, to minimize the detrimental effect of noisy ligand concentration.
Synopsis
Bacterial chemotaxis is a much-studied model for how small organisms sense chemicals in their environment and process information. The chemotaxis network includes membrane-bound receptors with different chemical specificities, flagellated rotary motors, and a network that transduces signals from the receptors to the motors, allowing bacteria to swim toward nutrient chemicals and away from toxic chemicals. The bacterial chemotaxis network is well studied, and involves autophosphorylation of the receptor-bound kinase CheA upon stimulation by receptors, followed by phospho-transfer to the diffusible response regulator CheY. Once phosphorylated, CheY molecules can bind to the motors favoring brief changes from counterclockwise to clockwise rotation, which induce the cell to tumble and change direction. The result is longer swimming runs when cells are moving in a favorable direction. Furthermore, cells can adapt the kinase activity (and therefore the motor behavior) by covalent modification/demodification of the chemoreceptors by the enzymes CheR/CheB. Importantly, in the bacterium Escherichia coli thousands of chemoreceptors form large polar and lateral clusters composed of trimers of receptor dimers. In these large clusters, receptors are thought to interact beyond trimers of dimers to form larger receptor signaling teams (Box 1). These receptor signaling teams lead to remarkable sensitivity to changes in chemical concentration, the ability to integrate multiple chemical signals, and precise adaptation to persistent stimulation by chemical ligands. Here, we report the discovery of a new level of organization and adaptation. On the basis of our new in vivo FRET data, the size of receptor signaling teams increases by up to a factor of three with receptor modification, such as that occurs during adaptation to an attractant. Our analysis establishes a powerful approach to correlated noise based on principal component analysis (PCA) that is widely applicable to quantitative data sets. Specifically, PCA is used to obtain tight confidence intervals on important quantities such as the size of receptor signaling teams. Finally, we present a theory that the observed variation in signaling-team size is a novel adaptive mechanism to optimally measure noisy chemical concentrations.
Acknowledgements
We thank Fred Hughson, Tom Shimizu, Monica Skoge, and Chris Wiggins for helpful suggestions. All authors (except CHH) acknowledge funding from the Human Frontier Science Program (HFSP).
