Bacterial chemotaxis | Receptor clusters | Conformational spread | Outstanding issues

Receptor Clusters

One of the best understood regions of cytoplasm is that associated with the cluster of chemotactic receptors in the plasma membrane of E. coli (Maddock & Shapiro, 1993; Eisenbach, 1996; Falke et al., 1997) This signaling complex detects serine, aspartate and other substances in the cell's environment and transmits information on their concentrations into the cell in the form of a phosphorylation signal that regulates flagellar rotation. The receptors are believed to exist in thermal equilibrium between two or more conformational states, and the output of the complex, usually measured in terms of phosphorylation levels of CheA or its phosphoacceptor CheY, is related to this equilibrium (Parkinson, 1993; Mowbray & Sandgren, 1998). From the standpoint of signal processing, the complex produces an amplified output that, because of associated adaptational machinery, is proportional to the rate of change of attractant concentration. The molecular structures of most of its components are now known at atomic resolution, and there is a wealth of biochemical and mutational data on its function. Over the past several years, we and others have developed detailed computer simulations of the molecular events accompanying this signal transduction scheme that provide an increasingly strong theoretical underpinning for the mass of data that have accumulated.

In a recent study, we proposed an atomic level structure for a lattice of serine (Tsr) receptors in coliform bacteria (Shimizu et al., 2000). The model was based on the atomic level structures of the individual proteins together with information from biochemical, genetic and behavioural experiments. Docking of individual proteins and their spatial arrangement was assessed by means of plastic models generated by 3-D printer technology.

Part of the proposed lattice. The structure is a regular two-dimensional lattice in which the cytoplasmic ends of chemotactic receptor dimers insert into a hexagonal array of CheA and CheW molecules.

Section through the proposed receptor cluster, parallel to the plasma membrane and about 30 nm from it. The lattice is viewed as if looking into the cell. The tails of receptor dimers, in sets of three, insert into the hexagonal lattice of CheA and CheW.

A unique feature of this model is that it creates a small compartment between the plasma membrane and an extended hexagonal lattice of the signaling proteins CheA and CheW. The proposed compartment, about 30 nm deep and 300-400 nm wide, contains a thicket of extended coiled-coils forming the cytoplasmic domains of the chemotactic receptors. The compartment is not closed, and should be freely accessible to cytoplasmic proteins diffusing in from the lateral borders or through 10 nm diameter pores in the hexagonal lattice. Despite the absence of sealed boundaries, however, there is reason to think that this minute volume of bacterial cytoplasm will be highly enriched in two diffusible proteins, CheR and CheB, which are responsible for adaptation in the bacterial system. These two enzymes control the methylation of chemotactic receptors at sites in the middle of the receptor tails and have also been shown to carry binding sites for the C-termini of the receptor tails. It seems reasonable to suppose that they will accumulate in the small compartment because of this binding, and that their elevated concentrations in that privileged volume will facilitate the methylation and demethylation reactions.

Conformational spread in the cluster

We have recently examined the consequences of a postulated mechanism in which conformational changes in a lattice of receptors can spread laterally (see Conformational spread). All of the analyses of this mechanism to date have been simplified in various ways to facilitate mathematical analysis. Thus, proteins have been assumed to be arranged in a closed ring or infinite square lattice and were assigned properties that were as simple as could be. We have been extending the analysis of conformational spread to a more realistic model of conformational changes in E. coli chemotactic receptors. Current simulations using the StochSim program include up to 4 methyl groups per receptor, significant dwell times for ligand and conformationally sensitive binding affinities for CheR and CheB. These analyses have so far shown that the enhanced sensitivity and range of response shown by the Duke-Bray model due to coupling between receptors is retained in the more complete description, but changed in significant ways. The presence of a range of methylation states means that there is not a single critical value of the coupling strength but rather a range of values over which effective performance is enhanced. The increase in sensitivity (chemotactic gain) obtained from a StochSim simulation is less dramatic than that in the idealized single methyl group model, but less "brittle" in the sense that a significant improvement in performance is obtained over a range of possible energy values. Most intriguingly, we have found that the dynamics of the situation lead to the spontaneous emergence of order in the receptor lattice, such that receptors with 4 methyl groups (fully methylated) and receptors with 0 methyl groups (fully unmethylated) tend to lie next to each other in the array. This is only a minimal degree of order, and could be an artifact of the simulation. However the spontaneous emergence of order within a stochastically fluctuating field of allosteric proteins is an intriguing and potentially important phenomenon.

The next step, we believe, is to incorporate specific structural information into the simulations of conformational spread. Until we do this, we will not be able to gain a credible picture of the integrated behaviour of the lattice. Of course, there are many uncertainties about the precise arrangement of proteins and the details of their conformational changes — it would be impossible at this stage to build a computer model that was correct in every respect. Nevertheless, we believe that if we can come reasonably close to the actual situation, then we may learn about the kinds of dynamic behaviour that are possible. For example, we might consider the three-chain assembly of receptors to be a conformational unit, whose twists and bends are closely coupled. Changes in this "unit" would then spread radially along three equally-spaced arms of protein out to adjoining CheW and CheA molecules in the lattice. Propagation of changes from one receptor complex to a neighbouring complex would take place through individual CheA molecules, from one monomer to an adjacent monomer. We might begin, therefore, by assigning different values of coupling energy to these different protein interactions and then examining the consequences for an array with the geometry of the receptor lattice. With any luck, the models we build will be sufficiently rich in detail (including realistic values of time delays) to generate realistic higher-order multiprotein behaviour.

References

Eisenbach, M. (1996). Control of bacterial chemotaxis. Mol. Microbiol. 20, 903-910.

Falke, J. J., Bass, R. B., Butler, S. L., Chervitz, S. A., & Danielson, M. A. (1997). The two-component signaling pathway of bacterial chemotaxis: a molecular view of signal transduction by receptors, kinases, and adaptation enzymes. Annu. Rev. Cell Dev. Biol. 13, 457-512.

Maddock, J. R., & Shapiro, L. (1993). Polar location of the chemoreceptor complex in the Escherichia coli cell. Science 259, 1717-1723.

Mowbray, S. L., Sandgren, M. O. J. (1998). Chemotaxis receptors: a progress report on structure and function. J. Struct. Biol. 124, 257-275.

Parkinson, J. S. (1993). Signal transduction schemes of bacteria. Cell 73, 857-871.

Shimizu, T. S., Le Novère, N., Levin, M. D., Beavil, A. J., Sutton, B. J., & Bray, D. (2000). Molecular model of a lattice of signalling proteins involved in bacterial chemotaxis. Nat. Cell Biol. 2, 792-796.

Site designed by Matthew Levin. This page updated 30 April 2007.