BioModels Database logo

BioModels Database

spacer

Bray and Bourret (1995), Bacterial Chemotaxis

September 2009, model of the month by Melanie I. Stefan
Original model: BIOMD0000000200

Bacteria such as E. coli can move along chemical gradients, a process known as chemotaxis. An E. coli cell switches back and forth between two modes of movement: A "straight" mode in which the bacterium swims straight in one direction and a "tumble" mode in which the bacterium turns around by some acute angle before taking up straight swimming again [1]. The bacterium can direct itself towards a chemical attractant by modulating the relative length of straight swimming and tumbling phases [1,2]. On a cellular level, the modes of movement are determined by the direction in which the flagella rotate: Counter-clockwise rotation results in straight swimming, while clockwise rotation results in tumbling [3] (see figure 1).

Direction of flagellar rotation controls motility in E. coli

Figure 1: The direction of flagellar rotation determines straight swimming or tumbling motion. Figure adapted from [4].

Chemotactic signalling pathway in E. coli

Figure 2: Chemotactic signalling pathway in E. coli. Figure taken from [5].

The direction of flagellar rotation is controlled by an intracellular signalling cascade triggered by binding of the chemotactic signal to a transmembrane receptor. In the case of aspartate chemotaxis (reviewed in [5]), aspartate binds to the dimeric receptor Tar (TT), which forms a complex with the two cytoplasmic proteins CheW (W) and CheA (AA). CheA can autophosphorylate depending on the ligand binding state of Tar: If aspartate is bound, autophosphorylation of CheA decreases. Phosphorylated CheA can transfer its phosphoryl group to CheY, which, when phosphorylated, binds to the flagellar motor protein FliM and induces a change in flagellar rotation from counterclockwise to clockwise. Thus, in the absence of a chemotactic attractor, the cell is more biased towards a tumbling motion than towards straight swimming. Alternatively, phosphorylated CheA can phosphorylate CheB, which modifies receptor methylation and thereby controls receptor strength. A schematic representation of the chemotactic signalling pathway is shown in figure 2.

Components of this pathway have been deleted or overexpressed in the past in order to learn from mutant phenotypes, some of which are quite surprising (reviewed in [5]). One might expect mutants overexpressing A to have higher levels of phosphorylated AA and hence show a stronger bias towards clockwise flagellar rotation (and hence, tumbling motion) than wildtype. In fact, the opposite is the case: Mutants overexpressing A show a stronger preference for counter-clockwise rotation (and hence, straight swimming) than wildtype. If T is deleted, the direction of flagellar rotation s almost entirely counterclockwise, leading to a predominantly straight-swimming phenotype. Deletion of W has a similar effect, indicating that correct assembly of the TTWWAA complex is crucial in order to maintain wildtype swimming behaviour. Intriguingly, overexpression of T has the same effect as deletion, and the same holds for overexpression of W. However, co-overexpression of both TT and W restores wildtype flagellar rotation and swimming patterns. In order to understand how these mutant phenotypes arise it was therefore necessary to investigate how exactly the TTWWAA receptor complex is assembled.

The model by Bray and Bourret (BIOMD0000000200, [5]) builds on an earlier model (MODEL6929313478, [6]), but places particular attention to the formation of the TTWWAA complex. In order to uncover the sequence of binding steps by which this complex is formed, the authors compiled a list of all theoretically possible sequences (see figure 3) and then used a genetic algorithm to find those sequences and associated dissociation constants that were compatible with experimental data. This approach yielded four different possible reaction networks, which differ in the numbers of reactions and in the activities of some of the complexes. Simulations showed that all of these networks agreed with available experimental data, and that further experimental work was necessary to differentiate between them.

Importantly, all of the four possible reaction networks could reproduce the effects of both TT and W deletion and overexpression and provide an explanation of this effect: If T is overexpressed, it will mainly form the inactive complex TTW, thus hindering W from participating in active complex formation. In a similar manner, overexpression of W leads to an accumulation of TTWW and WWAA complexes, and hence a sequestration of all available T and A into inactive complexes. If both T and W are overexpressed, complexes between them (TTW and TTWW) do form, but this does not affect the formation of wildtype-like levels of TTWWAA. The formation of inactive complexes at the expense of the full TTWWAA complex also explains the phenotype associated with overexpression of A: The autophosphorylation state of AA is mainly controlled by the full TTWWAA complex, so although more AA is available for phosphorylation, less of it is actually phosphorylated, thus increasing the bias towards counter-clockwise flagellar rotation and thus, straight swimming.

Possible sequences of TTWWAA receptor complex assembly

Figure 3: Possible sequences of assembly of the TTWWAA complex. Figure taken from [5].

The model presented here was but one in a series of models addressing chemotaxis in E. coli. Many other models have been designed and published since, elucidating all aspects of bacterial chemotaxis, from the reaction and diffusion patterns of single proteins (e.g. [7,8]) to the chemotactic behaviour of an entire virtual bacterium (e.g. [9,10]). Taken together, these models provide us with a comprehensive understanding of bacterial chemotaxis and a way of relating changes at the molecular level to behaviours of entire organisms or even populations. The work on bacterial chemotaxis thus provides us with a prime example of how computational systems biology can bridge various levels of experimental analysis and thus greatly improve our understanding of a given biological system.

Bibliographic References

  1. H.C. Berg and D.A. Brown. Chemotaxis in Escherichia coli analysed by three-dimensional tracking. Nature, 239(5374):500-504, 1972. [SRS@EBI]
  2. R.M. Macnab and D.E. Koshland, Jr. The Gradient-Sensing Mechanism in Bacterial Chemotaxis. Proc Natl Acad Sci U S A, 69(9): 2509-2512, 1972. [SRS@EBI]
  3. H.C. Berg and P.M. Tedesco. Transient response to chemotactic stimuli in Escherichia coli. Proc Natl Acad Sci U S A, 72(8):3235-3239, 1975. [SRS@EBI]
  4. S.M. Butler and A. Camilli. Going against the grain: chemotaxis and infection in Vibrio cholerae. Nat Rev Microbiol, 3(8):611-620, 2005. [SRS@EBI]
  5. D. Bray and R.B. Bourret. Computer analysis of the binding reactions leading to a transmembrane receptor-linked multiprotein complex involved in bacterial chemotaxis. Mol Biol Cell, 6(10):1367-1380, 1995. [SRS@EBI]
  6. D. Bray, R.B. Bourret, and M.I. Simon. Computer simulation of the phosphorylation cascade controlling bacterial chemotaxis. Mol Biol Cell, 4(5):469-482, 1993. [SRS@EBI]
  7. S.S. Andrews and D. Bray. Stochastic simulation of chemical reactions with spatial resolution and single molecule detail. Phys Biol, 1(3-4):137-151, 2004. [SRS@EBI]
  8. K. Lipkow, S.S. Andrews, and D. Bray. Simulated diffusion of phosphorylated CheY through the cytoplasm of Escherichia coli. J Bacteriol, 187(1):45-53, 2005. [SRS@EBI]
  9. D. Bray, M.D. Levin, K. Lipkow. The chemotactic behavior of computer-based surrogate bacteria. Curr Biol, 17(1):12-19, 2007. [SRS@EBI]
  10. L. Zonia, D. Bray. Swimming patterns and dynamics of simulated Escherichia coli bacteria. J R Soc Interface, 6(40):1035-1046, 2009. [SRS@EBI]
spacer
spacer