Fold-change detection as a chemotaxis model discrimination tool

Abstract

Fold-change detection (FCD) is the property that a dynamical system with an adapting output will exhibit identical transient output responses when its input signals are scaled. This feature was recently demonstrated in the chemotactic response of the bacterium Escherichia coli, confirming earlier theoretical predictions. The chemotaxis pathway of the bacterium Rhodobacter sphaeroides has the same modular structure as that in E. coli but is significantly more complex in that it has multiple homologues of the latter's chemotaxis proteins and features two, rather than one, chemosensory cluster. Recent experimental results suggest that R. sphaeroides may also exhibit FCD. In this paper, we present a set of theoretical assumptions on the dynamics of the R. sphaeroides chemosensory system, and use these to fit an integrated chemotaxis model to experimental data. We then show that the assumptions we place are sufficient to make FCD a robust property of this chemotaxis pathway, in agreement with preliminary experimental evidence. We argue that the fact that the model we present here is able to reproduce this transient dynamic property whilst earlier models cannot makes FCD a useful tool for model discrimination on the basis of transient dynamic response. This is in contrast to earlier model discrimination methods which tested the validity of models based on their ability to reproduce a finite set of experimental data. Further experiments that can provide additional validation of our theoretical assumptions are suggested.