30 March 2007
Yuhai Tu
IBM T J Watson Research Centre
Over the past 10 years, a great deal has been learnt about bacterial chemotaxis pathway at the quantitative level, propelled by new quantitative measurements (such as those using FRET in the Berg lab) and the corresponding quantitative modeling efforts. Yet, one major challenge remains to understand/predict the responses of the system to complex temporal signals, e.g., the temporal signal a cell experiences as it performs chemotaxis in a spatial gradient, at the molecular level. The subject of this talk reflects part of our efforts to link molecular level description of the pathway to the behavior of the cell.
Based on our current understanding of the molecular details of the pathway, we propose a simple kinetic model of the pathway to describe E. coli's response to arbitrary temporal signals. In our model, the description of the slow adaptation dynamics is based on a general feedback mechanism from receptor kinase activity, whose dependence on receptor methylation levels is in turn described by a cooperative allosteric model. We will use our model to explain quantitatively the response of E. coli to exponential ramps and exponentiated sine wave measured more than 20 years ago in [1], the beautiful (and still only) experiments with controlled non-trivial temporal stimuli. Our model also provides a natural solution/mechanism for the long standing (and forgotten?) puzzle observed more than 30 years ago by Spudich and Koshland in [2] and Berg and Tedesco in [3] on the additivity of the relaxation times for large amplitude stimuli.
If time permits, we will talk about the property of the pathway as a signal processor/filter to noisy signal; the implications of our findings on understanding the molecular level adaptation kinetics; the internal dynamics of the pathway as the cell perform chemotaxis in spatial gradients; and other future directions in our effort in linking molecular level description of the pathway to the behavior of the cell.