Dynamic modeling of the central metabolism of E. coli - Linking metabolite and regulatory networks

Abstract

Coupling complex regulatory and metabolic networks for the purpose of dynamic modeling requires knowledge of the quantitative kinetics of the participating reactions as well as the variation of parameters in the context of the physiological state of the system. This chapter aims at demonstrating the integration of the different networks for E. coli exposed to an increasing carbon limitation of a fed-batch process with constant feeding of the carbon and energy source glucose. Starting from a global observation of the response of the bacteria in terms of flux distribution and gene expression in the central metabolism, emphasis is given to the dynamic modeling of regulation phenomena in the catabolism. The cra regulon which is linked to the dynamic response of the metabolite fructose 1,6-bis(phosphate) serves as an example to introduce a new concept, in which the binding constants are estimated from DNA-binding site sequences of the regulatory proteins. By comparison of the nucleotide frequencies within the DNA-binding sites for the individual target genes of the regulon, it is possible to perform a reasonable estimation of the kinetic parameters. Results of these estimations are compared with experimentally observed transcript concentrations measured with the aid of quantitative PCR. In addition it is shown how these outputs of the regulatory networks can be linked to the maximal rates of the enzymes for the metabolic system of interest. The discussion of this issue is embedded within a critical assessment of different conceptual frameworks for modeling the metabolic network, which covers the spectrum of dynamic modeling at different levels of complexity, such as genome scale, modular approaches and reduced models. © 2009 Springer Netherlands.