Control of DNA structure and gene expression.

In the usual metabolic control theory, the concentrations of enzymes are considered to be parameters rather than variables, i.e., they remain constant as the system relaxes to a new steady state. They can only be reset by interventions. This type of control analysis is useful for understanding principles of metabolic control, and for understanding metabolic changes that are too quick or in too limited a metabolic system to involve changes in gene expression. In actual living systems, metabolic changes are often accompanied by changes in gene expression. In this contribution we shall illustrate how metabolic control analysis is enriched when gene expression is variable. To discuss the new principles emerging in control analysis with variable gene expression, we shall first discuss theoretical model systems. In the first, the number of genes is fixed, but the concentrations of mRNA and enzymes are determined by the activities of RNA polymerase, RNAases, ribosomes and proteases. In a second, there is feedback repression by a metabolite at the level of translation. New coefficients quantifying the strength of regulatory loops will be defined. Also coefficients that indicate to what extent these regulatory strengths themselves are controlled by system parameters, are defined and provided with a summation theorem. The experimental model system we employ, addresses the phenomenon that in prokaryotes, transcription rates are influenced by the extent of supercoiling of the DNA. This includes the transcription of the genes encoding the two enzymes (DNA gyrase and topoisomerase I) involved in the regulation of DNA supercoiling. In vitro the activity of DNA gyrase is influenced by the hydrolytic free energy of ATP. We shall present experimental evidence that the cellular free-energy state influences DNA supercoiling. We shall also discuss experiments inspecting the effect of active transcription on active DNA supercoiling. Also this system will be analyzed in terms of the control analysis with variable gene expression; here the four hierarchical levels (DNA, RNA, enzymes, metabolites) interact, adding complexity to the control analysis.