Metabolic Engineering from a Cybernetic Perspective. 2. Qualitative Investigation of Nodal Architechtures and Their Response to Genetic Perturbation

Abstract

A cybernetic representation of the branch point development of Stephanopoulos and Vallino is formulated. The model systems are employed to translate the qualitative properties of the nodal control architectures characterized by Stephanopoulos and Vallino into a mathematical context. It is shown that a cybernetic model in which the objective is the independent maximization of the levels of branch point products is consistent with the characterization of a flexible node. In contrast, the rigid control architecture is shown to be equivalent to the maximization of the mathematical product of the branch point products. It has been demonstrated subsequently that cybernetic metabolic network models are capable of predicting the system response to enzymatic amplification. However, given the complicated nature of the subsequent models, a clear illustration of the basic mechanism by which such predictions are manifested is not forthwith. Thus, a second objective of the present work is the examination of the response of the flexible and rigid control architectures to genetic perturbation, specifically enzymatic overexpression, with the expressed aim of elucidating the mechanism by which a cybernetic model predicts metabolic network responsiveness. It is shown that the ramifications of genetic perturbation are transmitted through the cybernetic representation of a metabolic network via the resource allocation structure which acts as the conduit by which regulatory signals are transmitted to seemingly unconnected portions of the network. It is postulated that enzymatic overexpression under an artificial promoter represents, from the perspective of the microorganism, an uncontrollable resource drain that forces the metabolic network control architecture to reevaluate the standing resource allocation policy as implemented via the cybernetic control variables. In biological terms, the reevaluation of allocation policy implies a shift in the level and activity of network enzymes yielding, in some cases, qualitatively different network function. It is our position that, conceptually, this is equivalent to the conventional wisdom that genetic manipulation of a metabolic network is the impetus for shifts in the network functionality, i.e., enzyme levels as well as activity. Thus, this development provides a necessary intellectual precursor for the formulation and analysis of the model systems that follow.

A hybrid model of anaerobic E. coli GJT001: Combination of elementary flux modes and cybernetic variables

“…To overcome this problem, the cybernetic model framework has been adapted to simulate more detailed metabolic networks to predict most fermentation byproducts through metabolism. In adapting the framework to a large network, Varner and Ramkrishna 12,13 developed an approach requiring identification of metabolic subunits each comprising elementary pathways as conceived by Straight and Ramkrishna. 14 The subunits were postulated to compete for the total resources available for synthesizing enzymes.…”

Mathematical Modelling as a Tool for Optimized PHA Production

“…That is why various kinetic and mechanistic simplifications are introduced in models. Another way to involve regulatory mechanisms related to cell metabolism is to apply cybernetic principles of modeling [125][126][127][128][129] . Thus, cybernetic variables introduced into a kinetic model substitute the unknown mechanistic factors of the cell regulatory system; they represent the different cellular processes: induction, repression, inhibition, and activation.…”

Mathematical Modeling of Plant Metabolic Pathways

“…When combined with genetic and environmental perturbations, it can yield valuable insights concerning the rigidity of flux splits at metabolic branch points and the existence of alternative pathways and can aid in the calculation of maximum theoretical yields (Stephanopoulos et al, 1998). Its chief limitations are that it may require restrictive assumptions concerning cofactor dependence, it does not address the degree of reversibility of reactions, and it is unable to compute fluxes via metabolic cycles (Stephanopoulos et al, 1998;Varner and Ramkrishna, 1999b).…”