InterJournal Complex Systems, 100
Status: Accepted
Manuscript Number: [100]
Submission Date: 971009
Systems properties of metabolic networks
Author(s): David Fell

Subject(s): CX.30

Category: Review Article


Metabolism is the collection of chemical interconversions carried out by living organisms as they feed, grow and reproduce. It is the overall sum of several thousand individual chemical reactions brought about by catalytic proteins called enzymes. This set of reactions forms a highly branched network in which the cross-connections are increased because the majority of reactions involve at least two reactants. Although each enzyme reaction consists of a number of elementary steps obeying linear chemical kinetics, its overall kinetic behaviour is nonlinear in the concentration of the chemical intermediates of the metabolic net. Thus a possible description of the dynamics of the network is as a system of nonlinear ordinary differential equations, though severe practical problems confront attempts to model even small parts of the metabolic net. In spite of this complexity, the structure of the network imposes linear constraints on its attainable steady states, and these constraints can be analysed independently of the nonlinear kinetics, for example to give maximum conversion efficiencies. Further, even though algebraic solution of the nonlinear equation system is intractable even for short sequences of steps, it has proved possible to establish some general principles governing the sensitivity of steady state properties to the catalytic activities of individual enzymes. These studies in 'metabolic control analysis' have shown that the total amount of control that can be exerted by enzymes on the steady state fluxes through the network is limited and distributed between the enzymes. This distribution of control between the enzymes of the network can be related to their kinetic responses in the vicinity of the steady state even without having a global description of each enzyme's kinetics. By such means, it is them possible to derive results governing the control, genetics and evolution of metabolic pathways, and also to determine the patterns of interactions within the network that maintain the best homoeostasis. Within the totality of metabolism, there is a hierarchical organisation in which relatively isolated modules have partially separable time scales. It has been shown that in principle this partitioning of the network can be reflected in partitioning of the control properties.

Retrieve Manuscript
Retrieve Previous Revision's Abstract
Submit referee report/comment

Public Comments: