Amino Acid Metabolism and Fluxes

G. WESLEY HATFIELD

[SECTION EDITOR: G. WESLEY HATFIELD]

Posted August 20, 2008

Microbiology and Molecular Genetics, University of California—Irvine, Irvine, CA 92697-4025

Mailing address: Microbiology and Molecular Genetics, C-237 Medical Sciences I, University of California—Irvine, Irvine, CA 92697-4025. Phone: (949) 824-5344, E-mail: This e-mail address is being protected from spambots. You need JavaScript enabled to view it

INTRODUCTION

By the mid-1960s, the pioneering work of Umbarger (10) and Gerhart and Pardee (2) had shown us that carbon flow through a biosynthetic pathway was controlled by allosteric inhibition (5) of the first enzyme of the pathway by its end product; and, studies of the lac operon by Jacob and Monod (4) had established that genes were controlled by an operator-repressor mechanism. Thus, in 1965 it seemed that it should take only a few more years to gain a complete understanding of the integration and regulation of metabolic pathways, or, as we say today, the systems biology of Escherichia coli. This, of course, was not to be the case. During the intervening forty-plus years, knowledge and technologies have continued to explode in unanticipated ways. We have learned new mechanisms and hierarchical levels of gene regulation. We have learned how changes in environmental, physical, and nutritional conditions affect gene expression and metabolism. Today, thanks to the lifetime dedication of scientific pioneers like Bruce Ames, Georges Cohen, Laszlo Csonka, Luigi Gorini, Nicholas Kredich, Werner Maas, Boris Magasanik, Harris Moyed, Ronald Somerville, Earl and Thressa Stadtman, H. E. Umbarger, Herbert Weissbach, and Charles Yanofsky, among others, and their many scientist progeny, we understand in great detail the molecular mechanisms of the many levels of metabolic and genetic regulation that control carbon flow through the amino acid biosynthetic pathways described in this section. However, as evidenced by the fact that each amino acid remains treated as a separate chapter, it is apparent that we still have not gained a satisfactory appreciation of how these biosynthetic pathways with their multiple levels of metabolic and genetic controls are integrated into the overall metabolism of a cell. This is our current challenge.

It is now clear that traditional experimental approaches are not sufficient for the integration and reconstruction of complex biological systems using the vast amounts of data mostly generated by high-throughput experiments. Only with computational methods and adequate modeling tools will we be able to reconstruct and query these large and complicated systems. Today, computational modeling approaches can be divided broadly into two categories. The first, which is the current focus of much systems biology research, is a top-down, discovery-oriented study of high-dimensional data sets to infer relationships and identify participating components in a network. The second, mainly a bottom-up approach that can be applied to metabolism, relies on generating models that can best describe experimental data and use them to generate hypotheses that can be tested in the laboratory. In other words, available modeling tools for metabolic pathways encompass different levels of complexity, ranging from top-down tools that describe topological interactions of the components of a metabolic network to bottom-up mechanistic models that simulate regulation of metabolic systems at the enzyme mechanism level.

Because of complicated enzyme reaction mechanisms and the frequent lack of rate constant measurements needed for solving differential equations, most investigators have turned their attention to the development of abstract, top-down modeling tools. For example, Palsson and colleagues have used metabolic flux balance analysis (FBA) methods to simulate steady-state metabolite flux through E. coli pathways representing hundreds of enzyme steps (8). An advantage of FBA is that it can make predictions about large metabolic networks without knowing anything about individual kinetic parameters or pathway regulation. It assumes that for each node of the network the entrance and exit rates must equal one another. In this case, a simple set of equations can be generated stating that total flux into a node minus the total flux out of the node must equal zero. Thus, FBA can provide valuable information about biomass conversions. However, since FBA does not consider pathway-specific regulation patterns it is less suited for the simulation of biochemical or genetic perturbations that require detailed knowledge of enzyme kinetic and regulatory mechanisms. These limitations have been addressed by more advanced flux-based modeling systems that consider the effects of enzyme concentrations, feedback regulators, and reversible enzyme reactions (1, 3).

More recently, Yang et al. (7, 11, 12) have developed a bottom-up, enzyme mechanism modeling language, kMech (kinetic mechanism), for the mathematical simulation of metabolic pathways. In contrast to FBA, this is a nonlinear, dynamic, enzyme-centric, mathematical modeling system that is focused on enzyme mechanisms rather than metabolic fluxes. Each enzyme mechanism is parsed into a set of fundamental association–dissociation reactions that are translated into a system of ordinary differential equations (ODEs) (9). kMech is built in a modular manner such that: (i) complex enzyme mechanisms can be modified or extended from simpler ones, (ii) these enzyme models can be assembled into pathways, and (iii) these pathways can be integrated into larger biological networks. This modular approach provides sufficient versatility to accommodate complex enzyme catalytic and regulatory mechanisms of individual enzymes and multienzyme complexes found in biological systems. The use of these kMech mathematical models to simulate the enzyme mechanisms for the biosynthesis of the branched chain and the threonine amino acid pathways is illustrated in Chapter Biosynthesis and Regulation of the Branched-Chain Amino Acids. Most recently, Najdi (6) has completed models for the enzyme and multienzyme complexes of the pathways of central metabolism that accurately simulate carbon flow from glucose to the aspartate and pyruvate family amino acids.

CONCLUSIONS

As we read through the chapters in this section summarizing the past 50 years of research elucidating the physiology, biochemistry, and genetics of amino acid biosynthesis in E. coli and Salmonella spp., we are overwhelmed by the wealth of detailed descriptions of metabolic and operon-, regulon-, and stimulon-specific genetic regulatory mechanisms that control and coordinate the functions of these individual pathways with the overall metabolism of the cell. Today, a new generation of interdisciplinary investigators from the computational and the life sciences are beginning to work together to assimilate this information into coherent systems. Nowhere is this systems biology approach initially more appropriate than for the central metabolism and amino acid biosynthesis systems of E. coli so thoroughly described in the chapters in this section. As new high-throughput tools for obtaining even more information from E. coli cells, such as the in vivo concentrations of enzymes and metabolites in individual cells during different nutritional and environmental growth conditions, and computational modeling tools become more sophisticated and powerful, this goal will become ever more attainable.