The dynamics of microbial growth on mixtures of substrates

Atul Narang, Purdue University

Abstract

This work is concerned with the experimental investigation and mathematical modelling of bacterial growth on mixtures of substitutable substrates. The model system for the experiments consists of a batch culture of E. coli growing on a synthetic medium containing a binary mixture of carbon and energy sources. Most of the carbon and energy sources considered here belong to the class of compounds referred to as organic acids or carboxylates. These mixed substrate experiments show that the spectrum of substrate utilization patterns is richer than commonly believed. Past work in the literature has generally focussed on the diauxic growth pattern wherein the two substrates are utilized sequentially. The experiments reported in this work show that the substrates in the mixture are often utilized simultaneously. The experiments also show the existence of multiple physiological steady states. The growth rate of a culture in a medium containing glucose and pyruvate depends on the conditions under which the inoculum is precultured. The goal of the mathematical model is to capture the diauxie and the new patterns of substrate utilization revealed by the experiments. The cybernetic framework is first analyzed. It is shown that the cybernetic model fails to capture the diauxic growth pattern. It displays a dependence on the initial conditions that is inconsistent with experimental data. The modification that must be made to remedy this defect is identified. It is then shown that even though the dynamics of the modified cybernetic model agree with diauxic behavior, the model cannot accommodate the more complex patterns of substrate utilization shown by the experiments. A kinetic model is then formulated, which is able to capture the dynamics of the sequential and simultaneous substrate utilization patterns. The consequences of the model are discussed. It is shown in particular that it is not necessary to postulate regulation in order to obtain the all-or-none dynamics of the diauxie; it suffices to account for the autocatalytic nature of inducible enzyme synthesis. Finally, the deficiencies of the kinetic model are discussed. Inclusion of energetics and the ribosomal machinery in the model are suggested as plausible ways of extending the model.

Degree

Ph.D.

Advisors

Ramkrishna, Purdue University.

Subject Area

Chemical engineering|Microbiology|Mathematics

COinS