STOICHIOMETRIC INTERPRETATION OF ESCHERICHIA-COLI GLUCOSE CATABOLISM UNDER VARIOUS OXYGENATION RATES

Metabolic by-product secretion is commonly observed in oxygen-limited cultures. Oxygen limitations occur because of limits in the capacity of the respiratory system or because of the oxygenation limits of the cultivation method used. The latter restriction is of considerable practical importance since it results in a critical cell concentration above which oxygenation is insufficient, leading to by-product secretion. In this study we used a flux balance approach to determine optimal metabolic performance of Escherichia coli under variable oxygen limitations. This method uses linear optimization to find optimal metabolic flux patterns with respect to cell growth. Cell growth was defined as precursor requirements on the basis of a composition analysis. A growth-associated maintenance requirement of 23 mmol of ATP per g of biomass and a non-growth-associated maintenance value of 5.87 mmol of ATP per g (dry weight)-h were incorporated on the basis of a comparison with experimental data. From computations of optimal growth increased oxygen limitations were found to result in the secretion of acetate, formate, and ethanol in that order. Consistent with the experimental data in the literature, by-product secretion rates increased linearly with the growth rate. The computed optimal growth under increasing oxygen limitation revealed four critical growth rates at which changes in the by-product secretion pattern were observed. Concomitant with by-product secretion under oxygen limitations were changes in metabolic pathway utilization. The shifts in metabolism were characterized by changes in the metabolic values (computed as shadow prices) of the various redox carriers. The redox potential was thus identified as a likely trigger that leads to metabolic shifts. Below a critical oxygen limitation of 0.88 mmol/g (dry weight)-h the anaerobic by-products were secreted at all growth rates, thus defining the upper limits of anaerobic growth. The general agreement between the optimal metabolic results derived here and experimental observations may prove to have evolutionary and ecological significance for stoichiometrically optimal metabolism.