Elementary Mass Action Stoichiometric Simulation Models Predict Non-Negligible Concentrations of Enzyme-Bound Metabolites

Improvement of cell factory performance, to the point where production of chosen molecules becomes economically profitable, requires extensive metabolic modification. Metabolic models have been used to both provide new insights into the inner workings of metabolism and new directions for strain engineering. Kinetic models in particular are key to model the dynamics of metabolism and substrate-level enzyme regulation.

Here, we use the elementary Mass Action Stoichiometric Simulation (eMASS) framework to build a prototype kinetic model ensemble for glycolysis in E. coli. Following eMASS, we first build an individual kinetic model for each reaction by decomposing it into elementary reactions, according to the respective reaction mechanism. Next, the associated elementary rate constants are fitted to kinetic data obtained from the literature, e.g. k_cat, K_m, K_i, while satisfying the respective Haldane relationship. Finally, we assemble the system’s kinetic model by combining the individual reaction models and integrating fluxomics and metabolomics data. Since we model both free and enzyme-bound metabolite concentrations explicitly, we drop the common assumption that enzyme-bound metabolite concentrations are negligible (x_tot ≈ x_free) and predict the free metabolite concentrations.

Our preliminary results show that the percentage of enzyme-bound metabolite vs. total metabolite can be as high as 40%, i.e. the amount of enzyme-bound metabolite may not be negligible even in glycolysis. These results derive from the fact that in some cases the concentration of metabolites is in the same order of magnitude as the enzymes with which they interact.