How Much Is It? Predicting Enzyme Costs with a Genome Scale Model of Yeast: Applications to Metabolic Engineering

Genome scale modelling (GEM) is a recurrent and successful approach for modelling metabolic fluxes of Saccharomyces cerevisiae (budding yeast) and predicting genetic modifications for increased performance of mutant strains¹. However, when considering the production of a metabolite of interest, these models typically make the major assumption that the uptake rate of the carbon source (e.g. glucose) is what limits production. This is hardly the case, since experimental yields are usually considerably lower than the maximum theoretical yields. Instead, what is likely happening in several cases is that there is a limitation of the amount of enzymes from the relevant pathways. GEMs of yeast so far are not able to encapsulate this phenomenon, because they do not include a direct way of connecting enzyme levels to metabolic fluxes. The latter should give insight into cellular capabilities at conditions of high enzymatic demand, such as yeast growing at a high growth rate or under stress, consuming non-typical carbon sources and/or overexpressing a given pathway.

Here we present the enhancement of a yeast GEM² to account for enzymatic constraints. It is done with the constraint-based approach³ extended to include enzymes as part of reactions, using mass balances for both metabolites and enzymes⁴. With this formalism reaction fluxes are limited by the maximum flux the enzyme is able to catalyse (v_max), which can be estimated from the enzyme’s abundance inside the cell together with the enzyme’s turnover number (k_cat)⁵. Using this approach with different levels of data, we show that we can predict metabolic strategies that regular constrained-based approaches cannot and significantly reduce variability of flux predictions. Most importantly, with different case studies we show that we gain insight into the distribution of enzyme costs inside any metabolic pathway and under different conditions. This way we can understand which could be the key steps that limit yeast growth/metabolite production in metabolic engineering applications.

References:

1         B. J. Sánchez and J. Nielsen, Integr. Biol., 2015, 7, 846–858.

2         H. W. Aung, S. A. Henry and L. P. Walker, Ind. Biotechnol., 2013, 9, 215–228.

3         N. E. Lewis, H. Nagarajan and B. Ø. Palsson, Nat. Rev. Microbiol., 2012, 10, 291–305.

4         E. J. O’Brien and B. Ø. Palsson, Curr. Opin. Biotechnol., 2015, 34, 125–134.

5         R. Adadi, B. Volkmer, R. Milo, M. Heinemann and T. Shlomi, PLoS Comput. Biol., 2012, 8.