Investigating the Warburg Effect Using a Core Metabolic Network with Kinetic Parameters

Cancer cells show enhanced metabolic activity, most notably through the aerobic production of lactate, the Warburg effect. This effect has been reproduced by others [1] through a genome scale metabolic model (GEM), that takes the cellular capacity for metabolic enzymes in to account. That approach requires that the turnover number (kcat) of each protein is known. Unfortunately kcat’s are only available for a fraction of the enzymes in a GEM (17.5% in the case of [1]).

We therefore developed a method that reduces a GEM to a core metabolic network (e.g. glycolysis, pentose phosphate pathway, TCA, oxidative phosphorylation and amino acid metabolism), whilst retaining the ability to produce and consume macromolecules through lumped reactions.

Using this method we generated a model with high kcat coverage (87% of the reactions from mammalian enzymes, out of which 75% from human). The model is used to answer questions about shifts in resource allocation under increased growth rate. Specifically focusing on nitrogen balance and the regeneration of ATP, NADH, NADPH and acetyl-CoA.

Preliminary results show that the Warburg effect can be predicted by the model, in agreement with [1], and that the relative distribution of mass amongst the enzyme complexes in the electron transport chain can be predicted by the model.

[1] Shlomi, T., Benyamini, T., Gottlieb, E., Sharan, R., & Ruppin, E. (2011). Genome-scale metabolic modeling elucidates the role of proliferative adaptation in causing the Warburg effect. PLoS Computational Biology, 7.