2.1. Genome-scale metabolic models
In this study, the genome-scale models of iMM904 (Mo, Palsson, & HerrgÄrd, 2009), iEM439 (E Motamedian, Saeidi, & Shojaosadati, 2016), and iJO1366 (Orth et al., 2011) for S. cerevisiae, Z. mobilis,and E. coli were used, respectively. The intracellular reversible reaction fluxes were constrained between -1000 to 1000 mmol/gDCW/h, while intracellular irreversible reaction fluxes had zero lower limits. For the simulation of growth on minimal glucose media by the three metabolic models, lower bounds of exchange reactions were set to zero except for glucose, NH4, H2O, SO4, O2, H+, phosphate and inorganic ions exchange reactions. Growth on glucose was simulated by the maximum uptake rate of 10 mmol/gDCW/h. Flux balance analysis (FBA) was used to simulate the growth, and the biomass reaction was used as an objective function to be maximized in all in silicoexperiments. MATLAB (R2017b) was used for modeling using the COBRA Toolbox, and the GLPK (GNU Linear Programming Kit) package was used to solve linear programming problems. The reconstructed metabolic models were presented at pH=7 and so, the metabolic models were modified at other pH levels according to the method presented in the next section. In this research, pH levels of 5, 6, and 7 were selected for simulation based on the reported intracellular pH range for S. cerevisiae(Valli et al., 2005), Z. mobilis (Kalnenieks, Pankova, & Shvinka, 1987), and E. coli (Martinez et al., 2012). The unrealistic intracellular pH level of 5 for E. coli only was also assumed for the intracellular medium of the three cells to study the effect of high acidity.