Improving ethanol production by studying the effect of pH using a
modified metabolic model and a systemic approach
pH is an important factor affecting the growth and production of
microorganisms; especially, it is effective on the efficiency of
ethanologenic microorganisms. It can change the ionization state of
metabolites via the change in the charge of their functional groups that
may lead to metabolic alteration. Here, we estimated the ionization
state of metabolites and balanced the charge of reactions in
genome-scale metabolic models of Saccharomyces cerevisiae, Escherichia
coli, and Zymomonas mobilis at pH levels 5, 6, and 7. The robustness
analysis was first implemented to anticipate the effect of proton
exchange flux on growth rates for the constructed metabolic models at
various pH. In accordance with previous experimental reports, the models
predict that Z. mobilis is more sensitive to pH rather than S.
cerevisiae and the yeast is more regulated by pH rather than E. coli.
Then, a systemic approach was proposed to predict the pH effect on
metabolic change and to find effective reactions on ethanol production
in S. cerevisiae. The correlated reactions with ethanol production at
predicted optimal pH in a range of proton exchange rates determined by
robustness analysis were identified using the Pearson correlation
coefficient. Then, fluxes of these reactions were applied to cluster the
various pHs by principal component analysis and to identify the role of
these reactions on metabolic differentiation because of pH change.
Finally, 12 reactions were selected for up and down-regulation to
improve ethanol production. Enzyme Regulators of the selected reactions
were identified using the Brenda database and 11 selected regulators
were screened and optimized via Plackett-Burman and 2-level full
factorial designs, respectively. The proposed approach has enhanced
yields of ethanol from 0.18 to 0.36 mol/mol carbon. Hence, not only a
comprehensive approach for understanding the effect of pH on metabolism
was proposed in this work, but also it successfully introduced key
manipulations for ethanol overproduction.