Using control bias to identify initial targets for bioproduction improvement

Abstract

Sensitivity analysis of bioprocess metabolic reaction networks analysis allows the prediction of system parameters such as those associated with the enzyme activity of certain reaction steps which significantly affect the overall production. However, uncertainties in kinetic rate expressions and in the resulting steady-state flux distributions limit the accuracy of these predictions. Starting from minimal information (reaction stoichiometry, and external fluxes in/out of the system and potentially identification of steps at equilibrium) a new preliminary method is proposed using sampling of elasticities and metabolic fluxes to calculate the control bias. The calculated control bias identifies steps which are likely to have positive control, negative control or negligible/uncertain control. This is intended to give initial guidance before further detailed investigation is carried out, identifying targets for any organism to enhance production of valuable chemicals. As a case study, this methodology is applied to succinic acid bioproduction using Actinobacillus succinogenes and analysis successfully reveals the reaction steps having the greatest positive and negative influence on biosuccinic acid production.