A computational approach to design and evaluate enzymatic reaction pathways: Application to 1-butanol production from pyruvate

We present a new computational strategy for the design and evaluation of novel enzymatic pathways for the biosynthesis of fuels and chemicals. The approach combines the use of the Biochemical Network Integrated Computational Explorer (BNICE) framework and a structure-based screening method for rapid generation and evaluation of novel enzymatic reactions and pathways. The strategy is applied to a case study of 1-butanol production from pyruvate, which yielded nine novel biosynthetic pathways. Using screening criteria based on pathway length, thermodynamic feasibility, and metabolic flux analysis, all nine novel pathways were deemed to be attractive candidates. To further assess their feasibility of implementation, we introduced a new screening criterion based on structural complementarity using molecular docking methods. We show that this approach correctly reproduces the native binding poses for a wide range of enzymes in key classes related to 1-butanol production and provides qualitative agreement with experimental measures of catalytic activity for different substrates. In addition, we show that the structure-based methods can be used to select specific proteins that may be promising candidates to catalyze novel reactions.