Kinetic Modeling and Isotopic Investigation of Isobutanol Fermentation by Two Engineered Escherichia coli Strains

We constructed an Escherichia coli BL21 strain with the Ehrlich pathway (the low-performance strain for isobutanol production). We also obtained a high isobutanol-producing E. coli strain JCL260 from the James Liao group (University of California). To compare the fermentation performances of the two engineered strains, we employed a general Monod-based model coupled with mixed-growth-associated isobutanol formation kinetics to simulate glucose consumption, biomass growth, and product secretion/loss under different cultivation conditions. On the basis of both kinetic data and additional 13 C-isotopic investigation, we found that the low-performance strain demonstrated robust biomass growth in the minimal growth medium (20 g/L glucose), achieving isobutanol production (up to 0.95 g/L). It utilized significant amounts of yeast extract to synthesize isobutanol when it grew in the rich medium. The rich medium also enhanced waste product secretion, and thus reduced the glucose-based isobutanol yield. In contrast, JCL260 had poor biomass growth in the minimal medium due to an inflated Monod constant ( K S ), while the rich medium greatly promoted both biomass growth and isobutanol productivity (∼60% of the theoretical isobutanol yield). With the optimized keto-acid pathway, JCL260 synthesized isobutanol mostly from glucose even in the presence of sufficient yeast extract. This study not only provided a kinetic model for scaled-up isobutanol fermentation but also offered metabolic insights into the performance trade-off between the two engineered E. coli strains.