Systematic Modeling and Identifiability Analysis of 2,3-Butanediol Biosynthesis by Metabolically Engineered Klebsiella oxytoca Using Glucose/Xylose Cosubstrates.

Comprehensive mathematical models for 2,3-butanediol (2,3-BDO) production by metabolically engineered Klebsiella oxytoca KMS006 were developed by integrating a modified Monod model with the Luedeking-Piret equation to describe microbial growth kinetics, substrate consumption, and metabolite formation (2,3-BDO, succinate, and acetate) during batch fermentation with varying glucose and xylose concentrations. Both stochastic and deterministic simulations were performed, with model calibration achieved through a least-squares minimization algorithm. A total of 18 kinetic parameters were estimated and rigorously assessed using sensitivity and identifiability analyses, confirming their robustness and predictive reliability. The models exhibited strong agreement with experimental data, accurately capturing fermentation dynamics across a range of substrate conditions. For cosubstrate utilization, the previously optimized parameters were incorporated into an extended kinetic framework, which successfully simulated simultaneous glucose and xylose metabolism. This study establishes one of the first validated kinetic models for mixed-sugar fermentation in K. oxytoca, providing a robust tool for process optimization, metabolic engineering, and the scalable production of 2,3-BDO from lignocellulosic hydrolysates.