Optimal Control of Fed‐Batch Fermentation with Autoinduction of Metabolite Production

Abstract

Induction of metabolite production is a common feature of many cultures obtained by recombinant DNA technology. Synthesis of a product in these cultures is triggered by a sharp change in environmental conditions or by autoinduction in which the cells, under conditions of growth limitation, produce a species that initiates the synthesis. If product synthesis is the objective of the fermentation, then the time of induction should be optimized so that enough biomass is produced prior to induction while ample nutrient is left in the broth to sustain product synthesis after induction. In this study, recombinant luminous Escherichia coli has been used to portray an autoinductive culture in batch and fed-batch fermentations. A model describing the cell density and substrate and inducer concentrations has been constructed, and its parameters were estimated. An optimal control strategy with three stages of the culture state (growth, inducer synthesis, and growth+product synthesis) was determined and validated experimentally by optimization of the substrate flow rate in a fed-batch fermentation.

Enhancement of Candida rugosa lipase production by using different control fed‐batch operational strategies

“…Yeasts' fed-batch processes, especially Saccharomyces cerevisiae, have been widely referenced in industrial production of heterologous proteins (Mendoza-Vega et al, 1994) and analysis of the respiro-fermentative growth in fed-batch fermentation has been applied for accurate parameter estimation (Ejiofor et al, 1994). Also, fedbatch operation is widely used in recombinant Escherichia coli cultures when it is necessary to maintain a specific growth rate below a critical value to avoid acetate production (Korz et al, 1995;Sakamoto et al, 1994;Tartakovsky et al, 1995). Despite these advantages, most of the lipase production processes are still performed in batch cultures where medium optimization and operation conditions are the main parameters studied Ohnishi et al, 1994;Rapp, 1995;Shimada et al, 1992) meanwhile fed-batch processes are scarcely used.…”

Stochastic Dynamic Optimization of Batch and Semicontinuous Bioprocesses

“…The determination of the optimal control profile of fedbatch bioreactors is an interesting and challenging dynamic optimization problem that has received major attention in recent years (see, for example, San and Stephanopoulos, 1984;Hong, 1986;Lim et al, 1986;Modak et al, 1986;Menawat et al, 1987;Johnson, 1987;Park and Ramirez, 1988;Cazzador, 1988;Modak and Lim, 1989;San and Stephanopoulos, 1989;Queinnec et al, 1991;Meszaros and Bales, 1992;Modak and Lim, 1992;Parulekar, 1992;Van Impe et al, 1992;Hansen et al, 1993;Modak, 1993;Lee and Ramirez, 1994;Kazakov and Ganipol'skiy, 1994;Queinnec and Dahhou, 1994;Tartakovsky et al, 1995;Bibila and Robinson, 1995). Though the feed rate profile is the control variable most frequently used, sometimes additional control variables such as pH or temperature must be considered.…”

Control of fermenters – a review

“…Tartakovsky and coworkers [44] studied various control aspects of fed-batch auto-inductive fermentation process for metabolite production. An unstructured model of the process describing cell density, substrate and inducer concentrations was ®rst derived and then used to determine the optimal open-loop control strategy composed of three stages, namely growth, inducer synthesis and product synthesis.…”