| Summary: | Industrial-sized fermenters differ from the laboratory environment in which bioprocess development initially took place. One of the issues that can lead to reduced productivity on a large scale or even early termination of the process is the presence of bioreactor heterogeneities. This work proposes and adopts a design–build–test–learn-type workflow that estimates the substrate, oxygen, and resulting growth heterogeneities through a compartmental modelling approach and maps <i>Yarrowia lipolytica</i>-specific behavior in this relevant range of conditions. The results indicate that at a growth rate of 0.1 h<sup>−1</sup>, the largest simulated volume (90 m<sup>3</sup>) reached partial oxygen limitation. Throughout the fed-batch, the cells experienced dissolved oxygen values from 0 to 75% and grew at rates of 0 to 0.2 h<sup>−1</sup>. These simulated large-scale conditions were tested in small-scale cultivations, which elucidated a transcriptome with a strong downregulation of various transporter and central carbon metabolism genes during oxygen limitation. The relation between oxygen availability and differential gene expression was dynamic and did not show a simple on–off behavior. This indicates that <i>Y. lipolytica</i> can differentiate between different available oxygen concentrations and adjust its transcription accordingly. The workflow presented can be used for <i>Y. lipolytica</i>-based strain engineering, thereby accelerating bioprocess development.
|
|---|