Robustness quantification in yeast: A methodology to study phenotypic, evolutionary, and genomic aspects of microbial robustness.

Doctoral thesis, 2024

Bioprocesses contribute to the shift towards a more sustainable economy. In bioprocesses, valuable chemicals can be generated from renewable resources while, at the same time, reducing carbon emissions. A major hurdle in bringing bio-based products to market is the time and cost involved in designing efficient cell factories. Cell factories developed in controlled laboratory settings achieve high yields and productivities, but often fail at a larger scale because of unforeseen perturbations. Microbial robustness, i.e., the ability to maintain functionality despite perturbations, is critical for designing cell factories but remains poorly studied, particularly with respect to quantification as well as evolutionary and genetic aspects.

In this thesis work, mathematical evaluation, phenotypic characterization, evolution and genomics were applied to address the lack of quantification methods and explore robustness in yeast. A Fano factor-based approach for measuring robustness across multiple parameters and perturbations was created. Measurement of physiological data revealed trade-offs between robustness and performance in yeast. Moreover, when screening yeast deletion libraries, it pointed to the MET28 gene, which encodes a transcription factor regulating sulfur metabolism, as a mediator of robustness. Finally, evolution in fluctuating environments improved robustness in the industrial strain Ethanol Red, but not in two laboratory strains, contrasting with fitness trends.

Altogether, applying robustness quantification to various experimental set-ups, enabled the identification of key genes and metabolic processes linked to enhanced robustness. This thesis thereby contributes to the field of physiology, particularly in the context of robustness. The developed techniques have the potential to advance design optimization and testing of robust strains in laboratory settings, thereby enabling a faster scale-up to industrial environments.