Metabolic Engineering and Molecular Tool Development in Yeast for Production of Bulk Chemicals

Abstract

The brewers’ yeast Saccharomyces cerevisiae has long been a close friend of humanity. Yeast has been used in food and beverage production for thousands of years, and our close relationship with yeast has led to it becoming a model organism in research. Yeast was the first eukaryote to have its genome sequenced, and has helped in the study of human diseases, such as cancer. This extensive characterisation, along with its routine use in bioprocesses makes it an attractive target for metabolic engineering efforts. The work presented in this thesis encompasses some of the work that has been undertaken as part of the Ph.D. programme. Herein, I present work towards the development of tools to improve the process of engineering S. cerevisiae, as well as to apply those and other existing metabolic engineering tools to gain insights into the metabolism of engineered yeast strains. This research focuses on the production of hydroxy acids, a class of carboxylic acid. In the first research study, we developed a new set of integrative yeast expression vectors based on the previously developed EasyClone system. We adapted this system for use with CRISPR-Cas9 to insert expression cassettes into the yeast genome without the need for also integrating markers. This increases the speed of the engineering cycle as no marker loop-out is required between transformations. It also negates the need for markers, both auxotrophic and dominant, which can alter native host physiology and potentially affect experimental results. We have made this vector toolkit available through the Addgene platform, and hope many will use it to facilitate their research in S. cerevisiae. Using our vector toolkit, we investigated different acetyl-CoA supply strategies and their impact on the production of 3-hydroxypropionic acid (3HP), a bulk chemical that is a common target for metabolic engineering. We discovered that expression of a bacterial pyruvate dehydrogenase complex was the most successful strategy, which increased 3HP titres by almost 100% when grown in a simulated fed-batch medium. In the second study, focus is shifted away from tool development and towards metabolic engineering. Here we engineered a Crabtree-negative strain of S. cerevisiae to produce three different hydroxy acids; lactate, malate, and 3HP. We characterised these strains, and sought to gain an understanding of how these heterologous pathways influence the metabolism and physiology of the host strain. We performed 13C-based metabolic flux analysis to quantify the carbon flux distributions through the central carbon metabolism. Between the strains, we identified large differences in flux distribution, but each strain showed surprisingly high levels of flux through the pentose phosphate pathway, and only low levels of flux through the tricarboxylic acid (TCA) cycle. The third study attempts to use metabolic flux and transcriptomic analysis to identify key gene deletion strategies that can boost the production of 3HP in a host strain (ST938) capable of producing high levels of this hydroxy acid. Transcriptomic analysis revealed that much of central carbon metabolism was upregulated in ST938 compared to the WT strain, but metabolic flux analysis revealed lower proportional carbon flux through the TCA cycle. Metabolic flux analysis also revealed that only a small proportion of the available carbon flux was diverted into 3HP production. Computational approaches and insights from the transcriptomic analysis identified several gene deletion strategies that were then implemented in vivo. Of these strategies, the deletion of PRY1 was able to increase final 3HP titres by 27%.