Exploration of the potential of nonconventional yeasts as cell factories by physiological characterisation and development of new expression tools

Yeasts are everywhere and are used in a wide range of different industrial production processes for production of food, beverages, biofuels, chemicals, industrial enzymes and therapeutic proteins. The yeast Saccharomyces cerevisiae, is the most commonly used yeast species, and sees abundant use both in scientific and industrial applications. However, many diverse yeast species exist that possess widely different physiological capabilities compared to S. cerevisiae, these yeasts are known as the nonconventional yeasts (NCYs). The NCYs have been embraced as useful alternatives to S. cerevisiae, but the full potential of NCYs and the diversity they represent has not yet been fully explored, i.e. when designing a new bio-process the focus of most groups is often on a single yeast species. Frequently, it is decided to shape the organism that one is most experienced with into a host for a novel process that its metabolism may not be favourable for.
This thesis is borne from a different mind-set, instead of attempting to force a yeast strain to conform to a production process that it is unfit for, it should instead be endeavoured to find yeast strains with a physiology that is physiologically fit for the process in question. Being fit for a process both concerns itself with how well a strain is able to produce the product of interest, but also how well it tolerates the conditions it is exposed to during the production process. This thesis concerns itself with; physiologically characterising NCYs to enable informed decisions when picking a yeast for a bio-process, screening NCYs for their potential as cell factories for heterologous protein production, and attempts at development of a novel molecular tool for stable high-level gene expression in NCYs.
A diverse set of 21 yeast species were physiologically characterised for their ability to utilise various carbon sources, their growth at various initial pH values, and their growth in the presence of redox agents using a highly automated quantifiable aerobic microcultivation-based growth assay. The growth assay was found to provide precise information on the growth-derived parameters maximum growth rate, lag phase duration, and biomass yield. Based on the maximum growth rate the pH optima could be estimated. Next, the yeasts tolerance towards redox agents was studied, a dose-dependent relationship between redox agent concentration and lag phase duration was discovered. This finding was used to quantify the yeasts tolerance towards oxidative and reductive stress, and was used to cluster the yeasts based on their tolerance profile. A possible model to explain the relationship between redox agent concentration and lag phase duration was discussed. The used growth assay proved to be applicable to a wide array of conditions and the ability to study stress in submerged cultivation instead of on solid media provide an opportunity to study yeast physiology in high-throughput under conditions more closely resembling those seen in production processes.
Next, a simple plasmid-based screening method was developed to enable exploring the NCYs for potential as cell factories for heterologous protein expression and secretion. In a test of the developed method, eight diverse yeast species were found to be transformable using these plasmids. In the majority of the tested yeast strains a proportional relationship was found between promoter strength and fluorescence from a reporter gene. This was used to assess promoter strength in the studied yeasts. Following this finding, the yeasts were screened for their ability to heterologously produce a difficult-toproduce protein, Armillaria mellea peptidyl-Lys metallopeptidase. Based on the estimated promoter activity and protease activity a non-obvious NCY, Naomovozyma castellii, was identified that could prove to be an interesting alternative production host of the tested protein. This demonstrated the simplicity and applicability of the method for screening for interesting production hosts among the NCYs.
Finally, it was investigated whether the functionality of the S. cerevisiae endogenous 2µ plasmid could be expanded to NCYs. The 2µ plasmid known for being highly stable and present in high copy-number in S. cerevisiae, traits that would be highly desirable to transfer to NCYs for heterologous expression. 2µ hybrid plasmids were constructed that allowed for testing whether the presence of a broadly applicable autonomously replicating sequence (ARS) or a S. cerevisiae centromere could restore the desirable traits of the original 2µ plasmid. ThreeS. cerevisiae strains and six NCY species were transformed with the constructed plasmids. The resulting strains were tested for their stability, where it was found that all plasmids were present as non-integrated episomal entities. In Kluyveromyces marxianus, it was found that inclusion of the broadly applicable ARS sequence in the 2µ hybrid plasmid appeared to restore some of the 2µ functionality, when compared to the reference plasmid which did not contain the 2µ sequence. Thus, it appears that plasmid engineering efforts could enable restoration of 2µ-like activity 2µ hybrid plasmids in NCYs.
Collectively, the studies presented in this thesis shows that NCYs are a highly diverse resource that can be characterised and screened for their potential as cell factories using relatively simple tools and methods.