Investigation of osmotolerant yeasts for very-high-gravity bioethanol production

Osmotolerance is an essential trait of Saccharomyces cerevisiae for several industrial applications, such as high-sugar dough fermentation, high-gravity brewing or wine making and also during high-gravity industrial bioethanol production in which typically high concentrations of sugars are used. Commercial application of very high-gravity (VHG) fermentation technology for bioethanol production has immense potential to increase final ethanol titer, and reduce the energy consumption for distillation and the amount of distillage residue. However, implementation of this technology requires a robust yeast strain that can efficiently ferment under high-gravity conditions. This work tried to help remediating this industrial bottleneck by exploring three different avenues. First, the natural biodiversity of Saccharomyces species has been extensively explored to identify one or more osmotolerant strains that have potential as a starter culture in bioethanol fermentation. This allowed the identification of several interesting yeast strains that performed well in fermentation experiments mimicking industrial bioethanol production. Moreover, several isolates were identified that were extremely tolerant to specific stressors and are good candidates for inverse metabolic engineering or further strain improvement through genetic engineering, experimental evolution, or breeding. In the second approach, the S. cerevisiae strain that showed most efficient very high-gravity fermentation performance (X6003) was subjected to polygenic analysis to identify the causative genetic elements responsible for its superior osmotolerance. For this purpose, pooled-segregant whole-genome sequence analysis, a powerful technology for efficient polygenic analysis, was employed. In this method, a haploid segregant (X6003-4D) of the osmotolerant yeast strain with the most superior trait-of-interest has been crossed with a haploid segregant (ER-42A) from an unrelated industrial strain (Ethanol Red) with a comparatively inferior phenotype.Subsequently, the resulting diploid hybrid strain X6003-4D/ER-42A has been sporulated and 32 segregants with the superior phenotype have been selected to construct the superior pool. 32 segregants were also randomly selected regardless of their phenotype to construct the unselected pool. Pooled genomic DNA extraction was performed for both pools separately and submitted to whole-genome sequence analysis. A total of ten quantitative trait loci (QTLs) were identified with this approach, all of which had a statistically significant linkage. Dissection of QTL2, QTL3 and QTL7 revealed three causative genes, namely RBK1, GPD1 and DUS3. Further analysis is required to identify the remainder of the causative alleles in the other QTLs. This work demonstrated the high genetic complexity of osmotolerance. Moreover, it confirmed the potential of polygenic analysis by pooled-segregant whole-genome sequencing to identify causative gene(s) with a subtle influence on the phenotype. Finally, in the third approach, the huge, yet barely exploited resource of non-conventional yeast biodiversity was explored to identify potential stress-tolerant alternatives for S. cerevisiae. This work resulted in the identification of several promising candidates that may replace or complement S. cerevisiae in second generation bioethanol fermentation, including Pichia kudriavzevii. Moreover, this work yielded a multipurpose database, providing novel insight in the properties of non-conventional yeasts and a resource for future studies to understand the molecular mechanisms underlying the extreme phenotypes of some of these yeasts. Altogether, the study supplemented our existing knowledge on the phenotypic diversity within different yeast species and within different strains of the same yeast species. It also gave us a better knowledge of the tolerance limits of different yeast species and strains for traits relevant to bioethanol production. Finally, the polygenic analysis by pooled-segregant whole-genome sequence analysis expanded our current knowledge on S. cerevisiaegenetic elements that contribute to superior fermentation performance under VHG conditions.