Colin McBryde

Application of the reuseable, KanMX selectable marker to industrial yeast: construction and evaluation of heterothallic wine strains of Saccharomyces cerevisiae, possessing minimal foreign DNA sequences

The characterisation of wine yeasts and the complex metabolic processes influencing wine fermentation and the quality of wine might best be achieved by exploiting the standard classical and recombinant genetic techniques which have been successfully used with laboratory strains. However, application of these techniques to industrial strains has been restricted because such strains are typically prototrophic and often polyploid. To overcome this problem, we have identified commercial wine strains with good mating and sporulation properties from which heterothallic derivatives were constructed by disruption of the HO gene. Consequently, these haploids are amenable to genetic analysis, whilst retaining desirable wine-making properties. The approach used was an adaptation of a previously published gene disruption procedure for laboratory yeast and is based on the acquisition of geneticin resistance from a removable KanMX marker. The present work is the first report of the application of a construct of this type to the disruption of the HO gene in wine yeasts that are in common commercial use. Most of the 4.9-kb disruption construct was successfully removed from the genome of the haploid derivative strains by loop-out of the KanMX marker through meiotic recombination. Sequencing of the HO region confirmed the reduction of foreign sequences to a 582-bp fragment comprised largely of a single direct repeat at the target gene. The removal of the active foreign gene (conferring antibiotic resistance) allows the application of other constructs based on the KanMX module without the need to resort to other selectable marker systems. Laboratory-scale fermentation trials typically showed minimal differences between the HO disruptants and the parental wine strains in terms of fermentation kinetics and formation of key metabolites.

Optimisation of industrial strains of Saccharomyces cerevisiae using recombinant and non-recombinant methods

The production of fermented beverages such as beer and wine is typically achieved using strains of Saccharomyces cerevisiae, which have been selected for their reliable fermentative properties and desirable sensory contribution to the end product. Even so, these strains are not without shortcomings and new strains are constantly being sought. In aiming to provide such strains, working with industrial isolates has its own set of challenges. Wine strains of S. cerevisiae are typically of higher ploidy or aneuploid, frequently lack convenient selectable markers and are homothallic. Laboratory strains, along with their comprehensive set of genetic tools, might be a more convenient research vehicle, but the fact remains that such strains fare poorly under the highly stressful conditions of industrial fermentations: sugar contents of over 200 g/L; low nutrient availability and pH; and high final ethanol concentrations. More importantly, the wines produced by laboratory strains of yeast frequently display poor sensory properties. Put simply, laboratory yeast lack the desirable phenotypic characteristics needed to make wine of quality to enable industrially relevant studies.