Identification and Understanding of Saccharomyces and Oenococcus Interactions in Wine Fermentation

Abstract

Winemakers are now more frequently choosing to inoculate yeast and bacteria together in a co-inoculation strategy to achieve faster, more efficient fermentations. However, this can be potentially problematic due to yeast-lactic acid bacteria (LAB) incompatibility that can result in stuck fermentations. This PhD thesis examined yeast-LAB compatibility using commercially available strains in co-inoculated fermentations to further understand the complexities of yeast-LAB interactions in wine. Commercial yeast-LAB pairs (72 in total) were initially screened in a synthetic juice to determine compatible (yeast and LAB able to complete alcoholic and malolactic fermentation) and incompatible (LAB unable to complete malolactic fermentation) pairs. The 72 yeast-LAB pairs were ranked based on fermentation performance, with additional in-depth analysis of the top four and bottom four pairs in a Shiraz juice. Fermentation kinetics and a number of fermentation relevant compounds were measured to elucidate reasons for differences in LAB fermentation performance. This experiment revealed differences in concentrations of H2S, esters and succinic acid between yeast-alone control fermentations and yeast-LAB co-inoculated fermentations. In parallel with these studies, a yeast quantitative trait loci (QTL) library was used to determine yeast specific traits that could impact LAB fermentation ability. A QTL was identified which spanned a genomic region containing the gene SSU1, known to encode a sulfite exporter (Ssu1p). Follow-up work using hemizygote strains revealed that yeast with SSU1 haploinsufficiency allowed LAB to perform malolactic fermentation faster than when co-inoculated with wild-type yeast. Considering the difference in H2S production and the influence of SSU1, a final experiment was performed to assess yeast and LAB sulfur pathway gene regulation in response to co-inoculation. Quantitative PCR was used to study metabolic links to yeast-LAB compatibility, as well as measurement of glutathione and H2S. This work involved RNA extraction from mixed yeast-LAB fermentation samples and measurements of H2S and glutathione over time. When assessing genes involved in sulfur metabolism, differences were observed between yeast only and yeast-LAB fermentations. There were also differences between yeast strains. Additionally, it was observed that there were higher concentrations of glutathione in co-inoculations compared to yeast-only fermentations. Intriguingly, there was a lack of correlation between H2S production and CYS3, CYS4, MET5 and MET10 gene expression. Overall the studies carried out in this thesis have highlighted the complexity of yeast-LAB interactions in wine fermentation. This work has provided a starting point for future work investigating yeast-LAB compatibility and the potential role of sulfur in compatibility outcomes.