Project 3.2.1

Putting microbial diversity to work in shaping wine style

Project summary

The AWRI has a long history in microbial strain development and the successful deployment and commercialisation of newly-developed strains. This project aims to build on that history.
There is substantial scope to both refine existing strains (such as the ‘rose’ yeast and Saccharomyces bayanus hybrids) and optimise their use through varietal or wine style pairing. For example, building on knowledge gained through production of the ‘rose’ yeast, this project will explore the degree to which yeast aromatic profiles can be modulated by modification of carbon flux through amino acid biosynthetic pathways.

Molecular technology will help to determine what is possible, and will provide foundational knowledge on biosynthetic pathways and markers for strain selection. Identification of molecular markers for specific aromatic traits will permit targeted selection of strains isolated during bioprospecting work.

In addition, this project will tap into resources identified in other AWRI projects that have a bioprospecting focus. Novel microorganisms, after initial screening, will be put to work directly through use of appropriate winemaking techniques and/or be used as source material for introducing greater genetic diversity into existing wine strains. Together these breeding and selection strategies will deliver non-genetically modified germplasm that can be used by industry, and will provide new microorganisms for winemakers seeking a point of differentiation in their wines.

Latest information

Rebuilding the rose – bringing floral aromas to different yeast backgrounds
For some time amino acid analogues have been used as selective agents to drive alterations in amino acid metabolism, and hence aroma compound production, in wine yeast. The first generation of these yeast produced very high concentrations of phenylethyl alcohol (PE), phenylethyl acetate (PEA), tryptophol and tyrosol. The overt ‘rose’ aroma associated with those very high concentrations of PE was perceived as being out of balance and the high concentrations of tryptophol and tyrosol drove bitterness, particularly in red wines.

In an attempt to address these issues, trials were undertaken with a second generation of PE-enhancing yeasts which exhibit a spectrum of PE and PEA production potential. Chardonnay wines made with these yeasts were assessed by a trained sensory panel. Despite this second generation of strains producing between two and fourteen times the concentration of PE and five to thirty-two times the concentration of PEA, only the wine with the highest concentration was perceived as been more ‘floral’. Wines with concentrations of PE between two and eight times higher than the control wine were not perceived as having increased ‘floral’ aroma.

Another factor influencing the performance of these PE-enhancing yeasts is the parent strain from which they were originally derived. With poor pH and SO2 tolerance, the original parent is not optimal for white wine production. A third generation of amino acid analogue resistant strains was therefore produced this year. These isolates have not only been screened for their PE and PEA production potential but also their capacity to produce tryptophol and tyrosol. Together with one of the moderate PE producers trialled in vintage 2018, two of the third-generation isolates were applied in pilot winemaking trials to assess their suitability for production of sparkling wine, for both primary and secondary fermentation. Final evaluations of these wines will be undertaken in the second half of calendar year 2019.

Digging into the details of wine yeast hybrids
A previous project generated a Saccharomyces cerevisiae x Saccharomyces uvarum hybrid that exhibited low acetate production in high sugar conditions. The concentration of acetate produced by this hybrid was not intermediate between the two parents, but significantly less than both, suggesting an unusual genetic interaction. To investigate this further, a large collection of progeny resulting from the sporulation of the hybrid was generated. This collection shows variance in the concentration of acetate produced, which can be used to map the determinants of the low acetate trait. This mapping is being undertaken using a method known as quantitative trait loci analysis. It is hoped that this work will allow a better understanding of how genomes as different from each other as S. cerevisiae and S. uvarum respond when they come together in the same individual.

In related work, the contribution of hybrid yeast to non-volatile wine composition is being explored in laboratory-scale red grape (Tempranillo) fermentation trials. Hybrid strains with different combinations of S. cerevisiae wine yeast and non-cerevisiae parents were included in the trial in order to evaluate the impact of differing S. cerevisiae genomic heritage on interspecific hybrid winemaking capabilities. Hybrid-dependent variation in total polysaccharides was observed, driven primarily by differences in the low molecular weight polysaccharide fraction. These are known to be released from pectins found in the grape cell wall. The majority of the variation in low molecular weight polysaccharides was related to the S. cerevisiae parent of the hybrid, with only minor contributions from the non-cerevisiae parents evident. This work is providing the foundations for a more nuanced approach to selecting parental strains for the generation of wine yeast hybrids.

A vintage trial comparing seven different interspecific hybrids, all products of rare mating with diverse non-cerevisiae species, was undertaken in 2018. This was the first ‘head-to-head’ comparison of such a diverse array of hybrids. The resultant wines underwent sensory assessment and chemical analyses for volatile flavour-active fermentation products, polysaccharides and haze-forming proteins. Sensorially, wines produced from the hybrid strains were rated higher in ‘fruit’ and ‘banana’ aroma and flavours (S. cariocanus, S. kudriavzevii and S. mikatae) as well as ‘floral’ aroma and acidity (S. eubayanus and S. arboricola), while the S. uvarum hybrid wines were rated highly for ‘honey’ aroma and yellow colour. These results aligned with the chemical analyses for compounds known to produce ‘fruit’ aroma (ethyl butanoate), ‘banana’ aroma (3-methyl butyl acetate) and ‘floral’ aroma (PEA). Differences were also observed in wine polysaccharide composition (S. uvarum) and haze-forming protein concentration (S. kudriavzevii, S. uvarum, S. eubayanus and S. arboricola) but the winemaking significance of these findings is yet to be determined.