Developing indicators of wine yeast performance: an overview of the impact of ethanol stress

Wine yeasts have been with us for thousands of years and over this time have been selected, inadvertently and systematically, for their robustness and the quality of wine that they produce. This has led to the generation of large collections of wine yeasts that can be accessed by winemakers to deliver reliability of performance and diversity of wines, but there is room for improvement in the yeasts that are available. The problem of suboptimal (stuck and sluggish) fermentations is a case in point. While some yeasts are quite reliable performers others can be diffi cult to get started, are sluggish in performance, or are prone to becoming stuck before fermentation is complete, and these problems are very costly to the wine industry. One option to minimise the risk of generating such problems is for winemakers to simply choose only the most reliable strains of yeast available to them, but this would limit the diversity of options available for delivery of different fl avour profi les. Thus it would benefi t the industry if we could fi nd a means of improving the robustness of wine yeasts, particularly those that deliver desirable fl avour profi les but currently have a propensity to be suboptimal performers. This article describes the fi rst stages of work at the AWRI that is attempting to achieve this end.While considerable research has gone into studying suboptimal fermentations and progress has been made in identifying features of grape must that are likely to compromise yeast performance (Alexandre and Charpentier 1998; Bisson 1999; Blateyron and Sablayrolles 2001; Bell and Henschke 2005), other than Zuzuarregui and del Olmo (2004), there are few reports in the literature of research directed to assessing the relative robustness of different wine yeasts with the aim of improving their performance and/or reliability. Most of the information we have on the relative effi ciency and reliability of different wine yeasts is from anecdotal reports; most winemakers know which of the strains available to them are likely to perform well and which are less reliable. We are using a more systematic approach to assessing the relative robustness of many commonly used wine yeasts with the aim of improving those that are most susceptible to suboptimal performance but are regarded as delivering good sensory properties to wines.Developing a laboratory-based approach to assessing resilience of wine yeasts is, however, not a straightforward exercise; particularly when one is interested in their robustness in a commercial winemaking setting. First, one has to identify ‘indicators’ of robustness, and for this we have decided to use ethanol-tolerance. We have chosen this criterion because it is widely believed in the industry and in yeast research (Ough 1992; Ivorra