Effect of fermentation temperature and culture media on the yeast lipid composition and wine volatile compounds Gemma Beltran, Maite Novo, Jos�� M. Guillam��n, Albert Mas, Nicolas Roz��s ��� Unitat d'Enologia del Centre de Refer��ncia de Tecnologia d'Aliments, Dept. Bioqu��mica i Biotecnologia, Facultat d'Enologia de Tarragona, Universitat Rovira i Virgili, c/ Marcel l�� Domingo, s/n. 43007 Tarragona, Spain Received 30 January 2007 received in revised form 10 August 2007 accepted 6 November 2007 Abstract The temperature of a wine fermentation strongly affects lipid metabolism and thus, aromatic profiles. Most of the metabolic studies are done in well-controlled laboratory conditions, yet wine is produced in less-reproducible industrial conditions. The aim of this study is to analyse the effect of fermentation temperature (13 ��C and 25��C) and culture media (synthetic media and grape must) on yeast lipid composition and volatile compounds in wine. Our results show that yeast viability was better at 13��C than at 25��C whichever growth medium is used, but that the complexity of the grape must enabled cells to reach higher viable population size. Viability was also related to the incorporation of linoleic acid and ��-sitosterol, which were present in the grape must. A lower temperature modified the cellular lipid composition of yeast, increasing the degree of unsaturation at the beginning of fermentation and decreasing the chain length as fermentation progressed. We also found that medium-chain fatty acids, mainly dodecanoic acid, were present in the cell phospholipids. Wines produced from grape must were more aromatic and had a lower volatile acidity content than those derived from a synthetic medium. Fermentations that were performed at the lower temperature also emphasized this feature. �� 2007 Elsevier B.V. All rights reserved. Keywords: Fatty acid Sterol Phospholipid Low temperature Aromatic compounds Grape must 1. Introduction Low temperature alcoholic fermentations are becoming more frequent due to the winemaker's tendency to produce wines with more pronounced aromatic profile. Wines produced at low temperatures (10���15 ��C) develop improved characteristics of taste and aroma (Feuillat et al., 1997). The temperature of fermentation affects both the retention of some varietal compounds and the production of fermentative metabolites. The improved quality of wines produced at low temperatures can be attributed to a greater retention of terpens, a reduction in higher alcohols and an increase in the proportion of ethyl and acetate esters in the total volatile compounds (Llaurad�� et al., 2002 Torija et al., 2003). The complexity of the fermentative medium also influences the yeast metabolism and thus, the fermentation performance and the final quality of the wine. Synthetic media simulating grape must are widely used in laboratory fermentations to keep a constant composition for the study of wine fermentation and wine yeast metabolism (Riou et al., 1997). However, wines produced from natural grape must have some varietal compounds and precursors, which could be incorporated to yeasts and modify some aspects of its metabolism and physiology, and consequently, the final wine composition. These changes can be more pronounced in specific conditions, such as fermentations performed at lower temperatures. Although low temperature fermentation has interesting applications in the enological industry, this practice also has some disadvantages. The optimal growth temperature for Sac- charomyces cerevisiae is 25 ��C, whereas 13 ��C is a restrictive temperature that increases the risk of stuck or sluggish fermentations as reviewed by Bisson (1999). In these Available online at www.sciencedirect.com International Journal of Food Microbiology 121 (2008) 169���177 www.elsevier.com/locate/ijfoodmicro ��� Corresponding author. Tel.: +34 977 55 82 80 fax: +34 977 55 82 32. E-mail address: nicolasrozes@urv.net (N. Roz��s). 0168-1605/$ - see front matter �� 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ijfoodmicro.2007.11.030

circumstances, the survival of cells can depend on their ability to adapt quickly to the changing environment. Changes in plasma membrane composition may be an adaptative response by the yeast since it is highly variable and clearly influenced by environmental factors such as temperature, oxygen, nutrient limitation and growth rate (Ratledge and Evans, 1989). These changes in plasma membrane are mainly accounted for by modifications in the lipid composition. The main lipid components of eukaryotic membranes are phospholipids, sterols, sphingolipids and glycerolipids. Altera- tions in fatty acid, phospholipid and sterol levels are needed to maintain ethanol tolerance (Bisson, 1999). The membrane fatty- acyl composition of yeast, like that of many other microorgan- isms, also changes with temperature: the lower the temperature, the more unsaturated the membrane fatty-acyl composition (Watson, 1987 Torija et al., 2003). However, the fatty acid composition of a cell can also be influenced by the en- vironment's lipid composition since it can include fatty acids from the medium in its own phospholipids (Thurston et al., 1981 Benchekroun and Bonaly, 1992 Rosi and Bertuccioli, 1992). In grapes, unsaturated fatty acids (UFA) represent the main lipid component. The most abundant of these is linoleic acid, followed by oleic, linolenic and palmitoleic acids (Castela et al., 1985). Of the saturated fatty acids (SFA), palmitic acid is the most abundant. However, the initial fatty acid content of the must depends on which technological procedures, such as pressing, maceration or clarification, are applied (Delfini and Costa, 1993). Regarding sterol content of natural musts, the main phytosterol in these musts was ��-sitosterol (Le Fur et al., 1994). In white wine production, the absence of oxygen suppresses fatty acid desaturation and the sterol biosynthesis by yeast, which reduces the capacity to synthesize unsaturated fatty acids and ergosterol, both essential for protecting the yeast against ethanol stress (Alexandre et al., 1994). In these circumstances, yeasts can incorporate exogenous sterols (Luparia et al., 2004) and unsaturated fatty acids (Chen, 1980), as it might be the case in industrial fermentations. On the other hand the reduced oxygen presence during the fermentation induces the synthesis of medium-chain fatty acid (C6 to C14) and their corresponding ethyl esters, which are toxic to the cells (Lafon-Lafourcade et al., 1984). The fermentation temperature and the culture media affect therefore the lipid metabolism, which is related to cell development, membrane integrity and the production of several by-products, especially those directly related to wine aroma. The aim of this study is to analyse the lipid metabolism response of S. cerevisiae fermenting in different media: one without lipids such as the synthetic must and another with the presence of lipids (grape must), that can be incorporated into the yeast membrane. During fermentation lipid metabolism is impaired and, thus, the response in the synthetic media will reflect the yeast metabolic response by itself. Furthermore, these differ- ences between the two media are analysed at lower temperature (13 ��C), where the response in lipid metabolism is highly needed for the cells to adapt their membrane fluidity. Also, the effects of both conditions (media and temperature) upon wine volatile compounds closely related to lipid metabolism and that affect strongly wine quality are analysed. 2. Materials and methods 2.1. Yeast strain and culture conditions A commercial S. cerevisiae wine strain (QA23, Lallemand S.A., Toulouse, France) was inoculated at an initial popula- tion of 2��106 cell ml��� 1 after rehydration in water at 37 ��C for 30 min according to the manufacturer's instructions. The imposition of the inoculated strain was checked by RFLP of mtDNA throughout the fermentation (Querol et al., 1992). This strain always represented more than 85% of the total analysed colonies all along the grape must fermentations. Two media were used in this experiment. One was Moscatell grape must (obtained from the experimental fields of the Faculty of Enology in Tarragona, Spain). This was clarified by settling for 12 hours at 20 ��C to separate the clear juice from the sediment in presence of 50 mg l��� 1 of sulphur dioxide. The final reducing sugar concentration was 205 g l��� 1. Yeast assimilable nitrogen (YAN) was determined by the Aerny method (Aerny, 1996) and then the grape must was made up to 300 mg l��� 1 with diammonium sulphate. The grape must lipid composition was analysed and is summarised in Table 1. The other was a synthetic grape must (Riou et al., 1997) with 200 g l��� 1 of reducing sugars (100 g l��� 1 Glucose and 100 g l��� 1 Fructose), 300 mg YAN l��� 1 and without lipids or anaerobic factors. In the wine cellar, the fermentations were performed in 100- liter tanks filled with 80 l of grape must. The tanks were equipped with efficient temperature control. In the laboratory the fermentations were done in 2-liter bottles filled with 1.8 l of the synthetic medium and fitted with closures that enabled the carbon dioxide to escape. The fermentations were performed at 13 and 25 ��C on both the laboratory scale and the semi- industrial scale by duplicate. 2.2. Yeast growth and fermentation kinetics Yeast growth was estimated by spreading the appropiate dilutions on YPD agar (glucose, 20 g l��� 1 bacteriological Table 1 Lipid composition of grape must (��g ml��� 1 ) a Fatty acids MCFA 0.34��0.07 LCFA 6.37��0.35 UFA b 7.47��0.56 Phospholipids 6.57��1.09 Neutral lipids Diacylglycerides 8.48��0.84 Triacylglycerides 2.01��0.16 Total sterols c 12.75��0.60 Sterol esters 1.29��0.13 MCFA: Medium Chain Fatty Acids (C6-C14) LCFA: Long Chain Fatty Acids (C16-C18) UFA: Unsaturated Fatty Acids (C14:1+16:1+C18:1+C18:2+C18:3). a Mean��Standard Deviation of three independent replicates. b Linoleic acid (C18:2) represented 60% of the total. c Stigmasterol and ��-sitosterol (phytosterols) represented 55% of the total. 170 G. Beltran et al. / International Journal of Food Microbiology 121 (2008) 169���177