Flor Yeasts of Saccharomyces Cerevisiae—Their Ecology, Genetics and Metabolism

International Journal of Food Microbiology 167 (2013) 269–275

Contents lists available at ScienceDirect

International Journal of Food Microbiology

journal homepage: www.elsevier.com/locate/ijfoodmicro

Review Flor yeasts of Saccharomyces cerevisiae—Their ecology, genetics and metabolism

Hervé Alexandre ⁎

UMR PAM Université de Bourgogne-AgroSup Dijon Laboratoire VALMIS Institut Universitaire de la Vigne et du Vin Jules Guyot, Université de Bourgogne, 21078 Dijon Cedex, France article info abstract

Article history: The aging of certain white wines is dependent on the presence of yeast strains that develop a biofilm on the wine Received 3 April 2013 surface after the alcoholic fermentation. These strains belong to the genus Saccharomyces and are called flor Received in revised form 12 August 2013 yeasts. These strains possess distinctive characteristics compared with Saccharomyces cerevisiae fermenting Accepted 31 August 2013 strain. The most important one is their capacity to form a biofilm on the air–liquid interface of the wine. The Available online 10 September 2013 major gene involved in this phenotype is FLO11, however other genes are also involved in velum formation by these yeast and will be detailed. Other striking features presented in this review are their aneuploidy, and Keywords: fl Sherry their mitochondrial DNA polymorphism which seems to re ect adaptive evolution of the yeast to a stressful en- Vin Jaune vironment where acetaldehyde and ethanol are present at elevated concentration. The biofilm assures access to Flor yeast oxygen and therefore permits continued growth on non-fermentable ethanol. This specific metabolism explains Saccharomyces the peculiar organoleptic profile of these wines, especially their content in acetaldehyde and sotolon. This review Wine deals with these different specificities of flor yeasts and will also underline the existing gaps regarding these as- Velum formation tonishing yeasts. © 2013 Elsevier B.V. All rights reserved.

Contents

1. Introduction...... 269 2. Floryeastecology...... 270 3. Distinctivegeneticfeaturesofvelumyeast...... 270 4. Why do ﬂor yeast ﬂoat?...... 271 5. Floryeastmetabolism...... 272 6. Theevolutionofchemicalcompoundsduringbiologicalaging...... 272 7. Factorsaffectingvelumformation...... 273 8. Conclusion...... 274 References...... 274

1. Introduction yeasts) develops naturally on the surface of the wine which is in contact with air. The production of French flor sherry wine (Vin Jaune)inthe Several types of wines are characterized by the development of a Jura region of France, where film-forming yeast (Flor yeast) develops film of yeast at the surface; this is known as flor velum yeast or “Flor on the surface of the wine, is similar to the process used to produce yeast”. Sherry is the most well-known of this group of wines and its pro- Fino Sherry in the Xeres and Montilla-Moriles regions in Spain duction process has been explained in detail by Pozo-Bayón and (Charpentier et al., 2002) except that wines are not fortified after the Moreno-Arribas (2011) and Benitez et al. (2011). After fermentation end of alcoholic fermentation and do not undergo sherry wine aging of Palomino fino grapes, the fermented wine is fortified with wine alco- system. Film-forming yeast strains can be present on the surface of hol to 15% (v/v), clarified by natural sedimentation and transferred into aging dry and sweet Szamorodni wines (belonging to the well known oak casks for storage. The storage period is minimum two years and can Tokay wine group). Vernaccia di Oristano is a Sardinian wine undergoing last more than 10 years and, during this time, a yeast velum (Flor biological aging which unlike sherry wines is not fortified and which does not undergo the “Solera” system. Aging with flor yeasts to produce fl ⁎ Tel.: +33 80 39 63 93; fax: +33 80 39 62 65. wines similar to or sherry is a method also used in areas such as South E-mail address: [email protected]. Africa, Armenia, California and southern Australia.

0168-1605/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.ijfoodmicro.2013.08.021 270 H. Alexandre / International Journal of Food Microbiology 167 (2013) 269–275

In all of these wines, a buoyant yeast biofilm develops at the surface relatedness. Regarding S. montuliensis, it is now considered as Torul- during aging, and protects the wine from oxidation. The oxidative me- aspora delbrueckii. Molecular analysis tools have been developed to tabolism of the flor yeasts facilitates changes in the characteristics of identify these yeasts (Esteve-Zarzoso et al., 2001). RFLP analysis of the the wine as the biofilm assures access to oxygen allowing their growth 5.8S rRNA gene and the 5.8S-ITS region allows S. cerevisiae velum on non-fermentable ethanol. Film-forming and oxidative metabolism is yeast to be differentiated from other S. cerevisiae (Fernandez-Espinar adaptive mechanisms which allow cells to survive under such condi- et al., 2000) as velum yeast is generally characterized by a 24 bp deletions. Flor yeasts have recently been reviewed (Pozo-Bayón and tion in the ITS1 (Internal Transcribed Spacer) (Esteve-Zarzoso et al., Moreno-Arribas, 2011; Benitez et al., 2011). However, the first study 2004). S. cerevisiae from Vin Jaune has no 24 bp deletion in the ITS1 re- mainly focuses on sherry wine with different aspects such as sherry gion (Charpentier et al., 2009) but has an additional C residue resulting wine process and chemical composition. The second study also deals in the creation of a HaeIII site in this sequence (Fig. 2). An additional with sherry wines and especially with the aging method, the physiolog- 370 bp sequence in this region was revealed by RFLP analysis using ical characteristics of both fermentation and aging yeasts. The authors CfoI (Charpentier et al., 2009). The 24 bp deletion has been proposed to give an in-depth overview regarding the role of FLO11 in velum forming. be related to a nuclear gene involved in ethanol tolerance (Fernandez- In this review the involvement of other genes is also discussed. The aim Espinar et al., 2000). Thus, the difference between yeasts associated of the present study is not to review the process but to describe in de- with Vin Jaune and Spanish flor could be due to differences in alcohol con- tails the specificity of flor yeasts encountered not only in sherry wines tent although Charpentier et al. (2009) suggest that the differences as described previously but also in Vernaccia di Oristano or “Vin between these yeasts are more likely to be linked to their phylogenetic Jaune”. Here, we provide an updated review as to what is known and origin. unknown about this unusual type of yeast, with a focus on their ecology, Two yeast groups have been identified in a biodiversity study on Vin and specific aspects of their genetics and metabolic activities. Jaune velum yeast. These groups, isolated from geographically separate areas, were shown by interdelta sequence typing to have different genetic structures (Charpentier et al., 2009); the velum aspect appears 2. Flor yeast ecology to be correlated with the interdelta profile. Yeasts characterized by the production of a thin velum have similar interdelta profile, which are dif- During aging of flor sherry wines, a yeast biofilm (Fig. 1) develops ferent to that of yeasts producing a thick velum. under harsh conditions: low oxygen concentration, high ethanol con- The origin of these yeasts is unclear; we do not know if they are pres- centration (from 14%v/v to 16%v/v), low pH and the presence of ent on grapes or found in the cellar. In Vin Jaune it is still to be deter- sulphites (around 30 mg/L total SO2). Few microorganisms can survive mined whether the yeasts responsible for alcoholic fermentation are in such conditions. Consequently, more than 95% of the flor may be the same as those present in the velum. For sherry wines, one study composed of Saccharomyces cerevisiae as this is the most adapted has followed the population dynamics of yeasts during alcoholic fer- yeast to grow in this environment (Pozo-Bayón and Moreno-Arribas, mentation and biological aging (Esteve-Zarzoso et al., 2001). Sherry 2011). Although the velum (Flor) is mainly composed of Saccharomyces, wines are fortified (the addition of alcohol) after the initial fermentation other yeast and bacteria can also be present. and it is only after fortification that flor yeasts were detected in the me- Analysis of 54 yeast strains isolated from the velum of French flor dium. These yeasts were not found during alcoholic fermentation, prob- sherry (Vin Jaune) has shown that all the strains belong to S. cerevisiae ably because their cell concentration is too low. During aging it is mainly (Charpentier et al., 2009). Other species such as Pichia, Candida and flor yeasts belonging to S. cerevisiae species that were present, although Hansenula have also been found in velum (Suarez-Lepez and Inigo- these are sometimes found together with Brettanomyces (previously Leal, 2004). The presence of the spoilage yeast Brettanomyces has been shown by Ibeas et al. (1996)). The study concludes that yeasts responsi- shown to be responsible for increased volatile acidity in wine during bi- ble for alcoholic fermentation are different from velum yeast. However, ological aging (Ibeas et al., 1996). A physiological study by Martinez the must in this study was inoculated with commercial yeast (Esteve- et al. (1995) reported that there are four main races of yeast involved Zarzoso et al., 2001) which could have influenced the development of in the formation of velum: Saccharomyces beticus, Saccharomyces indigenous yeast. Further studies are required, particularly for Vin cheresiensis, Saccharomyces montuliensis and Saccharomyces rouxii Jaune and Vernaccia wines where there are currently no data available. (also known as Zygosaccharomyces rouxii). This classification was As the aging process is different from sherry wine, there may also be dif- based on the ability of yeasts to ferment different sugars (galactose, dex- ferences between the biological aging processes. trose, lactose, maltose, melibiose, raffinose and sucrose). Other studies Contradictory results have been reported regarding the dynamics of confirm the presence of different yeasts in sherry flor (Mesa et al., yeast populations during biological aging. Martinez et al. (1997b) have 2000)aswellasinVin Jaune flor (Charpentier et al., 2009). However, shown that different yeasts succeed during aging: S. cerevisiae beticus S. beticus and S. cheresiensis are no longer considered as races or subspe- (now S. cerevisiae) is faster at forming velum and predominates in youn- cies of S. cerevisiae according to the last taxonomic study (Kurtzman ger wines, whereas S. cerevisiae montuliensis (now T. delbrueckii) pro- et al., 2011). Indeed, all these previously named races or subspecies duces and resists higher acetaldehyde concentrations and so appears are now considered as S. cerevisiae synonyms based on nuclear DNA later. However, Ibeas et al. (1997) reported that in different wine cellars, different flor yeasts dominate and persist during aging and this was confirmed by Esteve-Zarzoso et al. (2001). In view of the latest yeast taxon- omy it is important to underline that there are no more races (Kurtzman et al., 2011).

3. Distinctive genetic features of velum yeast

Some authors suggest that velum yeast possesses distinct genetic features that are different to that of other wine yeasts (for example, low chromosomal polymorphism (Martinez et al., 1995; Ibeas and Jimenez, 1996)). Elevated ethanol concentration found in wines aged under yeast velum may be responsible for such genetic differences (Ibeas et al., 1997; Martınez et al., 1998; Mesa et al., 1999, 2000). Genet- Fig. 1. Velum yeast on French ﬂor sherry wine Vin Jaune. ic heterogeneity could be associated with adaptation to a wine medium H. Alexandre / International Journal of Food Microbiology 167 (2013) 269–275 271

Fig. 2. Comparison of the partial rDNA ITS1 sequence of a “Vin Jaune” flor yeast strain (Mac51, FM177898), a classical wine yeast strain (CLIB227, AM900404) and a sherry flor yeast strain (CECT 11757, AJ275936). The Vin Jaune flor yeast strain possesses a C insertion (yellow) which is not present in the wine yeast strain sequence (green). The sherry flor yeast strain shows a 24 bp deletion (—). Adapted from Charpentier et al. (2009).

which is depleted in fermentable sugars and has high levels of ethanol comparing the FLO11 DNA sequence of a flor forming yeast (YNN295 and acetaldehyde (Martınez et al., 1998). Acetaldehyde can cause dou- laboratory strain) versus a non-flor forming yeast (133d flor yeast), ble strand DNA breaks and has been reported to be responsible for the Fidalgo et al. (2006) found that the FLO11 promoter is 0.1 kb shorter high mitochondrial DNA polymorphism observed in flor yeast (Ristow and the coding sequence is 1 kb larger in flor-forming yeast. Several et al., 1995; Castrejon et al., 2002). Acetaldehyde could also cause point mutations, deletions and rearrangements in both the promoter large-scale chromosomal rearrangement which may result in amplifica- and the ORF of the FLO11 gene were also observed (Fidalgo et al., tion of chromosomal segments (Infante et al., 2003). An increased ex- 2006). Hence, the ability to form a velum and float may be due to dereg- pression of genes present on these amplicons might be responsible for ulation of the promoter which leads to higher FLO11p synthesis and, the specific physiological traits of these yeasts. Wine yeast and flor consequently, greater hydrophobicity. yeast have different mitochondrial patterns (as described by Ibeas et al. Indeed, the longer ORF could be explained by an increased number (1996) and Martinez et al. (1995)) and this high polymorphism was of the repeated sequences in the central domain which has been first attributed to ethanol (Martinez et al., 1995). However, according shown to be involved in yeast flor floatability (Fidalgo et al., 2006). to Ibeas and Jimenez (1997), the ethanol-induced respiration-deficient However, comparisons of FLO11 DNA sequences in other flor strains mutant (petite mutant) and mitochondrial DNA polymorphism are have shown that the length of the repeats region differs depending on more likely to be due to small genomic lesions than exposure to ethanol. the strain (Fidalgo et al., 2006). Interestingly, Fidalgo et al. (2008) re- Later, it was shown that the mutagenic effect of ethanol and acetaldehyde ported that the repeats domain of the FLO11 gene was very unstable strongly contributes to the high frequency of mtDNA variability in flor under non-selective conditions. This study allowed the authors to dem- yeasts (Castrejon et al., 2002). Flor yeasts possess a high degree of hetero- onstrate that the ratio and/or distribution of the different repetitive zygosis (Budroni et al., 2000) and are characterized by a very low sporu- units, rather than expansions in the repeat domain of FLO11, may play lation frequency and viability. a role in the acquisition of floatability in wild flor yeasts (Fidalgo et al., A further specific feature of flor yeast is their aneuploidy (Bakalinsky 2008). An in depth analysis of FLO11 gene sequence has been done on and Snow, 1990; Guijo et al., 1997). In their study, Bakalinsky and Snow 20 different flor yeast strains (Zara et al., 2009). Their results support observed extra copies of chromosomes V, VII, and XIII. Using pulsed field the conclusion of previous studies regarding the high polymorphism electrophoresis of flor yeast nuclear DNA, Guijo et al. (1997) showed the of FLO11 gene sequence. Thirteen different alleles were identified presence of two, three or four copies of chromosomes. While most chro- which size varies from 3 to 6.1 kb. Using a multiple linear regression mosomes are present in two copies, chromosomes V, VII, and XI, are model that includes transcription level and the size of repeats of present in three or even four copies depending on the strain according FLO11, Zara et al. (2009) demonstrated that biofilm formation is not to Guijo et al. (1997). Polysomy (4 copies) was observed on Chromosome only influenced by the length of FLO11 repeats but also by the transcrip- XIII, where the ADH2 and ADH3 loci encoding alcohol dehydrogenases are tion level of the gene. Differences in transcription levels could not be found. Interestingly, these enzymes are involved in the oxidative conver- explained solely neither by the presence or absence of the 111 bp se- sion of ethanol to acetaldehyde during biological aging. Using CGH analy- quence within an upstream repression sequence of the FLO11 promoter sis, Infante et al. (2003) confirmed that flor yeast strains are aneuploidy as described previously (Fidalgo et al., 2006) nor by the presence of two for several chromosomes. short repeats sequences (Zara et al., 2009). It is suggested that FLO11 While most wine yeasts are homothallic, flor yeast showed great transcription level might be under epigenetic regulation (Zara et al., heterogeneity in their life cycle. A heterothallic life cycle was found in 2009). 10% (Mortimer et al., 1994) and a third semi-homothallic life cycle has FLO11 is thought to be repressed as when glucose is present in the also been suggested (Budroni et al., 2005; Pirino et al., 2004). medium (Verstrepen and Klis, 2006). This supports the notion that utilization of non-fermentable carbon sources (mainly ethanol) instead of 4. Why do flor yeast float? fermentable sources (such as glucose) is a prerequisite for flor formation (Iimura et al., 1980). However, Ishigami et al. (2006) found that One of the main characteristics of flor yeast is their ability to float on FLO11 was highly expressed in a flor yeast when using either glucose the surface of the wine and form a velum. This property was first attrib- or ethanol as the sole carbon source. Despite the presence of sugar uted to a high cell surface hydrophobicity (Iimura et al., 1980; Martinez (40 g/L), the development of a yeast biofilm on the surface is also ob- et al., 1997a,c) which would be due to specific cell wall composition. In- servedforTokajiSzamorodniwine(Kovacs et al., 2008). deed, treatment of cells with β-glucanase induces a large decrease in The ability to form a velum in a glucose medium may be related to cell surface hydrophobicity (Martinez et al., 1997c; Alexandre et al., the repression of the regulatory mechanisms of FLO11.Amongthemu- 1998); mannoproteins are also thought to be involved in hydrophobic- tations observed in FLO11 gene, one is a deletion in the promoter which ity (Alexandre et al., 2000). A classical genetic analysis concluded that relieves the gene from repression (Fidalgo et al., 2006). this phenotypic trait was linked to the presence of one or two genes. Taking into account the high polymorphism within FLO11 gene to- Reynolds and Fink (2001) showed that the FLO11 gene is involved in gether with a possible epigenetic regulation, the influence of glucose yeast biofilm formation and its role in yeast velum formation has since on FLO11 expression still needs to be clarified. been confirmed by other studies (Zara et al., 2005; Ishigami et al., FLO11 expression control is complex. The promoter contains four 2004; Fidalgo et al., 2006; Purevdorj-Gage et al., 2007). Deletion of known activation sequences and nine repression domains. Most of this gene in a flor forming yeast prevented a velum from forming these regulatory regions are targets for the MAPK pathway, the (Zara et al., 2005; Fidalgo et al., 2006). However, not all S. cerevisiae cAMPcascade, and the Gnc4p-controlled signaling pathway (Barrales strains that possess the gene are able to form a velum. FLO11 gene et al., 2008). The pH response pathway is also involved in controlling sequence comparison reveals numerous differences. For example, FLO11 expression together with the chromatin-remodeling complexes 272 H. Alexandre / International Journal of Food Microbiology 167 (2013) 269–275 which are central elements involved in FLO11 activation Barrales et al., tolerance: the role of Btn2p in amino acid transport may be linked to 2008, 2012). ethanol resistance (Espinazo-Romeu et al., 2008). Furthermore, another aspect poorly studied for flor yeast is the glycosylation of FLO11. Recently, it was shown using a mutant defective in 5. Flor yeast metabolism Flo11 glycosylation, that the glycosylation of Flo11 is required for biofilm formation (Meem and Cullen, 2012). Gluconeogenesis, the synthesis of sugars for macromolecule biosyn- Although FLO11 plays a major role in film formation, other genes thesis, is required to enable yeast growth when their sole carbon source such as HSP12 and NRG1 are also involved (Ishigami et al., 2004; Zara is ethanol (Turcotte et al., 2010). After fermentation, the yeasts are in an et al., 2002). A point mutation in HSP12, or deletion of the entire gene, alcohol rich environment which has a low pH and low dissolved oxygen results in an inability of the yeast to form a film (Zara et al., 2002). The level. The velum-forming yeast present in wine has a diauxic growth C-terminal truncated form of Nrg1p affects FLO11 repression, resulting pattern: during anaerobic respiration there is an increased expression in FLO11 expression and flor formation (Ishigami et al., 2004). of FLO11 gene and cell hydrophobicity (Zara et al., 2005). The cells can

BTN2, encoding a v-snare interacting protein involved in intracellu- then aggregate, trap bubbles of CO2 and reach the wine surface where lar protein trafficking, may indirectly affect velum flor formation as de- oxygen is available. Once at the wine surface, the oxidative metabolism letion of this gene in flor yeast affects biofilm formation (Espinazo- of yeast reduces the concentrations of ethanol, acetic acid, ethyl acetate, Romeu et al., 2008). glycerol, amino acids and organic acids. The properties of the cell wall protein Ccw7p/Pir2p/Hsp150p differ However, development of S. cerevisiae biofilm is not restricted to between laboratory and film-forming yeast strains. Film-forming strains aerobic growth on ethanol, since biofilm could be formed on other re- contain uniform proteins but are much smaller in size (87 kDa) than the duced non-fermentable carbon sources such as glycerol and ethyl ace- laboratory strains (117 kDa) (Kovacs et al., 2008). This difference is due tate for example (Zara et al., 2010). to deletions of two distinct parts within the repetitive region of the In parallel to the reduction of the above mentioned metabolites, an CCW7 gene; as a result of deletions, only eight of the 11 repeating increase in the concentrations of acetaldehyde, higher alcohols, acetoin units are found in film-forming strains. It is still not clear if this plays a and 2,3-butandediol is observed (Blandino et al., 1997; Mauricio and role in cell surface properties and velum formation. In addition to Ortega, 1993). Ethanol is the main source of energy for the cell and it their ability to float (linked to FLO11, NRG1 and HSP12 genes), flor yeasts is transformed into acetaldehyde using alcohol dehydrogenase (Del are also unusual in their tolerance of ethanol and acetaldehyde. HSP12, Carmen Plata et al., 1998). Some of the acetaldehyde is oxidized to acetic HSP82, HSP104 and, in particular, HSP26 are involved in this resistance: acid and subsequently transformed into acetyl-CoA; this then enters ei- a clear correlation between resistance to ethanol and acetaldehyde and ther the glyoxylate cycle (to form succinic acid) or the Krebs cycle. high induction of HSP genes has been demonstrated (Aranda et al., Once formed, the flor yeast film blocks access of oxygen and the cell 2002). It is worth noting that the genes encoding aldehyde dehydroge- medium becomes strongly reducing (Mauricio et al., 1997; Pozo-Bayón nase (ALD2/3, ALD4, ALD6) are differentially expressed in flor yeast com- and Moreno-Arribas, 2011). Yeast metabolism then oxidizes coenzymes pared to fermentative yeast (Aranda and del Olmo, 2003). The basal to maintain the intracellular redox balance (Berlanga et al., 2006). Pro- level of ALD6 in flor strain is higher than in fermentative strain. Further- duction of ethanol and higher alcohols favors the regeneration of more the aldehyde dehydrogenase activity of flor strain is also higher NAD(P)+; high alcohol dehydrogenase activity has been observed in compared to fermentative strain (Aranda and del Olmo, 2003). These flor forming yeast under biological aging (Mauricio et al., 2001). It was results are consistent with the need for flor yeast to acquire energy suggested that NADPH+ can be re-oxidized during amino acid synthe- from ethanol as the main carbon source. Interestingly, it seems that re- sis. This notion is supported by the release of amino acids such as L-thre- sponse to acetaldehyde stress is dependent on the general stress re- onine, L-tryptophan, L-cysteine, and L-methionine by yeast under low sponse mechanism involving the transcription factors Msn2/4p and oxygen level (Mauricio et al., 2001). Hsf1p (Aranda and del Olmo, 2003). During aging, there is a reduction in the medium content of proline, In aerobic conditions, ethanol affects respiratory chain function in the most abundant amino acid (Dos Santos et al., 2000). Proline is used yeast mitochondria, leading to substantial increases in the amounts of by flor yeast as a nitrogen source but not as a carbon source (Martinez reactive oxygen species (ROS). Ethanol and acetaldehyde may cause et al., 1995). Proline metabolism requires a permease and proline oxi- more changes to mtDNA sequences than nuclear DNA due to the prox- dase, both of which are inducible and oxygen dependant (Ingledew imity between mtDNA and the major site of endogenous ROS produc- et al., 1987); proline utilization is not restricted during the aging process tion (Castrejon et al., 2002). Flor yeast resistance to these effects since flor yeasts have an oxidative metabolism. might be due to their particularly efficient antioxidant defense system. There are metabolic differences between the different flor yeast. For example, SOD1 (a superoxide dismutase) expression is much higher The formerly called S. montuliensis (reclassified T. delbrueckii) and in flor yeast strains than in laboratory strains (Penate et al., 2001). Re- S. rouxii (reclassified Z. rouxii) consume the most ethanol and, there- cently, it has been shown that overexpression of genes involved in oxi- fore, produce greater quantities of acetaldehyde (Martinez et al., dative stress resistance in flor yeast strain allows a more rapid velum 1995, 1997c). development compared to the parental strain (Fierro-Risco et al., 2013). Overexpression of SOD1 and SOD2 leads to an increase superox- 6. The evolution of chemical compounds during biological aging ide dismutase activity and indirectly to increase catalase, glutathione reductase, and glutathioneperoxidase activities. HSP12 overexpression The biological aging of Vin Jaune takes six years while aging of Xeres showed higher levels of glutathione peroxidase and reductase activities. (also known as sherry) can take much longer. There can be many more Whatever the overexpression system studied, Fierro-Risco et al. (2013) changes in wine composition during this time and, consequently, differ- observed a higher intracellular glutathione content, a reduction in ences in sensorial perception. These changes are mainly due to the accu- peroxidized lipid concentration, and higher resistance to oxidative mulation of acetaldehyde, butanediol, acetoin, diacetyl and other stress conditions. These changes lead to a faster growth of the velum compounds during flor yeast metabolism (Martınez et al., 1998; and a better survival of the transformants compared to the wild type Charpentier et al., 2002). yeast. These results underlined the importance of the antioxidant de- Ethanol, the main compound of the wine, changes in concentration fense in the ability of flor strain to develop and survive. during aging. This will depend on the amount which is transformed Many genes are involved in ethanol resistance in yeasts, however into acetaldehyde and on its evaporation. Numerous factors can influ- the importance of these genes in flor yeast has not been fully investigat- ence ethanol concentration; for example, the yeast species, cellar tem- ed. The BTN2 gene is involved in both velum formation and ethanol perature, ratio between velum surface and wine volume and the ratio H. Alexandre / International Journal of Food Microbiology 167 (2013) 269–275 273 between velum surface and air volume in the cask. The alcohol concen- spontaneously formed by chemical condensation of acetaldehyde with tration in Xeres can reduce by an average of 0.2 to 0.3% (v/v) per year. α-ketobutyric acid during biological aging. The olfactive detection The rate is greater at the beginning of aging, especially during velum for- threshold of this compound is low (10 μg/L) (Pham et al., 1996). A mation (Martinez de la Ossa et al., 1987a,b; Martınez et al., 1998). This is high correlation exists between the level of sotolon and the aromatic in- logical as the energy needs of yeast are greater at the beginning of tensity of Vin Jaune (Pham et al., 1996). Sotolon level increases during velum formation. In contrast, the ethanol concentration in Vin Jaune in- aging and can be found in concentrations below 40 μg/L in one-year- creases with age; this is probably due to low hygrometry in the cellar old wines but more than 80 μg/L for four-year-old wines. Its concentra- which leads to a greater water evaporation than ethanol evaporation tion is not homogenous in casks and is lower under the velum. (Arbault, 1977). Collin et al. (2012) identified other new compounds which are pres- The volatile acidity of sherry wines (mainly acetic acid) can decrease ent in Jura flor sherry wine in addition to sotolon. Abhexon accounts for considerably, by more than 0.1 g/L (Martinez de la Ossa et al., 1987a,b; the sweet-caramel note of coffee beverages whereas the oxidation Martınez et al., 1998), during the first months. A similar decrease is ob- product of theaspirane has a strong grenadine-like odor. It is suggested served for Vin Jaune. However, the level of acetic acid fluctuates during that abhexon is produced by the reaction between α-ketobutyric acid aging. If the velum sinks during aging, acetic acid bacteria can develop and propanal. Dihydrodehydro-β-ionone and 4-hydroxy-7,8-dihydro- and increase the volatile acidity. Alternatively, acetic acid could be me- β-ionone are thought to be generated from theaspirane oxidation. Fur- tabolized during velum formation to form acetyl-CoA. Ethyl acetate is ther work is needed to study the existence of these compounds in other also present in wine; its concentration tends to decrease during biolog- flor sherry wines as this has currently not been reported. ical aging (Martinez de la Ossa et al., 1987a,b). Glycerol is a non-volatile by-product of alcoholic fermentation which 7. Factors affecting velum formation contributes to the organoleptic properties of wine (Noble and Bursick, 1984). During biological aging, glycerol level decreases and it may also The air–liquid biofilm is thought to be an adaptive characteristic that act as a carbon source for flor yeast (Charpentier et al., 2002). The glycerol allows cells to access oxygen and grow on non-fermentable carbon concentration in Xeres is initially around 6–7g/Lbutaround0.3g/Latthe sources such as ethanol (Zara et al., 2005). The biofilm therefore de- end of the process, influencing the characteristics of the wine produced. velops at the end of alcoholic fermentation when cell growth becomes The characteristics of wine that develop with aging are largely de- dependent on oxygen availability. The biofilm may also develop on pendent on acetaldehyde (systemically ethanal). During this process, non-fermentable substrates other than ethanol (Zara et al., 2010). sherries generally accumulate 300 to 400 mg/L ethanal but 700– Based on the dry weight of bio film formed per mg of substrate, it ap- 800 mg/L acetaldehyde has been reported in some cases (Martinez pears that the best carbon sources are glycerol, ethyl acetate and then et al., 1997c). The mean concentration in Vin Jaune is around 400– ethanol: the weight of biofilm formed with glycerol is 3.8 times greater 500 mg/L (Charpentier et al., 2004). Vernaccia di Oristano (from than with ethanol and 2.7 times greater than in the presence of ethyl ac- Sardinia), is also aged under velum yeast but has lower acetaldehyde etate (per mg of available carbon). Biofilm formation is inconsistent on concentrations (100–200 mg/L) (Galletti and Carnacini, 1996). Acetal- acetic acid, succinic acid and lactic acid. Unlike other microbial biofilms, dehyde concentration can reach higher values during aging, but de- those formed with wine strains have long been thought to float without crease due to its high volatility or by reacting with other wine the need of an extracellular polysaccharide matrix (Zara et al., 2002; constituent (such as ethanol) to form acetals. The chemical reaction of Ishigami et al., 2006). More recently, the existence of an extracellular acetaldehyde with ethanol produces 1,1-diethoxy ethane; a reaction matrix has been revealed (Zara et al., 2009). The presence of such a ma- with acetaldehyde and 2,3-butanediol gives 2,3,5-trimethyl-1,3- trix is intriguing and deserves further investigation in order to clarify dioxolane (Webb and Noble, 1976). Other acetals have been identified the role of this matrix together with its composition. by William and Stauss (1978) and Etievant (1979). Factors linked to the winemaking process could influence the devel- The production of acetaldehyde has been shown to be affected by opment of the velum in Xeres, Vin Jaune or other flor sherry-type wines. aerobic growth, medium composition, the nature of insoluble materials Difficulties in producing flor type sherries using the traditional film used to clarify the musts, aging, sulfur dioxide concentration in grape process were first reported by Crowther and Truscott (1957).They must and aeration (Romano et al., 1994; Berlanga et al., 2001). However, found that agitation before inoculation favors the development of the acetaldehyde concentration is mainly affected by the yeast strain pres- velum whereas ammonium inhibited the formation of the velum. ent during aging (Martinez et al., 1995, 1997c; Ibeas et al., 1997). Theroleofnitrogenonbiofilm formation still needs to be clarified. During aging, Vin Jaune is exposed to varying climate conditions: Some reports claim that no velum could form on old wine because of ni- 25 °C in summer and 10 °C during winter. The velum develops during trogen deficiency. However, other studies report that amino acid addi- the warmer season and sinks during winter. According to Charpentier tion has no effect on velum formation (Berlanga et al., 2006). FLO11 et al. (2002), the sinking of velum is correlated with an increase in poly- expression, a key gene for velum formation, is regulated by nitrogen sta- saccharides released by dead yeast. This phenomenon is attributed to tus (Zara et al., 2011). Biofilm formation was shown to favor ammonium yeast autolysis. Analysis of macromolecules released during the aging sulphate concentrations ranging from 0 to 37.5 mM and an inhibition of of Vin Jaune revealed that they contained 73.3–78.5% neutral sugars yeast film formation was observed at concentrations between 150 and and 6–7% proteins. The amino acid composition did not change during 450 mM. Consistent with this observation, FLO11 gene expression is re- aging. A high serine and threonine content was found, which would duced under conditions of high nitrogen availability. be expected as these amino acids are involved in the O-glycosidic link- Velum formation is dependent on FLO11 expression and fatty acid ages present in yeast mannoproteins. As aging progressed, the mannose biosynthesis. Zara et al. (2012) demonstrated that biofilm formation and glucose contents of macromolecules increased, but the ratio of poly- was affected by the presence of cerulenin, an antibiotic that inhibits meric mannose to glucose decreased. The mannoproteins released in de-novo fatty acid biosynthesis. The addition of fatty acids restored bio- wine during aging were partially hydrolyzed (Charpentier et al., film formation. Inositol availability is also thought to affect biofilm for- 2004), which is the results of β-glucanase activity. mation, possibly due to its role in the assembly of the GPI-anchor of A large increase in mannose content has also been detected in sherry Flo11p (Zara et al., 2012). wine during aging which reflects the enrichment in polysaccharides due Velum formation in sherry wines can be accelerated by inoculation to autolysis (Villamiel et al., 2008). with cells that are pre-adapted to ethanol and by adding an easily as- Sotolon (4,5-dimethyl-3-hydroxy-2(5)H-furanone) is an important similable carbon source or vitamins (Fatichenti et al., 1983; Martinez compound of sherry-like wine which evokes strong spicy-curry-nut et al., 1997c; Munoz et al., 2005). Ibeas et al. (1997) reported that sherry notes (Dubois et al., 1976; Martin et al., 1992). This compound is wine ethanol (15%), combined with temperatures beyond a critical 274 H. Alexandre / International Journal of Food Microbiology 167 (2013) 269–275 value of 22.5°, resulted in velum deterioration and correlated with a sig- Budroni, M., Giordano, G., Pinna, G., Farris, G.A., 2000. A genetic study of natural flor fi strains of Saccharomyces cerevisiae isolated during biological ageing from Sardinian ni cant increase in petite mutants (by 20 to 30% at 25 °C). wines. J. Appl. Microbiol. 89, 657–662. To improve the control of flor film formation, the kinetics of its Budroni, M., Zara, S., Zara, G., Pirino,G.,Mannazzu,I.,2005.Peculiarities of flor growth has been modeled in conditions which mimic flor sherry aging strains adapted to Sardinian sherry-like wine ageing conditions. FEMS Yeast Res. 5, 951–958. (Gutierrez et al., 2010). The authors suggest that their model can be Castrejon, F., Codon, A.C., Cubero, B., Benıtez, T., 2002. Acetaldehyde and ethanol are re- used for optimization and to control the process of biological aging of sponsible for mitochondrial DNA restriction fragment length polymorphism in flor wines. yeasts. Syst. Appl. Microbiol. 25, 462–467. Lactic acid bacteria (LAB) often develop in the velum and can lead to Charpentier, C., Dos Santos, A.M., Feuillat, M., 2002. Contribution à l'étude du métabolisme des levures à voile dans l'élaborationdes vins jaunes du Jura. Rev. Fr. d'œnologie 195, organoleptic deviation and deterioration of wine (Roldán et al., 2012). 33–36. To prevent the development of LAB, lysozyme could be used as its Charpentier, C., Dos Santos, A.M., Feuillat, M., 2004. Release of macromolecules by Saccha- fl “ ” muramidase activity has an antibacterial action (Roldán et al., 2012). romyces cerevisiae during ageing of French or sherry wine Vin Jaune .Int.J.Food Microbiol. 96, 253–262. However, if inoculation is used, lysozyme affects cell multiplication Charpentier, C., Colin, A., Alais, A., Legras, J.-L., 2009. French Jura flor yeasts: genotype and and the membrane hydrophobicity of the yeast. This inhibits their ag- technological diversity. Anton Leeuw. 95, 263–273. gregation, flotation and subsequent development of flor velum. When Collin, S., Nizet, S., Claeys Bouuaert, T., Despatures, P.-M., 2012. Main odorants in jura flor- sherry wines. Relative contributions of sotolon, abhexon, and theaspirane-derived lysozyme is used, CO2 cannot be trapped and the cells are unable to compounds. J. Agric. Food Chem. 60, 380–387. reach the surface (Roldán et al., 2012). Crowther, R.F., Truscott, J.H.L., 1957. The Use of Agitation in the Making of Flor Type Sher- ry. American Society Meeting in Enology, Asilomar, California (July 1956). Del Carmen Plata, M., Mauricio, J.C., Millan, C., Ortega, J.M., 1998. In vivo specific activity of alcohol acetyltransferase and esterase in two flor yeast strains during biological aging 8. Conclusion of sherry wines. J. Ferment. Bioeng. 85, 369–374. Dos Santos, A.M., Feuillat, M., Charpentier, C., 2000. Flor yeast metabolism in a model sys- Yeasts which are able to form a velum are likely to have acquired the tem similar to cellar ageing of the 3French Vin Jaune”: evolution of some by-products, – fl nitrogen compounds and polysaccharide. Vitis 39, 129 134. property to oat as an adaptive mechanism to cope with harsh environ- Dubois, P., Rigaud, J., Dekimpe, J., 1976. Identification de la diméthyl-4,5-tétrahydrofurane- mental conditions; the specific phenotypic traits of flor yeast seem to dione-2,3 dans le vin Jaune du Jura. Lebensm. Wiss. Technol. 9, 366–368. have been driven by their environmental constraints. Saccharomyces Espinazo-Romeu, M., Cantoral, J.M., Matallana, E., Aranda, A., 2008. Btn2p is involved in ethanol tolerance and biofilm formation in flor yeast. FEMS Yeast Res. 8, 1127–1136. yeast is a good model to study yeast domestication and adaptive mech- Esteve-Zarzoso, B., Peris-Toran, M.J., Garcia-Maiquez, E., Uruburu, F., Querol, A., 2001. anisms. Flor yeast is also a unique model for mixed biofilm and interspe- Yeasts population dynamics during the fermentation and biological aging of sherry cies interaction studies, as bacteria and other yeast can also be present wines. Appl. Environ. Microbiol. 67, 2056–2061. fi in the velum. The origin of this yeast still needs to be confirmed. Further Esteve-Zarzoso, B., Fernandez-Espinar, M.T., Querol, A., 2004. Authentication and identi - cation of Saccharomyces cerevisiae “fl or” yeast races involved in sherry ageing. Anton work could explore why this yeast is only naturally found in few places, Leeuw. 85, 151–158. when they appeared in each location and what the phylogenetic rela- Etievant, P., 1979. Constituants volatils du Vin Jaune: Identification d'acétals dérivés du – tionships between these flor yeasts are. glycérol. Lebensm. Wiss. Technol. 12, 115 120. Fatichenti, F., Farris, G.A., Deiana, P., 1983. Improved production of a Spanish-type sherry The ecology of these yeasts is still unknown. Outstanding questions by using selected indigenous film-forming yeasts as starters. Am. J. Enol. Vitic. 34, include i) whether flor yeast could be found on grapes or are only pres- 216–220. ent in the cellar and ii) whether flor yeast only appear after alcoholic fer- Fernandez-Espinar, M.T., Esteve-Zarzoso, B., Querol, A., Barrio, E., 2000. RFLP analysis of the ribosomal internal transcribed spacer and the 5.8S rRNA gene region of the mentation or are present both during alcoholic fermentation and aging. genus Saccharomyces: a fast method for species identification and the differentiation The factors that favor velum formation, and its persistence during of flor yeasts. Anton Leeuw. 78, 87–97. aging, are particularly relevant for the production of wine quality and Fidalgo, M., Barrales, R.R., Ibeas, J., Jimenez, J., 2006. Adaptive evolution by mutations in – fl the FLO11 gene. Proc. Natl. Acad. Sci. 103, 11228 11233. require further investigation. Wine quality is in uenced by sinking Fidalgo, M., Barrales, R.R., Jimenez, J., 2008. Coding repeat instability in the FLO11 gene of of the velum and this phenomenon also requires further study Saccharomyces yeasts. Yeast 25, 879–889. (Charpentier et al., 2004; Villamiel et al., 2008). Fierro-Risco, J., Rincon, A.M., Benitez, T., Codon, A.C., 2013. Overexpression of stress- related genes enhances cell viability and velum formation in sherry wine yeasts. Appl. Genet. Mol. Biotechnol. http://dx.doi.org/10.1007/s00253-013-4850-9. Galletti, G.C., Carnacini, A., 1996. Volatile composition of Vernaccia di Oristano sherry-like References wine as affected by biological ageing. J. Sci. Food Agric. 70, 44–50. Guijo, S., Mauricio, J.C., Salmon, J.M., Ortega, J.M., 1997. Determination of the relative ploi- Alexandre, H., Bertrand, F., Charpentier, C., 1998. Ethanol induced yeast film formation dy in different Saccharomyces cerevisiae strains used for fermentation and ‘flor’ film with cell surface hydrophobicity as a major determinant. Food Technol. Biotechnol. ageing of dry sherry-type wines. Yeast 13, 101–117. 36, 27–30. Gutierrez, P., Roldan, A., Caro, I., Perez, L., 2010. Kinetic study of the velum formation by Alexandre, H., Blanchet, S., Charpentier, C., 2000. Identification of a 49 kDa hydrophobic Saccharomyces cerevisiae (beticus ssp.) during the biological aging of wines. Process. cell wall mannoprotein present in velum yeast which may be implicated in velum Biochem. 45, 493–499. formation. FEMS Microbiol. Lett. 185, 147–150. Ibeas, J.I., Jimenez, J., 1996. Genomic complexity and chromosomal rearrangements in Aranda, A., del Olmo, M.L., 2003. Response to acetaldehyde stress in the yeast Saccharo- wine-laboratory yeast hybrids. Curr. Genet. 30, 410–416. myces cerevisiae involves a strain-dependent regulation of several ALD genes and is Ibeas, J.I., Jimenez, J., 1997. Mitochondrial DNA loss caused by ethanol in Saccharomyces mediated by the general stress response pathway. Yeast 20, 747–759. cerevisiae flor yeasts. Appl. Environ. Microbiol. 63, 7–12. Aranda, A., del Querol, A., Olmo, M.L., 2002. Correlation between acetaldehyde and etha- Ibeas, J.I., Lozano, I., Perdigones, F., Jimenez, J., 1996. Detection of Dekkera-Brettanomyces nol resistance and expression of HSP genes in yeast strains isolated during the biolog- strains in sherry by a nested PCR method. Appl. Environ. Microbiol. 62, 998–1003. ical aging of sherry wines. Arch. Microbiol. 177, 304–312. Ibeas, J.I., Lozano, I., Perdigones, F., Jimenez, J., 1997. Effects of ethanol and temperature on Arbault, B., 1977. Les Vins Jaunes. Rapport d'étude. Société de Viticulture du Jura, Arbois, the biological aging of sherry wines. Am. J. Enol. Vitic. 48, 71–74. France. Iimura, Y., Hara, S., Otsuka, K., 1980. Cell surface hydrophobicity as a pellicle formation Bakalinsky, A.T., Snow, R., 1990. The chromosomal constitution of wine strains of Saccha- factor in film strain of Saccharomyces. Agric. Biol. Chem. 44, 1215–1222. romyces cerevisiae. Yeast 6, 367–382. Infante, J.J., Dombek, K.M., Rebordinos, L., Cantoral, J.M., Young, E.T., 2003. Genome-wide Barrales, R.R., Jimenez, J., Ibeas, J.I., 2008. Identification of novel activation mechanisms for amplifications caused by chromosomal rearrangements play a major role in the adap- FLO11 regulation in Saccharomyces cerevisiae. Genetics 178, 145–156. tive evolution of natural yeast. Genetics 165, 1745–1759. Barrales, R.R., Jimenez, J., Ibeas, J.I., 2012. Chromatin Modulation at the FLO11 Promoter of Ingledew, W.M., Magnus, C.A., Sosulski, F.W., 1987. Influence of oxygen on proline utiliza- Saccharomyces cerevisiae by HDAC and Swi/Snf Complexes. Genetics 191, 791–803. tion during the wine fermentation. Am. J. Enol. Vitic. 38, 246–248. Benitez, T., Rincon, A.M., Codon, A.C., 2011. Yeasts used in biologically aged wines. In: Ishigami, M., Nakagawa, Y., Hayakawa, M., Iimura, Y., 2004. FLO11 is essential for flor for- Carrascosa Santiago, A.V., Munoz, R., Gonzalez Garcia, R. (Eds.), Molecular Wine Mi- mation caused by the C-terminal deletion of NRG1 in Saccharomyces cerevisiae. FEMS crobiology. Academic Press, pp. 51–84. Microbiol. Lett. 237, 425–430. Berlanga, T.M., Atanasio, C., Mauricio, J.C., Ortega, J.M., 2001. Influence of aeration on the Ishigami, M., Nakagawa, Y., Hayakawa, M., Iimura, Y., 2006. FLO11 is the primary factor in physiological activity of flor yeasts. J. Agric. Food Chem. 49, 3378–3384. flor formation caused by cell surface hydrophobicity in wild-type flor yeast. Biosci. Berlanga, T.M., Millàn, C., Mauricio, J.C., Ortega, J.M., 2006. Influence of nitrogen on the bi- Biotechnol. Biochem. 70, 660–666. ological aging of sherry wine. J. Sci. Food Agric. 86, 2113–2118. Kovacs, M., Stuparevic, I., Mrsa, V., Maraz, A., 2008. Characterization of Ccw7p cell wall Blandino, A., Caro, I., Cantero, D., 1997. Comparative study of alcohol dehydrogenase ac- proteins and the encoding genes of Saccharomyces cerevisiae wine yeast strains: tivity in flor yeast extracts. Biotechnol. Lett. 19, 651–654. relevance for flor formation. FEMS Yeast Res. 8, 1115–1126. H. Alexandre / International Journal of Food Microbiology 167 (2013) 269–275 275

Kurtzman, C.P., Fell, J.W., Boekhout, T., 2011. The Yeasts, A Taxonomic Study, 5th ed. Pirino, G., Zara, S., Pinna, G., Farris, G.A., Budroni, M., 2004. Diversity of Y region at HML Elsevier B.V. locus in a Saccharomyces cerevisiae strain isolated from a Sardinian wine. Anton Martin, B., Etiévant, P., LeQuéré, J.L., Schlich, P., 1992. More clues about sensory impact of Leeuw. 85, 29–36. sotolon in some flor-sherry wines. J. Agric. Food Chem. 40, 475–478. Pozo-Bayón, A., Moreno-Arribas, V., 2011. Sherry wines. Adv. Food Nutr. Res. 63, 17–40. Martinez de la Ossa, E., Caro, I., Bonat, M., Perez, L., Domecq, B., 1987a. Dry extract in sher- Purevdorj-Gage, B., Orr, M.E., Stoodley, P., Sheehan, K.B., Hyman, L.E., 2007. The role of ry and its evolution in the aging of sherry. Am. J. Enol. Vitic. 38, 321–325. FLO11 in Saccharomyces cerevisiae biofilm development in a laboratory based flow- Martinez de la Ossa, E., Perez, L., Caro, I., 1987b. Variations of the major volatiles through cell system. FEMS Yeast Res. 7, 372–379. ageing of sherry. Am. J. Enol. Vitic. 38, 293–297. Reynolds, T.B., Fink, G.R., 2001. Bakers' yeast, a model for fungal biofilm formation. Sci- Martinez, P., Codon, A.C., Pérez, L., Benitez, T., 1995. Physiological and molecular charac- ence 291, 878–881. terisation of flor yeasts: polymorphism of flor yeast populations. Yeast 11, Ristow, H., Seyfarth, A., Lochmann, E.R., 1995. Chromosomal damages by ethanol and ac- 1399–1411. etaldehyde in Saccharomyces cerevisiae as studied by pulsed field gel electrophoresis. Martínez, P., Pérez Rodríguez, L., Benítez, T., 1997a. Factors which affect velum formation Mutat. Res. 326, 165–170. by flor yeasts isolated from sherry wine. Syst. Appl. Microbiol. 20, 154–157. Roldán, A., Lasanta, C., Caro, I., Palacios, V., 2012. Effect of lysozyme on “flor” velum yeasts Martínez, P., Pérez Rodríguez, L., Benítez, T., 1997b. Evolution of flor yeast population dur- in the biological aging of sherry wines. Food Microbiol. 30, 245–252. ing biological aging of fino sherry wine. Am. J. Enol. Vitic. 48, 160–168. Romano, P., Suzzi, G., Turbanti, L., Polsinelli, M., 1994. Acetaldehyde production in Saccha- Martinez, P., Perez-Rodriguez, L., Benitez, T., 1997. Velum formation by flor yeasts isolated romyces cerevisiae wine yeasts. FEMS Microbiol. Lett. 118, 213–218. from sherry wines. Am. J. Enol. Vitic. 48, 55–62. Suarez-Lepez, J.A., Inigo-Leal, B., 2004. Microbiologia enological. Fundamentos de Martınez, P., Valcarcel, M.J., Perez, L., Benıtez, T., 1998. Metabolism of Saccharomyces vinificaciòn. Ediciones Mundi-Prensa, Madrid, Spain 673–716. cerevisiae flor yeasts during fermentation and biological aging of fino sherry: by- Turcotte, B., Liang, X.B., Robert, F., Soontorngun, N., 2010. Transcriptional regulation of products and aroma compounds. Am. J. Enol. Vitic. 49, 240–250. nonfermentable carbon sources in budding yeast. FEMS Yeast Res. 10, 2–13. Mauricio, J.C., Ortega, J.M., 1993. Influence of fermentation conditions on specificactivityof Verstrepen, K.J., Klis, F.M., 2006. Flocculation, adhesion and biofilm formation in yeasts. the enzymes alcohol and aldehyde dehydrogenase from yeasts. Microbios 75, 95–106. Mol. Microbiol. 60, 5–15. Mauricio, J.C., Moreno, J.J., Ortega, J.M., 1997. In vitro specific activities of alcohol and al- Villamiel, M., Carmen-Polo, M., Moreno-Arribas, M.V., 2008. Nitrogen compounds and dehyde dehydrogenases from two flor yeasts during controlled wine aging. J. Agric. polysaccharides changes during the biological ageing of sherry wines. Lebensm. Food Chem. 45, 1967–1971. Wiss. Technol. 41, 1842–1846. Mauricio, J.C., Valero, E., Millan, C., Ortega, J.M., 2001. Changes in nitrogen compounds in Webb, D.A., Noble, A.C., 1976. Aroma sherry wine. Biotechnol. Bioeng. 18, 939–952. must and wine during fermentation and biological aging by flor yeasts. J. Agric. Food William, P.J., Stauss, C.R., 1978. The influence of film activity on the aroma volatiles of flor- Chem. 49, 3310–3315. sherries. A study of volatiles isolated by headspace sampling. J. Inst. Brew. 84, Meem, M.H., Cullen, P.J., 2012. The impact of protein glycosylation on Flo11-dependent 148–152. adherence in Saccharomyces cerevisiae. FEMS Yeast Res. 12, 809– 818. Zara, S., Antonio Farris, G., Budroni, M., Bakalinsky, A.T., 2002. HSP12 is essential for bio- Mesa, J.J., Infante, J.J., Rebordinos, L., Cantoral, J.M., 1999. Characterization of yeasts in- film formation by a Sardinian sherry strain of S. cerevisiae. Yeast 19, 269–276. volved in the biological ageing of sherry wines. Food Sci. Technol. Res. 32, 114–120. Zara, S., Bakalinsky, A.T., Zara, G., Pirino, G., Demontis, M.A., Budroni, M., 2005. FLO11- Mesa, J.J., Infante, J.J., Rebordinos, L., Sanchez, J.A., Cantoral, J.M., 2000. Influence of the yeast based model for air–liquid interfacial biofilm formation by Saccharomyces cerevisiae. genotypes on enological characteristics of sherry wines. Am. J. Enol. Vitic. 51, 15–21. Appl. Environ. Microbiol. 71, 2934–2939. Mortimer, R.K., Romano, P., Suzzi, G., Polsinelli, M., 1994. Genome renewal: a new phe- Zara, G., Zara, S., Pinna, C., Marceddu, S., Budroni, M., 2009. FLO11 gene length and tran- nomenon revealed from a genetic study of 43 strains of Saccharomyces cerevisiae de- scriptional level affect biofilm-forming ability of wild flor strains of Saccharomyces rived from natural fermentation of grape must. Yeast 10, 1543–1552. cerevisiae. Microbiology 159, 3838–3846. Munoz, D., Peinado, R.A., Medina, M., Moreno, J., 2005. Biological aging of sherry wines Zara, S., Gross, M.K., Zara, G., Budroni, M., Bakalinsky, A.T., 2010. Ethanol-independent bio- using pure cultures of two flor yeast strains under controlled microaeration. film formation by a flor wine yeast strain of Saccharomyces cerevisiae. Appl. Environ. J. Agric. Food Chem. 53, 5258–5264. Microbiol. 76, 4089–4091. Noble, A.C., Bursick, A.C., 1984. The contribution to glycerol to perceived viscosity and Zara, G., Budroni, M., Mannazzu, I., Zara, S., 2011. Air–liquid biofilm formation is depen- sweetness in white wine. Am. J. Enol. Vitic. 35, 110–112. dent on ammonium depletion in a Saccharomyces cerevisiae flor strain. Yeast 28, Penate, X., Codon, A.C., Castrejon, F., Benitez, T., 2001. Papel de la superóxido dismutasa en 809–8014. el velo de flor. II Congreso de laSociedad Española de Genética, Sevilla. Zara, G., Goffrini, P., Lodi, T., Zara, S., Manazzu, I., 2012. FLO11 expression and lipid biosyn- Pham, T.T., Guichard, E., Charpentier, C., 1996. Quantification of sotolon in French Vins- thesis are required for air–liquid biofilm formation in a Saccharomyces cerevisiae flor jaunes during aging under the yeast. Sci. Aliment. 16, 281287. strain. FEMS Yeast Res. 12, 864–866.