Modulation of aroma and chemical composition of Albariño semi-synthetic wines by non-wine Saccharomyces yeasts and bottle aging.

Saccharomyces yeasts from different origins and species fermented in a semi-synthetic must containing aroma precursor of cv. Albariño and polyfunctional mercaptans precursors. The resulting wines were subjected to accelerate anoxic aging. Afterward, aroma profiles were analyzed by distinct gas chromatography methodologies. Cryotolerant strains showed better fermentation performances with significant differences in volatile and non-volatile fermentation products than Saccharomyces cerevisiae (S. cerevisiae). We suggested that the highest levels γ-butyrolactone and diethyl succinate in Saccharomyces uvarum (S. uvarum) strains, together with their substantial succinic acid yields, could be related to greater flux through the GABA shunt. These strains also had the highest production of ß-phenylethyl acetate, geraniol, and branched-chain ethyl esters. The latter compounds were highly increased by aging, while acetates and some terpenes decreased. S. kudriavzevii strains showed a remarkable ability to release polyfunctional mercaptans, with SK1 strain yielding up to 47-fold and 8-fold more 4-methyl-4-mercaptopentan-2-one (4MMP) than S. cerevisiae and S. uvarum strains, respectively. The wild S. cerevisiae beer isolate showed a particular aroma profile due to the highest production of ethyl 4-methylvalerate (lactic and fruity notes), γ-octalactone (coconut), and furfurylthiol (roasted coffee). The latter compound is possibly produced from the pentose phosphate pathway (PPP). Since erythritol, another PPP intermediate was largely produced by this strain.

Metabolic differences between a wild and a wine strain of Saccharomyces cerevisiae during fermentation unveiled by multi-omic analysis.

Saccharomyces cerevisiae, a widespread yeast present both in the wild and in fermentative processes, like winemaking. During the colonization of these human-associated fermentative environments, certain strains of S. cerevisiae acquired differential adaptive traits that enhanced their physiological properties to cope with the challenges imposed by these new ecological niches. The advent of omics technologies allowed unveiling some details of the molecular bases responsible for the peculiar traits of S. cerevisiae wine strains. However, the metabolic diversity within yeasts remained poorly explored, in particular that existing between wine and wild strains of S. cerevisiae. For this purpose, we performed a dual transcriptomic and metabolomic comparative analysis between a wild and a wine S. cerevisiae strains during wine fermentations performed at high and low temperatures. By using this approach, we could correlate the differential expression of genes involved in metabolic pathways, such as sulfur, arginine and thiamine metabolisms, with differences in the amounts of key metabolites that can explain some important differences in the fermentation performance between the wine and wild strains.

Metabolome segregation of four strains of Saccharomyces cerevisiae, Saccharomyces uvarum and Saccharomyces kudriavzevii conducted under low temperature oenological conditions.

The monitoring of fermentation at low temperatures (12-15°C) is a current practice in the winery for retention and enhancement of the flavour volatile content of wines. Among Saccharomyces species, Saccharomyces uvarum and Saccharomyces kudriavzevii have revealed interesting industrial properties, including better adaptation at low temperatures. To gather deeper knowledge of the fermentative metabolism at a low temperature of these species together with S. cerevisiae, we performed a comparative metabolomic analysis using four representative strains. We used batch cultures to obtain an exhaustive and dynamic image of the metabolome of strains passing through the sequential stresses related to the winemaking environment. A great variety of intra- and extracellular metabolites (>500 compounds) were quantified across fermentation using distinct chromatographic methods. Besides a global decrease in the lipid composition of the four strains when they entered into the stationary phase, we reported some strain-specific high magnitude changes. Examples of these differences included divergent patterns of production of short-chain fatty acids and erythritol in the S. uvarum strain. Strains also differed in expression for aromatic amino acid biosynthesis and sulphur metabolism, including the glutathione pathway. These data will allow us to refine and obtain the most value of fermentations with this alternative Saccharomyces species.

A time course metabolism comparison among Saccharomyces cerevisiae, S. uvarum and S. kudriavzevii species in wine fermentation.

In this study, we presented the first metabolome time course analysis performed among a set of S. uvarum, S. kudriavzevii and S. cerevisiae strains under winemaking conditions. Extracellular and intracellular metabolites, as well as physiological parameters of yeast cells, were monitored along the process to find evidence of different metabolic strategies among species to perform alcoholic fermentation. A thorough inspection of time trends revealed several differences in utilization or accumulation of fermentation by-products. We confirmed the ability of S. uvarum and S. kudriavzevii strains to produce higher amounts of glycerol, succinate or some fusel alcohols and their corresponding esters. We also reported differences in the yields of less common fermentative by-products involved in redox homeostasis, namely 2,3 butanediol and erythritol. 2,3 butanediol yield was higher in must ferment with cryophilic strains and erythritol, a pentose phosphate pathway derivative, was particularly overproduced by S. uvarum strains. Contrary to S. cerevisiae, a singular production-consumption rate of acetate was also observed in S. uvarum and S. kudriavzevii fermentations. Since acetate is a precursor for acetyl-CoA production which is involved in the biosynthesis of membrane lipids, cryophilc strains might take advantage of extracellular acetate to remodel cell membrane as ethanol content increased during fermentation.

Differences in metabolism among Saccharomyces species and their hybrids during wine fermentation.

This study aimed to investigate how parental genomes contribute to yeast hybrid metabolism using a metabolomic approach. Previous studies have explored central carbon and nitrogen metabolism in Saccharomyces species during wine fermentation, but this study analyses the metabolomes of Saccharomyces hybrids for the first time. We evaluated the oenological performance and intra- and extracellular metabolomes, and we compared the strains according to nutrient consumption and production of the main fermentative by-products. Surprisingly, no common pattern was observed for hybrid genome influence; each strain behaved differently during wine fermentation. However, this study suggests that the genome of the S. cerevisiae species may play a more relevant role in fermentative metabolism. Variations in biomass/nitrogen ratios were also noted, potentially linked to S. kudriavzevii and S. uvarum genome contributions. These results open up possibilities for further research using different "omics" approaches to comprehend better metabolic regulation in hybrid strains with genomes from different species.

Understanding the role of GRE3 in the erythritol biosynthesis pathway in Saccharomyces uvarum and its implication in osmoregulation and redox homeostasis.

Erythritol is produced in yeasts via the reduction of erythrose into erythritol by erythrose reductases (ERs). However, the genes codifying for the ERs involved in this reaction have not been described in any Saccharomyces species yet. In our laboratory, we recently showed that, during alcoholic fermentation, erythritol is differentially produced by Saccharomyces cerevisiae and S. uvarum species, the latter being the largest producer. In this study, by using BLAST analysis and phylogenetic approaches the genes GRE3, GCY1, YPR1, ARA1 and YJR096W were identified as putative ERs in Saccharomyces cerevisiae Then, these genes were knocked out in our S. uvarum strain (BMV58) with higher erythritol biosynthesis compared to control S. cerevisiae wine strain, to evaluate their impact on erythritol synthesis and global metabolism. Among the mutants, the single deletion of GRE3 markedly impacts erythritol production, although ΔYPR1ΔGCY1ΔGRE3 was the combination that most decreased erythritol synthesis. Consistent with the increased production of fermentative by-products involved in redox balance in the Saccharomyces uvarum strain BMV58, erythritol synthesis increases at higher sugar concentrations, hinting it might be a response to osmotic stress. However, the expression of GRE3 in the S. uvarum strain was found to peak just before the start of the stationary phase, being consistent with the observation that erythritol increases at the start of the stationary phase, when there is low sugar in the medium and nitrogen sources are depleted. This suggests that GRE3 plays its primary function to help the yeast cells to maintain the redox balance during the last phases of fermentation.

Modelling the physiological status of yeast during wine fermentation enables the prediction of secondary metabolism.

Saccharomyces non-cerevisiae yeasts are gaining momentum in wine fermentation due to their potential to reduce ethanol content and achieve attractive aroma profiles. However, the design of the fermentation process for new species requires intensive experimentation. The use of mechanistic models could automate process design, yet to date, most fermentation models have focused on primary metabolism. Therefore, these models do not provide insight into the production of secondary metabolites essential for wine quality, such as aromas. In this work, we formulate a continuous model that accounts for the physiological status of yeast, that is, exponential growth, growth under nitrogen starvation and transition to stationary or decay phases. To do so, we assumed that nitrogen starvation is associated with carbohydrate accumulation and the induction of a set of transcriptional changes associated with the stationary phase. The model accurately described the dynamics of time series data for biomass and primary and secondary metabolites obtained for various yeast species in single culture fermentations. We also used the proposed model to explore different process designs, showing how the addition of nitrogen could affect the aromatic profile of wine. This study underlines the potential of incorporating yeast physiology into batch fermentation modelling and provides a new means of automating process design.

A Dynamic Genome-Scale Model Identifies Metabolic Pathways Associated with Cold Tolerance in Saccharomyces kudriavzevii.

Saccharomyces kudriavzevii is a cold-tolerant species identified as a good alternative for industrial winemaking. Although S. kudriavzevii has never been found in winemaking, its co-occurrence with Saccharomyces cerevisiae in Mediterranean oaks is well documented. This sympatric association is believed to be possible due to the different growth temperatures of the two yeast species. However, the mechanisms behind the cold tolerance of S. kudriavzevii are not well understood. In this work, we propose the use of a dynamic genome-scale model to compare the metabolic routes used by S. kudriavzevii at two temperatures, 25°C and 12°C, to decipher pathways relevant to cold tolerance. The model successfully recovered the dynamics of biomass and external metabolites and allowed us to link the observed phenotype with exact intracellular pathways. The model predicted fluxes that are consistent with previous findings, but it also led to novel results which we further confirmed with intracellular metabolomics and transcriptomic data. The proposed model (along with the corresponding code) provides a comprehensive picture of the mechanisms of cold tolerance that occur within S. kudriavzevii. The proposed strategy offers a systematic approach to explore microbial diversity from extracellular fermentation data at low temperatures. IMPORTANCE Nonconventional yeasts promise to provide new metabolic pathways for producing industrially relevant compounds and tolerating specific stressors such as cold temperatures. The mechanisms behind the cold tolerance of S. kudriavzevii or its sympatric relationship with S. cerevisiae in Mediterranean oaks are not well understood. This study proposes a dynamic genome-scale model to investigate metabolic pathways relevant to cold tolerance. The predictions of the model would indicate the ability of S. kudriavzevii to produce assimilable nitrogen sources from extracellular proteins present in its natural niche. These predictions were further confirmed with metabolomics and transcriptomic data. This finding suggests that not only the different growth temperature preferences but also this proteolytic activity may contribute to the sympatric association with S. cerevisiae. Further exploration of these natural adaptations could lead to novel engineering targets for the biotechnological industry.

A Multiphase Multiobjective Dynamic Genome-Scale Model Shows Different Redox Balancing among Yeast Species of the Saccharomyces Genus in Fermentation.

Yeasts constitute over 1,500 species with great potential for biotechnology. Still, the yeast Saccharomyces cerevisiae dominates industrial applications, and many alternative physiological capabilities of lesser-known yeasts are not being fully exploited. While comparative genomics receives substantial attention, little is known about yeasts' metabolic specificity in batch cultures. Here, we propose a multiphase multiobjective dynamic genome-scale model of yeast batch cultures that describes the uptake of carbon and nitrogen sources and the production of primary and secondary metabolites. The model integrates a specific metabolic reconstruction, based on the consensus Yeast8, and a kinetic model describing the time-varying culture environment. In addition, we proposed a multiphase multiobjective flux balance analysis to compute the dynamics of intracellular fluxes. We then compared the metabolism of S. cerevisiae and Saccharomyces uvarum strains in a rich medium fermentation. The model successfully explained the experimental data and brought novel insights into how cryotolerant strains achieve redox balance. The proposed model (along with the corresponding code) provides a comprehensive picture of the main steps occurring inside the cell during batch cultures and offers a systematic approach to prospect or metabolically engineering novel yeast cell factories. IMPORTANCE Nonconventional yeast species hold the promise to provide novel metabolic routes to produce industrially relevant compounds and tolerate specific stressors, such as cold temperatures. This work validated the first multiphase multiobjective genome-scale dynamic model to describe carbon and nitrogen metabolism throughout batch fermentation. To test and illustrate its performance, we considered the comparative metabolism of three yeast strains of the Saccharomyces genus in rich medium fermentation. The study revealed that cryotolerant Saccharomyces species might use the γ-aminobutyric acid (GABA) shunt and the production of reducing equivalents as alternative routes to achieve redox balance, a novel biological insight worth being explored further. The proposed model (along with the provided code) can be applied to a wide range of batch processes started with different yeast species and media, offering a systematic and rational approach to prospect nonconventional yeast species metabolism and engineering novel cell factories.

Molecular profiling of beer wort fermentation diversity across natural Saccharomyces eubayanus isolates.

The utilization of S. eubayanus has recently become a topic of interest due to the novel organoleptic properties imparted to beer. However, the utilization of S. eubayanus in brewing requires the comprehension of the mechanisms that underlie fermentative differences generated from its natural genetic variability. Here, we evaluated fermentation performance and volatile compound production in ten genetically distinct S. eubayanus strains in a brewing fermentative context. The evaluated strains showed a broad phenotypic spectrum, some of them exhibiting a high fermentation capacity and high levels of volatile esters and/or higher alcohols. Subsequently, we obtained molecular profiles by generating 'end-to-end' genome assemblies, as well as metabolome and transcriptome profiling of two Patagonian isolates exhibiting significant differences in beer aroma profiles. These strains showed clear differences in concentrations of intracellular metabolites, including amino acids, such as valine, leucine and isoleucine, likely impacting the production of 2-methylpropanol and 3-methylbutanol. These differences in the production of volatile compounds are attributed to gene expression variation, where the most profound differentiation is attributed to genes involved in assimilatory sulfate reduction, which in turn validates phenotypic differences in H2 S production. This study lays a solid foundation for future research to improve fermentation performance and select strains for new lager styles based on aroma and metabolic profiles.

Aroma production and fermentation performance of S. cerevisiae × S. kudriavzevii natural hybrids under cold oenological conditions.

This work aims to describe the wine fermentation characteristics of 23 natural S. cerevisiae × S. kudriavzevii hybrid yeasts related to fermentative environments isolated from different regions and their significance for the aroma spectra of the produced wines. Fermentations were performed at 12 °C in artificial must, and S. cerevisiae and S. kudriavzevii pure species strains were used for comparison purposes. We determined the relevant kinetic parameters of fermentation, the concentration of the main metabolites and the main aroma-related compounds produced after fermentation. The results revealed that some strains that show well-rounded characteristics could be profitable yeast starters for low-temperature fermentation in winemaking, such as wine hybrid SPG172 but, surprisingly, also beer hybrid CECT11002, adding the efficient fermentative kinetics to the high production of aroma-related compounds. In addition, a novel metabolic correlation between fermentation performance and aroma production is described.

New Trends in the Uses of Yeasts in Oenology.

The most important factor in winemaking is the quality of the final product and the new trends in oenology are dictated by wine consumers and producers. Traditionally the red wine is the most consumed and more popular; however, in the last times, the wine companies try to attract other groups of populations, especially young people and women that prefer sweet, whites or rosé wines, very fruity and with low alcohol content. Besides the new trends in consumer preferences, there are also increased concerns on the effects of alcohol consumption on health and the effects of global climate change on grape ripening and wine composition producing wines with high alcohol content. Although S. cerevisiae is the most frequent species in wines, and the subject of most studies, S. uvarum and hybrids between Saccharomyces species such as S. cerevisiae×S. kudriavzevii and S. cerevisiae×S. uvarum are also involved in wine fermentations and can be preponderant in certain wine regions. New yeast starters of non-cerevisiae strains (S. uvarum) or hybrids (S. cerevisiae×S. uvarum and S. cerevisiae×S. kudriavzevii) can contribute to solve some problems of the wineries. They exhibit good fermentative capabilities at low temperatures, producing wines with lower alcohol and higher glycerol amounts, while fulfilling the requirements of the commercial yeasts, such as a good fermentative performance and aromatic profiles that are of great interest for the wine industry. In this review, we will analyze different applications of nonconventional yeasts to solve the current winemaking demands.