Advancements in Yeast Genetics Enhance Aromatic Profile of White Wines

Polyfunctional mercaptans, such as varietal thiols, are crucial in the aromatic profile of white wines. These compounds, linked to tropical and herbaceous aromas, exist in grapes as non-volatile precursors. Yeast enzymes during fermentation release the volatile fraction, creating the associated aromas. Factors like grape variety, viticultural practices, and environmental conditions directly affect the availability of these precursors and their transformation into volatile aromas during winemaking. Yeast metabolism, particularly its ability to internalize and process these precursors, is key to determining the final thiol content in wine. Genetic variations among yeasts and differences in winemaking practices can enhance or limit the release of these aromatic compounds. Biotechnological strategies, such as selecting strains with specific genetic traits or modifying them through genetic improvement techniques, have proven effective in increasing thiol concentrations in wines. Additionally, using microbial consortia and new technologies like synthetic biology offer promising prospects for improving thiol production while maintaining an optimal balance between fermentative capacity and aroma generation. As the industry faces challenges like climate change and shifts in consumer preferences, developing these innovative solutions promises significant improvements in the sensory quality of wines.

Polyfunctional mercaptans are sulfurous aromatic compounds that play a key role in the sensory differentiation of wines, mainly whites, especially those made from varieties like Sauvignon Blanc, Semillon, or Verdejo. Among them, varietal thiols significantly contribute to the aromatic quality of wines, providing characteristic tropical fruit and herbaceous aromas, making them one of the most studied aspects of wine chemistry in recent decades. Despite advances in understanding the genetic and metabolic basis of varietal thiol precursor synthesis in vines and their release by yeasts, many questions remain unanswered. Identifying the genes responsible for thiol precursor production in vines and how their expression is regulated in response to different environmental stimuli are areas needing further research. Additionally, elucidating genetic variations among different grape varieties could offer new tools for selecting cultivars with greater potential for producing aromatic wines rich in thiols.

The main compounds forming the thiolic varietal fraction of wines are 4MSP (4-methyl-4-sulfanyl-pentan-2-one), 3MH (3-mercaptohexan-1-ol), and its acetylated derivative, 3MHA (3-mercaptohexyl acetate), which, due to their low perception threshold, have a significant sensory impact on the wine's aromatic profile. These thiols are not found in their volatile form in grapes but are present as precursors conjugated with amino acids or oligopeptides (mainly cysteine and glutathione) that require yeast enzyme action during fermentation to release the volatile compounds responsible for aroma. However, thiol release is not just about yeast enzymatic capacity; it is also influenced by the quantity and availability of precursors in the grape must. The availability of these precursors depends on grape variety, viticultural practices, and environmental conditions, as well as biotic factors like Botrytis cinerea infection. These stimuli can induce the expression of certain genes, increasing thiol precursor accumulation in berries, directly affecting yeast's ability to release thiols during fermentation.

The process of thiol release during fermentation depends on yeasts capable of internalizing their precursors and expressing β-lyase enzymes that break cysteine and glutathione conjugates, releasing the volatile thiols responsible for wine aroma. Besides S. cerevisiae, other yeast species can release thiols and have proven useful tools for increasing these compounds' concentration in wine, either by having genetic variants on the involved genes or by having different transcriptional regulation than S. cerevisiae or by triggering interspecific interaction processes that result in synergistic effects among active yeasts during fermentation. However, it has been observed that although yeasts can release thiols, only a small fraction of the precursors present in the must is effectively transformed into volatile compounds. This is partly due to the complex regulation of gene expression involved in thiol internalization and subsequent release in yeasts, highlighting the need to continue understanding the molecular bases of thiol release in winemaking to develop microbiological tools or enological itineraries that maximize their release.

The final thiol content in wine initially depends on the presence and concentration of their precursors in the must, but yeast metabolism during fermentation is also a key aspect in releasing thiolic aromas. Yeasts, both S. cerevisiae and other non-Saccharomyces species, internalize thiolic precursors and enzymatically hydrolyze them, using the derived ammonium and pyruvate for their metabolism and releasing the precursor's volatile fraction as free thiols. In S. cerevisiae, the genes involved in this process are known. Thiolic precursors are internalized through general amino acid and oligopeptide transporters. OPT1 is described as the main transporter of glutathionylated precursors, while the general amino acid transporter GAP1 is responsible for most cysteinylated precursor internalization. Once in the cytoplasm, β-lyase activity enzymes break the C-S bond of cysteinylated precursors. The genes BNA3, CYS3, GLO1, and mainly IRC7 encode the enzymes responsible for releasing 4MSP from Cys-4MSP. On the other hand, the STR3 gene has been identified as responsible for releasing 3SH from Cys-3SH, although its specificity for this substrate is low. Regarding glutathionylated precursors, once internalized, they transform into cysteinylated precursors through a complex vacuolar pathway involving vacuolar transporters and enzymes encoded by genes like DUG1, DUG2, DUG3, ECM38, among others, and then follow hydrolysis pathways catalyzed by β-lyase activity enzymes. As for the thiol 3SHA, the acetylated derivative of 3SH, its formation from 3SH is catalyzed by the enzyme encoded by the ATF1 gene.

Knowledge about the genetic determinants of thiol production in yeasts has driven the exploration of genetic variants in the genes involved in this metabolism, especially in IRC7. Two allelic variants of this gene in S. cerevisiae have been described: a complete variant (IRC7F) and another with a 38 bp deletion (IRC7S). This deletion alters the open reading frame, generating an early stop codon that produces a shorter enzyme with lower catalytic activity. Studies confirmed that S. cerevisiae strains homozygous for the IRC7S allele have reduced β-lyase activity, resulting in low or no 4MSP thiol production. Moreover, both studies showed that most wine strains of S. cerevisiae are homozygous for the deleted IRC7S allele. This somewhat counterintuitive situation, where the less functional allelic variant of IRC7 is widely distributed in wine strains, has been explained by the association of the IRC7S allele with phenotypic and genomic particularities that favor its prevalence in domesticated habitats, especially in wine.

Additionally, the deletion previously described in the IRC7 gene does not fully explain the differences in Irc7p enzyme activity. Several single nucleotide polymorphisms in the IRC7 sequence have been identified, which, along with allele length, more accurately explain the differences in this enzyme's activity and, therefore, in the ability to release thiols among S. cerevisiae strains. Although the impact of genetic variants in thiolic precursor transporter genes, like OPT1 or GAP1, on thiol production has not been explored in depth, various oligopeptide transporter families have been identified in different S. cerevisiae strains. These families show specific preferences for peptides of different sizes (such as thiolic precursors) and could represent an adaptive advantage in environments with low nitrogen availability, like grape must. Furthermore, there are differences in the preference of different species for different types of thiolic precursors present in grape must, determining their direct contribution to the thiolic aroma profile in wines. Future studies should determine the genetic determinants (allelic variants, gene copy number variation, etc.) that justify different species' ability to transport thiolic precursors, as well as their response to transcriptional regulation mechanisms, key in this process, such as Nitrogen Catabolite Repression (NCR).

Certain transporters located in the plasma membrane play an essential role in capturing thiolic precursors from the extracellular medium (grape must) into yeast cells. The expression of the main involved genes (OPT1 -to a lesser extent, OPT2– and GAP1) is regulated by the availability of nitrogenous nutrients in the medium through NCR mechanisms. This regulation mechanism depends on transcription factors like Gat1p and Gln3p, which induce the expression of genes related to the metabolism of non-preferential nitrogen sources. These transcription factors are regulated by the Ure2p protein, active in conditions of availability of preferential nitrogen sources (i.e., ammonium and some amino acids like glutamine) and whose presence blocks Gat1p and Gln3p action, inhibiting the transcription of genes encoding transporters and enzymes involved in the catabolism of less preferential nitrogen sources, such as amino acids and oligopeptides conjugated with thiolic aromas. When the availability of preferential nitrogen sources decreases, Gln3p and Gat1p are released and translocate to the nucleus, activating the expression of genes for alternative nitrogen source metabolism. Additionally, IRC7, the main gene responsible for producing β-lyase activity enzymes on 4-MSP precursors, is regulated by NCR, and its expression is inhibited in conditions of high availability of preferential nitrogen sources in the must. Moreover, in this case, the availability of sulfur and copper in the medium also plays a crucial role in regulating the expression of β-lyase activity enzymes, like CYS3, whose expression spikes in sulfur-deficient conditions, or IRC7, whose expression decreases at high copper concentrations. It is noteworthy that gene expression levels, besides responding to environmental factors, vary among yeast strains, which, along with the existence of allelic variants, can explain different strains or species' varying ability to release thiols during fermentation. Additionally, different yeast strains exhibit varying sensitivity to NCR mechanisms, a key aspect in explaining different levels of volatile thiol release among strains.

In the last two decades, the molecular mechanisms determining thiolic compound release in wine fermentations have been described, and key aspects have been defined for bioprospecting S. cerevisiae strains with optimal enological characteristics, exploiting the species' natural diversity. Thanks to the knowledge outlined above, there are currently yeast strains with a high capacity for releasing varietal thiols in wines, and certain enological strategies necessary to maximize it are known (i.e., optimizing nitrogen nutrition). Despite this, microbial biotechnology has introduced alternative tools that can contribute to improving this process's efficiency. These include using genetic engineering, microbial consortia, or synthetic biology as key tools to enhance thiolic aromatic compound production.

The use of genetic engineering techniques is regulated by specific regulations that limit their industrial application, although, at an experimental level, dozens of mutants have been developed to better understand the genetic determinants of thiol release. The first work using genetic engineering to increase thiol release in wines introduced the tnaA gene from the bacterium Escherichia coli, encoding a tryptophanase with cystathionine β-lyase activity, into S. cerevisiae in two different ways: integrating it into the yeast genome and as a multicopy plasmid (extrachromosomal). The results showed that genome integration increased the production of volatile thiols 4MSP and 3SH in concentrations up to 25 times higher than unmodified strains and greater than those obtained with multicopy plasmids, highlighting the importance of gene stability and expression in enzymatic activity for thiol release.

In subsequent works, a similar approach was used to insert tnaA, but this time accompanied by S. cerevisiae genes involved in glutathione metabolism, like GSH1 (glutathione synthase) and GTT1 (glutathione S-transferase), to increase yeast glutathione metabolism. Thus, combining both genes with tnaA multiplied 3SH production by 10 compared to the strain carrying only the tnaA gene. Additionally, during the 21st century, extremely versatile and effective cloning techniques have been developed. Gene editing tools like CRISPR-Cas9 (and similar) have also been used for S. cerevisiae genetic editing. For example, mutants have been generated to increase 3SH production by introducing the tnaA gene. In this case, introducing the tnaA gene naturally increases yeast activity of native genes encoding acetyltransferases ATF1 and ATF2, enzymes that transform 3SH into its acetylated derivative, 3-SHA. These strains reached final concentrations of up to 7,000 ng/L of 3-SHA in Sauvignon Blanc (100 times more than in the control fermentation), intensifying the wine's aromatic profile.

However, other techniques, such as random mutagenesis combined with classical genetics approaches, have allowed the development of strains with optimized characteristics compatible with their industrial applications and compliant with current regulations and industry requirements. In relation to these approaches, studies using molecular breeding, a method combining sexual reproduction and recombination between commercial S. cerevisiae strains and strains carrying specific mutations (like in the URE2 gene), have obtained promising results in increasing thiol production. By crossing strains with beneficial mutations in URE2 but low enological fitness with commercial strains, followed by backcrossing their offspring with the commercial parent for four generations, S. cerevisiae strains were generated that maintained the optimal technological characteristics of the commercial strain while carrying the URE2 gene mutation from the non-commercial strain. This increased volatile thiol concentration between 2 and 7 times compared to the control fermentation.

On the other hand, using selected strains of different non-Saccharomyces species has also shown relevant results in thiol production in wine fermentations. Species like Metschnikowia pulcherrima and Torulaspora delbrueckii stand out for their ability to release 3SH and its acetylated derivative, depending on the strain and fermentation conditions. Various studies have demonstrated that their use in sequential fermentations with S. cerevisiae increases thiol release. The positive role of Lachancea thermotolerans in thiol release has also been recently described, due to a greater capacity for consuming and transforming thiolic precursors, especially 3-SH from its glutathionylated precursor.

In this line, the design of complex microbial consortia (with multiple species) that exploit different traits of S. cerevisiae and non-Saccharomyces species to optimize aromatic compound release while maintaining adequate fermentation kinetics is noteworthy. Combining five different species: M. pulcherrima and T. delbrueckii to increase thiol release, Hanseniaspora uvarum and Candida zemplinina to delay S. cerevisiae growth and thus favor the action of the first two, and S. cerevisiae, to adequately complete the fermentation process. These approaches are not without complications in their industrial practice, both due to the complexity of producing industrial inocula of multiple species and the subsequent control of complex inocula behavior. In this sense, it has been demonstrated that the behavior of complex consortia (containing 2 to 6 species), besides presenting emergent properties that can lead to new metabolic functions of industrial interest, is predictable based on the ecology of the different species forming part of the consortium, making the use of microbial consortia in enology promising.

Finally, in the future, synthetic biology offers a promising approach, and it can be anticipated that this discipline will allow, within the framework of the enological industry, redesigning current yeast strains or creating completely new biological systems to optimize wine sensory quality. These systems should combine S. cerevisiae's inherent characteristics, such as its high fermentative capacity, with those of other non-Saccharomyces species, which, thanks to their metabolism and extensive enzymatic repertoire, increase wine sensory complexity. This will generate chimeric strains, more versatile in aromatic production, but maintaining a robust fermentative profile against current industry challenges; challenges mainly arising from changing climatic conditions (wines with high potential alcohol levels, nutrient-deficient, and with acidity imbalances) and consumer demands.