Sulphur dioxide (SO₂) is mostly used as an antioxidant additive in winemaking, but excessive levels may be harmful to both wine quality and consumers health. During fermentation, yeast Saccharomyces cerevisiae contributes significantly to final SO₂ levels, and low-producing strains become especially interesting for the wine industry. Recent evidence implicating the impairment of sulphate transport in the SO₂ decrease prompted us to further investigate the sulphate/sulphite metabolic connection in multiple winery yeast strains. Here, we inactivated by CRISPR/Cas9 the high-affinity sulphate permeases (Sul1p and Sul2p) in four strains normally used in winemaking, selected by their different abilities to produce SO₂. Mutant strains were then used to perform fermentation assays in different types of natural must, and the final levels of SO₂ and other secondary metabolites, crucial for wine organoleptic properties, were further determined for all fermentation products. Overall, data demonstrated the double ΔSUL1/ΔSUL2 inactivation in winery strains significantly decreases the levels of SO₂ produced by mutant cells, without however altering both yeast fermentative properties and the ability to release relevant metabolites. Since similar effects were observed in diverse must types for strains with different features, the data strongly support that sulphate assimilation is the determining factor in SO₂ production during oenological fermentations.

Saccharomyces cerevisiae plays a fundamental role in winemaking, not only driving alcoholic fermentation but also producing secondary metabolites that contribute to the organoleptic properties of wine. To ensure consistent quality and process efficiency, wineries commonly employ selected starter strains. Accordingly, the ability to control strain purity and traceability is of critical importance. Currently, the inter-delta PCR method is widely used for the strain-specific genotyping of S. cerevisiae. However, its resolution diminishes when analyzing genetically similar strains, such as those isolated from related grape types or during genotyping of large yeast collections. To address this limitation, we developed a novel strategy that integrates computational and experimental approaches to identify highly specific allelic variants (single nucleotide polymorphisms, SNPs) within the S. cerevisiae genome. Comparative genomic analysis of twenty-eight different strains led to the identification of multiple strain-specific SNPs. From these, nine SNPs spanning five strains were selected and validated through targeted PCR assays. These assays confirmed the feasibility of using SNPs as reliable genetic markers for strain discrimination and traceability. Overall, our findings demonstrate that this SNP-based approach, implemented via multiplex allele-specific (AS) PCR assays, offers a rapid, cost-effective, and highly discriminatory alternative to current genotyping methods, particularly for differentiating closely related strains.

Efficient yeast breeding remains a major bottleneck in industrial biotechnology, particularly in food and beverage applications, where strain improvement must balance phenotypic innovation with regulatory constraints. Traditional hybridization procedures for natural Saccharomyces strains are experimentally limited by their diploid and homothallic nature, often resulting in labor-intensive processes of unpredictable duration. Here, we developed and validated a CRISPR/Cas9-assisted breeding framework that enables rapid, systematic, and informative hybridization by transiently converting homothallic strains into stable heterothallic mating partners through targeted HO gene inactivation. Using this strategy, we generated and characterized multiple intraspecific and interspecific hybrids challenging some industrially relevant traits, as sulfur dioxide release, acetic acid production, and the ability to restart stuck fermentations. In all cases, we generated the hybrids and assessed their performance within three months. Hybrid phenotypes followed inheritance patterns including parental-like, intermediate, and combinatorial traits, enabling rapid evaluation of cross utility. Moreover, cisgenic hybrids were generated by the restoration of the native HO locus, maintaining identical fermentative performance. Application to Saccharomyces cerevisiae × Saccharomyces uvarum interspecific crosses further demonstrated the versatility of the approach while revealing intrinsic biological constraints on trait combination. Overall, CRISPR/Cas9-assisted heterothallic conversion emerges as a versatile platform for accelerated yeast breeding, providing decision-enabling insights that support informed strain development across fermentation-based industries.