AIMS: Release of hydrogen sulfide by fermenting yeast is a potential problem in wine production, because of its strong organoleptic impact. To identify the genetic determinants of sulfide production, we compared the transcriptomes of two wine yeast strains with similar oenological properties, but with very different levels of sulfide production, UDC522 (high sulfide producer) and P29 (low producer). METHODS AND RESULTS: Oenological microfermentations were sampled at the peak production of sulfide. Transcription profiles of the two strains were analysed by three methods, a cDNA-based array, an oligonucleotide-based array and qRT-PCR analysis of selected transcripts. Less than 10% of yeast genes showed significant differences between the two strains. High sulfide production correlated with a general overexpression of thiamine biosynthesis genes, whereas genes linked to the catabolism of sulfur-containing compounds (like amino acids) showed no significant expression differences between both strains. CONCLUSIONS: Our data suggest a relationship between the thiamine biosynthetic pathway and sulfide production during wine fermentation. SIGNIFICANCE AND IMPACT OF THE STUDY: This study provides a first hint which indicates that for some yeast strains, biosynthesis of thiamine (and perhaps of other sulfur-containing compounds) may be more relevant than the general nitrogen metabolism in explaining sulfide production by some yeast strains during vinification, defining new targets for genetic improvement of wine yeast strains.
AIMS: Release of hydrogen sulfide by fermenting yeast is a potential problem in wine production, because of its strong organoleptic impact. To identify the genetic determinants of sulfide production, we compared the transcriptomes of two wine yeast strains with similar oenological properties, but with very different levels of sulfide production, UDC522 (high sulfide producer) and P29 (low producer). METHODS AND RESULTS: Oenological microfermentations were sampled at the peak production of sulfide. Transcription profiles of the two strains were analysed by three methods, a cDNA-based array, an oligonucleotide-based array and qRT-PCR analysis of selected transcripts. Less than 10% of yeast genes showed significant differences between the two strains. High sulfide production correlated with a general overexpression of thiamine biosynthesis genes, whereas genes linked to the catabolism of sulfur-containing compounds (like amino acids) showed no significant expression differences between both strains. CONCLUSIONS: Our data suggest a relationship between the thiamine biosynthetic pathway and sulfide production during wine fermentation. SIGNIFICANCE AND IMPACT OF THE STUDY: This study provides a first hint which indicates that for some yeast strains, biosynthesis of thiamine (and perhaps of other sulfur-containing compounds) may be more relevant than the general nitrogen metabolism in explaining sulfide production by some yeast strains during vinification, defining new targets for genetic improvement of wine yeast strains.