Pathway Analysis of Methionol Production During Fermentation in Synthetic Grape Medium

Abstract

Methionol (3-methylthiol-propanol) represents one of a family of volatile sulfur-containing organic compounds produced by Saccharomyces cerevisiae during winemaking, which contribute to the complexity of wine aroma and flavour. Being the fusel alcohol of methionine, methionol is believed to be the end-product of methionine catabolism in S. cerevisiae via a serial enzymatic degradation including transamination, decarboxylation, and dehydrogenation, via a process called the Ehrlich pathway. Methionine can also be catabolised via the demethiolation pathway where methionine undergoes γ-cleavage to produce methanethiol and other thiols. Previous reports have identified a handful of genes encoding enzymes in the Ehrlich pathway. However, a detailed sketch of how these enzymes participate in methionol production during fermentation has not been characterised. In this study, the impact of potential pathway genes, including ARO8, ARO9, ARO10, YER152C, PDC1, PDC5, PDC6 and STR3 in yeast on methionol production during fermentation was evaluated using single-gene-deleted BY4743 laboratory yeast strains as well as genetically-modified F15 commercial wine strains typically used in Sauvignon blanc production. Laboratory fermentation was carried out by inoculating yeast isolates into chemically-defined synthetic grape medium (SGM) mimicking New Zealand Sauvignon blanc juice used in industrial practice. Time-course analyses of yeast growth and methionol production revealed remarkable differences in terms of fermentation kinetics between BY4743 and F15 strains. Finished wines were analysed for methionol concentrations using headspace solid phase microextraction coupled with gas chromatography mass spectrometry (HS-SPME/GC-MS). Results from the finished wines showed that methionol production strongly responds to methionine input. Although none of the genes analysed proved indispensable, deletions of aro8 and pdc1 were identified to significantly impact methionol production. For the first time, single- and multiple-gene-disruption of ARO8, ARO9 and ARO10 in F15 strains were generated, and their synergistic effects on yeast growth and methionol production during fermentation were investigated. The aro8aro9 double-disrupted strain albeit viable, was observed to suffer significant growth retardation. Consistent to the result in the BY4743 deletion, aro8 was found to significantly reduce methionol production, whereas the aro9aro10 double-disruption showed a similar effect. Overall, our preliminary results suggest a more profound in vivo complexity of the Ehrlich pathway than previously characterised, as well as the existence of a potential alternative pathway resulting in a compensation in methionol production, providing the basis for further study.