Project 3.5.3

Formation and fate of positive and negative sulfur compounds

Project summary

Volatile sulfur compounds (VSCs) can contribute both positive and negative attributes to wines, and their control in a winery environment is an important avenue to increasing wine value either by increasing positive sensory attributes or through the reduction of negative characters. The occurrence of VSCs can be influenced by factors including: yeast selection and fermentation conditions; the nature and quantity of precursor compounds; the availability or absence of oxygen at different points of the winemaking process; and availability and speciation of transition metal ions such as copper.

Winemakers are familiar with the practice of adding copper sulfate to wines that exhibit ‘rotten egg’ or other ‘reduced’ characters when in tank. Many winemakers have also seen these ‘reduced’ characters disappear from their wines in the short term, only to see them return at a later date, sometimes after bottling. By exploring the chemistry of VSC formation and the important role played by metals, these common winemaking observations can be better understood, potentially leading to recommendations for ways to reduce the risk of undesirable ‘reduced’ aromas.

Latest information

A new way to measure sulfur compounds
A new HPLC-MS method for the analysis of thiols and disulfides in wine has been developed and validated which quantifies a wider range of VSCs than previous methods. The new method also allows for a comparison between the VSCs present in the headspace of wine and the VSCs present in the liquid phase to gain better insight into the fate of VSCs in wine and to possibly predict the way a wine will age in bottle. The new HPLC-MS method will be used in upcoming projects where the formation and fate of hydrogen sulfide (H2S) and other VSCs will be followed through the winemaking process and during storage of bottled wine.

Understanding the source of sulfur compounds
Drawing upon fermentations conducted in the AWRI’s research on yeast, quantitative data on the production of low molecular weight sulfur compounds and volatile thiols has been obtained for 99 yeast strains in synthetic grape juice. A large range of concentrations was observed for H2S, methyl thioacetate (MeSAc), ethyl thioacetate (EtSAc), 3MH, and 3MHA. This analysis revealed one strain that produced exceptionally high levels of all measured VSCs. Semi-quantitative data were also obtained for 4MMP, which was shown to correlate with 3MH. These studies have been reproduced in real grape juice and analysis to date indicates a strong correlation between VSC results from model studies and fermentations with grape juice.

Twenty strains producing high or low concentrations of VSCs were screened for carbon-sulfur (C-S) lyase activity. C-S lyase activity was found to be significantly correlated with 3MH release, although some strains with low C-S lyase activity also produced high levels of 3MH in the model ferments and vice versa, suggesting that in some strains C-S lyase activity is not the major limitation. Analysis of genomic datasets for these strains revealed that C-S lyase activity correlated with certain alleles (versions) of the known 4MMP-releasing gene IRC7.

Another approach to identifying genes important to thiol release from cysteine conjugates, and H2S release from cysteine, has involved construction of strains expressing different pyridoxal 5’-phosphate (PLP)-dependent enzymes at high levels. This has been achieved for six of the highest priority candidates by replacing their native promoters. None of these strains produced higher levels of sulfide in the presence of cysteine as a sole sulfur source. Another known thiol-releasing enzyme (Str3p) was, however, shown to have cysteine desulfhydrase activity, able to release H2S from cysteine.

The fate and formation of H2S and other VSCs
The formation of important volatile sulfur compounds that negatively affect wine aroma has been studied in red wine, white wine and model wine, and key precursors to H2S, methanethiol (MeSH), ethanethiol (EtSH) and dimethylsulfide (DMS) have been identified. Overall nine possible precursors have been investigated, as well as the wine conditions that may promote the release of the compounds associated with reductive aromas. These wine conditions include the effect of various metal concentrations in wine (Cu, Al, and Zn), the effect of wine pH and the interaction between wine pH and Cu in modulating the formation of VSCs. From the experiments conducted it is clear that S-methylmethionine (SMM) is a major precursor for DMS and that Al and Zn may increase the formation of DMS from SMM. The major precursors for MeSH are dimethyldisulfide (DMDS) and MeSAc, with Cu playing an important role in the release of MeSH from both DMDS and MeSAc. Both MeSAc and EtSAc act as key precursors to their corresponding thiols (MeSH and EtSH) with lower pH associated with significant increases in thiol formation from these precursors. The major contributing factor to the formation of H2S in wines remains elevated residual Cu concentration.

Copper is not just copper
Copper in wine can exist in a number of oxidation states and in complexes with a range of other wine compounds. Results from recent trials have indicated that the form copper is in can have a major effect on the development of VSCs. This applies also to other metals involved in VSC formation. In a collaborative project with Dr Andrew Clarke at Charles Sturt University on metal ion speciation, several methods to quantify the different copper species in wine have been investigated. Methods have been developed to distinguish between Cu(I) and Cu(II) in white wine. In addition, experiments have demonstrated that the complexes that Cu forms with other wine compounds are likely to be as important, if not more so, than the oxidation state of Cu(I) and Cu(II) at a given point in time.

The interaction of copper and sulfur dioxide
Three wines from the 2014 vintage were used in trials looking at the impact of timing of CuSO4 and SO2 additions as commonly used in wineries to understand their possible contribution to the formation of VSCs. A Chardonnay wine was sourced immediately after its primary fermentation was complete. The remaining two wines were finished Chardonnay and Shiraz wines which were treated with hydrogen peroxide to allow the adjustment of SO2 content. The experiment with the first Chardonnay wine showed that early copper addition in the presence of yeast lees resulted in the copper being removed from the wine when it was racked off lees. Late addition of copper resulted in an elevated residual copper concentration post-bottling. The second Chardonnay wine showed an increase in the concentration of MeSH when no copper was added. The Shiraz wine also showed an increase in MeSH when no copper was added, while the addition of copper resulted in an increase in H2S concentrations. These wines will continue to be analysed as they mature in bottle; however it has been noted that VSC formation has been significantly modified for the wines treated with hydrogen peroxide, perhaps as a result of the unintentional modification of other wine components involved in the process. The work was extended in the 2015 vintage using a Shiraz and a Cabernet Sauvignon wine collected immediately after primary fermentation and before sulfite was added.

Isolating copper from the VSC formation process
Given the role of metals in the formation of VSCs and the difficulties in producing wines free of metals, it is important to understand if their impact can be modulated by the addition of strong chelating compounds to deactivate the metals. Two wines were sourced from commercial wineries in the 2014 vintage and were treated with a series of known metal chelating compounds. So far, analysis for volatile sulfur compounds seven months post-treatment has shown that only one chelator is effective when used at five times the concentration of copper and iron present.

Project Team

Markus Herderich
Eric Wilkes
Marlize Bekker
Toni Garcia Cordente
Mark Smith
Paul Smith