Project Number
WW 10-21
Project title
The use of Torulaspora delbrueckii for wine production
Project leader
Augustyn, O P H
Team members
Jolly, N P
Van Breda, V
Chotoomia, S
Adonis, P
Paulse, C
Augustyn, O P H
EXECUTIVE SUMMARY
It is generally accepted that the wealth of yeast biodiversity with hidden potential, especially for oenology, is largely untapped and it is especially non-Saccharomyces yeasts that fall into this category. The non-Saccharomyces yeast, Torulaspora delbrueckii (anamorph: Candida colliculosa) is a fermentative yeast formerly classified as Saccharomyces rosei. It was previously suggested for vinification of musts low in sugar and acid and has been used for the commercial production of red and rosé wines in Italy, albeit some time back (1950’s). In more recent times it was found that T. delbrueckii in pure culture produced lower levels of volatile acidity than the S. cerevisiae strains tested. This yeast has also been used together with S. cerevisiae for the production of small-scale Chardonnay, Sauvignon blanc and Chenin blanc wines. The Sauvignon blanc and Chenin blanc wines were both judged to be better than the respective S. cerevisiae reference wines. The use of T. delbrueckii with and without co-inoculation of S. cerevisiae also holds potential for production of wines lower in alcohol. This may possibly be achieved due to a less efficient fermentative ability than the wine yeast, S. cerevisiae. Competition for sugars by T. delbrueckii may also result in less sugar available for ethanol production.
In this project forty-three South African Torulaspora delbrueckii yeast isolates from the ARC Infruitec-Nietvoorbij yeast culture collection, the T. delbrueckii type strain, commercial T. delbrueckii reference strains and a commercial Saccharomyces cerevisiae yeast had their identities confirmed and were characterised using conventional and molecular microbiological techniques. The performances of the yeasts were monitored by means of laboratory-scale fermentations in grape must at 15 ºC and 22 ºC. The fermentation kinetic data showed that the yeasts had varying speeds of fermentations. Biochemical and physiological grouping did not coincide with the fermentation abilities and good fermenters could be found in more than one group. Principle Component Analyses of chemical analytical data (alcohol, volatile acidity, glycerol, total SO2, residual sugar) showed different groupings. The yeasts that grouped close to the S. cerevisiae reference strain showed more acceptable wine chemical profiles, while those further away displayed less acceptable profiles. Three locally isolated strains and one commercial T. delbrueckii yeast strain, produced wines with acceptable chemical profiles at both temperatures. These strains also had comparable fermentation kinetics to the S. cerevisiae reference.
Subsequently the best nine natural T. delbrueckii isolates and reference strains were used in the production of small-scale Chenin blanc and Pinotage wines (2009). The small-scale cellar fermentations were carried out at 15 ºC and 24 ºC. Three of the T. delbrueckii yeast isolates and the commercial T. delbrueckii strain were used as single inoculants, while the S. cerevisiae yeast strain was co-inoculated with the other four T. delbrueckii yeast isolates at 0 h, 24 h and 48 h, respectively. The wines were chemically and sensorially evaluated and the data statistically analysed. Yeast lees samples of the must from the single inoculant white wines were collected at the end of the fermentations to determine the survival of the T. delbrueckii yeast inoculants. A principle component analysis of the chemical analysis, together with the sensory evaluation data, were used to identify the yeast strains that potentially could produce novel wines, either on their own or as a component of co-inoculated fermentations. The results for the white wine trial showed that two natural T. delbrueckii isolates had the potential for producing wines that compared well and even exceeded the quality of that produced by the S. cerevisiae reference strain with regards to chemical composition and overall sensory quality. The 48 hour treatment did not give good results and was discontinued. The sensory analyses data of the red wines compared well to the S. cerevisiae reference wine, but showed no increase in wine quality as seen as in the white wine trial. Furthermore, no pattern regarding inoculation times was observed.
The two best T. delbrueckii yeast strains were subsequently evaluated in Chenin blanc over four vintages (2010-1013) utilising T. delbrueckii-only and co-inoculated fermentations with S. cerevisiae at 0 and 24 h. Distinct stylistic differences were observed and dependant on the desired wine style, a suitable yeast inoculation treatment can be chosen. T. delbrueckii-only treatments lead to semi-sweet wines, and co-inoculation treatments lead to dry wines. The S. cerevisiae yeast choice makes a critical difference to the fermentation kinetics and wine style. Excessive volatile acidity production by S. cerevisiae can be tempered by the use of T. delbrueckii. Overall, two South African T. delbrueckii isolates fared very well and in some cases were better than the commercial strains. Generally the red wine evaluations showed that stylistic differences could be obtained, but not necessarily an increase in quality. Preliminary observations on polyphenolic analytic data of red wines lead to a separate project investigating chemical profiles of non-Saccharomyces wines. The resultant data is not contained in this report, but in the WW 10-25 progress report.
Improved T. delbrueckii strains by adaptive selection and the evaluation of a potential hybrid met with limited success and was not pursued further, due to the focus of the project on the best initial isolates. Determination of extra-cellular enzymatic activity showed no activity for the enzymes tested.
In conclusion, two T. delbrueckii isolates from an initial 43 isolates can be used singly or in co-inoculation with a S. cerevisiae yeast for changing styles of wine, producing novel wines and in some instances improving quality. Lowering of ethanol is only practical where a semi-sweet wine is desired. A 0.3% alcohol reduction can be achieved in dry wine, but is probably of no practical use. The two T. delbrueckii isolates compare well to the commercial strains and can be considered for commercialisation. Depending on the fermentation temperature, different T. delbrueckii strains will be suitable for specific wine styles and some can even be considered for single inoculations (without S. cerevisiae) for industrial fermentations.
Poster(s)
Van Breda, V and Jolly, N P. 2009. Torulaspora delbrueckii: A preliminary investigation for wine production. Poster. 4th International Congress of the South African Society for Enology and Viticulture. 28-30 July, Cape Town, South Africa.
Van Breda, V, Jolly, N P and Van Wyk, J. 2010. A preliminary investigation into the use of Torulaspora delbrueckii for wine production at lower and higher temperatures. Poster. 15th World Congress of Food Science and Technology. 22-26 August, Cape Town, South Africa.
Jolly, N P, Van Breda, V and Van Wyk, J. 2010. Torulaspora delbrueckii: the new wine yeast. Poster. 32nd National Congress of the South African Society for Enology and Viticulture. 18-19 November, Somerset West, South Africa.
Jolly, N P, Van Breda, V and Booyse, M. 2011. Comparison between South African Torulaspora delbrueckii isolates and commercial strains for use as a single inoculants in wine production. Poster. Oeno 2011: 9th International Enology Symposium. 15-17 June, Bordeaux, France.
Jolly, N P, Van Breda, V and Booyse, M. 2011. Wine production with Torulaspora delbrueckii yeast strains. Poster. 18th Conference of the South African Society for Microbiology. 24-27 November, Warmbaths, South Africa.
Van Breda, V, Jolly, NP and Van Wyk, J. 2014. The use of Torulaspora delbrueckii yeast strains for the production of small scale wine. Poster. 36th South African Society for Enology and Viticulture Conference. 12-14 November, Somerset West, South Africa.
Presentation(s)
Jolly, N P, Van Breda, V and Van Wyk, J. 2010. Torulaspora delbrueckii: The new wine yeast. Paper presented at the 32nd National Congress of the South African Society of Enology and Viticulture. 18-19 November, Somerset West, South Africa.
Article
Van Breda, V, Jolly, N P, Van Wyk, J. 2013. Characterization of commercial and natural Torulaspora delbrueckii wine yeast strains, International Journal of Food Microbiology, Mnth May v. 163 (p. 80-88)
http://www.sciencedirect.com/science/article/pii/S0168160513001098
Van Breda, V, Jolly, N P, Booyse, M, van Wyk, J. 2018. Torulaspora delbrueckii yeast strains for small-scale Chenin blanc and Pinotage vinifications, South African Journal of Enology and Viticulture, v. 39 (1) (p. 47-57)
http://www.journals.ac.za/index.php/sajev/article/view/1652