WINETECH Technical Yearbook 2019

were used in mixed culture fermentations in combination with one S. cerevisiae wine strain. Shiraz wines produced with S. cerevisiae on its own with or without MLF, served as the reference treatments. Yeast strains were commercially available cultures or were obtained from the ARC Infruitec- Nietvoorbij microorganism culture collection. Each of the eight yeast treatments were also evaluated in combination with the three MLF strategies (no MLF, co-inoculation and sequential inoculation). For co-inoculation, LAB were inoculated 24 hours after yeast and for sequential inoculation, LAB were added after alcoholic fermentation. In total, 24 treatments were evaluated in triplicate. A commercial Oenococcus oeni culture was used to induce MLF. A standardised small-scale (20 L) winemaking protocol was followed at an ambient temperature of 24°C. After completion of MLF, wines were bottled and stored at 15°C. Wines were subjected to descriptive sensory evaluations 24 months later. RESULTS AND DISCUSSION All the Shiraz wines fermented to dryness (residual sugar ˂4 g/L) within 18 days. In most cases, wines produced with non- Saccharomyces yeast had lower alcohol levels (15.49-15.94% v/v) than wines produced with S. cerevisiae only (~16% v/v). Wines produced with C. zemplinina in combination with S. cerevisiae contained

the lowest alcohol levels (15.49% v/v). Wines produced with non- Saccharomyces yeasts had higher VA levels than wines produced with S. cerevisiae only, but were well below the legal limits. The difference in VA levels was not detected by the sensory panel. Results show clearly that there were differences among the non- Saccharomyces strains with regard to their effect on LAB growth and progress of MLF. In most cases, wines produced with non- Saccharomyces yeasts completed MLF in a shorter period than wines produced with only S. cerevisiae. H. uvarum in combination with S. cerevisiae was most compatible with the inoculated LAB and progress of MLF. The C. zemplinina and one of the L. thermotolerans strains had a negative effect on LAB growth and therefore MLF in the co-inoculated MLF wines. However, the same effect was not observed with the yeasts when the MLF culture was sequentially inoculated. The inhibitory effect could be due to competition for essential nutrients or production of toxic metabolites. Wines produced with the eight yeast combinations and the three MLF strategies (no MLF, co- and sequential inoculation) resulted in all 24 treatments producing wines with significantly different volatile chemical profiles. The sensory data show that the different yeast combinations had a significant effect on the berry aroma of wines, while

MLF strategy (no MLF, co- and sequential inoculation) had a significant effect on berry aroma, acid balance and astringency. Wines with co-inoculated MLF scored slightly higher for berry aroma than wines without MLF, but both treatments scored significantly higher than wines with sequential MLF. With regard to acid balance, sequential MLF wines were perceived to be less balanced. However, this did not negatively affect the perceived quality of these wines, because they scored similar or better for overall quality than wines without MLF. Co-inoculated MLF wines were perceived to be more astringent than wines without MLF and significantly more astringent than sequential MLF wines. None of the treatments produced wines that were considered unacceptable with regard to astringency. Wines produced with non- Saccharomyces yeasts were different to the S. cerevisiae reference wines. The non- Saccharomyces yeast i so l ates showed potent i a l for producing wines with different styles and flavour profiles, but need further evaluation with different grape cultivars/varieties and in combination with different S. cerevisiae yeast strains. The yeast treatment and the stage of MLF induction had a significant effect on the standard chemical parameters and volatile composition of the wines, although the variation in wine composition did not always translate to perceivable sensory differences.

CONCLUSIONS Re s u l t s i nd i c a t e d t h a t mo s t non - Saccharomyces yeast strains investigated had a beneficial effect on the progress of MLF. Therefore, if MLF is required, it is important to choose Saccharomyces and non - Saccharomyces strains that are compatible and promote MLF. On the cont rar y, spontaneous and inoculated MLF can be delayed if yeast strains or combinations are used that negatively affect LAB growth. Selecting suitable non- Saccharomyces yeasts can enhance complexity and produce wines with lower alcohol levels. Wine flavour profiles can be modified by using different non- Saccharomyces yeast strains and MLF strategies. Induction of MLF as a co- or sequential inoculation can also result in significant changes to wine flavour profiles. The interactions between Saccharomyces , non- Saccharomyces and LAB are complex and the resulting changes to wine composition and flavour need further investigation. SUMMARY The use of non- Saccharomyces yeasts in combination with lactic acid bacteria (LAB) to improve the complexity and diversity of wine styles has received limited attention. Therefore, this study compared the interactions between seven non- Saccharomyces yeast strains and Saccharomyces cerevisiae in combination with three malolactic fermentation (MLF)