Materials and methods

Fermentations were carried out in small volumes in 15 L plastic containers using the Red Garnacha grape variety from the Mas dels Frares winery (Rovira i Virgili University, DO Tarragona). To carry out carbonic maceration (CM), the grapes were destemmed by hand, and whole uncrushed grape berries were placed in the containers, introducing CO₂ to achieve an anaerobiosis environment (Figure 1). The conditions were as follows: two sequential conditions in which two strains of T. delbrueckii were inoculated with Viniflora Prelude (Chr. Hansen Holding AS, Hoersholm, Denmark) and Viniferm NS Td (Agrovin, Alcázar de San Juan, Spain), and the control without inoculation. After 5 days of CM at 20°C, the grapes were pressed and the must fractions mixed. The fermented must from each condition was racked and inoculated with S. cerevisiae CLOS (Lallemand Inc., Montreal, QC, Canada) to complete the alcoholic fermentation. Finally, three malolactic fermentations (MLF) were carried out using two different strains of Oenococcus oeni (Lalvin VP41 from Lallemand Inc. and Viniflora CH11 from Chr. Hansen Holding AS) respectively for inoculation, and indigenous lactic acid bacteria (LAB) in a spontaneous MLF (Sp) (Figure 1). All fermentations (AF and MLF) were performed in biological triplicates. Statistical analysis of data was performed using ANOVA and the Tukey test with XLSTAT version 2022.5.1 software. A p-value of less than 0.05 was considered statistically significant.

Results

Effect of T. delbrueckii on wine colour

Interesting results were obtained regarding the colour and phenolic composition of the wines derived from inoculation with T. delbrueckii at the beginning of carbonic maceration. The presence of T. delbrueckii caused a significant increase in the concentration of anthocyanins in the final wines, especially with the TdV strain (Figure 2A). This higher concentration may be due to several causes, such as the greater oxidation of anthocyanins in the control wines, which took longer to begin FAL. On the other hand, this effect may also be associated with high pectolytic activity, which has been observed in certain non-Saccharomyces yeasts; however, this activity has not yet been demonstrated for T. delbrueckii. Differences were also found in pyranoanthocyanins, Vitisin A and Vitisin B (Figure 2B). These pigments play an important role in improving colour stability in wines, as they are less susceptible to pH changes and more resistant to discoloration caused by sulphur dioxide¹. The control samples at the end of the CM were significantly different from those inoculated with T. delbrueckii, since no pyranoanthocyanins were detected. By contrast, both the TdP and TdV samples showed the presence of these pigments after CM. However, after AF, pyranoanthocyanins were detected in all wines, their concentrations being higher in wines with T. delbrueckii than in Control. During the CM phase, there were no significant differences in colour intensity (CI) between the different conditions. However, after AF, the control wines showed a significant reduction in CI, and it was possible to visually differentiate the control wines from the ones inoculated with T. delbrueckii (Figure 2C).

Effect of T. delbrueckii on the aroma and organoleptic properties of wines

Inoculation using the two strains of T. delbrueckii significantly influenced the volatile profile and general organoleptic characteristics of the wines. The differences were mainly observed after AF using T. delbrueckii, with an increase in the families of ethyl esters, higher alcohols, short chain fatty acids (SCFAs) and higher alcohol acetates. The compounds that increased the most were 1-propanol, 2-methylpropanol, isoamyl acetate, 2-phenylethanol acetate, ethyl decanoate, ethyl dodecanoate and isopropanol; these were characterized by the descriptors alcohol, wine, banana, rose, grape and leaves. The TdP strain showed a significantly greater increase in the concentrations of 2-phenylethanol, 2-methylpropanol, and isoamyl acetate than the TdV strain. Isoamyl acetate is a characteristic compound of CM wines² whose descriptor is banana. This compound reached concentrations of 0.62 mg/L (Control), 0.63 mg/L (TdP) and 0.47 mg/L (TdV) after CM without any significant differences. However, the TdP and TdV wines showed a significant increase in isoamyl acetate concentration after AF, reaching concentrations of 3.37 mg/L and 1.32 mg/L, respectively, relative to the Control AF wine at 0.58 mg/L (Figure 3A).

Finally, to comprehensively understand the impact of T. delbrueckii inoculation during MC, a sensory analysis was performed after AF. A triangular test was conducted by 15 trained judges, using black glasses to minimize visual bias. The results showed significant differences between Control and TdP, as well as between Control and TdV, but no distinction between TdP and TdV. Subsequently, a descriptive tasting of the wines was carried out after the AF: the TdV wine showed intense aromas of red fruits, grass and banana, as well as higher overall perception values, while in the case of the TdP wine, the banana aroma and the general perception improved compared to the control wine. Surprisingly, banana aroma was the only descriptor responsible for a significant difference between the wines with T. delbrueckii and the control wine (Figure 3B). This suggests that consumers associated a more pronounced banana aroma with the presence of T. delbrueckii during CM.

Effect of T. delbrueckii on the malolactic fermentation performance

Once AF was completed, three MLF strategies were applied: inoculation with O. oeni strains OoVP41 and OoCH11, and spontaneous fermentation. Inoculation using O. oeni starter cultures was carried out in order to evaluate their potential under the competitive pressure of endogenous BAL.

Figure 4 shows the MLF kinetics of the three wine conditions with their respective MLF strategies. Inoculation with OoVP41 resulted in a shorter MLF duration than inoculation with OoCH11 and than in the spontaneous MLF. The TdP and TdV wines which were inoculated with OoVP41 completed the MLF two days earlier (i.e., within 8 days) than the control wines with the same strain (within 10 days). Previous studies under laboratory conditions³ have also characterized the OoVP41 strain as a highly efficient fermentative strain, corroborating this result under these competitive conditions. The influence of T. delbrueckii on MLF was also observed when using the OoCH11 strain. Significantly greater differences were observed in the spontaneous MLF. The control wines took 7 days to initiate MLF, while TdP took 4 days and TdV only 3 days. Regarding the total time of MLF, TdP and TdV were also shorter: TdP lasted 15 days and TdV 16 days, which is notably less than Control, which required 20 days (Figure 4). The effect of T. delbrueckii on MLF is associated with various factors that contribute to improving the efficiency of O. oeni⁴. In the present study, despite no differences being found in ethanol or pH levels, the better performance of MLF can be attributed to a reduction in MCFAs (medium-chain fatty acids), known to be toxic to O. oeni⁵, or to an increase in ammonium and mannoproteins. In addition to nutritional aspects, the increase in mannoproteins could be related to the reduction in MCFAs, since they can absorb these lipids, detoxifying the environment⁶.

Conclusions

Although in the present study the general physicochemical parameters were kept constant, the presence of T. delbrueckii significantly influenced the colour and aroma parameters, increasing the anthocyanin content and isoamyl acetate levels, especially with the TdP strain. Furthermore, T. delbrueckii significantly incremented other compounds associated with floral and fruity aromas.

The sensory evaluation revealed a tendency for increased red fruits, banana, and fresh grass aromas in TdV wines, and an enhanced perception of banana aroma and overall sensory attributes in TdP wines. Of these descriptors, banana aroma was the only one to show a statistically significant difference in the sensory analysis.

Regarding MLF, T. delbrueckii improved performance, especially in spontaneous MLF, starting earlier and requiring less time. Inoculation with OoVP41 led to a significant two-day reduction in the duration of MLF. These results highlight the potential benefits of using T. delbrueckii strains to improve MLF efficacy and the sensory attributes of carbonic maceration wines.

The approach used here could be especially useful in northern latitude wine regions, which are characterized by colder climates and higher acidity levels. MLF may become necessary in these regions to achieve the desired characteristics of their wines. Furthermore, inoculating with T. delbrueckii would help to enrich the sensory profile of these wines, guaranteeing high quality and distinctive characteristics.

Funding: This work was supported by the grant PGC2018-503 101852-B-I00 awarded by the Spanish Research Agency. CRV is grateful for the predoctoral fellowship from the Catalan Government (2020FI).