A study shows that low doses of mannoproteins accelerate the malolactic fermentation of Tempranillo

Friday, September 18, 2026

A study published Thursday in OENO One showed that adding selected mannoprotein extracts to red wine helped accelerate malolactic fermentation in Tempranillo wines from two consecutive harvests, while also altering some aspects of the wine's colour-related phenolic profile without increasing volatile acidity or adversely altering pH.

The research was conducted by Paloma Toraño, Albert Bordons, Nicolas Rozès, and Cristina Reguant using Tempranillo wines from the 2022 and 2023 vintages produced at the Mas dels Frares experimental winery of the Universitat Rovira i Virgili in Tarragona, Spain. The team investigated how commercial extracts of mannoproteins influenced malolactic fermentation carried out by Oenococcus oeni PSU-1, a lactic acid bacterium widely used as a reference strain in wine research.

Malolactic fermentation is a key step, following alcoholic fermentation, in many red wines. During this phase, bacteria convert L-malic acid into L-lactic acid, which reduces harsh acidity and helps improve microbiological stability. However, the process can be slow or even stop completely, as wine presents a challenging environment for bacteria, with alcohol, acidity, and other stressors limiting their activity. The authors noted that this uncertainty partly explains winemakers' interest in tools that can make the process more reliable.

Researchers tested three commercial mannoprotein products derived from the cell walls of Saccharomyces cerevisiae. These extracts differed in their polysaccharide and protein composition, a point the team considered important because these structural differences can influence how the compounds interact with wine and microorganisms. Mannoproteins are already known to occur naturally in wine through fermentation and yeast degradation, but their quantity and characteristics vary depending on the yeast strain and winemaking practices.

The tests were conducted using actual wine, not just synthetic media. This is important because much of the previous work in this area had been carried out under simpler laboratory conditions. The wines from both vintages were chemically similar before the onset of malolactic fermentation. In 2022, the wine had a pH of 3.55, an alcohol content of 13.2%, and an L-malic acid level of 1.62 g/L. In 2023, the wine had a pH of 3.51, an alcohol content of 13.8%, and an L-malic acid level of 1.61 g/L.

In the 2022 vintage, the team added mannoproteins at a concentration of 2 g/L. Wines treated with the three extracts completed malolactic fermentation in 14 days, while the untreated control took 17 days. The article specifies that this difference was statistically significant. The authors also noted that by day 12, the treated wines had already completed the process, whereas the control still contained more than 0.2 g/L of L-malic acid.

In the 2023 vintage, researchers switched to lower doses, closer to practical use in wineries, testing 0.25 g/L and 0.50 g/L. According to the study, these lower additions nevertheless shortened fermentation. Depending on the extract and dose, the treated wines completed malolactic fermentation in approximately 11 to 12 days, compared to 14 days for the control. The article's abstract highlighted 0.25 g/L as an effective low dose for stimulating the process.

The study did not show that the addition of mannoproteins significantly increased the O. oeni population at the end of fermentation. Bacterial counts remained around 10⁶ CFU/mL for all treatments. This suggests that the benefit was not simply due to a higher final cell count. The authors indicated that the improved fermentation kinetics could be related to the consumption of mannoproteins by O. oeni, meaning that the added compounds may have served as a useful resource during the bacteria's metabolic activity.

The results concerning wine quality were more mixed, but nonetheless noteworthy. The researchers reported no negative effects on pH or volatile acidity, two parameters important for both stability and sensory quality. However, the addition of mannoproteins altered the phenolic composition. Total anthocyanins decreased, but stable acylated anthocyanins and vitisins increased. These compounds are associated with more stable color in red wines, suggesting that the additives may reshape the pigment system rather than simply maintaining all pigments at the same level.

This finding could be of interest to wineries, as malolactic fermentation is not only a technical step but also a matter of timing and stability. If a low-dose additive allows the process to be completed more quickly and consistently from one vintage to the next, this could aid in winery planning, microbiological control, and managing wine color during aging. While the study does not demonstrate a guaranteed commercial outcome, it opens up a practical avenue that producers can explore under normal winemaking conditions.

The authors also pointed out that the wines used in the trials were not particularly challenging for O. oeni. Their pH was above 3.5, their alcohol content was moderate for a red wine, and the other conditions were relatively favorable. Consequently, the gains in fermentation time were modest rather than dramatic. Nevertheless, the team considered the results relevant because they show that the effect can be clearly detected in real wine and at doses more realistic for cellar use.

This work is also part of a broader effort to understand why some mannoprotein-based products produce different results than others. In this study, the three extracts exhibited distinct protein and polysaccharide profiles. One had a very high polysaccharide content and a low protein content, while the others contained more protein and different molecular weight fractions. The article argues that these chemical differences help explain the variations in how the extracts influence bacterial metabolism and wine composition.

The authors performed the experiments in triplicate and used a statistical analysis including ANOVA and Tukey's test, with a significance threshold set at p.