Use of Yeast Mannoproteins by Oenococcus oeni during Malolactic Fermentation under Different Oenological Conditions

Oenococcus oeni is the main agent of malolactic fermentation in wine. This fermentation takes place after alcoholic fermentation, in a low nutrient medium where ethanol and other inhibitor compounds are present. In addition, some yeast-derived compounds such as mannoproteins can be stimulatory for O. oeni. The mannoprotein concentration in wine depends on the fermenting yeasts, and non-Saccharomyces in particular can increase it. As a result of the hydrolytic activity of O. oeni, these macromolecules can be degraded, and the released mannose can be taken up and used as an energy source by the bacterium. Here we look at mannoprotein consumption and the expression of four O. oeni genes related to mannose uptake (manA, manB, ptsI, and ptsH) in a wine-like medium supplemented with mannoproteins and in natural wines fermented with different yeasts. We observe a general gene upregulation in response to wine-like conditions and different consumption patterns in the studied media. O. oeni was able to consume mannoproteins in all the wines. This consumption was notably higher in natural wines, especially in T. delbrueckii and S. cerevisiae 3D wines, which presented the highest mannoprotein levels. Regardless of the general upregulation, it seems that mannoprotein degradation is more closely related to the fermenting medium.


Introduction
Malolactic fermentation (MLF) is a biotransformation undergone in fermented beverages with low nutrient composition by different lactic acid bacteria (LAB) [1][2][3]. It is usually performed in wine and cider and results in an increase in pH values because of the decarboxylation of L-malic acid into L-lactic acid, an improvement of organoleptic wine characteristics and an increased microbiological stability [4][5][6].
As it occurs in fermented media, the fermenting microbiota-mainly yeasts that carry out alcoholic fermentation (AF)-have a significant impact on LAB and on the development of the MLF, usually carried out after AF [7][8][9][10].
As a result of the transformation of grape must into wine (or apple juice into cider), the media in which the LAB ferments will have low concentrations of nutrients, high acidity, and high concentrations of ethanol and sulphur dioxide. All these changes select the LAB most suited to this stressful environment, of which Oenococcus oeni is the best adapted species [1,11,12].
These two fermentative processes are usually inoculated with the species best adapted to each fermentation, Saccharomyces cerevisiae and O. oeni for AF and MLF, respectively [13]. sequence the primers used here in the transcriptional study were deduced. Yeasts were maintained on YPD plates (2% glucose, 2% bacto-peptone, 1% yeast extract, 2% agar, w/v, Panreac Química SLU, Castellar del Vallès, Spain) and the bacteria on MRSmf [32] plates, and all of them were stored at 4 • C.

Experimental Fermentations
O. oeni PSU-1 was inoculated in two media to carry out MLF in oenological conditions. First, O. oeni was directly inoculated in a wine-like medium (WLM). Second, AFs were performed with different yeast combinations in natural grape must, and then in the inoculated O. oeni in the obtained wines.

Fermentations in Wine-Like Medium
They were performed in 50 mL tubes containing 50 mL of sterile WLM, which was prepared as in Bordas et al. [33] with half nitrogen composition (1.25 g/L of casamino acids and 1.25 g/L of peptone). This model wine had a concentration of around 110 mg N/L of yeast assimilable nitrogen, 2 g/L of L-malic acid, and a pH of 3.4. A stock solution (40 g/L) of the commercial mannoprotein extract Mannoplus (Agrovin, Spain) was prepared in WLM under sterile conditions and the appropriate volume was added to each wine to obtain final concentrations of 0, 100, 200, and 400 mg/L of mannoprotein extract. Each wine (50 mL) was then inoculated with O. oeni PSU-1 for a population of around 2 × 10 7 cells/mL. These fermentations were carried out in triplicate and incubated statically at 20 • C. Samples were taken every 24 h to monitor the evolution of L-malic acid consumption and the bacterial population. This was calculated by plating samples on MRSmf and incubating at 27 • C in a 10% CO 2 atmosphere for 7 days. MLF was considered finished when L-malic acid was below 0.1 g/L.

Fermentations in Natural Grape Must
These were performed with natural concentrated Airén must (Mostos S.A., Tomelloso, Spain) diluted with sterile Milli-Q water to a density of 1080 ± 1 g/L [23]. The must was supplemented with 0.4 g/L of Nutrient Vit NatureTM (Lallemand Inc.) and the pH was adjusted to 3.6. The must was then sterilized using 0.1% (v/v) of dimethyl dicarbonate (Santa Cruz Biotechnology, Inc., Dallas, TX, USA) and stored overnight at 4 • C.
In order to undergo the fermentations, precultured yeasts in YPD liquid medium were inoculated for a population of 2 × 10 6 cell/mL. In the case of non-Saccharomyces, after 48 h from the initial inoculation, ScQA23 was also inoculated for the same population. Fermentations were carried out statically in 500 mL of must, at 20 • C, and in triplicate. Every 48 h, density (Densito 30PX Portable Density Meter, Mettler Toledo, Spain) and yeast population were determined. YPD agar plates were used to calculate the total number of yeast cells present, and lysine agar medium (Oxoid LTD., Basingstoke, UK) was used to quantify the non-Saccharomyces yeasts, after incubation at 28 • C for 48 h. AF was considered finished when the sugar concentration was below 2 g/L. At this point the wines were centrifuged at 10,000× g for 5 min at 4 • C, filtered (0.22 µm Whatman, Thermo Fisher Scientific, Waltham MA, USA), and transferred to sterile 50 mL tubes. The wines were then inoculated with O. oeni PSU-1 for a population of 2 × 10 7 CFU/mL and incubated in the same conditions as the AFs. These MLF were also carried out in triplicate. Samples were taken every 24 h to monitor the consumption of L-malic acid and the evolution of the bacterial population, as described above.

Sampling and RNA Extraction
Cell pellets of O. oeni were collected during the fermentations. In the case of WLM, samples were collected at 1, 5, and 8 days after inoculation, corresponding to 24 h, end of MLF, and three days after the end of MLF (post-MLF), respectively. Sampling of the natural grape fermentations was performed at the end of MLF. Cells of O. oeni grown in MRSmf for 2 days were also collected. In all cases, 50 mL of wine was centrifuged (4250× g, 15 min) at 4 • C. The supernatant was discarded, and the cell pellet was frozen with liquid nitrogen and stored at −80 • C until analysis. The cell pellet was then washed twice with 1 mL of sterile 10 mM Tris-HCl buffer at pH 8 by centrifugating at 2000× g for 5 min at room temperature. Afterwards, the cell pellet was resuspended in 200 µL of the same Tris-HCl buffer with 50 mg/mL of lysozyme (Roche Life Science, Mannheim, Germany) and incubated for 30 min at 37 • C. Finally, the RNA was purified using a High Pure RNA Isolation Kit Version 13 (Roche Life Science) following the manufacturer's instructions. The RNA was stored at −80 • C until analysis.

RT-qPCR
The RNA samples were cleaned of contaminant DNA traces with a TURBO DNAfree Kit (Thermo Fisher Scientific) following the manufacturer's instructions. Reverse transcription (RT) and real-time qPCR were performed following Olguín et al. [35] using QuantStudio 5 Real-Time PCR Systems (Thermo Fisher). Four genes (manA, manB, ptsI, and ptsH) related to mannose intake by O. oeni were evaluated [30]. In addition, another four genes (dnaG, dpoIII, gyrA, and gyrB) were evaluated as internal controls. Of these, gyrA and gyrB presented the least variation within conditions (data not shown) and were therefore used as reference genes in this experimentation. The primers used for all the genes studied can be found in Supplementary Table S1. The ∆∆Ct method [36] was used to calculate the relative expression of each gene. The expression of O. oeni after growing for 2 days in MRSmf was used as the reference condition in WLM. To study the differences in this metabolism through the use of non-Saccharomyces in natural wines, the expression of O. oeni in ScQA23 wine was used as a control condition.

Statistical Analysis
All the statistical analyses of the results were performed using the statistics software XLSTAT version 2020.2.3 (Addinsoft, Paris, France). The analysis of variance was carried out by ANOVA with a subsequent Tukey HSD test to determine significant differences between the samples. The confidence interval used was 95% and the statistical level of significance was set at p ≤ 0.05.

Fermentations
All the MLF performed with O. oeni PSU-1 in WLM and in natural grape must were completed. Information about the development of AF in natural grape must can be found in Supplementary Figure S1.
The addition of mannoproteins in WLM did not change the duration or consumption rate of L-malic acid (Table 1). In all the conditions, MLF took 5 days to finish. As for the consumption rate, only the addition of 100 mg/L of mannoprotein extract showed a significant increase in this value (Table 1). Under the studied conditions, the hypothesized positive effect of mannoproteins on MLF [34] was not observed.
More differences were observed in natural grape wine MLFs. In these media, O. oeni showed a reduction in MLF length in T. delbrueckii and Sc3D wines with respect to the control ScQA23 wine ( Table 2). The L-malate consumption rate increased by more than 25% in these wines. The MLF of M. pulcherrima wine was similar to the control condition. In general, the use of non-Saccharomyces promoted the MLF of O. oeni [10]. However, the fermentation with MpFlavia did not show this effect.

Mannoprotein Utilization
We studied mannoprotein utilization by O. oeni during MLF in wine by precipitating the total polysaccharide fraction and quantifying it as the concentration in mannose equivalents after acidic hydrolysis before and after MLF. This procedure allowed us to estimate the concentration of mannoproteins [37] that were degraded during MLF [34].
As regards WLM, a very low concentration of mannose equivalents (mannose eq.) was detected ( Figure 1). This concentration was significantly increased by the addition of the commercial mannoprotein extract. We observed that the addition of this extract brought about a linear increase in the mannose concentration we quantified (data not shown). This corresponded to a mannose eq.: mannoprotein extract ratio (in mg/L) of around 0.23.
In all cases, the mannoprotein concentration detected decreased by the end of MLF ( Figure 1). We also quantified it 3 days after completion of MLF to better understand the metabolism in WLM. In this post-MLF sampling we observed a dramatic reduction compared to the previous samplings ( Figure 1). It seems that when L-malic acid is completely metabolized, the utilization of mannoproteins increases. As a result, we can relate the utilization of mannoproteins to a survival metabolism that is enhanced when the preferred substrate of O. oeni in wine, L-malic acid, is drained. It is also notable that this decrease the commercial mannoprotein extract. We observed that the addition of this extract brought about a linear increase in the mannose concentration we quantified (data not shown). This corresponded to a mannose eq.: mannoprotein extract ratio (in mg/L) of around 0.23. In all cases, the mannoprotein concentration detected decreased by the end of MLF ( Figure 1). We also quantified it 3 days after completion of MLF to better understand the metabolism in WLM. In this post-MLF sampling we observed a dramatic reduction compared to the previous samplings ( Figure 1). It seems that when L-malic acid is completely metabolized, the utilization of mannoproteins increases. As a result, we can relate the utilization of mannoproteins to a survival metabolism that is enhanced when the preferred substrate of O. oeni in wine, L-malic acid, is drained. It is also notable that this decrease was related to the initial mannoprotein concentration. The higher the initial mannoprotein concentration, the higher the degradation in this post-MLF sampling ( Figure 1).
The results for mannoprotein utilization during MLF in natural grape wine can be found in Figure 2. Considering ScQA23 wine as the control condition, we observed a significant increase in mannoprotein concentrations in T. delbrueckii wines and when inoculating Sc3D. Similar values of around 340 mg /L of mannose eq. were quantified in TdViniferm and in Sc3D wines. As regards TdZymaflore wine, the mannoprotein concentration detected was the highest: 440 mg/L eq. of mannose on average. Thus, T. delbrueckii is a non-Saccharomyces yeast related to an increase in mannoprotein concentrations [19,20,37] in its fermented wines and also when added as yeast lees in a synthetic medium [34]. In addition, Sc3D is a selected commercial strain that is an overproducer of these macromolecules [20]. The results for mannoprotein utilization during MLF in natural grape wine can be found in Figure 2. Considering ScQA23 wine as the control condition, we observed a significant increase in mannoprotein concentrations in T. delbrueckii wines and when inoculating Sc3D. Similar values of around 340 mg /L of mannose eq. were quantified in TdViniferm and in Sc3D wines. As regards TdZymaflore wine, the mannoprotein concentration detected was the highest: 440 mg/L eq. of mannose on average. Thus, T. delbrueckii is a non-Saccharomyces yeast related to an increase in mannoprotein concentrations [19,20,37] in its fermented wines and also when added as yeast lees in a synthetic medium [34]. In addition, Sc3D is a selected commercial strain that is an overproducer of these macromolecules [20].
In T. delbrueckii and Sc3D wines, in which the concentration of mannoproteins after AF and mannose utilization by O. oeni were the highest (Figure 2), the L-malic acid consumption rate was also the highest ( Table 2). We can therefore relate better MLF performance to enhanced mannoprotein utilization. It is interesting to note that the same O. oeni strains exhibited different use patterns depending on the yeast inoculated and the amount of mannoprotein released during AF.
In the case of MpFlavia, however, no differences were observed with respect to the control (280 mg/L of mannose eq. on average) even though this yeast-and this strain in particular-usually increases the mannoprotein concentration in wines [20,37] or produces similar concentrations to S. cerevisiae. The interactions of the yeast with the medium in which it is fermenting presumably determine the release of mannoproteins, probably due to the particular autolytic process in that medium [38]. In addition, M. pulcherrima and S. cerevisiae wines showed the lowest L-malic acid consumption rate (Table 2), which is also related to less mannoprotein release and utilization by O. oeni, supporting the idea that better MLF performances may be related to greater mannoprotein degradation (Figure 2). The use of mannoproteins was different in the two types of wine studied. In natural grape wines, the highest mannoprotein degradation was detected once the L-malic acid was exhausted. This is probably because mannoprotein concentration is higher in natural wines and also because other inhibitor compounds, such as yeast metabolites or less nutrients, may be present in these wines. In fact, the yeast assimilable nitrogen (Table 2) in natural wines was less than half of that in the WLM (around 110 mg N/L). Under these more limiting conditions, the use of mannoproteins is enhanced in O. oeni. Thus, the breakdown of mannoproteins would release amino acids, apart from mannose, which could eventually be used as a nitrogen source by O. oeni. In T. delbrueckii and Sc3D wines, in which the concentration of mannoproteins afte AF and mannose utilization by O. oeni were the highest (Figure 2), the L-malic acid con sumption rate was also the highest ( Table 2). We can therefore relate better MLF perfor mance to enhanced mannoprotein utilization. It is interesting to note that the same O. oen strains exhibited different use patterns depending on the yeast inoculated and the amoun of mannoprotein released during AF.
In the case of MpFlavia, however, no differences were observed with respect to th control (280 mg/L of mannose eq. on average) even though this yeast-and this strain i particular-usually increases the mannoprotein concentration in wines [20,37] or pro duces similar concentrations to S. cerevisiae. The interactions of the yeast with the medium in which it is fermenting presumably determine the release of mannoproteins, probabl due to the particular autolytic process in that medium [38]. In addition, M. pulcherrim and S. cerevisiae wines showed the lowest L-malic acid consumption rate (Table 2), whic is also related to less mannoprotein release and utilization by O. oeni, supporting the ide that better MLF performances may be related to greater mannoprotein degradation (Fig  ure 2).
The use of mannoproteins was different in the two types of wine studied. In natura grape wines, the highest mannoprotein degradation was detected once the L-malic acid was exhausted. This is probably because mannoprotein concentration is higher in natura wines and also because other inhibitor compounds, such as yeast metabolites or less nu trients, may be present in these wines. In fact, the yeast assimilable nitrogen ( Table 2) i natural wines was less than half of that in the WLM (around 110 mg N/L). Under thes more limiting conditions, the use of mannoproteins is enhanced in O. oeni. Thus, th breakdown of mannoproteins would release amino acids, apart from mannose, whic could eventually be used as a nitrogen source by O. oeni.

General Wine Chemical Compounds
The MLF undergone with O. oeni in WLM transformed L-malic acid into L-lactic acid analogously in all wines (data not shown). We analyzed the chemical composition at th end of MLF and after three days of the completion of MLF (Table 1). No differences wer observed within the conditions and sampling points as regards pH. Sugar concentratio

General Wine Chemical Compounds
The MLF undergone with O. oeni in WLM transformed L-malic acid into L-lactic acid analogously in all wines (data not shown). We analyzed the chemical composition at the end of MLF and after three days of the completion of MLF (Table 1). No differences were observed within the conditions and sampling points as regards pH. Sugar concentration decreased slightly as a consequence of MLF. Nevertheless, the behavior in all WLM was similar and no sugar decrease was observed in the post-MLF sampling. Similarly, citric acid was not consumed by O. oeni in these conditions, neither at the end of MLF nor post-MLF. Citric acid decreased slightly at the post-MLF sampling point as a consequence of the consumption of the main carbon source-L-malic acid-being drained. The consumption of citric acid was around 0.1 g/L in all cases (Table 1) and no significant changes were detected in acetic acid concentration, as this is usually related to sugar or citric acid consumption. D-lactic acid, which can be an end product of sugars such as glucose, fructose, or mannose-which we can also relate to mannoprotein metabolism-was detected in higher concentrations in those wines to which the mannoprotein extract was added. Even when the concentrations were low after MLF, the addition of the mannoprotein extract resulted in a doubling of the D-lactic acid detected (Table 1). Moreover, in the post-MLF sampling the concentrations increased in all wines, whereas no significant increase was observed in the WLM without the addition.
More differences were observed in the compounds studied in the natural wines (Table 2). First, all the wines finished AF with similar L-malic acid concentrations of around 1.5 g/L (data not shown). The T. delbrueckii wines produced the most acidic wines, which were significantly different from the M. pulcherrima wine. The pH differences were as high as 0.1. As a result of MLF, the pH value increased, and the observed differences were also similar since they had similar amounts of L-malic acid. The residual sugar concentration (glucose + fructose) after AF was always below 2 g/L. Nevertheless, the wine inoculated with ScQA23 completely drained the sugars, whereas in the others that were fermented with the other S. cerevisiae strain and sequential inoculations with non-Saccharomyces, sugar traces were detected. These residual sugars were slightly consumed by O. oeni by the end of MLF in those wines inoculated with non-Saccharomyces, particularly those fermented with T. delbrueckii. In these wines the citric acid concentration after AF was slightly different as a result of the yeasts' metabolism.
As mentioned earlier, the yeast assimilable nitrogen in the natural wines (Table 2) was lower than in WLM (110 mg N/L). The concentration of NOPA was significantly altered as a result of AF inoculation in the natural wines ( Table 2). The T. delbrueckii and Sc3D wines had lower NOPA concentrations than the ScQA23 and M. pulcherrima wines. This can be explained by the different amino acid consumption patterns and preferences [39], which are also the result of yeast-yeast interactions [40]. However, the differences were small, and all the wines had enough NOPA concentration to ensure MLF. In contrast, similar ammonium concentrations were also observed ( Table 2).
The use of non-Saccharomyces is usually related to higher citric acid concentrations after AF [10], although this has only been clearly observed with Starmerella bacillaris [41]. In the present study, the wines fermented with T. delbrueckii showed a significant increase in citric acid concentrations compared to the S. cerevisiae control wine. After MLF, this concentration was reduced as a consequence of the consumption by O. oeni. All the wines had around 0.11 g/L of this acid after MLF, with the exception of the ScQA23 wine, which had twice the concentration ( Table 2).
After AF, the acetic acid concentration was different in the obtained wines. As described in the literature, the use of some non-Saccharomyces can change the concentration of this compound [14,23]. Generally speaking, T. delbrueckii tends to decrease it while M. pulcherrima increases it. In our study, M. pulcherrima significantly increased the volatile acidity of the wine after AF (up to 0.59 g/L on average), and only TdViniferm decreased it significantly (0.17 g/L on average). As a result of the consumption of sugars and citric acid by O. oeni, acetic acid and D-lactic acid were produced. After MLF, the concentration of acetic acid also depended on the combination of yeast species used in AF. The S cerevisiaefermented wines had the intermediate concentration of acetic acid, while the T. delbrueckii wines had the lowest and M. pulcherrima had the highest. D-lactic acid is a sugar related to the LAB metabolism and, therefore, it was not detected after AF. It increased as a result of the O. oeni metabolism after MLF ( Table 2). The detected concentration was dependent on the mannoprotein concentration after AF (Figure 2), which resulted in higher D-lactic acid in the T. delbrueckii and Sc3D wines, 0.29 and 0.26 g/L on average, respectively.

Transcriptional Response of Mannose-Related Genes
The transcriptional regulation of four selected genes (Supplementary Table S1) previously related to mannose uptake in O. oeni [30] were evaluated in a wine-like medium with increasing concentrations of mannoprotein extract. This transcriptional regulation was compared to the expression of O. oeni prior to inoculation to determine the expression level of these genes under oenological conditions. The relative expression (RE) of the genes was quantified 24 h after inoculation, at the end of MLF, and 3 days after completion of the fermentation (Figure 3).
Studying the gene expression of mannose uptake-related genes under wine conditions is difficult since their expression is also dependent on the concentrations of other sugars or sugar alcohols [30,31,42]. These genes are generally activated when growing in the presence of sugars [31] and under ethanol stress [30]. Their relationship with the pH is more variable. Jamal et al. [30] observed that manA was expressed more in acidic conditions, whereas manB was more active in neutral pH, and ptsI and ptsH showed no variation.
In our study, the genes manA, ptsI, and ptsH were upregulated in response to WLM conditions 24 h after inoculation. The expression of manB did not show any change compared to the control condition (before inoculation). The expression at the end of MLF was variable depending on the gene and mannoprotein concentration. However, there was a general increase after the end of MLF in all the studied conditions. Three days after L-malic acid exhaustion, when most of the mannose consumption was detected, all the genes were upregulated compared to the control condition. It seems that when not many carbon sources are available, O. oeni increases the expression of these permeases to enable bacterial survival. Indeed, the bacterial population remained stable at around 10 7 CFU/mL 3 days after the completion of MLF (data not shown). This highlights the importance of using alternative energy sources after L-malic consumption in order to allow for bacterial survival at least 3 days after completion of MLF.
detected concentration was dependent on the mannoprotein concentration after AF (Figure 2), which resulted in higher D-lactic acid in the T. delbrueckii and Sc3D wines, 0.29 and 0.26 g/L on average, respectively.

Transcriptional Response of Mannose-Related Genes
The transcriptional regulation of four selected genes (Supplementary Table S1) previously related to mannose uptake in O. oeni [30] were evaluated in a wine-like medium with increasing concentrations of mannoprotein extract. This transcriptional regulation was compared to the expression of O. oeni prior to inoculation to determine the expression level of these genes under oenological conditions. The relative expression (RE) of the genes was quantified 24 h after inoculation, at the end of MLF, and 3 days after completion of the fermentation (Figure 3). Studying the gene expression of mannose uptake-related genes under wine conditions is difficult since their expression is also dependent on the concentrations of other sugars or sugar alcohols [30,31,42]. These genes are generally activated when growing in the presence of sugars [31] and under ethanol stress [30]. Their relationship with the pH is more variable. Jamal et al. [30] observed that manA was expressed more in acidic conditions, whereas manB was more active in neutral pH, and ptsI and ptsH showed no variation.
In our study, the genes manA, ptsI, and ptsH were upregulated in response to WLM conditions 24 h after inoculation. The expression of manB did not show any change compared to the control condition (before inoculation). The expression at the end of MLF was variable depending on the gene and mannoprotein concentration. However, there was a general increase after the end of MLF in all the studied conditions. Three days after Lmalic acid exhaustion, when most of the mannose consumption was detected, all the genes were upregulated compared to the control condition. It seems that when not many carbon We also studied these four genes in wine from natural grape must with different AF inoculation regimes by the end of MLF (Figure 4). Taking the expression of these genes in ScQA23 wine as the reference conditions, manB, ptsI, and ptsH were upregulated in the other wines. The gene manA did not show an expression pattern different from the reference condition with the exception of a slight upregulation in Sc3D wine.
It is interesting to see that the expression of these genes was upregulated in M. pulcherrima wine (Figure 4), which had low mannoprotein degradation levels (Figure 2), comparable to the reference condition. Therefore, the wine matrix must have an important effect on the expression of these genes, which makes it difficult to find any relation between RE and mannoprotein use. Moreover, it has to be remembered that these genes encode nonspecific hexose permeases, which are usually active [30] and respond not only to mannose. Thus, it should not be surprising that these genes may be upregulated in response to oenologically stressful conditions, related with yeast inhibitory metabolites for O. oeni. Besides, higher REs were observed in these genes when O. oeni was fermenting in natural grape-derived wines with respect to WLM (Figures 3 and 4). Nevertheless, no correlation was found between the RE and mannoprotein use patterns in O. oeni. This suggests a complex regulation dependent on the medium and not specifically linked to the expression of mannose uptake-related genes. response to oenologically stressful conditions, related with yeast inhibitory metabolites for O. oeni. Besides, higher REs were observed in these genes when O. oeni was fermenting in natural grape-derived wines with respect to WLM (Figures 3 and 4). Nevertheless, no correlation was found between the RE and mannoprotein use patterns in O. oeni. This suggests a complex regulation dependent on the medium and not specifically linked to the expression of mannose uptake-related genes.

Conclusions
This study presents new information on mannoprotein utilization by O. oeni during MLF in wine-like medium and in wine. Different mannoprotein concentrations were quantified following an AF inoculation strategy. T. delbrueckii wines together with Sc3D wines were those with the highest concentrations of mannoproteins released. Mannoprotein utilization by O. oeni was dependent on the fermenting media. Low degradation of mannoproteins was observed when fermenting in a low mannoprotein concentration medium (WLM), whereas this degradation was higher in natural wines, with a higher mannoprotein content. This greater utilization of mannoproteins may be associated with a faster MLF in T. delbrueckii and Sc3D wines. The genes manA, manB, ptsI, and ptsH-directly related to mannose uptake but also active with other sugars-were upregulated in response to oenological conditions. O. oeni showed an increased RE of manB, ptsI, and ptsH

Conclusions
This study presents new information on mannoprotein utilization by O. oeni during MLF in wine-like medium and in wine. Different mannoprotein concentrations were quantified following an AF inoculation strategy. T. delbrueckii wines together with Sc3D wines were those with the highest concentrations of mannoproteins released. Mannoprotein utilization by O. oeni was dependent on the fermenting media. Low degradation of mannoproteins was observed when fermenting in a low mannoprotein concentration medium (WLM), whereas this degradation was higher in natural wines, with a higher mannoprotein content. This greater utilization of mannoproteins may be associated with a faster MLF in T. delbrueckii and Sc3D wines. The genes manA, manB, ptsI, and ptsHdirectly related to mannose uptake but also active with other sugars-were upregulated in response to oenological conditions. O. oeni showed an increased RE of manB, ptsI, and ptsH in non-Saccharomyces and Sc3D wine compared to ScQA23 wine. Altogether, it seems that the mannoprotein metabolism is activated under wine conditions and that mannoprotein uptake is enhanced in stressful conditions. Further research is needed to clarify the regulation of mannoprotein metabolism in O. oeni, seeking out other possible genes/proteins involved in this metabolism. The use of different yeasts and mannoprotein extracts should be further evaluated with more O. oeni strains and conditions as potential activators of MLF.

Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.