(internal5.4.1. Control of Microbial Contamination

 data).

ML-PRIME™

As a result of the global warming, wines with a pH of over 3.5–3.7 are more and more frequent. At those pH levels, we can observe very fast growth of various indigenous microorganisms, some of which are spoilage bacteria that can cause a loss of wine quality or present health concerns. Co-inoculation is advantageous because it allows for the early dominance of a selected wine LAB strain and the faster onset and completion of MLF and early wine stabilization [42].

A more recent OIV regulation (OIV-Oeno-264-2014) on good vinicultural practices for controlling *Brettanomyces* proposed the co-inoculation of selected yeas<sup>t</sup> and selected wine bacteria to shorten the lag phase between the end of alcoholic fermentation and the start of malolactic fermentation and consequently limit the implantation and the growth of *Brettanomyces*, another wine spoilage microorganism with a detrimental impact on wine quality. It also states that the use of malolactic starters is a good way to limit the development of *Brettanomyces* and the production of the undesired compounds

4-ethylphenol, 4-ethylguaïacol, and 4-ethylcatechol. These volatile phenols are characterized by animal-like off-flavors described as horse and barnyard, and/or pharmaceutical odors characterized as medicinal. Because of the high initial vitality of *L. plantarum* ML Prime™, an immediate onset of malolactic fermentation can be observed, as evidenced in Figure 4 where malic acid was degraded during alcoholic fermentation. However, it is important to respect the windows of application summarized in Table 2, which is narrower than for an *O. oeni* starter culture.


**Table 2.** Optimum conditions for the use of *Lactobacillus plantarum* ML Prime™.

5.4.2. Malolactic Fermentation and Red Wine (Pinot Noir) Color

In Pinot Noir, MLF is often delayed because the resulting wines have anecdotally been reported to have superior color. Delayed MLF in a Pinot noir wine, for up to 4 months, showed improved wine color intensity [72]. Pinot noir wine color presents its own unique challenges, particularly because of its low tannin and anthocyanin content, with a bias towards the less stable acetylated form. The formation of wine color is a complex reaction with many different factors having and integral role. It is known that microbial metabolites, acetaldehyde, and pyruvic acid play a role in the formation of polymeric pigments [73]; however, the degradation of these compounds by *O. oeni* and the impact it could have on red wine color is not been well understood. A study in Pinot noir wine showed that there was no significant impact on color loss (A520) when MLF was delayed by up to six months; however, there was an impact on the formation of polymeric pigments [74] This study demonstrated the role of acetaldehyde and/or pyruvic acid degradation by *O. oeni* during MLF as a cause for reduced polymeric pigment formation independent of the pH change.

The results from a research collaboration with the Oregon State University and the team of James Osborn [75] showed different LAB species and strains can metabolize acetaldehyde at different rates (Table 3), which then in turn will affect red wine color post MLF (Figure 6). *Lactobacillus plantarum* (ML-Prime™) metabolizes acetaldehyde at a slower rate to *O. oeni* strains (*O. oeni* OM and *O. oeni* AL).

**Table 3.** Acetaldehyde concentration (mg/L) in a Pinot noir wine pre- and post-MLF and a control wine without MLF extracted from Bartowsky and Krieger-Weber [75].


**Figure 6.** Wine color (520 nm) of Pinot noir wine that did not undergo MLF (Control) or MLF with different malolactic bacteria strains; the wines were stored at 13 ◦C and extracted from Bartowsky and Krieger-Weber [75].

When color and polymeric pigment values were measured in the wines post-MLF (Day 0) and after 30 and 90 days storage at cellar temperatures (Figure 6), a reduction in color was observed in wines that underwent MLF with *O. oeni* AL or *O. oeni* OM, whereas less loss of color was noted in wines that underwent MLF with the *L. plantarum* (ML-Prime™). After 30 or 90 days of aging, no loss of polymeric pigment was noted in wines that underwent MLF with ML-Prime™.

The overall color of Pinot noir wine can be better managed by selecting a specific wine lactic acid bacteria with consideration of the acetaldehyde metabolism and timing of MLF inoculation. Delaying MLF can actually also promote the combination of tannins and anthocyanins, resulting in a lesser impact of SO2 on color. However, this approach to use a delayed MLF for more and stable color must be carefully weighed up against potential microbial spoilage, including Brettanomyces and biogenic amine formation (indigenous LAB).

#### *5.5. A New Concept of Lactobacillus plantarum Starter Cultures for Low pH White Wines*

Although it was out of the comfort zone for *L. plantarum*, ML-Prime (pH ≥ 3.4 and malic acid maximum 3 g/L) was also tested in white wines, and the results had been surprisingly good. Due to the optimized production process resulting in a very high de-acidification activity, it showed an excellent performance when added into the must (24 h after the yeast) or wine with initial low pH and high malic acid content. Even at a pH as low as 3.05 it allows a partial degradation of the malic acid in the white wine vinification process. The percentage of malic acid degradation depends on the specific must or wine conditions (pH, acid malic content, the total acidity, temperature, and the SO2 content) and the grapes varietals, and can vary between 20% and 90%. Figure 7 shows the kinetics of malic acid degradation with *L. plantarum* (ML-Prime™) in a 2017 Chardonnay from South of France with an initial malic acid concentration of 3.6 g/<sup>L</sup> adjusted to pH 3.1, pH 3.2, pH 3.3, pH 3.4, or pH 3.5, respectively. Chardonnay is known for its difficulties to undergo MLF. *L. plantarum* was able to degrade about 30% of the malic acid (between 1.1 and 1.5 g/<sup>L</sup> malic acid had been degraded) when the wine pH ranged from pH 3.1 to pH 3.5, whereas, at pH 3.5, a complete malic acid degradation was achieved.

**Figure 7.** Kinetics of malic acid degradation in a 2017 Chardonnay (South-France) after co-inoculation with *L. plantarum* (ML-Prime™) depending on the pH. Initial must analyses: malic acid concentration 3.6 g/L, total sugars 189 g/L, potential alcohol 11.2% (*v*/*v*), 7.56 g/<sup>L</sup> total acidity (in tartaric acid), nitrogen 193 mg/<sup>L</sup> (Internal laboratory trials Lallemand SAS).

As the wine matrix is very versatile, a precise prediction of how much malic acid will remain in the wine is not possible. Figure 8 shows the malic acid degradation in sequential inoculation in a 2019 Chardonnay from Germany. Again, the matrix was a difficult Chardonnay with a pH of 3.21, 13.2% (*v*/*v*) alcohol, and 55 mg/<sup>L</sup> total SO2, the temperature was at 17 ◦C. Under these challenging conditions, most of the *O. oeni* strains failed in sequential inoculation. Only one *O. oeni* strain OM started malolactic fermentation, but only after re-inoculation and with a very slow degradation. The *L. plantarum* culture (ML-Prime™) finished MLF when inoculated with the normal inoculation ratio after 25 days, and, when doubling the inoculation dosage MLF, it was finished within 10 days. In this experiment, the *L. plantarum* cultures maintained a high cell viability throughout the malolactic fermentation. Contrastingly, this Chardonnay wine must have contained a toxic compound, which had trigged *O. oeni*, as upon inoculation a sharp die-off of the *O. oeni* population could be observed. Only *O. oeni* Oo OM showed better survival and could finally implant, regrow, and induce MLF in this wine.

**Figure 8.** Kinetics of malic acid degradation in a 2019 Chardonnay (Germany) after sequential inoculation with *L. plantarum* (ML-Prime™), and different *O. oeni* strains. The kinetics show malic acid degradation with the *L. plantarum* culture inoculated with a single and double dosage and *O. oeni* strain Oo OM, and the average of 6 other *O. oeni* strains inoculated at a single dosage and re-inoculated at a double dosage. Wine analyses before inoculation: malic acid concentration 3.18 g/L, pH 3.21, 13.2% (*v*/*v*), 55 mg/<sup>L</sup> total SO2 (Internal data).

Knowing the specifications above, it can be a tool for winemakers who want to achieve only a partial MLF in case of white wines vinification or a biological de-acidification instead of chemical de-acidification. The window of application for *L. plantarum* ML-Prime™ is outlined below:


#### *5.6. Interesting Sensory Properties of Lactobacillus plantarum in Wine Application*

As outlined by Lerm et al. [71], *Lactobacillus plantarum* strains might have an added beneficial influence, as it has the genetic potential to influence the wine aroma profile to a larger extent than *O. oeni*, due to its cache of enzymes. Mtshali et al. [76] conducted a genetic screening for wine-related enzymes within *Lactobacillus* species isolated from South African wines. They found a range of genes encoding for b-glucosidase, protease, esterase, citrate lyase (a-, b- and c-subunits), and phenolic acid decarboxylase. These findings indicated a possible use of *L. plantarum* not only for conducting MLF but also as the potential source of enzymes to impact positively on wine aroma, but expression under wine conditions needs further investigation. The commercial starter strain *L. plantarum* strain V22 had been included in a genetic screening of winemaking LAB starter strains mainly belonging to the species *O. oeni* for wine-relevant enzymatic activities [11]. The enzymes of interest that were screened for included β- glucosidase, esterase, protease, and phenolic acid decarboxylase (PAD). The V22 strain was found to possess more diverse enzymatic profiles related to aroma than *O. oeni*. The biggest differences were observed for the presence of esterase, protease, and PAD. The findings of Iorizzo et al. [68] reported the release of free volatiles from odorless glyosidic aroma precursors by all 11 *L. plantarum*

strains in their study in a synthetic wine medium. Interestingly, the *L. plantarum* strain M10 was not only a major producer of 1-octanol, but also released a considerable amount of other odorant compounds with low odor thresholds. Still, these findings need to be validated in a real wine matrix. Spano et al. [77] reported that the expression of β-glucosidase gene in *L. plantarum* is regulated by abiotic stresses such as ethanol, temperature, and pH.

Still, the application of this species in grape must and wine is rather new since only recently commercial starter cultures had been made available, which can survive also at higher alcohol levels and can induce a reliable malolactic fermentation in wine. Further research is needed to elaborate the sensory contribution of these species to the wine aroma profile.

#### *5.7. Other Applications of Lactobacillus plantarum Apart from the Induction of Malolactic Fermentation*

However, the application of *L. plantarum* in vinification should not be restricted to de-acidification through malolactic fermentation only, Lucio et al. [78,79] have most recently proposed *L. plantarum* for the biological acidification of wines. Within the project CENIT CDTI 2008, they selected *Lactobacillus* strains, which show a high potential as biological acidification starters for winemaking when inoculated prior to the alcoholic fermentation into high pH grape must. WO 2015/110484A2 patent application [80] proposes reverse inoculation (inoculation prior to the inoculation with selected wine yeast) or the co-inoculation (inoculation together with the wine yeast) of selected homo-fermentative or facultative hetero-fermentative lactic acid bacteria strains to produce fermented fruit beverages, such as wine or cider, with a reduced alcohol level. Moreover, the international patent application WO 2015/110484 A2 [81] relates to the use of lactic acid bacteria as bio-protective agents against unwanted microorganisms, such as mold and gram-negative bacteria, such as acetic acid bacteria. The inventors propose a specific *L. plantarum* strain as an antimicrobial agen<sup>t</sup> in the process of winemaking.
