3.2.6. Others

Other inhibitory compounds include certain fungicides and pesticides, especially the former, which may have a detrimental effect on wine LAB. The residues of systemic fungicides are the inhibitoriest and they are often used in the later stage of grape maturation to control the fungus *Botrytis cinerea* [60]. More recently, mixes of medium-chain fatty acid or the addition of fumaric acid had been proposed to inhibit the growth of wine lactic acid bacteria and the resulting malolactic fermentation.

#### **4. Selected Wine Lactic Acid Bacteria Starter Cultures and the Timing of Inoculation**

Even when desirable malolactic acid bacteria are established in a winery, the onset of the MLF may take several months and may occur in some barrels and tanks but not in others. There are several options available to control MLF: firstly, the selection of conditions to encourage the growth of the indigenous malolactic flora; secondly, the induction of MLF in wines by inoculating with wines already undergoing MLF; or, thirdly, the induction of MLF by inoculation with either laboratory-prepared or commercial strains of LAB. Increased recognition of the influence of MLF on wine quality has led winemakers in recent years to seek better control over the occurrence and outcome of MLF. For this reason, the induction of the MLF by the use of selected LAB starter cultures is fast becoming the preferred option. Due to a massive inoculum with bacteria at 10<sup>6</sup> cfu/mL, less time is required for the bacteria to grow up to a cell density high enough to rapidly degrade the malate present in wine. Undesirable bacteria can be suppressed, which prevents wine alterations [1].

In the 1980s, commercially available strains of wine LAB from around the world became available to the wine industry. These LAB strains were selected from good spontaneous MLF and screened for good MLF kinetics, reliable performance under wine conditions, and desired sensory properties. They will tolerate the difficult growth and survival conditions found in wine and will produce compounds

that impart positive sensory impacts to the wine. In the early 1990s, direct inoculation MLB starter cultures became available, and the most effective are pre-acclimatized during the production process. This step allows them to survive being added directly to wine, with no decrease in viable cell numbers, and no loss of malolactic activity. These wine LAB preparations can be added directly to wine or rehydrated in water for a short time prior to their addition.

Traditionally, when selected cultures of known wine LAB are used, inoculation is performed at the completion of AF. That said, already in 1985, Beelman and Kunkee [61] explored the possibility of inoculating wine LAB into juice along with the yeas<sup>t</sup> used to conduct AF. Current thinking identifies the following times during wine production when selected wine LAB can be added (Figure 2).

**Figure 2.** Inoculation regimes for selected wine lactic acid bacteria Adapted from Bartowsky, Australian Wine Research Institute (AWRI), Entretiens Scientifiques Lallemand Dubrovnik 2011.

#### *4.1. Co-inoculation with Selected Yeast and Selected Wine Lactic Acid Bacteria*

Before 2003, researchers at the Université de Bordeaux recommended making the wine LAB addition only after the completion of AF. They felt this timing would avoid the production of acetic acid and d-lactic acid, compounds derived from the hetero-fermentative carbohydrate metabolism of LAB [62]. They proposed that wine LAB added at earlier points during AF may result in slow or stuck yeas<sup>t</sup> fermentation, or result in MLF inhibition due to yeas<sup>t</sup> antagonism. To date, none of these concerns have been observed when both AF and MLF have been properly managed.

But other researchers proposed the inoculation of selected wine LAB into juice along with yeas<sup>t</sup> because it was felt nutrient availability would be enhanced, and the absence of alcohol would allow wine LAB to better acclimatize to environmental conditions and grow more vigorously. Beelman and Kunkee [61] showed that MLF in the presence of fermentable sugars does not necessarily lead to the production of excessive amounts of acetic acid, as long as yeas<sup>t</sup> fermentation starts promptly and goes to completion [63,64]. For a successful co-inoculation, some parameters are crucial for its success—choosing the right wine yeast, correctly rehydrated, good temperature management, and the proper yeas<sup>t</sup> nutrition strategy are key points to integrate for any fermentation. Well-fed and healthy wine yeas<sup>t</sup> and bacteria leads to complete and regular alcoholic and malolactic fermentations.

Since 2003, co-inoculation has gained increasing interest across all winemaking regions. Today, co-inoculation is understood as the practice of inoculating selected wine lactic acid bacteria at the beginning of the winemaking process shortly after yeas<sup>t</sup> inoculation, usually 24 to 48 h after yeas<sup>t</sup> inoculation. This technique is advantageous because not only will it secure the malolactic fermentation under most difficult conditions, but also because there are definite advantages that are recognized by winemakers and professionals: First bullet

• MLF can be completed in between 3 days and 2 weeks depending on the type of musts and the bacteria used.


#### *4.2. Sequential Inoculation with Selected Wine Lactic Acid Bacteria Post-alcoholic Fermentation*

The traditional inoculation at the end of AF does not pose the risk of the bacterial decomposition of sugars and the resultant increase in VA, nor does the production of excessive amounts of lactic acid, known as "piqûre lactique," occur. Inoculation at this point avoids much of the toxicity attributed to some carboxylic acids, such as fumaric acid, as their concentration declines after AF [38]. The merit of inoculation at the end of AF can also be related to the availability of the bacterial nutrients, nitrogen-containing bases, peptides, amino acids, and vitamins that have arisen from yeas<sup>t</sup> death and subsequent autolysis [66]. Another advantage may be simply from a logistical point of view. When using sequential inoculation, the wines that should undergo MLF can be separated from the wines where acidity is to be conserved. The vinification process can be conducted so that only one type of fermentation at a time is monitored. Often, this is perceived as less risk for cross-contamination.

However, exposure to high levels of ethanol that are present may result in delayed MLF, especially in wines produced in hot climates. If wine conditions are not limiting, selected wine LAB added after the AF are able to achieve cell concentrations comparable to those inoculated into must. In cases of nutrient limitation or adverse wine chemical parameters, the addition of a bacterial nutrient will support MLF. In instances where alcohol levels exceed 14.5% (*v*/*v*), selected wine LAB strains tolerant to alcohol must be used or they must be acclimatized before inoculating into wine. The additive, inhibiting effect of ethanol, pH, and SO2 must be considered, and the strain best adapted to the conditions must be chosen.

#### **5. Advantages of** *Lactobacillus plantarum* **Starter Cultures**

*Lactobacillus plantarum* strains have shown most interesting results for their capacity to induce MLF under high pH conditions (Table 1), their facultative hetero-fermentative properties that avoid acetic acid production (Figure 3), and their more complex enzymatic profile compared to *O. oeni*, which could play an important role in the modification of wine aroma.


**Figure 3.** Sugar metabolism of wine lactic acid bacteria.

#### *5.1. Lactobacillus plantarum Starter Cultures for the Induction of Malolactic Fermentation in Must and Wine*

In 1988, the first malolactic starter culture was introduced to the wine industry. Prahl et al. [7] proposed to inoculate must before alcoholic fermentation using a facultative hetero-fermentative *L. plantarum* starter culture making use of non-proliferating cells. In EP 0398957B1, they disclosed a method of introducing a freeze-dried biomass of *L. plantarum* directly into must or fruit juice to induce MLF without significant consumption of sugars present in the must or fruit juice and substantially without any production of volatile acidity. The malolactic bacteria had been unable to survive in the fermented wine. The application under practical conditions asked for an inoculation 24 to 48 h prior to the addition of yeas<sup>t</sup> with 10 g/hl of the freeze-dried preparation, corresponding to an inoculation level of 5 × 10<sup>7</sup> cells/mL wine [8]. Malic acid degradation had initiated rapidly; it slowed down and stopped when alcohol levels reached about 5–8% (*v*/*v*). At this stage, the *Lactobacillus* cultures died <sup>o</sup>ff. Depending on the wine parameters, mainly pH, temperature, and the speed of yeas<sup>t</sup> fermentation, more or less malic acid is degraded. The advantage would have been partial malic acid degradation in low pH white wines. The disadvantages had been that microbial stability could not be achieved, since a part of the malic acid stayed in the wine as a source for the growth of other microorganisms. The amount of malic acid degraded is varying and not predictable. Furthermore, the application of this culture had only been recommended for low pH white wines, since above pH 3.5 the *L. plantarum* could grow and metabolize glucose producing l/d-lactic acid. This risk is quite high since lactic acid production would be significant and even yeas<sup>t</sup> growth could be triggered by the excessive growth of this microorganism under high pH conditions. The use of this starter culture was limited, as the degree of malic acid conversion was variable and rarely complete, and due to the limited application in low pH wines only, along with the risk to leave the wines for two days without SO2 and yeas<sup>t</sup> addition.

In 2004, Bou and Krieger [9] filed a patent on an "Alcohol-tolerant malolactic strains for the maturation of wines with average or high pH". The patent had been published in 2004 under the international application number PCT/FR2004/001421. The patent relates to LAB strains of the genera *Lactobacillus* and *Pediococcus* capable of initiating and carrying out a complete MLF upon direct inoculation in dried, frozen, or lyophilized state, without a previous acclimatization step at a concentration of between 106–5 × 10<sup>7</sup> cfu/mL, into a wine with an alcohol content of 10% (*v*/*v*) or more and an average high pH level. The resistance to alcohol is apparent with an excellent survival rate on inoculation and a rapid start of malic acid degradation. Claims had been supported by examples of the successful induction of MLF under various wine conditions with *Lactobacillus* strains DSM-9916 and CNCM I-2924. These strains had been chosen from a pool of LAB strains selected for their good tolerance to various limiting conditions and specifically for high alcohol tolerance.

A more recent Italian selection led to *L. plantarum* strain V22 [10]. As part of a European project, where chemical adjuvant, wine yeast, and LAB were screened for their ability to degrade Ochratoxin A (OTA) in must and wine, three *L. plantarum* strains were selected at the University Catolica Sacro Cuore in Piacenza (UCSC) [67]. Ochratoxin A is a mycotoxin suspected of being nephrotoxic, teratogenic, hepatotoxic, and carcinogenic. *Lactobacillus plantarum* V22 showed the highest degradation of OTA under the experimental conditions. The three strains were tested in freeze-dried MBR® form for the induction of MLF in wine. Dried MLF starter cultures in MBR® form will allow direct inoculation into wine without significant loss of MLF activity. The *L. plantarum* strain V22 was the most robust under the tested conditions. This strain had been tested during three vintages under various high pH (>p<sup>H</sup> 3.5) and high alcohol conditions (≥14% (*v*/*v*)), proving to be as fast as *O. oeni* starter cultures when inoculated after alcoholic fermentation. Due to its facultative hetero-fermentative properties, *L. plantarum* is most interesting for co-inoculation without the risk of volatile acid formation when inoculated in the presence of sugars. Again, it proved to have a faster fermentation rate compared to *O. oeni*, if wine pH was higher than pH 3.5.

#### *5.2. Specific Feature of Lactobacillus plantarum of Oenological Interest*

Iorizzo et al. [68] selected 11 *L. plantarum* isolates from spontaneous MLF in wines from Southern Italy and characterized them according to their oenological characteristics and for their potential use as starter cultures for MLF in wine. None of the 11 strains produced biogenic amines which is an important criteria for its potential use as MLF starter culture. Cappozzi et al. [69] studied the biogenic amine degradation by *L plantarum* and found two strains able to degrade putrescine and tyramine under wine-like conditions.

Knoll et al. [70] studied LAB isolated from South African red wines during alcoholic and MLFs and 9 commercial malolactic bacteria starter strains including *L. plantarum* V22 for antimicrobial activity. Of the entire screened isolates, 26 strains, belonging to *L. plantarum*, *L. paracasei*, *L. hilgardii,* and *O. oeni*, showed activity towards various wine-related and non-wine-related indicator strains on a synthetic medium. A PCR-based screening revealed the presence of the plantaricin encoding genes plnA, plnEF, plnJ, and plnK in five selected *L. plantarum* strains, including V22. These strains have also been screened for bacteriocin activity by plate assays, on normal MRS media, MRS pH 3.5 and MRS 10% (*v*/*v*) ethanol

(unpublished data). All 20 strains were tested against nine di fferent sensitive organisms. Seven strains, including *L. plantarum* V22, showed bacteriocin inhibitory activity against all of the sensitive strains tested under those pH and ethanol conditions, but under real wine conditions bacteriocin producing activity was not expressed. Iorizzo et al. [68] could not detect a bacteriocin-producing activity within their selection of 11 *L. plantarum* strains from South Italian wines.

#### *5.3. Mixed Oenococcus oeni and Lactobacillus plantarum Starter Cultures for the Induction of MLF*

Lerm et al. [71] studied various *Oenococcus oeni* and *Lactobacillus plantarum* strains isolated from the South African wine environment for their potential use as malolactic starter cultures. These strains were characterized with regards to their properties of oenological interest, including the genetic screening for enzyme-encoding genes (enzymes implicated in wine aroma modification, as well as the absence of enzyme negatively impacting on of the final wine quality or integrity such as biogenic amine formation or production of ethylcarbamate), the ability to survive in wine, their fermentation capabilities, as well as their volatile acidity production. A total of three *O. oeni* and three *L. plantarum* strains were selected at the completion of this study. These strains showed the most potential during the characterization and were able to successfully complete MLF in Pinotage wine. It was again found that *L. plantarum* strains displayed a more diverse enzyme profile than *O. oeni* strains, particularly with regards to the presence of the aroma-modifying enzymes β-glucosidase and phenolic acid decarboxylase (PAD), which implies the future use of this species in the modification of the wine aroma profile and use as commercial starter culture. It was concluded that *Lactobacillus plantarum* strains might have an added beneficial influence in that it has the genetic potential to influence the wine aroma profile to a larger extent than *O. oeni*, due to its cache of enzymes. Based on outcome of this study a mixed starter culture consisting of an *Oenococcus oeni* and a *Lactobacillus plantarum* strain has been introduced 2011 as "Co-Inoculant" for simultaneous inoculation together with the yeas<sup>t</sup> for induction of malolactic fermentation. This was the first commercially available blend of its kind in the world recommended for co-inoculation in high pH grape musts (>p<sup>H</sup> 3.4) only. Nowadays, a second blend *O. oeni*/*L. plantarum* is on the market, which can work at lower pH (>p<sup>H</sup> 3.2).

#### *5.4. A New Concept of Lactobacillus plantarum Starter Cultures for High pH Red Wines.*

Although co-inoculation (inoculation of selected wine lactic acid bacteria 24 to 48 h after the inoculation with selected wine yeast) is getting very popular and is more and more applied because of its various benefits outlined in chapter 4.1, some winemakers still consider co-inoculation with *O. oeni* as risky because of their obligatory hetero-fermentative properties. They wrongly fear co-inoculation, although this practice has more than proven itself to be a secure choice for high pH red wines (above pH 3.5) in which the native flora is even more critical. The biggest fear is to ge<sup>t</sup> a stuck alcoholic fermentation due to antagonism with the wine LAB, and having the bacteria taking over, resulting in an important increase in volatile acidity due to the hetero-fermentative metabolism of the residual sugars. More recently, a new starter culture called ML Prime ™ (a pure *Lactobacillus plantarum* ferment) was released. Due to its specific optimized production process, this *Lactobacillus plantarum* starter culture expresses extremely high malolactic enzymatic activity as soon as it is added to must. MLF is therefore completed in record time (3–7 days in average) during alcoholic fermentation (Figures 4 and 5), and, unlike classical inoculum with *O. oeni*, no further growth is needed, which explains the very rapid onset of MLF upon inoculation into the must without any impact on yeas<sup>t</sup> vitality and alcoholic fermentation. As explained before, *L. plantarum* degrades hexose sugars only via the homo-fermentative pathway, so there is no risk of acetic acid production from residual sugars that may be present in high pH wines, or still present, when MLF has finished before the end of the alcoholic fermentation.

**Figure 4.** Vinification with a new generation starter culture of *Lactobacillus plantarum.*

**Figure 5.** Duration of Malolactic fermentation (MLF) in various trials (days) inoculated with *L. plantarum*
