Production of a low-alcohol fruit beverage

Abstract

The present invention relates to the production of fermented fruit beverages, such as wine and cider, with a reduced level of alcohol. Specifically, the present invention is directed to a method for producing a beverage with a reduced content of alcohol comprising using reverse inoculation or co-inoculation of homofermentative or facultative heterofermentative lactic acid bacterium strain and a yeast strain.

Claims

1. A method for producing a fermented fruit beverage having a reduced alcohol content, comprising: (a) inoculating a fruit must with at least one Lactobacillus plantarum strain in an amount of at least 510.sup.6 CFU/ml and fermenting the fruit must with the at least one Lactobacillus plantarum strain for at least 24 hours; and (b) after fermenting the fruit must with the at least one Lactobacillus plantarum strain for at least 24 hours, fermenting the fruit must with at least one yeast strain to obtain a fermented fruit beverage, wherein at least a portion of sugar in the fruit must is converted by the fermenting with the at least one Lactobacillus plantarum strain prior to the fermenting with the at least one yeast strain, wherein the fermented fruit beverage is selected from red wine, white wine, sparkling wine and cider, and has an alcohol content after step (b) of at least 4% (v/v), wherein the alcohol content of the fermented fruit beverage after step (b) is at least 0.5% (v/v) less than the alcohol content of a fermented fruit beverage prepared by the same method but without inoculating and fermenting the fruit must with the Lactobacillus plantarum strain(s).

2. The method according to claim 1, wherein the at least one Lactobacillus plantarum strain is selected from the group consisting of the Lactobacillus plantarum strain CHCC12399 that was deposited with the German Collection of Microorganisms and Cell Cultures (DSMZ) under accession No. DSM 27565, and functionally equivalent mutant strains thereof, wherein the mutant strains are obtained by using the deposited strain as starting material and wherein less than 1% of the nucleotides of the mutant strain are changed as compared to the mother strain, and wherein the mutant strains, when used to inoculate and ferment a fruit must prior to or simultaneously with fermentation of the fruit must with at least one yeast strain, result in a fermented fruit beverage having an alcohol content less than that of a fermented fruit beverage prepared by the same method comprising fermenting with the same yeast strain(s) but without inoculating and fermenting with the mutant strain.

3. The method according to claim 1, wherein the at least one Lactobacillus plantarum strain comprises the Lactobacillus plantarum strain CHCC12399 that was deposited with the German Collection of Microorganisms and Cell Cultures (DSMZ) under accession No. DSM 27565.

4. The method according to claim 1, wherein the fruit must is fermented with the yeast strain within 72 hours after being inoculated with the at least one Lactobacillus plantarum strain.

5. The method according to claim 1, wherein the at least one yeast strain includes a yeast strain of a species selected from Pichia kluyveri, Saccharomyces cerevisiae, Saccharomyces pastorianus, Saccharomyces bayanus, Torulaspora delbrueckii, and Kluveromyces thermotolerans.

6. The method according to claim 1, wherein one yeast strain is used in step (b).

7. The method according to claim 1, wherein two or more yeast strains are used in step (b).

8. The method of claim 1, wherein at least 5 g/L sugar in the fruit must is converted by the fermenting with the at least one Lactobacillus plantarum strain prior to the fermenting with at least one yeast strain.

9. The method of claim 1, wherein the fermented fruit beverage contains less than 10 g/L sugar following the fermenting with the at least one yeast strain.

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIG. 1 shows the malolactic activity of different Lactobacillus plantarum strain CHCC12399 inoculation dosages with a Saccharomyces cerevisiae wine yeast in Riesling juice. As control, a fermentation with only Saccharomyces cerevisiae wine yeast is shown.

(2) FIG. 2 depicts final alcohol levels in Riesling wine produced with different Lactobacillus plantarum strain CHCC12399 inoculation regimes. A Saccharomyces cerevisiae wine yeast has been added after 24 hours of Lactobacillus plantarum inoculation. As control, a fermentation with only the Saccharomyces cerevisiae wine yeast is shown.

(3) FIG. 3 shows malolactic activity of the Lactobacillus plantarum strain CHCC12399 co-inoculated with a Saccharomyces cerevisiae wine yeast in Riesling juice. As control, a fermentation with only the Saccharomyces cerevisiae wine yeast is shown.

(4) FIG. 4 depicts alcohol production during the fermentation of Riesling juice with the Lactobacillus plantarum strain CHCC12399 co-inoculated with a Saccharomyces cerevisiae wine yeast. As control, a fermentation with only the Saccharomyces cerevisiae wine yeast is shown.

(5) FIG. 5 shows flavor analysis of final Chardonnay wines. Final levels of 11 flavor compounds are shown for both the control Chardonnay wine (only Saccharomyces wine yeast; Chardonnay con) and Chardonnay with addition of the Lactobacillus plantarum strain CHCC12399 (Chardonnay L. pl).

(6) FIG. 6 depicts flavor analysis of final Merlot wines. Final levels of 11 flavor compounds are shown for both the control Merlot wine (only Saccharomyces wine yeast; Merlot con) and Merlot with addition of the Lactobacillus plantarum strain CHCC12399 (Merlot L. pl).

(7) FIG. 7 shows a sensory analysis of final Chardonnay wines. Chardonnay 154 Test is the Chardonnay wine with addition of the Lactobacillus plantarum strain CHCC12399 and Chardonnay 157 Control is the control wine (only Saccharomyces wine yeast).

(8) FIG. 8 depicts a sensory descriptive analysis of final Chardonnay wines. Chardonnay 154 Test is the Chardonnay wine with addition of the Lactobacillus plantarum strain CHCC12399 and Chardonnay 157 Control is the control wine (only Saccharomyces wine yeast).

(9) FIG. 9 shows Lactobacillus cell counts during fermentation of Hungarian white grape juice with Lactobacillus plantarum strain CHCC12399 (NoVA) and Merit wine yeast added at different time points.

(10) FIG. 10 depicts yeast cell counts during fermentation of Hungarian white grape juice with Lactobacillus plantarum strain CHCC12399 (NoVA) and Merit wine yeast added at different time points.

(11) FIG. 11 shows glucose concentrations during fermentation of Hungarian white grape juice with Lactobacillus plantarum strain CHCC12399 (NoVA) and Merit wine yeast added at different time points.

(12) FIG. 12 depicts fructose concentrations during fermentation of Hungarian white grape juice with Lactobacillus plantarum strain CHCC12399 (NoVA) and Merit wine yeast added at different time points.

(13) FIG. 13 shows glucose and fructose concentrations during fermentation of Hungarian white grape juice with Lactobacillus plantarum strain CHCC12399 (NoVA) and Merit wine yeast added at different time points.

(14) FIG. 14 depicts malic acid concentrations during fermentation of Hungarian white grape juice with Lactobacillus plantarum strain CHCC12399 (NoVA) and Merit wine yeast added at different time points.

(15) FIG. 15 shows lactic acid concentrations during fermentation of Hungarian white grape juice with Lactobacillus plantarum strain CHCC12399 (NoVA) and Merit wine yeast added at different time points.

(16) FIG. 16 shows ethanol production during fermentation of Hungarian white grape juice with Lactobacillus plantarum strain CHCC12399 (NoVA) and Merit wine yeast added at different time points.

EXAMPLES

Materials and Methods

(17) Fermentation Set-Up

(18) Lab-scale fermentation trials were carried out in 200 ml of grape juice. Four different grape juices were used: Italian white grape juice, Italian red grape juice, Riesling juice and Hungarian white grape juice. All fermentations were performed at 20 C. until completion (sugar levels<10 g/L). Lactobacillus plantarum strain CHCC12399 isolated from South African grape juice, another Lactobacillus plantarum strain or a Lactobacillus paracasei strain was inoculated at different inoculation levels, ranging from 510.sup.6 to 510.sup.7 CFU/ml. Saccharomyces cerevisiae wine yeast was inoculated at 110.sup.6 CFU/ml.

(19) Headspace GC-FID Analysis

(20) Headspace gas chromatography coupled with flame ionisation detection (GC-FID) was used for the measurement of esters and higher alcohols in the fermentation products. Fermentation samples were centrifuged, after which 2 ml was collected in vials. Samples were then analyzed with a calibrated Perkin Elmer GC System with a headspace sampler. The GC was equipped with a DB-WAXETR column (length, 30 m; internal diameter, 0.25 mm; layer thickness, 0.5 m; Agilent Technologies, Germany). The split-splitless injector was used and held at 180 C. Samples were heated for 30 min at 70 C. in the headspace autosampler before injection (needle temperature: 110 C.). Helium was used as the carrier gas. After starting at 60 C., the oven temperature was raised after 2 min from 60 C. to 230 C. at 45 C./min and was finally held at 230 C. for 5 min. During the GC-program a constant flow rate (10 mL/min) of the carrier gas (He) was maintained. The FID temperature was kept constant at 220 C. respectively. The results were analyzed with Turbochrom software.

(21) Sensory Analysis

(22) Sensory analysis was performed with a professional tasting panel.

(23) Final Wine Parameter Analysis

(24) Ethanol, total sugar (glucose+fructose), total acid (TA), volatile acid (VA), pH and malic acid analysis was performed with Oenofoss equipment according to the manufacturer's protocol. Lactic acid, glucose, fructose and acetic acid concentrations were measured with HPLC by methods known to the skilled person and as described by e.g. Castellari et al. (2000. An improved HPLC method for the analysis of organic acids, carbohydrates, and alcohols in grape must and wines. Journal of Liquid Chromatography & Related Technologies 23 (13); p. 2047-2056).

(25) Results

Example 1: Reverse Inoculation of Grape Juice with Lactobacillus plantarum Strain CHCC12399 and a Wine Saccharomyces cerevisiae Yeast Strain

(26) The first experiment is performed with a Lactobacillus plantarum strain CHCC12399 isolated from grape juice, which was below pH 3.5. Lab trials with the Lactobacillus plantarum strain have been performed in both Italian red and white grape juice. The Lactobacillus plantarum strain used here comes as a product in a frozen format (frozen pellets).

(27) Lactobacillus plantarum was inoculated at several dosages (ranging from 510.sup.6 to 510.sup.7 CFU/ml) in a reverse inoculation with a wine Saccharomyces cerevisiae, inoculated after 72 h at level of 110.sup.6 CFU/ml. The fermentation volume was 200 ml. The initial parameters of the grape juice are described in Table 2.

(28) TABLE-US-00002 TABLE 2 Initial parameters of red and white grape juice Malic Ethanol Glu/Fru* acid TA** % VA*** Samples (g/L) pH (g/L) (g/L) (v/v) (g/l) Italian white 162.5 3.52 3.0 4.1 0 0.20 grape juice Italian red 162.5 3.52 3.0 4.1 0 0.20 grape juice *Glu/Fru = sum of glucose and fructose **TA = total acidity ***VA = volatile acidity

(29) In total, 4 conditions have been tested for each grape juice: Control: only adding Saccharomyces cerevisiae 510.sup.6 CFU/ml Lactobacillus plantarum CHCC12399+Saccharomyces cerevisiae after 72 h 110.sup.7 CFU/ml Lactobacillus plantarum CHCC12399+Saccharomyces cerevisiae after 72 h 510.sup.7 CFU/ml Lactobacillus plantarum CHCC12399+Saccharomyces cerevisiae after 72 h

(30) When the fermentations were finished (when sugar was depleted), alcohol level and 10 other relevant wine parameters were measured (Table 3 and 4).

(31) TABLE-US-00003 TABLE 3 Final wine parameters in the red wine Malic Lactic Acetic Red acid acid Glucose Fructose Ethanol acid wine (g/L) (g/L) (g/L) (g/L) (vol %) (g/L) Control 2.28 0.12 0.08 0.42 9.50 <LOD 5.00E+06 1.57 0.62 0.07 0.50 9.35 <LOD CFU/ml 1.00E+07 0.90 1.18 0.08 0.52 9.25 <LOD CFU/ml 5.00E+07 <LOD 4.23 0.07 0.28 8.40 0.11 CFU/ml

(32) TABLE-US-00004 TABLE 4 Final wine parameters in the white wine Malic Lactic Acetic White acid acid Glucose Fructose Ethanol acid wine (g/L) (g/L) (g/L) (g/L) (vol %) (g/L) Control 2.16 0.12 0.09 2.23 9.00 <LOD 5.00E+06 1.32 0.88 0.07 1.50 9.15 <LOD CFU/ml 1.00E+07 0.29 2.09 0.08 1.68 8.85 <LOD CFU/ml 5.00E+07 0.13 3.86 0.07 1.36 8.00 0.04 CFU/ml

(33) From Table 3 and 4, it is clear that the wine inoculated with 510.sup.7 CFU per ml Lactobacillus plantarum had 1% less alcohol (v/v), compared to the control wine and this was the case for both the red and white wine. Interestingly, the dosage of Lactobacillus plantarum seems of high importance to reach the lower alcohol percent, as inoculation of 110.sup.7 CFU/ml has no significant effect on the alcohol level. From Table 4, it is can also be seen that addition of Lactobacillus plantarum has no negative effect on the acetic acid production, as the concentrations are similar between the control and all dosages of Lactobacillus plantarum. However, it is believed that lower dosages of Lactobacillus plantarum will also reduce alcohol, depending on the time of inoculation of yeast afterwards; the longer Lactobacillus plantarum are allowed to ferment alone, the more sugar will be converted to lactic acid or other metabolites. It is also clear that no more acetic acid is produced when adding the Lactobacillus plantarum.

(34) As more lactic acid is produced than is theoretically possible from malic acid, part of the glucose and fructose is converted to lactic acid, leaving less sugar to be converted to ethanol (Table 5). As can be seen in Table 5, more than 16 g/L sugar (glucose and fructose) has been consumed by Lactobacillus plantarum in the first 24 hours. This was very surprising, as it was expected that only malic acid would have been consumed under these conditions. 1% of ethanol (v/v) corresponds to 16 g/L of sugar consumed by Lactobacillus plantarum. If all sugar would be converted to lactic acid, at least 16 g/L of lactic acid should be formed. As this is not the case, either other metabolic products are formed from glucose or the lactic acid reacts with other metabolic products present during fermentation to form yet another set of metabolic products.

(35) TABLE-US-00005 TABLE 5 Sugar measurements 24 h after inoculation of Lactobacillus plantarum, before adding the wine yeast Inoculation rate Glu/Fru* (g/L) Glu/Fru* (g/L) of L. plantarum after 24 h 24 h 5.00E+06 CFU/ml 179.7 2.6 1.00E+07 CFU/ml 170.7 11.6 5.00E+07 CFU/ml 165.0 17.4 Control (no 182.3 0.0 L, plantarum added) *Glu/Fru = sum of glucose and fructose

(36) To confirm that 510.sup.7 CFU/ml of Lactobacillus plantarum is needed to reduce the final alcohol percentage in wine with 1% (v/v), another set of experiments was performed. In this experiment, inoculation levels between 110.sup.7 CFU/ml and 510.sup.7 CFU/ml Lactobacillus plantarum were tested for their effect on ethanol production. As adding the wine yeast after 72 h is not commercially viable, the wine yeast was added after 24 h. In this case, Riesling juice was used for the fermentation experiments. The initial parameters of the Riesling juice are given in Table 6.

(37) TABLE-US-00006 TABLE 6 Initial parameters of Riesling grape juice Malic Ethanol Glu/Fru* acid TA** % VA*** Sample (g/L) pH (g/L) (g/L) (v/v) (g/L) Riesling 180.7 3.50 6.7 7.6 0 0.39 *Glu/Fru = sum of glucose and fructose **TA = total acidity ***VA = volatile acidity

(38) Lactobacillus plantarum was added in 6 different inoculation dosages: 210.sup.7 CFU/ml, 2.510.sup.7 CFU/m, 310.sup.7 CFU/ml, 3.510.sup.7 CFU/ml, 410.sup.7 CFU/ml and 510.sup.7 CFU/ml in the Riesling grape juice. After 24 h, a wine Saccharomyces cerevisiae strain (110.sup.6 CFU/ml) was added to the juice to finish the fermentation. The fermentation volume was 200 ml and the fermentation temperature was 20 C. The final wine parameters are shown in Table 6. The malolactic fermentation (MLF) performance and final ethanol levels are depicted in FIGS. 1 and 2.

(39) TABLE-US-00007 TABLE 7 Final wine parameters in the Riesling wine Ethanol Malic Glu/fru* TA** % acid VA*** Samples (g/L) (g/L) pH (v/v) (g/L) (g/L) Control 2.6 4.1 3.54 10.2 5.5 0.39 2E7 2.6 4.2 3.55 10.2 0 0.30 2.5E7 2.7 4.2 3.54 10.0 0 0.30 3E7 2.5 4.3 3.54 10.0 0 0.30 3.5E7 2.4 4.4 3.53 9.7 0 0.36 4E7 3.9 4.2 3.51 9.4 0 0.29 5E7 4.0 4.3 3.51 9.1 0 0.30 *Glu/Fru = sum of glucose and fructose **TA = total acidity ***VA = volatile acidity

(40) The results in Table 7 and FIG. 2 clearly show that 510.sup.7 CFU/ml Lactobacillus plantarum is needed to decrease the ethanol percent with 1% (v/v). However, from an inoculation level of 3.510.sup.7 CFU/ml, alcohol level is decreasing in the final wine. On top of that, FIG. 1 shows that MLF is performed during the first three days of fermentation by inoculating Lactobacillus plantarum in all dosages.

Example 2: Co-Inoculation of Grape Juice with Lactobacillus plantarum Strain CHCC12399 and a Wine Saccharomyces cerevisiae Yeast Strain

(41) As it is easier to add both Lactobacillus plantarum and the yeast together in a so called co-inoculation, it was tested if the ethanol percentage could be reduced in final wine by co-inoculating the Lactobacillus plantarum with a Saccharomyces wine yeast. In this way, winemakers can add both the Lactobacillus plantarum and the wine yeast at the start of a wine fermentation, which makes the new technique easier to use in a commercial way.

(42) For this experiment, the same Riesling juice was used as described in Table 5. Lactobacillus plantarum was added at an inoculation dosage of 510.sup.7 CFU/ml, together with the Saccharomyces cerevisiae wine yeast (110.sup.6 CFU/ml). The fermentation was performed at 20 C. Malolactic fermentation (MLF) activity and alcohol production are shown in FIGS. 3 and 4.

(43) FIGS. 3 and 4 clearly show that co-inoculation of Lactobacillus plantarum with a Saccharomyces cerevisiae wine yeast results in a final alcohol reduction of 1% (v/v). Also malic acid can be completely degraded when using co-inoculation. This means that co-inoculation of Lactobacillus plantarum together with Saccharomyces cerevisiae in the correct dosage results in a final alcohol reduction of 1% (v/v) with a completed MLF during the alcoholic fermentation.

Example 3: Reverse Inoculation of Grape Juice with Lactobacillus plantarum Strain CHCC12399, Lactobacillus plantarum Strain LPL-1 or Lactobacillus paracasei Strain LPA-1, Respectively, and a Wine Saccharomyces cerevisiae Yeast Strain

(44) To investigate if this property of lowering the alcohol content is strain- and/or species-specific, another strain of Lactobacillus plantarum, LPL-1, and a Lactobacillus paracasei strain, LPA-1, were tested for reducing alcohol (see Table 8). The same protocol was used as before, where the Saccharomyces cerevisiae wine yeast was added after 24 h. The Lactobacillus strains were inoculated in at a rate of 510.sup.7 CFU/ml and the yeast was added at a rate of 110.sup.6 CFU/ml.

(45) TABLE-US-00008 TABLE 8 Final wine parameters in Riesling wine fermented with 3 different Lactobacillus strains and a control Glu/Fru* TA** Ethanol % Malic acid VA*** Average (g/L) (g/L) (v/v) (g/L) (g/L) L. plantarum 6 5.75 9.1 0 0.38 CHCC12399 L. plantarum LPL-1 3.65 4.8 9.5 0 0.285 L. paracasei LPA-1 4.35 5.6 8.65 3.9 0.31 Only 7 6.35 10.05 4.5 0.31 Saccharomyces *Glu/Fru = sum of glucose and fructose **TA = total acidity ***VA = volatile acidity

(46) It is very clear from Table 8 that all the tested Lactobacillus sp. can reduce alcohol with at least 0.5%. However, it seems that only Lactobacillus plantarum has the combined property of reducing alcohol and degrading the malic acid completely.

Example 4: Reverse Inoculation with Lactobacillus plantarum Strain CHCC12399 and a Wine Saccharomyces cerevisiae Yeast Strain-Field Trials

(47) In these experiments, it was tested if this also works in the field. Two trials were set-up in wineries, where the wine Lactobacillus plantarum CHCC12399 strain was tested in two different grape varieties: a white grape variety, Chardonnay and a red grape variety, Merlot. In the field trials, it was also tested how the alcohol reduction with Lactobacillus sp. affect the flavor profile of the final wine, as this is the main negative issue with using either non-Saccharomyces yeast or aeration for reducing alcohol.

(48) In all cases, Lactobacillus plantarum was added on the grapes or in the grape juice at the start of fermentation at a dosage of 510.sup.6 CFU/ml and yeast was added after 24 hours. The Chardonnay trial was done in 450 hL tanks and the Merlot trial was performed in 200 hL tanks. In all cases, the initial sugar content was approximately the same and there was a control fermentation carried out with only adding Saccharomyces wine yeast. The results are shown in Table 9.

(49) TABLE-US-00009 TABLE 9 Final wine parameters in wine trials fermented with and without the wine Lactobacillus plantarum strain. Wine yeast was added after 24 h. The control fermentation is a fermentation with only Saccharomyces wine yeast. Initial Final Ethanol Grape Glu/Fru* Glu/Fru* % TA** VA*** variety Trial (g/L) (g/L) (v/v) (g/L) (g/L) Char- Control 226 0 14.14 4.58 0.37 donnay Char- L. 229 0 13.03 3.47 0.30 donnay plantarum Merlot Control 250 26 13.25 4.16 0.27 Merlot L. 244 16 12.35 4.21 0.31 plantarum *Glu/Fru = sum of glucose and fructose **TA = total acidity ***VA = volatile acidity

(50) The results in Table 9 show that in all field trials, the wine Lactobacillus plantarum was able to reduce the final alcohol percent with at least 0.8% (v/v) and this with a very low inoculation rate (510.sup.6 CFU/ml). This means the proposed technique of using a Lactobacillus strain to reduce alcohol with at least 0.5% also works in big scale.

(51) The wines from Table 9 were analyzed for flavor profile. Furthermore, the Chardonnay wines were analyzed by a professional sensory panel to investigate if the final wines are acceptable with regard to flavor profile compared to the control wines. The sensory panel consisted of 10 people who scored the wines 1 to 5 for each characteristic. The flavor analysis results, performed with headspace-GC-FID, are shown in FIGS. 5 and 6. Sensory analysis results are shown in FIGS. 7 and 8.

(52) As can be seen in both FIGS. 5 and 6, the flavor profile of the Chardonnay control compared with the Chardonnay inoculated with Lactobacillus plantarum look very similar. Especially ethyl acetate, which is a main trouble component when non-Saccharomyces strains are added, is very similar in both wines and even a bit less in the Chardonnay with addition of Lactobacillus plantarum. The same is true for the Merlot wines, also here, the flavor profiles look very similar.

(53) It is very clear from the sensory analysis that the Chardonnay wine with addition of Lactobacillus plantarum has certainly no defects compared to the control wine. The test wine is hotter, as shown in FIG. 7, but the overall aromatic intensity and complexity are more preferred in the test wines, compared to the control wine. This is confirmed in the descriptive analysis, where the Chardonnay with the addition of Lactobacillus plantarum has a more fruity character (grapefruit, pineapple, mango), compared to the control wine.

Example 5: Reverse Inoculation with Different Time Periods Between Inoculation with Lactobacillus plantarum Strain CHCC12399 and Inoculation with Merit Wine Yeast in Hungarian White Grape Juice

(54) To investigate the potential of Lactobacillus plantarum strain CHCC12399 to reduce alcohol by reducing glucose and/or fructose into lactic acid at the beginning of wine fermentation, experiments were performed in Hungarian white grape juice with the following characteristics (Table 10).

(55) TABLE-US-00010 TABLE 10 Hungarian white grape juice parameters: Malic Total Ethanol Glu/Fru acid acid % VA Parameters (g/l) pH (g/l) (g/l) (v/v) (g/l) Grape juice 166.2 3.29 1.5 3.1 0 0.16

(56) The fermentation set-up is described in Table 11. Here it is shown that 6 different set-ups were used: addition of Lactobacillus plantarum strain CHCC12399 (NoVA) at the start of fermentation with addition of yeast at different time points afterwards. As control fermentations only yeast were inoculated (exp 11 and 12). The yeast used is Merit inoculated at a concentration of 110.sup.6 CFU/ml. Fermentations were carried out at 20 C.

(57) TABLE-US-00011 TABLE 11 Fermentation set-up Exp Time of yeast number NoVa added inoculation Total volume 1 5E7 CFU/ml 0 h 200 ml 2 5E7 CFU/ml 0 h 200 ml 3 5E7 CFU/ml 24 h 200 ml 4 5E7 CFU/ml 24 h 200 ml 5 5E7 CFU/ml 48 h 200 ml 6 5E7 CFU/ml 48 h 200 ml 7 5E7 CFU/ml 72 h 200 ml 8 5E7 CFU/ml 72 h 200 ml 9 5E7 CFU/ml 200 ml 10 5E7 CFU/ml 200 ml 11 0 h 200 ml 12 0 h 200 ml

(58) Fermentations were carried out until sugar was depleted. During the fermentations, samples were taken for Oenofoss measurements, as well as HPLC measurements for sugars and acids.

(59) Viability of Lactobacillus was analysed by anaerobic plating on grape juice agar (GJ5; 77.5 g/L grape juice concentrate (K V Saft Val), 22.4 g/L yeast extract (Bio Springer), 0.6 g/L Tween 80 (Sigma-Aldrich), 0.1 g/L MnSO.sub.4, H.sub.2O (Merck), 15 g/L agar (SO-BI-GEL) in tap water) with natamyxin (0.075 g/L; Delvocide from DSM Food Specialities B.V.). The plates were incubated for 3 days at 30 C. and then counted.

(60) Viability of yeast was analysed by plating on standard YGC (yeast extract, glucose, chloramphenicol) agar plates. The plates were incubated for 2 days at 30 C. and then counted.

(61) The results shown do not contain experiment number 9 and 10, as spontaneous fermentation with yeast started in these samples and they are therefore considered to be inaccurate. Spontaneous fermentation also started in exp number 7 and 8, but was over seeded with Merit yeast at 72 hours after addition of Lactobacillus plantarum strain CHCC12399 (NoVa). The Lactobacillus and yeast cell counts are given in FIGS. 9 and 10, respectively. The sugar and acid measurements are shown in FIGS. 11-15.

(62) From the results shown in FIGS. 9 and 10, it is clear that Lactobacillus plantarum grew in all experiments, except for the experiment with co-inoculation, where the cell count does not change over time. This could be explained by the fact that yeast counts are already very high at day 1, meaning that Lactobacillus plantarum strain CHCC12399 does not have the time to grow. Lactobacillus plantarum strain CHCC12399 (NoVA) is still active, though, as malic acid is completely consumed after day 2 (FIG. 14).

(63) Ethanol concentrations during fermentation are shown in FIG. 16 and final ethanol concentrations are shown in Table 12.

(64) TABLE-US-00012 TABLE 12 Final ethanol concentrations of all wines EtOH % NoVA Yeast (v/v) + 0 h 8.3 + 24 h 8.2 + 48 h 8.05 + 72 h 7.85 0 h 8.45

(65) It is very clear from the final ethanol concentrations in Table 12 that adding Lactobacillus plantarum strain CHCC12399 (NoVA) in a concentration of 510.sup.7 CFU/ml and fermenting 2 or 3 days before the addition of yeast can reduce alcohol levels. Highest alcohol reduction was with yeast inoculation after 3 days in this case, where the final alcohol reduction is 0.6% ethanol (v/v).

(66) From FIGS. 11-14, it is also clear that malic acid needs to be consumed first, before Lactobacillus plantarum strain CHCC12399 (NoVA) starts eating sugar. Malic acid is consumed within 2 days for all different scenarios. Even with the co-inoculation of Lactobacillus plantarum strain CHCC12399 and yeast, malic acid gets consumed within 2 days after the start of fermentation. Lactic acid is produced from both malic acid and sugar (fructose), resulting in a higher lactic acid production when the yeast is added later. Almost no lactic acid is produced in the control fermentation, where only yeast is added.

(67) It is also clear from the single sugar results in Table 13 that Lactobacillus plantarum strain CHCC12399 (NoVA) prefers fructose over glucose.

(68) TABLE-US-00013 TABLE 13 Sugar consumption (in g/L) with NoVA, measured before yeast addition NoVa Yeast Day measured Glucose Fructose + 0 h 0 0 0 + 24 h 1 0 0 + 48 h 2 0 0.6 + 72 h 3 0.1 2.7

(69) Lactobacillus plantarum strain CHCC12399 consumes fructose and no glucose before the yeast is added, but must also consume more fructose (and/or glucose) during the start of alcoholic fermentation, in order to decrease the alcohol level with 0.6%.

CONCLUSION

(70) It is clear from this experiment that Lactobacillus plantarum strain CHCC12399 can reduce alcohol in final wines by reverse inoculation. The results show that malic acid is assimilated first, before Lactobacillus plantarum strain CHCC12399 starts assimilating sugar. It was also surprising to find that Lactobacillus plantarum strain CHCC12399 consumes fructose as the main sugar, and does not seem to consume glucose in this case.

DEPOSIT AND EXPERT SOLUTION

(71) The strain of Lactobacillus plantarum CHCC12399 was deposited with Deutsche Sammlung von Mikroorganismen and Zellkulturen GmbH, Inhoffenstr. 7B, D-38124 Braunschweig, Germany on 1 Aug. 2013 under the accession number DSM 27565.

(72) The deposit has been made under the conditions of the Budapest Treaty on the International Recognition of the Deposit of Microorganisms for the Purposes of Patent Procedure.

(73) The Applicant requests that a sample of the deposited microorganisms should be made available only to an expert approved by the Applicant.

REFERENCES

(74) Ciani, M., and Ferraro, L. (1996) Enhanced glycerol content in wines made with immobilized Candida stellata cells. Applied and Environmental Microbiology, 62:128-132.

(75) Comitini, F., Gobbi, M., Domizio, P., Romani, C., Lencioni, L., Mannazzu, I., and Ciani, M. (2011) Selected non-Saccharomyces wine yeasts in controlled multistarter fermentations with Saccharomyces cerevisiae. Food Microbiology, 28:873-882.

(76) Contreras, A., Hidalgo, C., Henschke, P. A., Chambers, P. J., Curtin, C., and Varela, C. (2013) Evaluation of non-Saccharomyces yeast for the reduction of alcohol content in wine. DOI:10.1128/AEM.03780-13.

(77) Di Maio, S., Genna, V., Gandolfo, V., Amore, G., Ciaccio, M., and Oliva, D. (2012) Presence of Candida zemplinina in Sicilian musts and selection of a strain for wine mixed fermentations. South African Journal of Enology and Viticulture, 33:80-87.

(78) Erten, H., and Campbell, I. (2001) The production of low-alcohol wines by aerobic yeasts. Journal of the Institute of Brewing, 107:207-215.

(79) Ferraro, L., Fatichenti, F. and Ciani, M. (2000) Pilot scale vinification process using immobilized Candida stellata cells and Saccharomyces cerevisiae. Process Biochemistry, 35:1125-1129.

(80) Garcia, V., Vasquez, H., Fonseca, F., Manzanares, P., Viana, F., Martinez, C. et al. (2010) Effects of using mixed wine yeast cultures in the production of Chardonnay wines. Revista Argentina de Microbiologia, 42:226-229.

(81) Grossmann, M., Kruse, R. and Heintz, W. (1991). Producing drinks of low to zero alcohol contentoxygen is added to liquid in controlled manner during fermentation to convert sugars to water and carbon dioxide. DE3939064 (A1).

(82) Kaeppeli, O. (1989). Production of sugar-free or low-sugar fruit juiceby fermentation with yeast to carbon dioxide and water stage, avoiding ethanol production. CH668887 (A5).

(83) Kolb, E., Wiesenberger, A., Stahl, U. and Harwart, K. (1993). Biological processes for the partial reduction of sugar in fruit juices. Proceedings of the International Federation of Fruit Juice Producers Symposium, Budapest, 7-20.

(84) Kutyna, D. R., Varela, C., Henschke, P. A., Chambers, P. A., and Stanley, G. A. (2010) Microbiological approaches to lowering ethanol concentrations in wine. Trends Food Science and Technology, 21:293-302.

(85) Maygar, I. and Toth, T. (2011) Comparative evaluation of some oenological properties in wine strains of Candida stellata Candida zemplinina, Saccharomyces uvarum and Saccharomyces cerevisiae. Food Microbiology, 28:94-100.

(86) Pickering G. J. (2000). Low and reduced-alcohol wines: a review. Journal of Wine Research, 11:129-144.

(87) Sadoudi, M., Tourdot-Marechal, R., Rousseaux, S., Steyer, D., Gallardo-Chacon, J. J., Ballester, J., Vichi, S., Guerin-Schneider, R., Caixach, J., and Alexandre, H. (2012) Yeast-yeast interactions revealed by aromatic profile analysis of Sauvignon Blanc wine fermented by single or co-culture of non-Saccharomyces and Saccharomyces yeasts. Food Microbiology, 32:243-253.

(88) Schmidtke, L. M., Blackman, J. W., and Agboola, S. O. (2012) Production technologies for reduced alcoholic wines. Journal of Food Science, 71:25-41.

(89) Smith, P. M. (1995) Biological processes for the reduction of alcohol in wines. M. Appl. Sci., dissertation, Lincoln University, New Zealand.