Introduction

A simplistic view of winemaking involves the biochemical transformation of hexose sugars present in grape must to ethanol and carbon dioxide by Saccharomyces cerevisiae and/or other Saccharomycetaceae. S. cerevisiae is generally the dominant fermentative species, as it is able to withstand fermentation-related stresses (especially ethanol), as well as inhibit other species, so that fermentation completes in a timely manner (Albergaria and Arneborg, 2016). As such, it is common practise for grape must to be inoculated with commercial monocultures to ensure fermentation reliability, although wines created in this way can lack sensorial complexity (Padilla et al., 2016). Conversely, wine made by “wild” fermentation, making use of the native microflora present on grapes, can provide a complexity not achieved through monoculture (reviewed in Belda et al., 2017). Whilst these uninoculated or ‘spontaneous/wild' fermentations are at risk of microbial failure if incorrectly managed, between 3–6 % of wines produced in Australia are made by this means, in particular, within the premium wine sector (The Australian Wine Research Institute, 2019). A more recent and reliable alternative is the use of selected non-Saccharomyces strains co-cultured with S. cerevisiae, which allow increased complexity and mouthfeel (attributes associated with wine quality) and consistent fermentation reliability (Padilla et al., 2016). The use of these techniques is reflected in the increasing commercial availability of non-Saccharomyces starter cultures (including Metschnikowia spp, Lachancea thermotolerans and Torulaspora delbrueckii), either as single strains or mixed with other non-Saccharomyces and/or S. cerevisiae.

Mixed fermentations involving S. cerevisiae and non-Saccharomyces yeast are generally performed using two different inoculation regimes: co-inoculation or sequential inoculation. Co-inoculation involves simultaneous inoculation with more than one yeast, whereas sequential inoculation involves an inoculation with the non-Saccharomyces species, followed by S. cerevisiae some days later to enable fermentation completion (Padilla et al., 2016). Co-inoculation offers benefits, such as acidification in the case of Lachancea thermotolerans (Comitini et al., 2011; Gobbi et al., 2013; Kapsopoulou et al., 2007), decreased acetic acid (Comitini et al., 2011; Zohre and Erten, 2002), increased acetate esters (Varela et al., 2017; Zohre and Erten, 2002) and lower alcohol contents by Metschnikowia pulcherrima (Varela et al., 2017). Whilst both methods can improve wine sensory complexity, sequential inoculation is generally favoured due to the additional control over S. cerevisiae populations through timing and inoculation rate, allowing the expression of non-Saccharomyces characteristics (Vilela, 2020).

Recent advancements in co-inoculation strategies include the use of slow growing S. cerevisiae strains to achieve greater non-Saccharomyces abundance. Albertin et al. (2017) utilised a long-lag-phase S. cerevisiae in co-fermentation with five non-Saccharomyces yeasts: Hanseniaspora uvarum, Candida zemplinina, Metschnikowia spp., Torulaspora delbrueckii and Pichia kluyveri. The long lag phase trait depended on the presence of sulfite, and arose due to reduced activity of the sulfite exporter, SSU1. The resulting wines exhibited increased complexity and fruitiness because of the extended presence of the non-Saccharomyces yeast. Whilst this is one example of a novel inoculation protocol, other yeast mutants that exhibit a long lag phase or slow growth independent of the presence of sulfite would be of merit.

This study targeted the disruption of SER1, encoding 3-phosphoserine aminotransferase, because of its slow-growth phenotype, which was hypothesised as being useful in a co-inoculation regime. The ser1Δ deletion was first associated with the slow initial growth in a heme-deficient laboratory strain (Reiner et al., 2006). A naturally occurring ‘loss of function' variant, Ser1p (G78R) in Sake yeast, resulted in reduced growth rate and biomass production (Jung et al., 2018). The introduction of Ser1p (G78R) into wine Saccharomyces genotypes using “self-cloning” techniques has commercial potential, since such strains are classified as non-recombinant in jurisdictions such as Japan and the US (reviewed in Hanlon and Sewalt, 2020). Accordingly, we introduced the Ser1p (G78R) variant into the commonly used wine strain EC1118 by CRISPR/Cas9 gene editing and assessed the fermentation outcomes in monoculture and co-culture with either Lachancea thermotolerans or Metschnikowia pulcherrima, in order to determine its potential for use as a wine starter culture.

Materials and methods

Results

1. Introduction of the SER1(G78R) variant into EC1118 via CRISPR/Cas9

SER1(G78R) is a naturally occurring ‘loss of function' variant identified in a Sake yeast strain by QTL analysis, which is associated with increased chronological age (Jung et al., 2018). The associated glycine to arginine substitution was introduced into the wine yeast EC1118 as a non-synonymous mutation at nucleotide position +232 (G > C i.e., GGG > CGG) by homology directed repair using CRISPR/Cas9 (Shaw et al., 2019). Both the SER1(G78R) naturally occurring ‘loss of function' variant (Jung et al., 2018) and a SER1 deletion in a ser1Δ heme deficient yeast (Reiner et al., 2006) have been found to reduce growth. Whilst serine biosynthesis via glycolysis is disrupted, the mutants are not auxotrophs, as serine can be produced from glycine via gluconeogenesis (Melcher and Entian, 1992). The reduced growth phenotype may be beneficial in wine co-inoculations, since the persistence of non-Saccharomyces in the later stages of fermentation may result in wine with greater aroma complexity and novelty because of the compounds derived from these yeasts.

The fermentation characteristics of SER1(G78R) were compared to wild type EC1118 in 100 mL fermentations of CDGJM. The mutant exhibited slower initial growth (Figure 1A), similar to that observed in ser1Δ heme-deficient strains (Reiner et al., 2006) and in other strains containing the SER1(G78R) mutation (Jung et al., 2018). EC1118 growth took 48 h to reach stationary phase, whereas SER1(G78R) took 144 h, achieved a reduced viable population (Figure 1A) and took an extra 120 h to complete fermentation (i.e., 288 h vs 168 h) (Figure 1B). Analysis of fermentation relevant metabolites by HPLC showed significantly higher acetic acid (1.90 g/L vs 1.61 g/L ), lactic acid (0.07 g/L vs 0.00 g/L ), succinic acid (1.32 g/L vs 0.98 g/L ), glycerol (4.62 g/L vs 4.30 g/L ) and ethanol (102.10 g/L vs 100.20 g/L; Table 3) in the mutant vs the wildtype. The observed increases in succinic acid, acetic acid and glycerol were expected, as they have previously been reported with ser1Δ mutants of the laboratory strain, BY4742, when grown in synthetic must (Chidi et al., 2016). Whilst statistically significant increases in lactic acid and ethanol were also observed, they are unlikely to be sensorially significant. The acetic acid concentrations in both strains were over the sensory threshold (1.19 g/L for white wines; Corison et al., 1979), explaining why they contributed to the wines having a harsh ‘vinegar' smell.

Figure 1. Evaluation of fermentation performance and viability of EC1118 and SER1(G78R) grown in CDGJM.

Save View full size Expand inline Collapse inline

Fermentations were conducted in 100 mL of CDGJM containing 200 g/L total sugars and 450 mg/L N (n = 3). Viable cell counts (A) and total sugars (B) were determined by flow cytometry and enzymatic analysis, respectively for EC1118 (●) and SER1(G78R) (■). Error bars denote standard deviation from the mean.

Table 3. Fermentation duration (h) and metabolite content (g/L) of ‘wines’ produced by EC1118 and SER1(G78R) when fermenting in CDGJM.

View popup Expand inline Collapse inline

	Duration	Acetic Acid	Lactic Acid	Malic Acid	Succinic Acid	Glycerol	Ethanol
EC1118	168 ± 0.00	1.61 ± 0.03	0.00 ± 0.00	3.25 ± 0.03	0.98 ± 0.01	4.30 ± 0.03	100.20 ± 0.49
SER1(G78R)	288 ± 0.00*	1.90 ± 0.03*	0.07 ± 0.00*	3.20 ± 0.02	1.32 ± 0.04*	4.62 ± 0.03*	102.10 ± 0.25*

*Denotes significant difference (Unpaired t-test; p < 0.01 between the strains for the given parameter).

2. Co-inoculated fermentations in Sauvignon blanc juice: fermentation performance and metabolite contribution

Yeasts with slowed initial growth have been shown to be potentially useful in co-inoculation experiments (Albertin et al., 2017). To investigate the effects of SER1(G78R) in co-inoculated winemaking, 100 mL fermentations were undertaken in Sauvignon blanc juice with both S. cerevisiae and non-Saccharomyces monocultures and co-inoculated cultures. Monocultures of EC1118 and SER1(G78R) completed fermentation, whereas MP2 was sluggish with 54.5 ± 5.0 g/L residual sugar when the other fermentations were concluded (Supplementary Table 1). This was likely due to a reduction in viable cells present in the fermentation after 264 h when the ethanol reached 55.7 g/L (5.5 % (w/v)) (Figure 2A, Supplementary Table 1). Concerto monocultures largely failed to consume sugars (Figure 2A) with the yeast dying 24 h after inoculation (Supplementary Table 1). These results support the recommendation of the manufacturer that Concerto be used in unsulfured grape juice, as the yeast is sulfite (SO₂)-sensitive at 40 mg/L total SO₂, which is within the range (20–50 mg/L), ordinarily considered non-inhibitory to non-Saccharomyces (Henick-Kling et al., 1998). As with CDGJM, the fermentation duration of SER1(G78R) was significantly extended (+216 h; Table 4) and had a lower population density compared to EC1118 (Figure 2A).

All co-inoculated fermentations completed fermentation. EC1118 quickly dominated the fermentations when co-inoculated with MP2, successfully outcompeting Metschnikowia regardless of the inoculation regime (Figure 2B) and with the fermentations having the same duration as the EC1118 monoculture (Table 4, Figure 2A). The population dynamics of EC1118 and Concerto mimicked that of MP2, with EC1118 rapidly dominating the fermentation resulting in no detectable Concerto cells by WL plating after 72 h (Figure 2C).

Interestingly, the slower-growing SER1(G78R) mutant permitted greater populations of non-Saccharomyces in co-culture than did EC1118. This was evident in SER1(G78R) fermentations co-inoculated with MP2 (Figure 2D). Thus, while EC1118 dominated the culture early on and no colonies of MP2 were noted under either inoculation regime (Figure 2B), this was not the case for the mutant (Figure 2D). The maximum cell number reflected the amount inoculated for the given inoculation ratio; MP2 CFUs declined after 168 h, eventually dropping to zero at the end of fermentation. The largely viable SER1(G78R) population was responsible for the MP2 co-inoculated fermentations finishing after 408 h (Figure 2D), in contrast to the MP2 monoculture. The duration of the mixed fermentation was identical to SER1(G78R) as a monoculture (Figure 2A). In both cases, the Metschnikowia population declined with increasing ethanol content as fermentation progressed (Figure 2A and 2D).

The growth behaviour of the Concerto Lachancea in monoculture and co-culture with SER1(G78R) was surprisingly different, with the presence of the slow-growing mutant enabling the growth of the SO₂-sensitive Lachancea in SO₂ conditions that were lethal in the Concerto monoculture (Figure 2A). This led to Concerto dominating the co-inoculated fermentation in numbers greater than observed with EC1118 (Figures 2C and 2E), and an earlier completion of fermentation than the SER1(G78R) monoculture (i.e., 240 h vs 408 h; Table 4). These results allude to a synergistic interaction between the two yeasts when in co-culture in sulfured Sauvignon blanc juice, the basis of which is unclear. Whilst the fermentation was extended in the co-inoculations with the SER1(G78R) mutant compared to EC1118, the presence of Concerto, and to a lesser extent MP2, was anticipated to influence the chemical composition of the resultant wines.

Figure 2. Viable yeast counts from Sauvignon blanc juice fermentations

Save View full size Expand inline Collapse inline

Viable yeast counts from WL agar plating of either monocultures (A) of EC1118 (●), SER1(G78R) (■), MP2 (▲) or Concerto (▼) or co-inoculated cultures of (B) EC1118 with MP2, (C) EC1118 with Concerto, (D) SER1(G78R) with MP2 or (E) SER1(G78R) with Concerto. Solid lines in co-inoculated fermentations denote S. cerevisiae populations, whereas dashed lines denote non-Saccharomyces populations. Square (■) markers in co-inoculated fermentations correspond to a 1:9 inoculation ratio, whereas circles (●) correspond to a 1:1 inoculation ratio.

HPLC analysis of the completed wines targeted organic acids, glycerol and ethanol, as well as lactic acid, which is reportedly produced by L. thermotolerans in a strain-dependent manner (Hranilovic et al., 2020; Vaquero et al., 2020). Interestingly, lactic acid did not differ between EC1118 or SER1(G78R) and Concerto in the monocultures and co-inoculated fermentations (Table 4). Furthermore, no differences were seen in any of the other quantified metabolites in the wines produced by EC1118 in monoculture and co-inoculated fermentations (Table 4). This can likely be attributed to the dominance of EC1118 in these fermentations (Figure 2A).

The comparison of monocultures of EC1118 and its SER1(G78R) mutant showed the impact of the ‘loss of function' mutation in glycolysis (Supplementary Figure 1). 3-phosphoglycerate feeds directly into serine biosynthesis with the SER1 encoded phosphoserine transaminase catalysing a bi-directional reaction between 3-phosphooxypyruvate and 3-phospho-L-serine. Disruption of SER1 is postulated to result in a feed-back by the accumulated precursors with the glycolytic flux favouring acetic acid production. This is corroborated by the large increase in acetic acid in wines produced by the SER1(G78R) mutant compared to EC1118 (1.56 g/L vs 0.09 g/L; Table 4). Such concentrations can be considered a fault, as they exceed the odour threshold (1.19 g/L for white wines; Corison et al., 1979) and contributed to an unpleasant ‘vinegar-like' aroma as seen in CDGJM. In contrast, the marginal change in succinic acid (3.85 g/L (EC1118) vs 3.55 g/L (SER1(G78R)); Table 4) indicated that flux through the tricarboxylic acid cycle was not greatly affected by the mutation. The 29.5 % increase in glycerol by the mutant (8.59 g/L vs 6.63 g/L; Table 4), also points to a modulation of glycolysis between 3-phosphoglycerate and glyceraldehyde-3-phosphate due to the SER1 disruption (Supplementary Figure 1). This increase would likely alter the perception of sweetness, but not viscosity (Noble and Bursick, 1984), as detectable changes are noted at 5.2 g/L and 25 g/L, respectively. The strain's classification in Australia as genetically modified prevented sensory analysis due to containment regulations.

The slow growth of SER1(G78R) enabled non-Saccharomyces to persist in co-inoculated fermentations, thereby producing significant differences between the monoculture and most co-fermentations in the resulting wines. Acetic acid and glycerol yields were affected by the non-Saccharomyces co-inoculated with SER1(G78R) and the inoculation regime (Table 4). Co-fermentation with Metschnikowia (MP2) resulted in 11 % more glycerol when inoculated at 9 times the Saccharomyces rate. This was associated with a 73 % decrease in acetic acid (0.41 vs 1.56 g/L) and a 2 % decrease in ethanol (-2.58 g/L) compared to the SER1(G78R) monoculture (Table 4). These findings mirror earlier observations (Hranilovic et al., 2020), where MP2 also increased glycerol and decreased ethanol and acetic acid in monocultures and sequentially inoculated fermentations. Sadoudi et al. (2017) also demonstrated decreased acetic acid (40 %) and increased glycerol (12 %) in a sequential culture of S. cerevisiae/M. pulcherrima (inoculated 1:10) compared to the pure S. cerevisiae culture.

Table 4. Fermentation duration (h) and metabolite content (g/L) of monoculture and co-inoculated fermentations in Sauvignon blanc juice.

View popup Expand inline Collapse inline

Strain(s)	Duration	Acetic Acid	Lactic Acid	Malic Acid	Succinic Acid	Glycerol	Ethanol
EC1118 (monoculture)	168 ± 0.00c	0.21 ± 0.07de	0.69 ± 0.02a	3.97 ± 0.01a	3.85 ± 0.01b	6.63 ± 0.03e	86.29 ± 0.11a
EC1118 + MP2 (1:1)	168 ± 0.00c	0.24 ± 0.02de	0.70 ± 0.15a	3.96 ± 0.02a	3.79 ± 0.03b	6.65 ± 0.06e	85.95 ± 0.45a
EC1118 + MP2 (1:9)	168 ± 0.00c	0.12 ± 0.11e	0.77 ± 0.04a	4.05 ± 0.02a	3.75 ± 0.04b	6.69 ± 0.01e	86.13 ± 0.34a
EC1118 + Concerto (1:1)	168 ± 0.00c	0.26 ± 0.02de	0.68 ± 0.05a	3.91 ± 0.02a	3.81 ± 0.03b	6.63 ± 0.00e	86.22 ± 0.05a
EC1118 + Concerto (1:9)	168 ± 0.00c	0.22 ± 0.03de	0.72 ± 0.02a	3.92 ± 0.03a	3.81 ± 0.03b	6.61 ± 0.01e	86.37 ± 0.15a
SER1(G78R) (monoculture)	408 ± 0.00a	1.56 ± 0.11a	0.55 ± 0.09a	4.12 ± 0.16a	3.55 ± 0.02c	8.59 ± 0.04c	85.87 ± 0.14a
SER1(G78R) + MP2 (1:1)	408 ± 0.00a	1.28 ± 0.03b	0.74 ± 0.06a	4.06 ± 0.06a	3.48 ± 0.02c	8.83 ± 0.03b	85.66 ± 0.06a
SER1(G78R) + MP2 (1:9)	408 ± 0.00a	0.41 ± 0.10cd	0.72 ± 0.12a	3.83 ± 0.11a	3.15 ± 0.08d	9.60 ± 0.07a	83.29 ± 0.38b
SER1(G78R) + Concerto (1:1)	240 ± 0.00b	0.62 ± 0.02c	0.58 ± 0.16a	3.31 ± 0.19b	3.79 ± 0.02b	7.36 ± 0.12d	86.22 ± 1.25a
SER1(G78R) + Concerto (1:9)	240 ± 0.00b	0.39 ± 0.01d	0.58 ± 0.01a	3.08 ± 0.04b	4.05 ± 0.01a	7.22 ± 0.05d	86.50 ± 0.60a

Letters denote significant difference between values in columns (Tukey’s HSD test; p < 0.01).

*Monocultures of Concerto and MP2 did not finish fermentation, having 166.20 ± 5.26 g/L and 54.46 ± 5.01 g/L residual sugar respectively. Refer to Supplementary Table 1.

In the co-fermentations with Concerto, L. thermotolerans dominated and glycerol levels fell to between those of the SER1(G78R) monoculture and the EC1118; with the lowest concentrations measured in the high (1:9) inoculum ratio (Table 4). These findings allude to Concerto being a high producer of glycerol, a phenotype commonly observed in L. thermotolerans yeast (Benito et al., 2016b; Comitini et al., 2011; Gobbi et al., 2013; Kapsopoulou et al., 2007).

Succinic acid differed between the two non-Saccharomyces and different inoculation ratios (Saccharomyces:non-Saccharomyces). Co-inoculation of SER1(G78R) with MP2, gave significantly lower succinic acid at the (1:9) ratio compared to the SER1(G78R) monoculture (3.15 g/L vs 3.55 g/L; Table 4). The small reduction in succinic acid at the 1:1 ratio with MP2 was not significant (3.48 g/L vs 3.55 g/L). This differs from Hranilovic et al. (2020), who reported MP2 to produce either more or comparable amounts of succinic acid to that of EC1118 grown in CDGJM or grape juice, respectively. Co-inoculation of SER1(G78R) with Concerto gave increased succinic acid. When inoculated equally, succinic acid (3.79 g/L) was similar to the EC1118 fermentations (3.81 g/L), whist the higher inoculum of Concerto (1:9) resulted in a 6.3 % increase (4.05 g/L). These concentrations are considerably higher than in the commercial white wines surveyed by Coulter et al. (2004) (0.1–1.6 g/L; ave. 0.6 g/L). Whilst increased succinic acid in wine can be identifiable as a salty sour taste (Coulter et al., 2004), sensory analysis was not undertaken here due to regulatory restrictions.

Malic acid degradation by wine yeast is a useful attribute providing there is an alternative to the malolactic fermentation undertaken by lactic acid bacteria to modulate acidity and aroma (Du Plessis et al., 2017). While EC1118 has minimal impact on malic acid (Redzepovic et al., 2003), degradation by non-Saccharomyces was checked for. As expected, concentrations in EC1118-dominated fermentations were unchanged in the Sauvignon blanc juice (Table 4). SER1(G78R) in monoculture vs 1:1 inoculation with MP2 changed malic acid little. However, significant reductions were seen, albeit only for co-inoculation of SER1(G78R) with Concerto at 1:1 or 1:9, these being 20 % and 25 %, respectively (Table 4). This range agrees with reports of L-malic acid degradation being strain-dependent (reviewed in Benito, 2018).

Acetate esters result from esterification of ethanol or higher alcohols (from Ehrlich degradation of amino acids; reviewed in Belda et al., 2017) with acetyl-CoA and are important contributors to wine aroma. MP2 is reported to increase acetate ester production (Hranilovic et al., 2020). The end-point ester analysis in this study revealed several differences. Monocultures of EC1118 and SER1(G78R) differed significantly in propyl acetate (pear-like), but values (Table 5) remained well below the sensory threshold of 4.7 mg/L for white wines (Miller, 2019). In terms of the comparison between EC1118 mono- and co-inoculated fermentations, no differences were found; this is probably because the latter were dominated by EC1118, which is “defined as a neutral strain” (Marcon et al., 2018). In the case of SER1(G78R) mono- vs co-cultures, many differences were evident. Co-culture with MP2 - particularly at the 1:9 ratio - lead to significant increases in propyl, isobutyl, isoamyl, 2-phenylethyl and ethyl acetates, ranging in magnitude from 1.7- to 5.9-fold (Table 5). In this case, the largest quantitative increase (in ethyl acetate) reached 8.8 mg/L, thereby exceeding the sensory threshold (7.5 mg/L; Guth, 1997) at which fruity aromas are perceived, but remaining well below 100 mg/L, at which a ‘nail polish remover' fault is evident (Sumby et al., 2010). Isobutyl acetate (up 5.9 x to 14 µg/L) has a sensory threshold of 1.6 µg/L, at which a fruity, apple- and banana-like scent is imparted (Haggerty, 2016). Isoamyl acetate (up 3.5x to 76 µg/L) exceeds the 30 µg/L sensory threshold of this ‘banana-like' acetate ester (Guth, 1997). 2-Phenylethyl acetate concentrations were below the sensory threshold of 250 µg/L (Guth, 1997). Overall, these results reflect the enhancement of acetate esters, as previously described for fermentations utilising M. pulcherrima (Hranilovic et al., 2020).

Table 5. Acetate ester content (µg/L) of monoculture and co-inoculated fermentations in Sauvignon blanc juice.

View popup Expand inline Collapse inline

Strain(s)*	Propyl Acetate	Isobutyl Acetate	Isoamyl Acetate	cis-3-Hexenyl Acetate	2-Phenylethyl Acetate	Ethyl Acetate
EC1118	2.66 ± 0.21d	1.58 ± 0.12cd	35.88 ± 7.17b	56.39 ± 2.05a	17.51 ± 1.36b	2184 ± 177.0c
EC1118 + MP2 (1:1)	2.65 ± 0.22d	1.51 ± 0.11d	35.38 ± 10.60b	60.95 ± 3.77a	19.59 ± 0.95b	2238 ± 230.1c
EC1118 + MP2 (1:9)	2.97 ± 0.52d	1.65 ± 0.23cd	36.23 ± 15.98b	53.86 ± 11.66a	20.53 ± 4.06b	2996 ± 419.2bc
EC1118 + Concerto (1:1)	2.76 ± 0.12d	1.59 ± 0.04cd	26.9 ± 8.41b	56.76 ± 3.09a	19.74 ± 1.24b	2432 ± 142.1bc
EC1118 + Concerto (1:9)	2.88 ± 0.71d	1.71 ± 0.55cd	36.22 ± 8.96b	53.39 ± 0.99a	18.74 ± 0.91b	2228 ± 334.1c
SER1(G78R)	8.64 ± 1.17c	2.37 ± 0.32bcd	21.84 ± 2.86b	63.18 ± 9.45a	18.64 ± 1.58b	2363 ± 189.2bc
SER1(G78R) + MP2 (1:1)	11.29 ± 1.34b	3.15 ± 0.40b	21.98 ± 5.76b	54.36 ± 11.30a	22.03 ± 2.10b	3598 ± 519.6b
SER1(G78R) + MP2 (1:9)	14.85 ± 0.46a	14.06 ± 0.32a	75.86 ± 22.13a	60.97 ± 5.62a	46.33 ± 2.38a	8801 ± 526.9a
SER1(G78R) + Concerto (1:1)	4.42 ± 0.18d	2.75 ± 0.16bc	15.21 ± 7.37b	8.96 ± 0.81b	2.31 ± 0.21c	3342 ± 135.6bc
SER1(G78R) + Concerto (1:9)	3.87 ± 0.50d	3.28 ± 0.59b	15.93 ± 5.66b	5.54 ± 1.03b	1.90 ± 0.19c	3275 ± 631.1bc

Letters denote significant difference between values in the same column (Tukey’s HSD test (p < 0.01).

*Monocultures of Concerto and MP2 did not finish fermentation; data available in Supplementary Table 1.

The significant differences between SER1(G78R) monocultures compared with co-cultures with Concerto comprised reductions in propyl, cis-3-hexenyl and 2-phenylethyl acetates (Table 5). Propyl acetate was reduced to below its sensory threshold, presumably modulating its pear-like aroma. Cis-3-hexenyl acetate (fruity, green tea aroma) does not have a reported sensory threshold in wine; therefore the influence of the ~10-fold reduction in this compound is unclear.

The Principal Component Analysis (PCA) of the HPLC and GC-MS data sought to define the relationship between fermentation regime and wine metabolites (Figure 3). Four distinct groups were identified within the samples. The first (top right) comprised solely the fermentations containing both monoculture and co-inoculated EC1118, the clustering likely due to a lack of significant differences between the sensory compounds analysed (Table 4 and 5). This is not surprising given the predominance of EC1118 in these fermentations (Figure 2A–2C). For the SER1(G78R) fermentations co-inoculated with MP2, there were two clusters: one (far left) comprised the 1:9 co-inoculation ratio of SER1(G78R) and MP2, which featured high acetate esters (propyl, isobutyl, isoamyl, 2-phenylethyl and ethyl acetates) and glycerol. The other group (centre left) comprised SER1(G78R)/MP2 1:1 co-inoculations and SER1(G78R) monoculture fermentations; this group is largely influenced by the production of acetic acid - a phenotype associated with SER1(G78R) fermentations.

Final wine volatile composition was dependent on the inoculation regime, with the higher inoculation ratio (i.e., greater MP2 inoculum; Figure 2D) resulting in largely different fermentations compared to lower inoculation densities, which resulted in fermentations more similar to the SER1(G78R) monocultures. The final group (bottom right) comprised all fermentations involving SER1(G78R)/Concerto co-inoculations, the Concerto monoculture being non-viable. The resultant wines reflect the dominance of the Concerto population in both co-fermentations, and thereby the characteristics of this non-Saccharomyces (Figure 2E). These fermentations were typically characterised by a lower production of some acetate esters, in particular cis-3-hexenyl acetate and 2-phenylethyl acetate.

Figure 3. Principal component analysis of acetate esters, organic acids and glycerol for individual fermentations in Sauvignon blanc wine fermentations.

Save View full size Expand inline Collapse inline

Red represents loadings of compounds analysed and blue represents the PC scores of individual fermentations. Samples suffixed with E = EC1118 monoculture, S = SER1(G78R) monoculture, EM = EC1118/MP2 co-inoculation, EL = EC1118/Concerto co-inoculation, SM = SER1(G78R)/MP2 co-inoculation and SL – SER1(G78R)/Concerto co-inoculation. 1:1 and 1:9 denote the inoculation ratio used.

3. S. cerevisiae/Concerto co-inoculated fermentations in Chardonnay juice confirm alleviation of SO2 sensitivity

The influence of co-culture with Saccharomyces on the SO₂ sensitivity of Concerto was investigated by carrying out a repeat of the fermentations in another white variety, Chardonnay, containing 34.4 mg/L total SO₂. Plating on lysine media revealed the persistence of L. thermotolerans in co-inoculation (EC1118 or SER1(G78R), compared to a loss of L. thermotolerans viability in the monoculture after 24 h (Figure 4A). By the 72 h time-point in the Sauvignon blanc fermentation, Concerto had reached 5.7 x 10⁷ CFU/mL when co-inoculated with SER1(G78R), whereas no L. thermotolerans colonies were seen when co-inoculated with EC1118. These results imply that the SO₂ resistance effect is due to the presence of S. cerevisiae biomass. To test this, an equivalent experiment was undertaken using heat-inactivated EC1118 or SER1(G78R) cells. In this instance, however, all fermentations lost viability after 24 h (data not shown), indicating that the SO₂ resistance phenotype experienced by Concerto when co-inoculated with S. cerevisiae strains involves biologically functional yeast and not just biomass. Further work is required to elucidate the precise mechanism involved.

Figure 4: Viable cell densities of Concerto or S. cerevisiae after 72 h in Chardonnay juice.

Save View full size Expand inline Collapse inline

Fermentations (100 mL) were conducted in Chardonnay juice containing 34.4 mg/L total SO₂. Concerto was present as a (A) monoculture (○) or co-inoculated 1:1 with EC1118 (□) or SERI(G78R) (△) with a cell density determined as CFU/mL by lysine plating. S. cerevisiae population counts (B) were determined by YEPD plating, for EC1118 (○) and SER1(G78R) (□) monocultures, or Concerto cultures co-inoculated 1:1 with EC1118 (△) or SER1(G78R) (◊). In mixed culture fermentations, values from lysine plating were subtracted from total values determined by YEPD plating.

Discussion

In winemaking, mixed fermentation by S. cerevisiae and non-Saccharomyces yeast is a common method for introducting complexity. Typically, the yeasts are added sequentially, starting with non-Saccharomyces to alter the aroma profile (and reduce ethanol), followed a few days later by the addition of the ethanol-tolerant Saccharomyces to finish the fermentation. As Saccharomyces implants quickly and outcompetes the non-Saccharomyces, focus has been on maximising the non-Saccharomyces population and persistence during fermentation, in particular when yeasts are inoculated together rather than sequentially. One approach is to extend the Saccharomyces lag phase prior to exponential growth. Albertin et al. (2017) were the first to modulate the lag phase using sulfite through the breeding of S. cerevisiae strains with short and long lag phases associated with the sulfite pump (SSU1) translocations. The authors reported improved wine fruitiness and complexity in co-inoculated fermentations with non-Saccharomyces when using the long-lag phase strains or low inoculation rates of the short-lag phase strains.

‘Loss of function' mutations that result in protracted growth independent of sulfite provide an alternative paradigm. Deletion of the SER1 (3-phosphoserine aminotransferase) required for serine and glycine biosynthesis from glycolysis (Supplementary Figure 1) was first demonstrated to negatively impact the initial growth phase of a heme-deficient S. cerevisiase strain (Reiner et al., 2006). A naturally occurring ‘loss of function' variant, SER1(G78R), was later identified in Sake yeast in a QTL study of chronological aging (Jung et al., 2018); the SER1(G78R) mutant exhibited slowed growth and reduced biomass in synthetic media (yeast nitrogen base, YNB) containing either glucose (0.5, 2 or 10 %) or galactose (2 %) as the carbon source. In the present study transfer of the SER1(G78R) mutation to a wine genotype, namely EC1118, also resulted in a slow-growth phenotype and reduced biomass (total cell number) when grown in Chemically Defined Grape Juice Medium (Figure 1). The use of unsulfured CDGJM confirmed the growth phenotype of SER1(G78R) as SO₂-independent in contrast to the ‘SSU1 translocation' strains described by Albertin et al. (2017). The SER1(G78R) mutant was evaluated against EC1118 in co-inoculated fermentations with L. thermotolerans strain Concerto or M. pulcherrima strain MP2 in Sauvignon blanc juice. The results (Figure 2) supported the hypothesis that the use of a slower-growing SER1(G78R) mutant results in a larger and more persistent population of non-Saccharomyces yeast; for example, SER1(G78R) was able to support a large population of MP2 at both inoculation ratios before a decline in MP2 occurred in the later stages, and with SER1(G78R) finishing the fermentation. This pattern was similar to the sequential fermentation reported by Contreras et al. (2014), where the M. pulcherrima CFUs declined and a subsequent inoculation with an S. cerevisiae strain after 50 % sugar consumption allowed the completion of the fermentation. The decline most likely reflects the low ethanol tolerance of M. pulcherrima (3–5 %), with only a few tolerating up to 9 % (Barbosa et al., 2018).

The influence of the wild-type and SER1(G78R) mutant on the expression of non-Saccharomyces attributes related to wine composition was investigated as part of the Sauvignon blanc juice fermentations. L. thermotolerans was chosen because of its ability to produce lactic acid (Benito et al., 2015; Gibson et al., 2018; Jolly et al., 2014; Kapsopoulou et al., 2007; Morata et al., 2018), a trait whose exploitation has proven beneficial for the bioacidification of low acid Merlot wine to improve sensory properties (Hranilovic et al., 2021). M. pulcherrima was chosen for its ability to produce high concentrations of some acetate esters and lower alcohol content during wine fermentations (Hranilovic et al., 2020; Varela et al., 2016).

The relative abundance of EC1118 in the co-inoculated fermentations was reflected in the lack of significant variation in the metabolite profiles of the resultant wines (Tables 4 and 5). This scenario is typical of co-inoculated fermentations and is one reason why winemakers rely on a sequential inoculation regime (Vilela, 2020) to ensure expression of non-Saccharomyces attributes (Dutraive et al., 2019; Englezos et al., 2018; Hranilovic et al., 2020; Lin et al., 2020; Lu et al., 2017). On the other hand, SER1(G78R) co-inoculated fermentations resulted in largely greater numbers of non-Saccharomyces yeasts, which is reflected in the concentrations of key metabolites (Table 4) and volatile compounds analysed (Table 5).

High acetic acid and volatile acidity are often tackled through mixed fermentation involving L. thermotolerans (Benito et al., 2015; Benito, 2018; Comitini et al., 2011) or M. pulcherrima (Hranilovic et al., 2020; Varela et al., 2016). Similar findings were obtained for the fermentations co-inoculated with SER1(G78R), in which the high amounts of acetic acid produced by SER1(G78R) (Supplementary Figure 1) decreased as a result of the uptake and metabolism of hexose by low acetic acid-producing non-Saccharomyces yeasts (Table 4). As such, concentrations of acetic acid were close to the legal limits ( > 1.5 g/L in Australia; Coultier et al., 2004) in fermentations using SER1(G78R) in monoculture and with MP2 (1:1), but were significantly reduced due to the high inoculation rates of non-Saccharomyces (9:1). Further work is required to evaluate different strain combinations and inoculation regimes in conditions more related to winemaking; e.g., non-sterile conditions and on larger scales to determine the degree of implantation and effect on wine composition (Lin et al., 2022). The influence of parameters such as sulfite, pH and temperature on microbial interaction and succession will also need to be examined.

The other organic acids of interest in terms of wine acidity were lactic acid and malic acid. Malic acid in wine is generally known for its harsh acidic quality and is decarboxylated to lactic acid by lactic acid bacteria, such as Oenococcus oeni or Lactobacillus species, to improve sensorial properties and improve wine stability (Bartowsky, 2014). Some yeasts, such as Schizosaccharomyces pombe, are known to completely degrade malic acid as they possess malate permease and malic acid decarboxylase (Benito et al., 2016a). Malic acid degradation is also associated with L. thermotolerans strains (Gobbi et al., 2013; Hranilovic et al., 2018; Kapsopoulou et al., 2005; Whitener et al., 2017), as observed here, although only L. thermotolerans-dominated fermentations reduced malic acid by up to ~25 %. Interestingly, this was not reflected in an increase in lactic acid, a phenotype typically associated with L. thermotolernas yeast; however, Concerto is considered a poor producer of lactic acid, as in some cases it will not produce significantly more of it than some S. cerevisiae strains (Vaquero et al., 2020).

Glycerol and ethanol modulation using non-Saccharomyces yeast has been extensively researched with a view to producing ‘reduced alcohol' wines. Both species used in this study are able to increase glycerol (Contreras et al., 2014; Gobbi et al., 2013; Hranilovic et al., 2020; Kapsopoulou et al., 2007) and decrease ethanol content (Contreras et al., 2014; Hranilovic et al., 2018; Hranilovic et al., 2020; Hranilovic et al., 2021; Varela et al., 2016). Glycerol production was improved when MP2 was co-cultured with SER1(G78R). Meanwhile, the influence of Concerto was more complicated, producing less glycerol than the SER1(G78R) monoculture, although it was still significantly higher than in all EC1118 fermentations (Table 4). SER1(G78R) can still be regarded as a superior glycerol producer compared to EC1118, but yield was dependent upon its abundance in co-fermentations (Figure 2E). Increased glycerol production in S. cerevisiae yeast occurs with a concurrent increase in acetic acid (Eglinton et al., 2002; Remize et al., 1999; van Wyk et al., 2020), which is needed to balance the NAD⁺/NADH ratio, which tends towards NAD⁺ by glycerol biosynthesis. An increase in acetic acid and glycerol was observed in the SER1(G78R) monoculture, but not in the non-Saccharomyces co-inoculated cultures, indicating the addition of Concerto and MP2 as a way of addressing redox balance. Previously, Sadoudi et al. (2017) demonstrated an increase in glycerol and decrease in acetic acid in M. pulcherrima/S. cerevisiae sequential fermentations, with the presence of M. pulcherrima altering the expression of the genes involved in acetic acid biosynthesis in S. cerevisiae. This could explain the reduction of acetic acid in the co-inoculated fermentations in this study. Another plausible explanation is that MP2-derived acetic acid is reduced through the production of acetate esters, a phenomenon observed in studies involving M. pulcherrima (Binati et al., 2020; Contreras et al., 2014; Hranilovic et al., 2020; Sadoudi et al., 2012; Varela et al., 2016). van Wyk et al. (2020) demonstrated that the overexpression of alcohol acetyltransferase (ATF1) in S. cerevisiae strains that accumulate both glycerol and acetic acid due to NAD-dependent glycerol-3-phosphate dehydrogenase (GPD1) overexpression resulted in lower acetic acid and higher acetate ester concentrations. Whilst outside the scope of this paper, it would be interesting to assess ATF1 and GPD1 transcript levels in M. pulcherrima strains to determine if this is also the case.

Many authors have noted specific increases in acetate esters in mixed fermentations involving M. pulcherrima. In this study, five esters (ethyl acetate, isoamyl acetate, 2-phenylethyl acetate, isobutyl acetate and cis-3-hexenyl acetate) were assessed based on prior reports of their increase with M. pulcherrima inclusion in fermentation (Contreras et al., 2014; Hranilovic et al., 2020; Sadoudi et al., 2012; Varela et al., 2016). Propyl acetate was also analysed. In most instances, there were statistically significant increases in acetate esters with the higher MP2 inoculations correlating to higher acetate esters (Table 5).

Hranilovic et al. (2018) also demonstrated an increase in acetate esters by Concerto when sequentially inoculated with S. cerevisiae in Shiraz fermentations. Specifically, significant increases were observed in ethyl acetate, ethyl phenylacetate, isoamyl acetate and isobutyl acetate. In this study, whilst higher mean concentrations of propyl acetate and isobutyl acetate were observed, and ethyl acetate and isobutyl acetate significantly increased at the higher inoculation ratio, isoamyl acetate, cis-3-hexenyl acetate and 2-phenylethyl acetate were reduced (Table 5). Differences between studies may be related to wine type (red vs white) and variety (Shiraz vs Sauvignon blanc). Hranilovic et al. (2018) further reported on vintage-dependent variability in acetate esters, in particular isoamyl acetate: one juice co-inoculated with Concerto had significantly higher concentrations compared to S. cerevisiae, but the inverse was found in a subsequent vintage. Such trends highlight the importance of juice composition in winemaking outcome.

Arguably, the most interesting observation of the present study was the alleviation of SO₂ sensitivity associated with Concerto in sulfured Sauvignon blanc and Chardonnay juice. The results for the SER1(G78R) co-fermentations are therefore promising, as they allow the use of SO₂ sensitive non-Saccharomyces yeasts - which might offer other interesting oenological properties - in what would otherwise be lethal concentrations of SO₂. Similar findings were made for the EC1118 co-inoculated cultures when plating on selective lysine media (Figure 4A), even though the Concerto concentrations were reduced. As it is well known that acetaldehyde accumulation generally occurs during the yeast growth phase (Amerine and Ough, 1964; Weeks, 1969), the binding of SO₂ by acetaldehyde is a probable cause of this observation. The fact that this phenomenon was observed with both wild-type and mutant S. cerevisiae strains implies that other slow-growing yeast (e.g., Albertin et al., 2017) possess this ability. Further research would be of benefit in order to both understand this phenotype and develop novel strains for use in co-inoculation and SO₂ sequestration.

Conclusion

The transfer of a novel and naturally-occurring SER1 variant (SER1(G78R)) to the wine yeast EC1118 via CRISPR/Cas9 results in a slower-growth phenotype. This mutation may also have the added advantage over previously reported slow-growing yeasts of SO₂ not needing to elicit slow growth, thereby giving it broader applicability. This phenotype benefited novel mixed cultures in which the mutant was co-inoculated with non-Saccharomyces as it resulted in a greater expression of positive sensory effects imparted by the non-Saccharomyces. Furthermore, this strain can be used with non-Saccharomyces strains that have low sulfite tolerance, thereby permitting the standard wine-making practice of using SO₂ to prevent oxidation and microbial spoilage. Further evaluation is required of this strain in combination with, for example, other non-Saccharomyces and inoculation ratios to determine whether it would be suitable for use as a ‘mixed’ starter culture comprising multiple yeast species. The construction of this strain is an example of ‘self-cloning’ and, as such, is likely to be permitted for use in countries such as Japan and USA where regulations are less stringent than in Europe and Australia (Hanlon and Sewalt, 2020). To date, Australia’s ‘clean green image’ in winemaking and regulation of GMOs prevents the use of genetically modified yeast and bacteria, apart from variants isolated from the vineyard and point mutations generated by mutagenesis.

Acknowledgements

The authors would like to thank Nick van Holst for his technical assistance and Tom Ellis and William Shaw from the Imperial College London for providing plasmids pWS173 and pWS082. We also thank Emily Nicholson and Sue Maffei at CSIRO for their assistance with the GC-MS analysis of the wine volatile compounds.

This project was supported by funding from Wine Australia [UA1803_2.1] and The Australian Research Council Training Centre for Innovative Wine Production (www.ARCwinecentre.org.au; project number IC170100008), which is funded by the Australian Government with additional support from Wine Australia and industry partners. Wine Australia invests in and manages research, development and extension on behalf of Australia’s winegrowers and winemakers and the Australian Government. Tom Lang is a recipient of a University of Adelaide Faculty of Sciences Divisional Scholarship and Wine Australia Supplementary Scholarship (AGW Ph1603).

The University of Adelaide and CSIRO are members of the Wine Innovation Cluster (http://www/thewaite.org/waite-partners/wine-innovation-cluster/).

References

• Adams, A., Gottschling, D., Kaiser, C., & Stearns, T. (1998). Methods in yeast genetics: A Cold Spring Harbor laboratory course manual (1997 ed.). Cold Spring Harbor Laboratory Press.
• Albergaria, H., & Arneborg, N. (2016). Dominance of Saccharomyces cerevisiae in alcoholic fermentation processes: role of physiological fitness and microbial interactions. Applied Microbiology and Biotechnology, 100(5), 2035-2046. https://doi.org/10.1007/s00253-015-7255-0
• Albertin, W., Zimmer, A., Miot-Sertier, C., Bernard, M., Coulon, J., Moine, V., . . . Masneuf-Pomarede, I. (2017). Combined effect of the Saccharomyces cerevisiae lag phase and the non-Saccharomyces consortium to enhance wine fruitiness and complexity. Applied Microbiology and Biotechnology, 101(20), 7603-7620. https://doi.org/10.1007/s00253-017-8492-1
• Amerine, M. A., & Ough, C. S. (1964). Studies with controlled fermentation. vIII. Factors affecting aldehyde accumulation. American Journal of Enology and Viticulture, 15(1), 23. The Australian Wine Research Institute. (2019). AWRI Vineyard & Winery Practices Survey. https://www.awri.com.au/wp-content/uploads/2019/05/AWRI_Practices_Survey_Final_Report.pdf
• Barbosa, C., Lage, P., Esteves, M., Chambel, L., Mendes-Faia, A., & Mendes-Ferreira, A. (2018). Molecular and phenotypic characterization of Metschnikowia pulcherrima strains from Douro wine region. Fermentation, 4(1). https://doi.org/10.3390/fermentation4010008
• Bartowsky, E. J. (2014). WINES | Malolactic Fermentation. In Encyclopedia of Food Microbiology (2nd ed., pp. 800-804). Oxford: Academic Press.
• Belda, I., Ruiz, J., Esteban-Fernández, A., Navascués, E., Marquina, D., Santos, A., & Moreno-Arribas, M. V. (2017). Microbial contribution to wine aroma and its intended use for wine quality improvement. Molecules, 22(2). https://doi.org/10.3390/molecules22020189
• Benito, Á., Calderón, F., & Benito, S. (2016a). New trends in Schizosaccharomyces use for winemaking. In Grape and Wine Biotechnology (1st ed., pp. 307-323): IntechOpen
• Benito, Á., Calderón, F., Palomero, F., & Benito, S. (2016b) Quality and composition of Airén wines fermented by sequential inoculation of Lachancea thermotolerans and Saccharomyces cerevisiae. Food Technology and Biotechnology, 54(2), 135-144. https://doi.org/10.17113/ftb.54.02.16.4220
• Benito, Á., Calderón, F., Palomero, F., & Benito, S. (2015). Combine use of selected Schizosaccharomyces pombe and Lachancea thermotolerans yeast strains as an alternative to the traditional malolactic fermentation in red wine production. Molecules, 20(6), 9510-9523. https://doi.org/10.3390/molecules20069510
• Benito, S. (2018). The impacts of Lachancea thermotolerans yeast strains on winemaking. Applied Microbiology and Biotechnology, 102(16), 6775-6790. https://doi.org/10.1007/s00253-018-9117-z
• Binati, R. L., Lemos Junior, W. J. F., Luzzini, G., Slaghenaufi, D., Ugliano, M., & Torriani, S. (2020). Contribution of non-Saccharomyces yeasts to wine volatile and sensory diversity: A study on Lachancea thermotolerans, Metschnikowia spp. and Starmerella bacillaris strains isolated in Italy. International Journal of Food Microbiology, 318, 108470. https://doi.org/10.1016/j.ijfoodmicro.2019.108470
• Chidi, B. S., Rossouw, D., & Bauer, F. F. (2016). Identifying and assessing the impact of wine acid-related genes in yeast. Current Genetics, 62(1), 149-164. https://doi.org/10.1007/s00294-015-0498-6
• Comitini, F., Gobbi, M., Domizio, P., Romani, C., Lencioni, L., Mannazzu, I., & Ciani, M. (2011). Selected non-Saccharomyces wine yeasts in controlled multistarter fermentations with Saccharomyces cerevisiae. Food Microbiology, 28(5), 873-882. https://doi.org/10.1016/j.fm.2010.12.001
• Contreras, A., Hidalgo, C., Henschke, P. A., Chambers, P. J., Curtin, C., & Varela, C. (2014). Evaluation of non-Saccharomyces yeasts for the reduction of alcohol content in wine. Applied and Environmental Microbiology, 80(5), 1670. https://doi.org/10.1128/AEM.03780-13
• Corison, C. A., Ough, C. S., Berg, H. W., & Nelson, K. E. (1979). Must acetic acid and ethyl acetate as mold and rot indicators in grapes. American Journal of Enology and Viticulture, 30(2), 130.
• Coulter, A., Godden, P. W., & Pretorius, I. (2004). Succinic acid-how is it formed, what is its effect on titratable acidity, and what factors influence its concentration in wine? Australian and New Zealand Wine Industry Journal, 19, 16-25.
• Du Plessis, H., Du Toit, M., Nieuwoudt, H., Van der Rijst, M., Kidd, M., & Jolly, N. (2017). Effect of Saccharomyces, non-Saccharomyces yeasts and malolactic fermentation strategies on fermentation kinetics and flavor of Shiraz wines. Fermentation, 3(4). https://doi.org/10.3390/fermentation3040064
• Dutraive, O., Benito, S., Fritsch, S., Beisert, B., Patz, C.-D., & Rauhut, D. (2019). Effect of sequential inoculation with non-Saccharomyces and Saccharomyces yeasts on Riesling wine chemical composition. Fermentation, 5(3), 79. https://doi.org/10.3390/fermentation5030079
• Eglinton, J. M., Heinrich, A. J., Pollnitz, A. P., Langridge, P., Henschke, P. A., & de Barros Lopes, M. (2002). Decreasing acetic acid accumulation by a glycerol overproducing strain of Saccharomyces cerevisiae by deleting the ALD6 aldehyde dehydrogenase gene. Yeast, 19(4), 295-301. https://doi.org/10.1002/yea.834
• Engler, C., Kandzia, R., & Marillonnet, S. (2008). A one pot, one step, precision cloning method with high throughput capability. PLOS One, 3(11), e3647. https://doi.org/10.1371/journal.pone.0003647
• Englezos, V., Rantsiou, K., Cravero, F., Torchio, F., Pollon, M., Fracassetti, D., . . . Cocolin, L. (2018). Volatile profile of white wines fermented with sequential inoculation of Starmerella bacillaris and Saccharomyces cerevisiae. Food Chemistry, 257, 350-360. https://doi.org/10.1016/j.foodchem.2018.03.018
• Fowell, R. R. (1965). The identification of wild yeast colonies on lysine agar. Journal of Applied Bacteriology, 28(3), 373-383. https://doi.org/10.1111/j.1365-2672.1965.tb02167.x
• Gardner, J. M., McBryde, C., Vystavelova, A., Lopes, M. D. B., & Jiranek, V. (2005). Identification of genes affecting glucose catabolism in nitrogen-limited fermentation. FEMS Yeast Research, 5(9), 791-800. https://doi.org/10.1016/j.femsyr.2005.02.008
• Gibson, B., Vidgren, V., Peddinti, G., & Krogerus, K. (2018). Diacetyl control during brewery fermentation via adaptive laboratory engineering of the lager yeast Saccharomyces pastorianus. Journal of Industrial Microbiology and Biotechnology, 45(12), 1103-1112. https://doi.org/10.1007/s10295-018-2087-4
• Gobbi, M., Comitini, F., Domizio, P., Romani, C., Lencioni, L., Mannazzu, I., & Ciani, M. (2013). Lachancea thermotolerans and Saccharomyces cerevisiae in simultaneous and sequential co-fermentation: A strategy to enhance acidity and improve the overall quality of wine. Food Microbiology, 33(2), 271-281. https://doi.org/10.1016/j.fm.2012.10.004
• Guth, H. (1997). Quantitation and sensory studies of character impact odorants of different white wine varieties. Journal of Agricultural and Food Chemistry, 45(8), 3027-3032. https://doi.org/10.1021/jf970280a
• Haggerty, J. (2016). Characterisation of the wine meta-metabolome: linking aroma profiles to yeast genotype. PhD Thesis, The University of Adelaide
• Hanlon, P., & Sewalt, V. (2020). GEMs: genetically engineered microorganisms and the regulatory oversight of their uses in modern food production. Critical Reviews in Food Science and Nutrition, 1-12. https://doi.org/10.1080/10408398.2020.1749026
• Henick-Kling, T., Edinger, W., Daniel, P., & Monk, P. (1998). Selective effects of sulfur dioxide and yeast starter culture addition on indigenous yeast populations and sensory characteristics of wine. Journal of Applied Microbiology, 84(5), 865-876. https://doi.org/10.1046/j.1365-2672.1998.00423.x
• Hranilovic, A., Albertin, W., Capone, D. L., Gallo, A., Grbin, P. R., Danner, L., . . . Jiranek, V. (2021). Impact of Lachancea thermotolerans on chemical composition and sensory profiles of Merlot wines. Food Chemistry, 349, 129015. https://doi.org/10.1016/j.foodchem.2021.129015
• Hranilovic, A., Gambetta, J. M., Jeffery, D. W., Grbin, P. R., & Jiranek, V. (2020). Lower-alcohol wines produced by Metschnikowia pulcherrima and Saccharomyces cerevisiae co-fermentations: The effect of sequential inoculation timing. International Journal of Food Microbiology, 329, 108651. https://doi.org/10.1016/j.ijfoodmicro.2020.108651
• Hranilovic, A., Li, S., Boss, P. K., Bindon, K., Ristic, R., Grbin, P. R., . . . Jiranek, V. (2018). Chemical and sensory profiling of Shiraz wines co-fermented with commercial non-Saccharomyces inocula. Australian Journal of Grape and Wine Research, 24(2), 166-180. https://doi.org/10.1111/ajgw.12320
• Jiranek, V., Langridge, P., & Henschke, P. A. (1995). Amino-acid and ammonium utilization by Saccharomyces cerevisiae wine yeasts from a chemically-defined medium. American Journal of Enology and Viticulture, 46(1), 75-83.
• Jolly, N. P., Varela, C., & Pretorius, I. S. (2014). Not your ordinary yeast: non-Saccharomyces yeasts in wine production uncovered. FEMS Yeast Research, 14(2), 215-237. https://doi.org/10.1111/1567-1364.12111
• Jung, P. P., Zhang, Z., Paczia, N., Jaeger, C., Ignac, T., May, P., & Linster, C. L. (2018). Natural variation of chronological aging in the Saccharomyces cerevisiae species reveals diet-dependent mechanisms of life span control. NPJ Aging and Mechanisms of Disease, 4, 3. https://doi.org/10.1038/s41514-018-0022-6
• Kapsopoulou, K., Kapaklis, A., & Spyropoulos, H. (2005). Growth and fermentation characteristics of a strain of the wine yeast Kluyveromyces thermotolerans isolated in Greece. World Journal of Microbiology and Biotechnology, 21(8), 1599-1602. https://doi.org/10.1007/s11274-005-8220-3
• Kapsopoulou, K., Mourtzini, A., Anthoulas, M., & Nerantzis, E. (2007). Biological acidification during grape must fermentation using mixed cultures of Kluyveromyces thermotolerans and Saccharomyces cerevisiae. World Journal of Microbiology and Biotechnology, 23(5), 735-739. https://doi.org/10.1007/s11274-006-9283-5
• Lang, T. A. Walker, M. E., & Jiranek, V. (2021) Disruption of ECM33 in diploid wine yeast EC1118: cell morphology and aggregation and their influence on fermentation performance, FEMS Yeast Research, 21(5), foab044. https://doi.org/10.1093/femsyr/foab044
• Lin, M. M.-H., Boss, P. K., Walker, M. E., Sumby, K. M., Grbin, P. R., & Jiranek, V. (2020). Evaluation of indigenous non-Saccharomyces yeasts isolated from a South Australian vineyard for their potential as wine starter cultures. International Journal of Food Microbiology, 312, 108373. https://doi.org/10.1016/j.ijfoodmicro.2019.108373
• Lin, M.M.H., Boss, P.K., Walker, M.E., Sumby, K.M., & Jiranek, V. (2022). Influence of Kazachstania spp. on the chemical and sensory profile of red wines. International Journal of Food Microbiology, 362, p.109496. https://doi.org/10.1016/j.ijfoodmicro.2021.109496
• Lu, Y., Peh, J. C. H., Lee, P.-R., & Liu, S.-Q. (2017). Modulation of grape wine flavor via the sequential inoculation of Williopsis saturnus and Saccharomyces cerevisiae. Food Biotechnology, 31(4), 245-263. https://doi.org/10.1080/08905436.2017.1369434
• Marcon, A. R., Schwarz, L. V., Dutra, S. V., Agostini, F, Delamare, A. P. L. & Echeverrigaray, S. (2018). Contribution of a Brazilian Torulaspora delbrueckii isolate and a commercial Saccharomyces cerevisiae to the aroma profile and sensory characteristics of Moscato Branco wines. Australian Journal of Grape and Wine Research, 24(4), 461-468. https://doi.org/10.1111/ajgw.12347
• Melcher, K., & Entian, K. D. (1992). Genetic analysis of serine biosynthesis and glucose repression in yeast. Current Genetics, 21(4-5), 295-300. https://doi.org/10.1007/bf00351686
• Miller, G. (2019). Whisky Science: A Condensed Distillation. (1st ed.): Springer.
• Morata, A., Loira, I., Tesfaye, W., Bañuelos, M. A., González, C., & Suárez Lepe, J. A. (2018). Lachancea thermotolerans applications in wine technology. Fermentation, 4(3). https://doi.org/10.3390/fermentation4030053
• Noble, A. C., & Bursick, G. F. (1984). The contribution of glycerol to perceived viscosity and sweetness in white wine. American Journal of Enology and Viticulture, 35(2), 110. Retrieved from http://www.ajevonline.org/content/35/2/110.abstract
• Padilla, B., Gil, J. V., & Manzanares, P. (2016). Past and future of non-Saccharomyces yeasts: From spoilage microorganisms to biotechnological tools for improving wine aroma complexity. Frontiers in Microbiology, 7, 411-411. https://doi.org/10.3389/fmicb.2016.00411
• Redzepovic, S., Orlic, S., Majdak, A., Kozina, B., Volschenk, H., & Viljoen-Bloom, M. (2003). Differential malic acid degradation by selected strains of Saccharomyces during alcoholic fermentation. International Journal of Food Microbiology, 83(1), 49-61. https://doi.org/10.1016/s0168-1605(02)00320-3
• Reiner, S., Micolod, D., Zellnig, G., & Schneiter, R. (2006). A genomewide screen reveals a role of mitochondria in anaerobic uptake of sterols in yeast. Molecular Biology of the Cell, 17(1), 90-103. https://doi.org/10.1091/mbc.e05-06-0515
• Remize, F., Roustan, J. L., Sablayrolles, J. M., Barre, P., & Dequin, S. (1999). Glycerol overproduction by engineered Saccharomyces cerevisiae wine yeast strains leads to substantial changes in by-product formation and to a stimulation of fermentation rate in stationary phase. Applied and Environmental Microbiology, 65(1), 143. https://doi.org/10.1128/AEM.65.1.143-149.1999
• Sadoudi, M., Rousseaux, S., David, V., Alexandre, H., & Tourdot-Maréchal, R. (2017). Metschnikowia pulcherrima influences the expression of genes involved in PDH bypass and glyceropyruvic fermentation in Saccharomyces cerevisiae. Frontiers in Microbiology, 8(1137). https://doi.org/10.3389/fmicb.2017.01137
• Sadoudi, M., Tourdot-Marechal, R., Rousseaux, S., Steyer, D., Gallardo-Chacon, J. J., Ballester, J., . . . 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(2), 243-253. https://doi.org/10.1016/j.fm.2012.06.006
• Shaw, W. M., Yamauchi, H., Mead, J., Gowers, G. F., Bell, D. J., Oling, D., . . . Ellis, T. (2019). Engineering a model cell for rational tuning of GPCR signaling. Cell, 177(3), 782-796 e727. https://doi.org/10.1016/j.cell.2019.02.023
• Sumby, K. M., Grbin, P. R., & Jiranek, V. (2010). Microbial modulation of aromatic esters in wine: Current knowledge and future prospects. Food Chemistry, 121(1), 1-16. https://doi.org/10.1016/j.foodchem.2009.12.004
• van Wyk, N., Kroukamp, H., Espinosa, M. I., von Wallbrunn, C., Wendland, J., & Pretorius, I. S. (2020). Blending wine yeast phenotypes with the aid of CRISPR DNA editing technologies. International Journal of Food Microbiology, 324, 108615. https://doi.org/10.1016/j.ijfoodmicro.2020.108615
• Vaquero, C., Loira, I., Bañuelos, M. A., Heras, J. M., Cuerda, R., & Morata, A. (2020). Industrial performance of several Lachancea thermotolerans strains for pH Control in white wines from warm areas. Microorganisms, 8(6), 830. https://doi.org/10.3390/microorganisms8060830
• Varela, C., Barker, A., Tran, T., Borneman, A., & Curtin, C. (2017). Sensory profile and volatile aroma composition of reduced alcohol Merlot wines fermented with Metschnikowia pulcherrima and Saccharomyces uvarum. International Journal of Food Microbiology, 252, 1-9. https://doi.org/10.1016/j.ijfoodmicro.2017.04.002
• Varela, C., Sengler, F., Solomon, M., & Curtin, C. (2016). Volatile flavour profile of reduced alcohol wines fermented with the non-conventional yeast species Metschnikowia pulcherrima and Saccharomyces uvarum. Food Chemistry, 209, 57-64. https://doi.org/10.1016/j.foodchem.2016.04.024
• Vilela, A. (2020). Modulating wine pleasantness throughout wine-yeast co-inoculation or sequential inoculation. Fermentation, 6(1). https://doi.org/10.3390/fermentation6010022
• Weeks, C. (1969). Production of sulfur dioxide-binding compounds and of sulfur dioxide by two Saccharomyces yeasts. American Journal of Enology and Viticulture, 20(1), 32.
• Whitener, M. E. B., Stanstrup, J., Carlin, S., Divol, B., Du Toit, M., & Vrhovsek, U. (2017). Effect of non-Saccharomyces yeasts on the volatile chemical profile of Shiraz wine. Australian Journal of Grape and Wine Research, 23(2), 179-192. https://doi.org/10.1111/ajgw.12269
• Zohre, D. E., & Erten, H. (2002). The influence of Kloeckera apiculata and Candida pulcherrima yeasts on wine fermentation. Process Biochemistry, 38(3), 319-324. https://doi.org/10.1016/S0032-9592(02)00086-9