Gene expression of specific enological traits in wine fermentation

Claudio Martínez*^1,2 · Verónica García^1,2 · David González¹ · Matías Jara¹ · Matías Aguilera¹ · Maria Angélica Ganga¹

¹Universidad de Santiago de Chile, Departamento de Ciencia y Tecnología de los Alimentos, Santiago, Chile
²Universidad de Santiago de Chile, Centro de Estudios en Ciencia y Tecnología de Alimentos, Santiago, Chile

*Corresponding author: claudio.martinez@usach.cl

Financial support: This research was supported by Grant 1100509 (Fondecyt). V. García was supported by a Fondecyt 3120107 Postdoctoral Fellowship.

Keywords: enological traits, gene expression, Saccharomyces cerevisiae, wine fermentation, wine yeast.

Background: Wine fermentation is a dynamic process and yeast has a precise genic regulation system that allow them to coordinate the gene transcription levels. However, the fermentation stage at which there is a great variation in the genic expression associated to a particular enological trait is often unknown and there are no simple experimental approaches to define it.

Results: To identify the most adequate stage in which to evaluate the expression of the genes associated to specific enological traits we identified three stages of fermentation using the industrial strain EC1118. These stages, called early (0-16.4 hrs), middle (16.4-88.7 hrs) and late (88.7 hrs onwards), were characterized according to changes in the speed of CO₂ production. The greatest nitrogen consumption velocity (1.016 g/L h) was achieved in the early stage. The most significant changes in the consumption of sugar and ethanol production acceleration occurred in the middle stage, and the greatest velocity of ethanol production (0.043%/h) and fructose consumption (0.338 g/L h) occurred in the late stage. The expression profiles for a set of genes characteristic of these processes were clearly defined during the stages of fermentation: identifying a peak for the genes related to the consumption of nitrogen during the early stage, followed by an increase in the expression of genes related to the consumption of sugars in the middle stage. Finally, during the late stage there is a decrease in the expression of genes involved in the consumption of sugar, except for HXT3 the levels of which remain high.

Conclusions: The kinetic and transcriptional profiles of the enological traits under study are clearly distinct and each is associated to a particular stage of the fermentation. Our results confirm that CO₂ production is a simple parameter to estimate the stages of the fermentation. Therefore, using the kinetics of CO₂ loss it is possible to select the most adequate moment to study the expression of genes associated to the main enological traits: sugar consumption and production of nitrogen and ethanol.

The study of genic expression profiles associated to traits of enological relevance is currently of central importance in wine making, and several works have analyzed these phenotypes using a high throughput approach. Thus, Rossignol et al. (2003) evaluated the changes in the transcription levels during fermentation and found that more than 2000 genes vary their expression in response to changes in the culture medium. Other transcriptional studies in wine yeast strains have been carried out to evaluate the effect of nitrogen availability (Marks et al. 2003; Mendes-Ferreira et al. 2004; Jimenez-Marti and Del Olmo, 2008), response to rehydration (Rossignol et al. 2003), biomass production (Gómez-Pastor et al. 2010) and ethanol tolerance (Stanley et al. 2010), amongst others. All these studies have focused on obtaining a global knowledge of the yeast transcriptome during fermentation; however, the stage at which there is a greater variation in the genic expression associated to a specific phenotype is often unknown particularly in industrial yeasts. Furthermore, in these studies the enological traits are evaluated at the end of the fermentation, however, the greatest phenotypic variation and the expression of genes associated to these occur in the early stages and therefore to consider the moment of greatest difference in the genic expression is highly relevant for its evaluation. Hence, it is necessary to develop an experimental approach that allows a definition of the fermentation stages and their relationship to the transcriptional and phenotypic changes.

In this work, a simple method based on the monitoring of CO₂ loss was implemented to determine the most adequate stage to study genic transcription in wine fermentation. Within this context, we analyzed the gene expression profiles and the phenotypes related to the production of ethanol and the consumption of sugar and nitrogen in the commercial strain EC1118 under synthetic wine fermentation conditions. The genes analyzed for each enological trait of interest were selected because on a global scale their genic products represent the main metabolic processes associated to these phenotypes.

Strains and fermentation conditions

The wine yeast strain EC1118 was used in this study (Lalvin, Zug, Switzerland). The fermentations were carried out in the synthetic must described by Rossignol et al. (2003) containing 125 g/L glucose, 125 g/L fructose and 300 mg/L of assimilable nitrogen. The fermentations were carried out in 50 mL Falcon tubes containing 40 mL of synthetic must at 25ºC for 15 days without shaking. The inoculum consisted of 1 x 10⁶ cells/mL obtained from a 16 hrs culture in the same must with shaking. CO₂ production was monitored by weighing the tubes to determine weight loss.

Fermentation stages

The points where each fermentation stage ends were defined as the moments where a change in the acceleration of CO₂ loss occurred. These points were obtained from the second derivative graph of CO₂ loss in time.

Enological parameters

The residual sugar and ethanol production were determined according to Nissen et al. (1997). The consumption of nitrogen as yeast assimilable nitrogen (YAN) was determined using the formaldehyde titration method described by Aerny (1996). The kinetic parameters of CO₂ release and ethanol production were adjusted to a sigmoid non-linear fit. The glucose and fructose consumption measurements and YAN uptake were adjusted to a four-parameter logistic equation and third order polynomial non-linear fit respectively. The adjustments were done using GraphPad Prism version 5.00 (San Diego California USA).

RNA preparation and QPCR

RNA extraction, reverse transcription and QPCR experiments were done according to Contreras et al. (2012). Results were analyzed using the LightCycler 4.0 software (Roche, Germany) and quantification of relative gene expression was done using the mathematical method described by Livak and Schmittgen (2001) and normalized with the ACT1 and SCR1 genes following Vandesompele et al. (2002). The relative expression trend of the genes associated to each phenotype at 6 hrs of fermentation was obtained by averaging the relative genic expression means for each point. The error associated to each measurement was also considered. This trend was graphed with the software Graph Pad Prism using the LOWESS function (Chambers et al. 1983). 

Stages of alcoholic fermentation

The production of CO₂ is easy to evaluate and is a widely used criterion for the study of fermentations in much the same way as the production of biomass, but a lot less complex. Although the production of biomass follows a similar sigmoidal pattern to the formation of CO₂ during alcoholic fermentation, they do not occur simultaneously (Rossignol et al. 2003; Varela et al. 2004). Nevertheless, the loss of CO₂ is an indicator of the fermentative processes making it more adequate to estimate the degree of metabolic activity of the cell. To establish the stages of the fermentation, a non-linear adjustment was carried out using the kinetics of CO₂ loss. The changes in acceleration were calculated from the second derivative of this function (Figure 1a) thus defining the limits of each stage of fermentation. Hence, in the early stage between 0 and 16.4 hrs, the velocity of CO₂ loss was 0.17 g/L h, whilst maximum velocity was achieved in the middle stage of fermentation, between 16.4 and 88.7 hrs with an average of 0.86 g/L h. Finally, in the late stage from 88.7 hrs onwards this parameter decreased again to 0.2 g/L h. This result shows that the early stage is characterized by an almost null CO₂ release, suggesting that it could reflect the onset of fermentation driven by the lag phase of cell growth and the progressive saturation of the medium with CO₂ as described previously (El Haloui et al. 1989). The middle stage is characterized by the maximum velocity and maximum acceleration of CO₂ release, a process that has been correlated to exponential cell growth (El Haloui et al. 1989). Finally the late stage is characterized by  a decrease in CO₂ release, reflecting the inhibition of the fermentative process (El Haloui et al. 1989). These criteria may be used in any alcoholic fermentation given that all the kinetics of CO₂ loss are similar (Rossignol et al. 2003, Marullo et al. 2006).

Enological parameters

During the fermentation, the enological parameters, glucose, fructose and YAN consumption, as well as the production of ethanol were measured (Figure 1). Non-linear adjustments to the kinetics of each parameter were carried out to determine the velocity of change of each of these phenotypes in the fermentation stages previously defined. The maximum average velocity of nitrogen consumption was obtained in the early stage of fermentation (1.016 g/L h). On the other hand, the lowest velocity of glucose and fructose consumption was observed during this stage (0.09 and 0.085 g/L h, respectively). This is reflected in a low velocity of ethanol production that reached a value of 0.007%/h.

The average consumption velocity of both sugars increased during the middle stage of the fermentation, however, the average velocity of fructose consumption was lower than that of glucose (0.197 and 0.290 g/L h, respectively). The same tendency was observed with the average velocity of ethanol production which increased to 0.038%/h. On the other hand, the average velocity of YAN consumption decreased to 0.871 g/L h. These results suggest that this was the most active stage of the fermentation, where the greatest number of changes occurred in the traits analyzed.

In the late stage of the fermentation, the average velocity of nitrogen source consumption decreased (an average of 0.160 g/L h). In addition, the highest velocity of glucose consumption was observed in this stage, reaching an average of 0.329 g/L h, however, there was no mayor difference with respect to the velocity observed during middle stage. On the other hand, the velocity of fructose consumption varied significantly between these two stages, reaching a maximum only during the late stage (0.338 g/L h), which explains why the majority of the residual sugar detected corresponds to this hexose. Finally, the highest production velocity of ethanol was obtained in this stage reaching 0.043%/h.

Gene expression during alcoholic fermentation

To evaluate the gene expression involved in the consumption of nitrogen we analyzed the genes GAP1, MEP2, TAT1 and GNP1 that correspond to the amino acids, ammonium, tryptophan and glutamine permeases, respectively (Figure 2b). These genes are involved in the transport of ammonium, glutamine and tryptophan that represent around 80% of the YAN consumed during alcoholic fermentation (Chiva et al. 2009). Furthermore, we analyzed the gene URE2 that encodes for the main negative regulator of the nitrogen catabolic repression mechanism. The maximum expression of genes related to this phenotype was observed at 6 hrs, the first point analyzed in the early stage of the fermentation. Subsequently, the transcript levels decreased and only halfway through the middle stage did the expression values equal those observed at 6 hrs post inoculation. The increase in expression of these genes in the middle stage would correspond to changes in expression of GAP1 and GNP1. The assimilation of nitrogen is regulated by the NCR mechanism (Magasanik, 2003), which allows the yeast to select the best nitrogen source in the culture medium. In the presence of preferred nitrogen sources, the transcription of genes involved in the consumption and assimilation of less preferred nitrogen sources is suppressed (Magasanik, 1992; Ter Schure et al. 2000). Of the genes analyzed, only URE2 increased its transcript levels during the middle stage of the fermentation. This gene encodes for the NCR effector system which keeps the transcriptional factor Gln3p in the cytoplasm which in turn activates the expression of genes related to the assimilation of less preferred nitrogen sources (Mitchell and Magasanik, 1984; Magasanik, 1992; Blinder et al. 1996). The increase in its transcript levels indicates that the yeast is in a state of catabolic repression by nitrogen, which is confirmed by the high levels of YAN in the medium during this stage and towards the end of the fermentation (Figure 1b). This suggests that measuring the velocity of nitrogen consumption and the expression of the genes involved must be carried out during the early stage of fermentation, since it is here where the most significant changes occur.

On the other hand, to analyze the expression of genes related to the consumption of carbon sources we selected the genes HXK1, HXK2, HXT1 and HXT3 that encode for hexokinases 1 and 2 and the low affinity hexose transporters 1 and 3, respectively. Furthermore, the transcriptional factors CAT8 and HAP4 and the glycolytic genes CDC19 and PGK1 which encode for pyruvate kinase and 3-phosphoglycerate kinase respectively were also analyzed. These genes showed a progressive increase in the levels of transcripts during the early stage and a maximum expression during the middle stage (Figure 2c), which correlates with the average consumption velocities of glucose and fructose. Finally, during the late stage of fermentation the levels of these gene transcripts decreased continuously.

The highest velocity of CO₂ loss occurred during the middle stage of the fermentation which is consistent with an increase in biomass during this stage. The expression of the genes related to the consumption of sugars also increased and they all showed a similar expression profile (Figure 2c). This phenomenon precedes the point of maximum velocity in the consumption of glucose (130 hrs) and fructose (161 hrs) that occurred during the late stage of the fermentation. This may be explained by the time lag that exists between the expression of a gene and the effects of its genic product (Belle et al. 2006; Siwiak and Zielenkiewicz, 2010). However, genes HXT1 and HXT3 that encode for low affinity permeases (Reifenberger et al. 1997), showed different expression profiles, where gene HXT1 was expressed in the first stage of the fermentation reaching a peak at 48 hrs. This is supported by Petit et al. (2000) who showed a positive regulation of this gene in response to the high concentrations of glucose and osmotic pressure present in the must initially. Subsequently, the expression of gene HXT1 decreases rapidly and coincides with the increase in the velocity of glucose consumption. On the other hand, gene HXT3 also showed a peak at 48 hrs of fermentation but its transcript levels remained high during the late stage, which could explain the consumption of fructose during the final stages of the fermentation as previously described (Pérez et al. 2005; Guillaume et al. 2007). These results show that the middle stage would be the most adequate to measure the expression of genes related to sugar consumption and evaluate changes in this enological trait.

To study the expression of genes related to ethanol production, we analyzed the genes PDC1, PDC6 and ADH1 that encode for the enzymes pyruvate decarboxylase 1 and 6 and alcohol dehydrogenase, respectively. These represent the most important isoforms of the enzymes involved in ethanol production (Figure 2a). During the early stage of the fermentation, the expression of the genes ADH1 and PDC1 decreased significantly with respect to the 6 hrs post-inoculation, a tendency which reverts during the middle stage where the maximum levels of transcripts were obtained for both genes. Finally, during the late stage of fermentation a sustained decrease in the transcript levels of these genes was observed. A different genic expression profile was observed for PDC6 since its transcript levels increased continuously during the early stage of fermentation, reaching their maximum value halfway during the middle stage and maintaining these until the end. Observations have shown that  under laboratory conditions the expression of gene ADH1 varies in response to the concentration of glucose and ethanol, remaining high during the first stages of growth (De Smidt et al. 2012). However, under our assay conditions the transcript levels of ADH1 were low in the initial stage of the fermentation. During the middle stage, when the velocity of glucose consumption increased and ethanol production began ADH1 reached its maximum level of expression (between 12-72 hrs, Figure 2a). In agreement with this observation, the velocity of ethanol production during the middle stage increased approximately 6 times in comparison to that in the early stage.

A similar expression profile to ADH1 was observed for the gene PDC1. This suggests a coordinated regulation of the two main genes involved in ethanol production and support the notion that halfway during the middle stage is the most adequate moment to carry out genic transcription studies associated to this phenotype.

Our results suggest that each enological phenotype and the associated gene expression must be evaluated in a particular stage of the fermentation, and these stages could be easily defined by the loss of CO₂.

AERNY, J. (1996). Composés azotés des moûts et desvins. Revue Suisse de Viticulture, Arboriculture, Horticulture, vol. 28, no. 3, p. 161-165.

BELLE, A.; TANAY, A.; BITINCKA, L.; SHAMIR, R. and O’SHEA, E.K. (2006). Quantification of protein half-lives in the budding yeast proteome. Proceedings of the National Academy of Sciences of the United States of America, vol. 103, no. 35, p. 13004-13009. [CrossRef]

BIDDICK, R.K.; LAW, G.L. and YOUNG, E.T. (2008). Adr1 and Cat8 mediate coactivator recruitment and chromatin remodeling at glucose-regulated genes. PLoS One, vol. 3, no. 1, p. e1436. [CrossRef]

BLINDER, D.; COSCHIGANO, P.W. and MAGASANIK, B. (1996). Interaction of the GATA factor Gln3p with the nitrogen regulator Ure2p in Saccharomyces cerevisiae. Journal of Bacteriology, vol. 178, no. 15, p. 4734-4736.

CONTRERAS, A.; GARCÍA, V.; SALINAS, F.; URZÚA, U.; GANGA, M.A. and MARTÍNEZ, C. (2012). Identification of genes related to nitrogen uptake in wine strains of Saccharomyces cerevisiae. World Journal of Microbiology and Biotechnology, vol. 28, no. 3, p. 1107-1113. [CrossRef]

CHAMBERS, J.M.; CLEVELAND, W.S.; KLEINER, B. and TUKEY, P.A. (1983). Graphical Methods for Data Analysis. Wadsworth International Group. 395 p. ISBN 0-534-98052-x.

CHIVA, R.; BAIGES, I.; MAS, A. and GUILLAMON, J.M. (2009). The role of GAP1 gene in the nitrogen metabolism of Saccharomyces cerevisiae during wine fermentation. Journal of Applied Microbiology, vol. 107, no. 1, p. 235-244. [CrossRef]

DE SMIDT, O.; DU PREEZ, J.C. and ALBERTYN, J. (2012). Molecular and physiological aspects of alcohol dehydrogenases in the ethanol metabolism of Saccharomyces cerevisiae. FEMS Yeast Research, vol. 12, no. 1, p. 33-47. [CrossRef]

DUEÑAS-SÁNCHEZ, R.; CODÓN, A.C.; RINCÓN, A.M. and BENÍTEZ, T. (2010). Increased biomass production of industrial bakers’ yeasts by overexpression of Hap4 gene. International Journal of Food Microbiology, vol. 143, no. 3, p. 150-160. [CrossRef]

EL HALOUI, N.E; CORRIEU, G.; CLERAN, I. and CHERUY, A. (1989). Method for on-line prediction of kinetics of alcoholic fermentation in wine making. Journal of Fermentation and Bioengineering, vol. 68, no. 2, p. 131-135. [CrossRef]

GÓMEZ-PASTOR, R.; PÉREZ-TORRADO, R.; CABISCOL, E. and MATALLANA, E. (2010). Transcriptomic and proteomic insights of the wine yeast biomass propagation process. FEMS Yeast Research, vol. 10, no. 7, p. 870-884. [CrossRef]

GUILLAUME, C.; DELOBEL, P.; SABLAYROLLES, J.M. and BLONDIN, B. (2007). Molecular basis of fructose utilization by the wine yeast Saccharomyces cerevisiae: A mutated HXT3 allele enhances fructose fermentation. Applied and Environmental Microbiology, vol. 73, no. 8, p. 2432-2439. [CrossRef]

JIMENEZ-MARTI, E. and DEL OLMO, M.L. (2008). Addition of ammonia or amino acids to a nitrogen-depleted medium affects gene expression patterns in yeast cells during alcoholic fermentation. FEMS Yeast Research, vol. 8, no. 2, p. 245-256. [CrossRef]

LIVAK, K.J. and SCHMITTGEN, T.D. (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2-^DDC_T Method. Methods, vol. 25, no. 4, p. 402-408. [CrossRef]

MA, M. and LIU, L.Z. (2010). Quantitative transcription dynamic analysis reveals candidate genes and key regulators for ethanol tolerance in Saccharomyces cerevisiae. BMC Microbiology, vol. 10, no. 1, p. 169-189. [CrossRef]

MAGASANIK, B. (1992). Regulation of nitrogen utilization. In: JONES, E.W.; PRINGLE, J.R. and BROACH, J.R. eds. The Molecular and Cellular Biology of the Yeast Saccharomyces: Gene Expression. New York, Cold Spring Harbor Publishers, p. 283-317.

MAGASANIK, B. (2003). Ammonia assimilation by Saccharomyces cerevisiae. Eukaryotic Cell, vol. 2, no. 5, p. 827-829. [CrossRef]

MARKS, V.D.; VAN DER MERWE, G.K. and VAN VUUREN, H.J.J. (2003). Transcriptional profiling of wine yeast in fermenting grape juice: Regulatory effect of diammonium phosphate. FEMS Yeast Research, vol. 3, no. 3, p. 269-287. [CrossRef]

MARULLO, P.; BELY, M.; MASNEUF-POMAREDE, I.; PONS, M.; AIGLE, M. and DUBOURDIEU, D. (2006). Breeding strategies for combining fermentative qualities and reducing off-flavor production in a wine yeast model. FEMS Yeast Research, vol. 6, no. 2, p. 268-279. [CrossRef]

MENDES-FERREIRA, A.; MENDES-FAIA, A. and LEAO, C. (2004). Growth and fermentation patterns of Saccharomyces cerevisiae under different ammonium concentrations and its implications in winemaking industry. Journal of Applied Microbiology, vol. 97, no. 3, p. 540-545. [CrossRef]

MITCHELL, A. and MAGASANIK, B. (1984). Three regulatory systems control production of glutamine synthetase in Saccharomyces cerevisiae. Molecular and Cellular Biology, vol. 4, no. 12, p. 2767-2773.

NISSEN, T.L.; SCHULZE, U.; NIELSEN, J. and VILLADSEN, J. (1997). Flux distributions in anaerobic, glucose-limited continuous cultures of Saccharomyces cerevisiae. Microbiology, vol. 143, no. 1, p. 203-218. [CrossRef]

PÉREZ, M.; LUYTEN, K.; MICHEL, R.; RIOU, C. and BLONDIN, B. (2005). Analysis of Saccharomyces cerevisiae hexose carrier expression during wine fermentation: Both low- and high-affinity Hxt transporters are expressed. FEMS Yeast Research, vol. 5, no. 4-5, p. 351-361. [CrossRef]

PETIT, T.; DIDERICH, J.A.; KRUCKEBERG, A.L.; GANCEDO, C. and VAN DAM, K. (2000). Hexokinase regulates kinetics of glucose transport and expression of genes encoding hexose transporters in Saccharomyces cerevisiae. Journal of Bacteriology, vol. 182, no. 23, p. 6815-6818. [CrossRef]

REIFENBERGER, E.; BOLES, E. and CIRIACY, M. (1997). Kinetic characterization of individual hexose transporters of Saccharomyces cerevisiae and their relation to the triggering mechanisms of glucose repression. European Journal of Biochemistry, vol. 245, no. 2, p. 324-333. [CrossRef]

ROSSIGNOL, T.; DULAU, L.; JULIEN, A. and BLONDIN, B. (2003). Genome-wide-monitoring of wine yeast gene expression during alcoholic fermentation. Yeast, vol. 20, no. 16, p. 1369-1385. [CrossRef]

SIWIAK, M. and ZIELENKIEWICZ, P. (2010). A comprehensive, quantitative, and genome-wide model of translation. PLoS Computational Biology, vol. 6, no. 7, p. e1000865. [CrossRef]

STANLEY, D.; CHAMBERS, P.J.; STANLEY, G.A.; BORNEMAN, A. and FRASER, S. (2010). Transcriptional changes associated with ethanol tolerance in Saccharomyces cerevisiae. Applied Microbiology and Biotechnology, vol. 88, no. 1, p. 231-239. [CrossRef]

TER SCHURE, E.G.; VAN RIEL, N.A.W. and VERRIPS, C.T. (2000). The role of ammonia metabolism in nitrogen catabolite repression in Saccharomyces cerevisiae. FEMS Microbiology Reviews, vol. 24, no. 1, p. 67-83. [CrossRef]

VANDESOMPELE, J.; DE PRETER, K.; PATTYN, F.; POPPE, B.; VAN ROY, N.; DE PAEPE, A. and SPELEMAN, F. (2002). Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biology, vol. 3, no. 7. [CrossRef]

VARELA, C.; PIZARRO, F. and AGOSIN, E. (2004). Biomass content governs fermentation rate in nitrogen-deficient wine musts. Applied and Environmental Microbiology, vol. 70, no. 6, p. 3392-3400. [CrossRef]

VAUDANO, E.; COSTANTINI, A.; NOTI, O. and GARCIA-MORUNO, E. (2010). An RT-qPCR approach to study the expression of genes responsible for sugar assimilation during rehydration of active dry yeast. Food Microbiology, vol. 27, no. 6, p. 802-808. [CrossRef]