OUP user menu

Overexpression of csc1-1. A plausible strategy to obtain wine yeast strains undergoing accelerated autolysis

Eduardo Cebollero, Adolfo Martinez-Rodriguez, Alfonso V. Carrascosa, Ramon Gonzalez
DOI: http://dx.doi.org/10.1016/j.femsle.2005.03.030 1-9 First published online: 1 May 2005


The potential of several alternative genetic engineering based strategies in order to accelerate Saccharomyces cerevisiae autolysis for wine production has been studied. Both constitutively autophagic and defective in autophagy strains have been studied. Although both alternatives lead to impaired survival under starvation conditions, only constitutively autophagic strains, carrying a multicopy plasmid with the csc1-1 allele under the control of the TDH3 promoter, undergo accelerated autolysis in the experimental conditions tested. Fermentation performance is impaired in the autolytic strains, but industrial strains carrying the above-mentioned construction are still able to complete second fermentation of a model base wine. We suggest the construction of industrial yeasts showing a constitutive autophagic phenotype as a way to obtain second fermentation yeast strains undergoing accelerated autolysis.

  • Sparkling wine
  • Autolysis
  • Genetic engineering
  • Autophagy
  • Saccharomyces cerevisiae

1 Introduction

Sparkling wines made by the traditional method (e.g. Champagne in France or Cava in Spain) undergo two main fermentative steps. The first one is similar to most wine fermentations and results in the base wine (usually a white wine made out of red or white grape cultivars, but “rosé” sparkling wines also exist). For the secondary fermentation this base wine is mixed with a combination of yeasts, sucrose, and bentonite called “liqueur de tirage”, bottled, and allowed to ferment and age. At the end of the aging period, yeast cells During the aging period yeast cells die and undergo autolysis, releasing nitrogen compounds, mostly peptides and amino acids, to the external medium. This process affects aroma, flavor and foaming properties of sparkling wines. These have been correlated with the products of the hydrolytic degradation of yeast cells, including free amino acids, peptides, mannoproteins, nucleic acid derivatives or lipids [[[]. Products released to the wine by living and autolytic yeast cells are also important for the quality of still wines, especially those aged on lees [[,[]. Several authors have reported a correlation between the autolytic capacity of yeast strains and the quality of the sparkling wines obtained [[,[,[0]. However, autolysis in enological conditions is a slow process that requires long aging periods in order to reach the desired results. This contributes to the relatively high production costs of these wines.

Mainly two methods have been tested in order to accelerate the acquisition of aging-like properties during sparkling wine production: adding yeast autolysates to the wine, or increasing the temperature of aging [[1]. Both techniques result in organoleptic defects in the final product, which are often described as toasty [[2]. The use of a combination of killer and sensitive yeast strains has been recently suggested as a way to accelerate the onset of yeast autolysis during sparkling wine production [[3]. The effect of this procedure on the sensory properties of the wine has not yet been evaluated. Tini et al. [[0] and Gonzalez et al. [[4] found interesting results by using autolytic strains derived through meiosis from an industrial second fermentation yeast, or mutant yeast strains obtained by UV irradiation, respectively.

Autophagy is a ubiquitous process in eukaryotic cells in which the bulk cytoplasm and, sometimes, entire organelles are delivered for degradation into a lytic compartment (e.g. lysosome or vacuole) by a vesicular transport system (reviewed in [[5]). In S. cerevisiae, autophagy is usually induced under carbon or nitrogen starvation conditions, and is necessary as a preliminary step for meiosis. Autophagy starts with the formation of autophagosomes, which are double membrane vesicles carrying fractions of the cytoplasm [[6]. After reaching the vacuole, the outer membrane of the autophagosome fuses with the vacuolar membrane and an autophagic body (corresponding to the inner membrane of the vesicle and its content) is released into the vacuolar lumen [[6,[7], finally the cargo is processed by the resident hydrolases (reviewed in [[8]).

The isolation of yeast mutants defective in autophagy allowed the identification of many genes involved in the process [[9[1], and some of them have been shown to be also involved in the cytosol to vacuole targeting (cvt) pathway [[2], a specific vesicular transport process that delivers precursors of the vacuolar enzymes amiopeptidase I (Ape1) and α-mannosidase to the vacuole (reviewed in [[3]). ATG1/APG1 is essential for the autophagic and cvt pathways, and codes for a serine/threonine protein kinase that takes part in a dynamic protein complex together with Atg13p and other proteins specific for the cvt or autophagic pathways [[4,[5]. This complex plays an important role in switching between the cvt and autophagy pathways in response to nutrient availability. The affinity between Atg1p and Atg13p is increased upon induction of autophagy [[5] but the role of the Atg1p kinase activity in autophagy is controversial. Matsuura et al. [[4] reported that the kinase activity of Apg1p is inhibited during the autophagic process, while Kamada et al. [[5] showed that this activity increased when autophagy was induced being essential for both autophagy and the cvt pathway. Recent results from Abeliovich et al. [[6], demonstrating that the kinase activity of Atg1p is not required for induction of autophagy, agree with the earlier observation.

ATG8/AUT7, also required for the cvt and autophagic pathways, was the first gene whose expression was shown to be increased during the induction of autophagy [[7,[8]. Increased levels of Atg8p during autophagy are required for the autophagosomes to grow into their correct size [[9]. Its recruitment to the membrane of the preautophagosomal structure, the putative precursor of the mature autophagosome, where most of the autophagy-related proteins co-localize [[5,[0,[1], is needed for the membrane dynamics during autophagy [[8,[0,[2].

So far a single gene has been identified whose mutation leads to constitutive autophagy. The csc1-1 allele (constitutive sequestration of cytosol by autophagy) was identified as responsible of the induction of autophagy in nutrient rich medium [[3]. CSC1 is allelic to VPS4/END13, an AAA-type ATPase (ATPase associated with variety of cellular activities) that plays an important role in multiple steps of membrane traffic through the endocytic pathway [[4[6], and is supposed to have a role in the late endosome. The “csc” phenotype is dominant in multicopy and is due to a missense gain-of-function point mutation in a region of the CSC1 open reading frame that is highly conserved among the AAA-type ATPase family members [[3]. In contrast to csc1-1, overexpression of the wild type CSC1 gene does not lead constitutive autophagy, although Δcsc1 cells show a severe defect in autophagy [[3].

We have recently shown that autophagy could play a role in the release of yeast compounds to the medium during autolysis in enological conditions [[7], and would like to use genetic engineering of genes involved in the autophagic process in order to improve the properties of industrial yeast strains used for second fermentation. Our working hypothesis was that accelerating autophagy would make yeast strains more suitable for the aging process in terms of time required and/or amount of compounds released to the external medium. However, since the release of autolysis compounds seems to come after cell death in enological conditions [[8,[9], we questioned the effect on autolysis of mutations that impair autophagy, but also lead to a quick loss of viability in stationary phase or under starvation conditions [[1]. Therefore, we decided to investigate the effect of promoting or stopping autophagy on yeast features that might be relevant for their enological performance, by using recombinant laboratory and industrial strains carrying mutations on ATG1/APG1, ATG8/AUT7 or CSC1/VPS4/END13.

2 Materials and methods

2.1 Strains

Escherichia coli DH5α (supE44, ΔlacU169 [?80 lacZΔM15], hsdR17, recA1, endA1, gyrA96, thi-1, relA1) [[0] was used for the production of the plasmids constructed in this study. The following S. cerevisiae strains were obtained from Euroscarf (European S. cerevisiae Archive for Functional Analysis) and used in this work, BY4741 (MATα, his3Δ1, leu2Δ0, met15Δ0, ura3Δ0) and its isogenic strains deleted for ATG1/APG1 (ΔATG1: MATα, his3Δ1, leu2Δ0, met15Δ0, ura3Δ0, ygl180w::kanMX4) or ATG8/AUT7 (ΔATG8: MATα, his3Δ1, leu2Δ0, met15Δ0, ura3Δ0, ybl078c::kanMX4). S. cerevisiae T73-4[[1], an ura3Δ derivative of S. cerevisiae T73 (Lallemand), which in turn is a commercial strain isolated in Alicante (Spain), was used for the construction of recombinant industrial yeast strains.

2.2 Molecular biology techniques

Unless otherwise specified, all DNA manipulations were performed as described by Sambrook et al. [[2]. Polymerase chain reactions (PCR) were performed using AmpliTaq GOLD DNA polymerase (Applera Hispania, Tres Cantos, Spain). Restriction enzymes were from Roche Diagnostics SL Barcelona, Spain. Yeast genomic DNA was extracted by the method of Querol et al. [[3]. DNA fragments resolved in agarose gels were purified with the QIAquick Gel Extraction Kit (QIAGEN GmbH, Hilden, Germany). E. coli was transformed by electroporation [[2] and plasmids were purified from E. coli cells by using the High Pure Plasmid Isolation Kit (Roche). Transformation of S. cerevisiae BY4741 and T73-4 followed the lithium acetate method described by Ito et al. [[4] as modified Gietz and Woods [[5]. Transformants were selected in SMM plates without uridine. Yeast transformation was verified by plasmid extraction followed by E. coli transformation using the method described by Robzyk and Kassir [[6].

2.3 Construction of CSC1 and CSC1-1 strains

The ORF of S. cerevisiae CSC1 was amplified by PCR using genomic DNA from S. cerevisiae BY4741 as template, and the following primers: 5′-TAGGGATCCTATAGTAGATGGGGTAC-3′ and 5′-AGAATTGATAATGCTAGGGTATCC-3′. A BamHI target sequence (underlined) was introduced near the 5′ end of the upstream primer. This PCR fragment was cloned in the plasmid vector pGEM-T (Promega Corporation, Southampton, UK) following the instructions of the provider. The resulting plasmid (pAM1) served as template for site-directed mutagenesis with the QuickChange Mutagenesis Kit (Stratagene, La Jolla, CA), following the instructions of the provider and using the following primers: 5′-GGACAGTGCCATCAGGAGAAGATTTAAAAGAAGAATATATATACC-3′ and 5′-GGTATATATATTCTTCTTTTAAATCTTCTCCTGATGGCACTGTCC-3′ (the nucleotide corresponding to the point mutation is underlined in both primers). This resulted in plasmid pAM2 carrying a csc1-1 allele, with the sequence described by Shirahama et al. [[3] (a G to A transition at position 871 of the ORF). The CSC1 and csc1-1 inserts of these plasmids were transferred to a yeast expression plasmid as follows: pAM1 or pAM2 were digested with SalI, the 5′ protruding ends were then filled in with Klenow (Roche), and the insert released by subsequent digestion with BamHI. The purified inserts were then ligated to pYES2 (Invitrogen BV, Groningen, Netherlands) previously digested with EcoRI, filled in with Klenow and digested with BamHI. This resulted in plasmids pAM3 and pAM4, for the wild type or the mutant ORF, respectively. Exchange of the GAL1 promoter by the TDH3 promoter in pAM3 and pAM4 was done by the method of Wang and Malcolm [[7] as follows. The TDH3 promoter was amplified by PCR, using yeast genomic DNA as template and primers carrying 5′ extensions complementary to the regions flanking the GAL1 promotor in pYES2: 5′-CGTGGGGATGATCCACTAGTCCCGTACATGCCCAAAATAGG-3′ and 5′-CCTTAGTTAAAAAATCTCCCGTGCTCATTTTGTTTGTTTATGTGTGTTT-3′ (5′ extensions underlined). The PCR product was used as a mutagenic primer pair for site directed mutagenesis with the QuikChange Site-directed Mutagenesis kit (Stratagene) using pAM3 or pAM4 as templates, resulting in pCG1 and pCG2, respectively. The most relevant parts of the constructions were verified by sequencing, before yeast transformation.

2.4 Media

Yeast strains were maintained on YPD plates, 1% yeast extract, 2% peptone, 2% glucose and 2% agar. Supplemented minimal medium (SMM) consist of 0.67% yeast nitrogen base, 2% glucose and the auxotrofic requirements specific to each strain. 1.67% agar was added for solid medium. SD-N contains 0.17% of yeast nitrogen base without amino acids and ammonium sulphate with 2% glucose. SD-C contains 0.67% of yeast nitrogen base supplemented with the auxotrofic requirements. Synthetic base wine contained tartaric acid 3 g/l, malic acid 6 g/l, citric acid 0.3 g/l, yeast nitrogen base without ammonium sulphate and amino acids 1.7 g/l, ammonium sulphate 0.5 g/l, sucrose 20 g/l, the appropriate percentage of ethanol and auxotrophic requirements as needed. pH was adjusted to 3.5 with potassium hydroxide. Model wine buffer for accelerated autolysis experiments contained tartaric acid 4 g/l, malic acid 3 g/l, acetic acid 0.1 g/l, potassium sulphate 0.1 g/l, magnesium sulphate 0.025 g/l and ethanol 10% (v/v). pH was adjusted to 3.0 with sodium hydroxide.

2.5 Survival in starvation media

Cells from an overnight culture in SMM medium grown at 30 °C and 200 rpm were centrifuged for 5 min to 3000g, washed twice with a sterile 0.9% NaCl solution and inoculated in SD-N or SD-C to a final concentration of 107 cells per ml. Samples from these cultures, incubated at 30 °C and 200 rpm, were taken at different time points and spread in the appropriate dilutions on YPD or SMM plates without uridine. The plates were incubated for 2–3 days at 30 °C. Confirmation of plasmid loss was obtained by replica-plating colonies grown in YPD plates in both YPD and SMM plates.

2.6 Analysis of autolytic capacity in accelerated conditions

Assays of accelerated autolysis were performed as described by Feuillat [[8] with minor modifications. The yeasts were cultured in SMM at 30 °C and 200 rpm for 12–72 h, depending on the strain, in order to have the cells in the early exponential growth phase. Yeast cells were harvested by centrifugation at 5000g and 4 °C for 5 min, and washed three times with 0.9% NaCl. The washed yeast was suspended in the model wine buffer to a final concentration of 108 cells per milliliter. Autolysis was conducted by incubating the cell suspensions at 30 °C and 100 rpm. Samples were taken at 0, 2, 4, 6, 8, 24 and 48 h and filtered through a 0.45 μm membrane. The cell-free autolysates were stored at −70 °C until analysis. The survival of the different strains in the model wine buffer was calculated by inoculating on YPD and SMM plates. Amino acid concentration was estimated by the method of Doi et al. [[9]. These experiments were repeated three times and data analyzed by one-way ANOVA and Student–Newman–Keuls test for means comparisons. Differences were considered significant for p < 0.05. The calculations were carried out by means of the SPSS program for Windows, release 11.5, run on a personal computer.

2.7 Fermentation experiments

Cells from an overnight culture in SMM were centrifuged for 5 min at 3000g, washed twice with a sterile 0.9% NaCl solution and suspended in sterile synthetic base wine. Erlenmeyer flasks with 50 ml of synthetic base wine, containing different amounts of ethanol were inoculated to a final concentration of 106 cells/ml. The flasks, closed with Müller valves filled with sulphuric acid, were weighted at time zero before statically incubating either 30 or 17 °C. The flasks were weighted for several days until stationary values were reached. Ethanol production was calculated as described by Vaughan-Martini and Martini [[0]. Alternatively the flasks were incubated without a Müller valve, and samples were taken periodically. Yeast cells were removed by centrifugation and the supernatant was stored at −20 °C until residual reducing sugars were quantified with DNS (3,5-dinitro-salicylic acid) as described by Bernfeld [[1]. Before quantification of residual sugars the samples were incubated for 2 h at 37 °C with invertase (Sigma–Aldrich Química, Tres Cantos, Spain) to ensure complete hydrolysis of residual sucrose.

3 Results

3.1 Construction of CSC1 and CSC1-1 strains

The “constitutive sequestration of cytosol by autophagy” phenotype conferred by the overexpression of the csc1-1 allele is recessive, but dominant in multicopy. This was originally shown by studying the phenotypes of yeast strains expressing csc1-1 under the control of the GAL1 promoter in the multicopy plasmid pYES2 [[3]. However, the use of the conditional promoter made necessary a galactose induction step. This hampered the study of the effect of the overexpression of the csc1-1 allele under carbon starvation conditions or during accelerated autolysis assays. In addition, the GAL1 promoter would be inactive in industrial strains under enological conditions. For these reasons we decided to exchange the inducible GAL1 promoter by the constitutive TDH3 promoter as described in Section 2. The resulting plasmids, pCG1 and pCG2, as well as pYES2, used as a control, were employed for transforming S. cerevisiae BY4741, resulting in strains YES2, CSC1 and CSC1-1, respectively. These strains, together with BY4741 and the strains defective in autophagy ΔATG1 and ΔATG8, were used in the following experiments.

3.2 Survival under starvation conditions

Concerning survival in nitrogen or carbon starvation conditions the effect of mutations leading to a defect in autophagy (ygl180w::kanMX4 in ΔATG1 and ybl078c::kanMX4 in ΔATG8) is totally different of that of promoting autophagy by overexpression of csc1-1. During a 26 days incubation in either carbon or nitrogen starvation conditions the constitutively autophagic strain CSC1-1 shows a reduced survival rate, as compared with control YES2 and CSC1 strains, in both carbon (Fig. 1(a)) and nitrogen (Fig. 1 (b)) starvation conditions. The relative loss of viability is higher in the case of carbon starvation, 4 log instead of 2 log for nitrogen starvation. Because the csc1-1 allele is coded in a multicopy plasmid in the CSC1-1 strain, and all viable counts were performed in YPD medium, the question arises of what happens to the plasmid under starvation conditions. This was estimated by comparing viable counts in YPD and SMM media. After 26 h of incubation the number of cells harboring the pCG2 plasmid was reduced by about two or three orders of magnitude in nitrogen or carbon starvation conditions, respectively (data not shown). It was concluded that, at least the CSC1-1 strain, undergoes plasmid loss during incubation in starvation conditions. This was further confirmed by replica plating colonies previously grown in YPD medium in both YPD and SMM media. After 15 days of incubation the plasmid was almost completely lost in the CSC1-1 strain (92–100% loss, data not shown). Interestingly, approximately a 20% plasmid loss was also observed for strain CSC1, indicating a weak detrimental effect of the overexpression of the wild type CSC1 allele. Plasmid loss in the YES2 strain represented about 1–2% (data not shown). Results were similar for carbon and nitrogen starvation.


Loss of viability under carbon (a) or nitrogen (b) starvation conditions of the yeast strains BY4741 (diamonds), ΔATG1 (squares), ΔATG8 (triangles), YES2 (open circles), CSC1 (stars), CSC1-1 (black circles). The number of viables was calculated by platting in YPD medium.

In contrast to the similar response to both starvation conditions found for CSC1-1, strains defective in autophagy behave quite differently in response to carbon or nitrogen starvation. Actually, under carbon starvation conditions both ΔATG1 and ΔATG8 are almost indistinguishable from the control BY4741 strain in terms of viable counts (Fig. 1(a)). On the other side, nitrogen starvation has a severe effect on the viability of both strains. This effect is stronger than seen for the CSC1-1 strain, and leads to a complete loss of viability after 20 days of incubation under nitrogen starvation conditions for the ΔATG1 strain (Fig. 1 (b)). The relative sensitivity to nitrogen starvation of both strains (ΔATG1 and ΔATG8) seems to be correlated with their autophagic phenotype. It has been shown that mutants defective in ATG8 form autophagosomes of reduced size and result in a low autophagic rate [[9], while those defective in ATG1 are completely unable to form autophagosomes. Accordingly, viability loss of the ΔATG1 strain is faster than that of ΔATG8 under nitrogen starvation conditions.

3.3 Autolytic capacity in accelerated conditions

The amount of amino acids released to the medium in accelerated autolysis conditions has been shown to be a good predictor of the performance of second fermentation wine yeast strains during aging of sparkling wines [[]. The results of the experiments of accelerated autolysis shown in Fig. 2 were unequivocal; a single strain, CSC1-1, undergoes accelerated autolysis much faster than the others. After only 2 h of incubation in autolysis medium the amount of amino acids released by CSC1-1 was at least five times higher than any other strain (Fig. 2). After 8 h this amount is still at least twice higher than the control YES2 strain. These differences were statistically significant (p < 0.05). After 24 h of incubation a stationary value was reached and these differences were completely diluted. Thus, it seems that overexpression of the csc1-1 allele leads to acceleration in the autolytic process in the CSC1-1 strain, but not to an increased yield of autolytic products (i.e. autolysis proceeds faster, but it does not go further, in the CSC1-1 strain as compared to the control).


Release of amino acids in accelerated autolysis assays for the six yeast strains analyzed. Results are expressed as milligrams of leucine equivalents per ml per gram of initial dry weight.

Mutants defective in autophagy did not show any acceleration in the release of amino acids to the external medium. Actually, data shown in Fig. 2 for time points 6 and 8 h show an apparently slower release of amino acids by ΔATG1 and ΔATG8 strains as compared to the control BY4741 strain, but statistical analysis indicated that these differences were not significant (p > 0.05).

3.4 Fermentation of synthetic base wine

From the results described above, it was concluded that overexpression of the csc1-1 allele could be an interesting strategy in order to induce accelerated autolysis in second fermentation industrial wine yeast strains. However, industrial second fermentation conditions are quite stressful for yeasts, and we wondered to what extent the “weakness” induced in the strain by the csc1-1 allele would also impair its fermentation performance. The ability to ferment 20 g/l of sucrose added to a synthetic base wine containing different amounts of ethanol was tested for YES2, CSC1 and CSC1-1 strains. The time required for completing fermentation for each strain and growth condition is shown in Table 1. At 17 °C and in the absence of ethanol, the end of the fermentation process is already delayed for CSC1-1 as compared to CSC1 and YES2 strains. This detrimental effect on the fermentation ability becomes more evident with increasing initial ethanol concentration, especially when combined with a higher incubation temperature (30 °C). Indeed, strain CSC1-1 is unable to complete the fermentation process with 4% ethanol at 17 °C or with 2% ethanol at 30 °C, even though ethanol production is faster at the highest temperature. The effect of higher fermentation temperatures is especially dramatic with 4% ethanol. In these conditions neither CSC1 nor CSC1-1 are able to grow or to initiate fermentation. It is worth noting that CSC1 was not distinguishable from the control strain for fermentation at 17 °C.

View this table:

Time required for completing fermentation by different recombinant laboratory strains in the presence of varying amounts of ethanol and at different temperatures

3.5 Fermentative capacity of industrial strains

The encouraging results obtained with strain CSC1-1 in the viability and autolysis experiments must be analysed with care, because of the strong reduction in the fermentative power experienced by this strain. In case industrial strains expressing the csc1-1 allele were unable to complete a second fermentation, this would preclude any practical application of industrial recombinant strains for the purpose of shortening the aging period. It was therefore necessary to test the effect of overexpression of the csc1-1 allele on the fermentative power of industrial strains. To do that, yeast strain T73-4 was transformed with plasmids pYES2, pCG1 and pCG2. This strain was chosen because it is an auxotrophic derivative of a commercial wine yeast strain that can be transformed with the same constructions we used for laboratory strains. The resulting recombinant strains were named T73-YES2, T73-CSC1 and T73-CSC1-1. Second fermentation experiments in synthetic base wine at 17 °C were performed with the three strains. The amount of residual sugar was recorded until total exhaustion, which was reached after 12 days for T73-YES2, 16 days for T73-CSC1 and 20 days for T73-CSC1-1 (data not shown). In fermentation experiments performed at 30 °C with strain T73-CSC1-1, total consumption of sugar took 4 days in the absence of ethanol, 6 days with 7% ethanol and 11 days with 10% ethanol (Fig. 3). These results highlight an undeniable detrimental effect of overexpression of csc1-1 (and to a lesser extent of CSC1) on the fermentation performance of industrial strains, but the recombinant strain is still able to completely ferment the sugars added to the base wine (at a concentration similar to that used in the manufacture of sparkling wines) at either 17 or 30 °C.


Sugar consumption by strain T73-CSC1-1 at 30 °C in synthetic base wine containing different amounts of ethanol, 0% (squares), 7% (triangles) or 10% (diamonds).

4 Discussion

One clear conclusion of our experimentation is that slowing down autophagy does not seem to be a promising way to get to the objective of accelerating autolysis, at least by using strains carrying deletions of ATG1 or ATG8. It should be noted that while deletion of ATG1 or ATG8 has a marked effect on the survival of yeast to nitrogen starvation conditions, no effect could be appreciated concerning survival to carbon starvation. In contrast, all the results presented above point to the acceleration of autophagy, by overexpression of the csc1-1 allele, as an interesting strategy in order to construct recombinant second fermentation wine yeast strains able to autolyse faster than the original strains, as a potential way to shorten the time necessary to obtain quality sparkling wines. This conclusion is mainly sustained by the results shown in Fig. 2, in which a great acceleration of autolysis is observed for CSC1-1 as compared to the control strain.

The doubts raised by the poor fermentation performance of the laboratory strains over-expressing the csc1-1 allele were acceptably solved by the ability of industrial strains transformed with this construction to completely ferment the sucrose added in a model wine system with 10% ethanol, albeit fermentation was slower.

No effect of the overexpression of the wild type allele CSC1 on the autolytic capacity could be observed. Despite this, it would be interesting to test the effect on the quality of sparkling wines of industrial strains expressing not only csc1-1 but also the wild type allele CSC1, because it cannot be excluded that minor effects, that could not be detected in the accelerated autolysis test, had an appreciable effect on the quality of wines made in industrial conditions. The existence of minor effects is mainly supported by three observations: a moderate plasmid loss rate under carbon or nitrogen starvation conditions, inability to ferment sucrose in a model base wine containing 4% ethanol at 30 °C, and fermentation time for T73-CSC1 is intermediate between T73-YES2 and T73-CSC1-1.

We have not found a good explanation to the mechanism of plasmid loss by the laboratory strains under carbon or nitrogen starvation conditions, when cell cycle is apparently arrested. However, this genetic instability should be taken into account for the construction of industrial second fermentation strains intended to be used in real industrial wine making conditions. A genetic engineering strategy for industrial strains should consider either the insertion of one or several copies of the construction in the yeast genome, or the replacement of the wild type alleles by csc1-1 (twice or more, depending on the particular genotype of the industrial strain chosen). Probably this should be enough to confer the “constitutive sequestration of cytosol by autophagy” phenotype, which was first described in a haploid strain carrying a csc1-1 allele in its original locus[[3].


We are grateful to Daniel G. Ramos and Ma Victoria Santamaría for technical assistance, and Bernd Hilhorst for critical reading of the manuscript. This work was supported by the Spanish Ministerio de Educación y Ciencia. (AGL2003-01762 and AGL2000-01569). E. Cebollero is the recipient of a FPI fellowship from the Spanish Ministerio de Educación y Ciencia.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
  34. [34].
  35. [35].
  36. [36].
  37. [37].
  38. [38].
  39. [39].
  40. [40].
  41. [41].
  42. [42].
  43. [43].
  44. [44].
  45. [45].
  46. [46].
  47. [47].
  48. [48].
  49. [49].
  50. [50].
  51. [51].
View Abstract