OUP user menu

Monitoring of Saccharomyces and Hanseniaspora populations during alcoholic fermentation by real-time quantitative PCR

Núria Hierro, Braulio Esteve-Zarzoso, Albert Mas, Jose M. Guillamón
DOI: http://dx.doi.org/10.1111/j.1567-1364.2007.00304.x 1340-1349 First published online: 1 December 2007

Abstract

Real-time, or quantitative, PCR (QPCR) was developed for the rapid quantification of two of the most important yeast groups in alcoholic fermentation (Saccharomyces spp. and Hanseniaspora spp.). Specific primers were designed from the region spanning the internal transcribed spacer 2 (ITS2) and the 5.8S rRNA gene. To confirm the specificity of these primers, they were tested with different yeast species, acetic acid bacteria and lactic acid bacteria. The designed primers only amplified for the intended group of species and none of the PCR assays was positive for any other wine microorganisms. This technique was performed on reference yeast strains from pure cultures and validated with both artificially contaminated wines and real wine fermentation samples. To determine the effectiveness of the technique, the QPCR results were compared with those obtained by plating. The design of new primers for other important wine yeast species will enable to monitor yeast diversity during industrial wine fermentation and to detect the main spoilage yeasts in wine.

Key words
  • Saccharomyces
  • Hanseniaspora
  • yeast quantification
  • real-time PCR
  • wine fermentation

Introduction

Alcoholic fermentation is a complex microbial process that is characterized by a succession of different species of yeasts (Fleet, 1993). Yeasts with a low fermentation activity, such as Candida spp., Hanseniaspora spp., Pichia spp., Rhodotorula spp. and Kluyveromyces spp., are predominant in grape musts and during the early stages of fermentation. Subsequently, Saccharomyces cerevisiae proliferates, dominating and completing the wine fermentation (Fleet & Heard, 1993; Beltran et al., 2002). Hanseniaspora species have been widely reported as being the major non-Saccharomyces yeasts associated with grape microbiota (Sabaté, 2002) and during the initial stages of wine fermentation (Pretorius, 2000). Some strains of these apiculate yeasts have shown positive oenological properties and it has been suggested that they can be used in grape-must fermentations to enhance the aroma and flavour profiles of the wines (Romano et al., 2003; Capece et al., 2005). It was thought, however, that this contribution was limited to the first few days of fermentation because the growth of these yeasts was suppressed by the rapid development of the Saccharomyces strains (Pretorius, 2000). The dominance of this fermentative species in the middle and final stages of the process was so overwhelming that it was highly unlikely that other non-Saccharomyces species would be isolated.

Pérez-Nevado (2006) have studied the mechanism involved in the cellular death of two Hanseniaspora wine strains (Hanseniaspora guillermondii and Hanseniaspora uvarum) during mixed fermentations with S. cerevisiae under oenological growth conditions. When S. cerevisiae reached cell densities of around 107 CFU mL−1, a strong reduction in the Hanseniaspora population was observed regardless of the ethanol concentration. The authors hypothesized that one or more toxic compounds produced by S. cerevisiae triggers the early death of the Hanseniaspora cells, although it has not yet been possible to identify the nature of these compounds. Several authors had previously demonstrated that viable cells of the yeast S. cerevisiae at a high density cause the growth arrest of two non-Saccharomyces yeast species in mixed cultures (Nissen & Arneborg, 2003; Nissen et al., 2003; Arneborg et al., 2005). However, these authors postulated that the early death seemed not to be due to the presence of toxic compounds. Rather, they seemed to be mediated by a cell–cell contact mechanism. In spite of these reports, using a direct molecular method, Mills (2002) detected an active population of Hanseniaspora strains throughout fermentation that went unnoticed by the plating method. The authors also recently detected a permanent population of H. uvarum and Hanseniaspora osmophila strains throughout different fermentations (Hierro et al., 2006a). In the authors' opinion, however, a still unanswered question in most studies of yeast is to what extent these minor species contribute to the total yeast population during wine fermentation. The application of plating-independent molecular methods may reveal the real diversity during wine fermentation.

In the last few years, researchers have used methods to directly identify yeasts from wine without the need for plating (Ibeas et al., 1996; Cocolin et al., 2000; Mills et al., 2002; Phister & Mills, 2003; Delaherche et al., 2004). Most of these methods rely on the direct amplification of yeast DNA from wine by PCR. Phister & Mills (2003) highlighted two main advantages of the direct characterization of microorganisms over yeast enrichment and plating: first, regardless of their capacity to grow in a plate, all the yeast populations are detected and, second, analysis is fast.

Because of its specificity and sensitivity, one of the most promising PCR techniques in food control is real-time, or quantitative, PCR (QPCR) (Bleve et al., 2003). QPCR assays have been developed to detect and enumerate various bacteria and fungi in food (Hein et al., 2001; Blackstone et al., 2003; Bleve et al., 2003; Haarman & Knol, 2005), but also to detect microorganisms in wine (Phister & Mills, 2003; Delaherche et al., 2004; Pinzani et al., 2004; Martorell et al., 2005; González et al., 2006; Rawsthorne & Phister, 2006; Hierro et al., 2006b).

In this study, a method is developed for rapidly (<8 h) and accurately quantifying two of the main populations of yeasts during alcoholic fermentation: Saccharomyces and Hanseniaspora strains. For the quantification of both Saccharomyces and Hanseniaspora strains, the primers were designed from the region spanning the internal transcribed spacer 2 (ITS2) and the 5.8S rRNA gene. The QPCR assays were first conducted on reference yeast strains from pure cultures to detect the specificity of the primers. Later, this technique was validated with both artificially contaminated and natural samples of wine fermentation and, to determine effectiveness, the results were compared with those obtained from plating.

Materials and methods

Yeast strains

The strains used in this study, all of which were obtained from the Spanish Type Culture Collection (CECT), are listed in Table 1. All yeast strains were grown overnight in yeast extract peptone dextros (YEPD) media (1% yeast extract; 2% peptone and 2% glucose, w/v) at 28°C. Yeast and bacterial strains were used to test the specificity of the primers (Table 1). The strain Lipomyces kononenkoae 1967T was used as an external reference because this species has been never isolated from a wine source. Aliquots of known cell quantities (105 cells) of L. kononenkoae 1967T were added to each cell dilution of different yeast strains used to generate standard curves.

View this table:
1

Table 1. Reference strains used in this study

MicroorganismCECT DesignationOther designationsIsolation sourceSaccharomyces-specific PCR resultsHanseniaspora-specific PCR results
Yeasts
Candida boidinii1014TCBS 2428Tanning fluid
10029MCYC 113Milk
10035MCYC 124Amygdalus communis
Candida sake1044CBS 617Lambic beer
Candida stellata11046CBS 2649Grape juice
11110CBS 843Wine grapes
Dekkera bruxellensis1009CBS 72Lambic beer
1451TCBS 74Lambic beer
Hanseniaspora uvarum1444TCBS 314Muscat grape+
10389MCYC 1857Grape juice+
11105CBS 2589Grape must+
11106CBS 5073Wine grape+
11107CBS 8130Grapes+
Hanseniaspora guillermondii11027MCYC 2380Grape must+
11029CBS 465Infected nail+
11102CBS 1972Grape juice+
Hanseniaspora osmophila11206CBS 313Ripe Riesling grape+
11207CBS 105Grape+
Hanseniaspora vineae1471CBS 6555Grape juice+
Issatchenkia terricola11139CBS 4715Dregs of pressed grapes
11176CBS 2617Soil
Saccharomyces bayanus1941TCBS 300Beer+
1969CBS 395Juice of Ribes nigrum+
Saccharomyces cerevisiae1171CBS 1320+
1319ATCC 26602Sugar refinery+
1942NTCBS 1171Beer+
Saccharomycodes ludwigii1371IFI 979
1382IFI 982
Schizosaccharomyces pombe1378ATCC 24751Millet beer
1379ATCC 26760Grape must
10685TCBS 356
Torulaspora delbrueckii1880Wine
Zygosaccharomyces bailii11042CBS 3014Wine
11043CBS 4688Grape must
Zygosaccharomyces rouxii1230CBS 741Honey
1232TCBS 732Must of black grape
Lipomyces kononenkoae1967TCBS 2514Soil of citrus orchard
Bacteria
Acetobacter aceti298TATCC 15973Beech-wood shavings of vinegar plant
Oenoccocus oeni217TATCC 23279Wine
  • CECT, Spanish Type Culture Collection, Universidad de Valencia; CBS, Centralalbureau voor Schimmelcultures, culture collection in Delft, the Netherlands; MCYC, Microbiology Collection of Yeast Cultures, Universidad Politécnica de Madrid; ATCC, American Type Culture Collection, Manassas, VA; IFI, International Fabricare Institute.

Primer design

Saccharomyces cerevisiae-, H. uvarum- and L. kononenkoae- specific primer sets were designed by aligning the region spanning the ITS2 and the 5.8S rRNA gene. Three alignments were made that included sequences of the target species (S. cerevisiae, H. uvarum and L. kononenkoae, respectively) and those of the most common wine yeast species. Sequences were obtained from the GenBank database, and alignment was performed with the clustal w multiple sequence alignment (Thompson et al., 1994). The final selection of the primers was performed using the abi primer express program (Applied Biosystems, Foster City, CA). The blast search (Basic Alignment Search Tool, Internet address: http://www.ebi.ac.uk/blastall/nucleotide.html) was used to check the specificity of each primer. The sequence of primers and amplicon size generated are shown in Table 2.

View this table:
2

Primer sequences developed for RT-PCR assays

Target speciesPrimerSequence 5′–3′PCR product size (bp)
S. cerevisiaeSCER-RCGCAGAGAAACCTCTCTTTGGA175
CESP-FATCGAATTTTTGAACGCACATTG
H. uvarumHUV-RAACCCTGAGTATCGCCCACA121
CESP-FATCGAATTTTTGAACGCACATTG
L. kononenkoaeLIP-RTAGAAGCCAGATATGTGCCCATC117
CESP-FATCGAATTTTTGAACGCACATTG

Specificity of PCR assays

DNA samples from all yeasts were extracted according to Querol (1992). PCRs were carried out in 25 μL final volume containing 5 μL of DNA template (between 10 and 100 ng), 1 μM of each respective primer, 100 μM of each dNTP, 2.5 mM MgCl2, 10X buffer and 2.5 U of Taq DNA polymerase (ARK Scientific, Darmstadt, Germany).

The PCR conditions were an initial denaturation at 95°C for 5 min, followed by 35 cycles of denaturing at 95°C for 1 min, annealing at 60°C for 1 min and extension at 72°C for 1 min, with a final extension at 72°C for 5 min. All amplifications were performed in a GeneAmp PCR System 2700 (Applied Biosystems, Foster City, CA). The products of the PCR were analysed by electrophoresis on a 3% (w/v) agarose gel in 1 × TBE buffer stained with ethidium bromide and visualized under UV light. A 100-bp DNA ladder marker (Gibco BRL, Eggestein, Germany) was used as the size standard.

DNA extraction for the QPCR assay

Yeast cell suspensions were washed with sterile water and the pellets were resuspended in 700 μL of AP1 buffer (DNeasy Plant minikit, Qiagen, Valencia, CA) and transferred to a 2 mL conical-bottom microcentrifuge tube containing 1 g of 0.5 mm diameter glass beads. The tubes were shaken in a mini bead-beater (Biospec Products Inc., Bartlesville, OK) for 3 min at the maximum rate and then centrifuged at 9300g for 1 min. The DNA in the supernatant was transferred to a sterile microfuge tube and purified using the DNeasy Plant minikit (Qiagen, Valencia, CA) according to the manufacturer's instructions.

Standard curves

Standard curves were created by plotting the cycle threshold (CT) values of the QPCR performed on dilution series of DNA or yeast cells (106–1 cells mL−1) against the log input cells per millilitre. Yeast strains H. uvarum 1444T and 11107 and S. cerevisiae 1171, 1319 and 1942NT were used to achieve Hanseniaspora- and Saccharomyces-specific standard curves, respectively. These standard curves were generated from cells grown in YEPD and from cells incubated in wine. Aliquots of L. kononenkoae 1967T containing 105 cells were added to each dilution as an external reference.

Artificially contaminated wine

Different mixed cultures with known populations of S. cerevisiae (S.c) plus H. uvarum (H.u) were incubated for 24 h in Macabeo white wine (11% ethanol; 70 mg L−1 total SO2; pH 3.2) and Cabernet Sauvignon red wine (13.5% ethanol; 60 mg L−1 total SO2; pH 3.6) previously sterilized by filtration. The final cell concentrations of the mixed cultures, estimated by microscopic counting, were: in white wine (1) 2.5 × 106 cells mL−1S.c + 5.0 × 103 cells mL−1H.u, (2) 1.5 × 107 cells mL−1S.c + 1.0 × 105 cells mL−1H.u and in red wine (3) 1.0 × 105 cells mL−1S.c + 2.0 × 106 cells mL−1H.u, (4) 3.0 × 105 cells mL−1S.c + 1.0 × 104 cells mL−1H.u. Several dilutions of each mixed cultures were plated on two different media: the nonselective YEPD–agar medium and the selective lysine–agar medium (Oxoid), which does not support the growth of Saccharomyces species (Angelo & Siebert, 1987). DNA from each mixture was isolated as described above and quantified by QPCR.

Wine fermentations and sampling

Wine fermentations were carried out in the experimental cellar of the Oenology Faculty in Tarragona (Spain) during the 2006 vintage. Macabeo was the grape variety chosen for the vinifications. The grape must obtained was separated into two 100 L tanks after sulphitation (60 mg L−1) and settling. Both fermentations were performed spontaneously (no yeast inoculation). The only difference was that one was fermented at a low temperature (13°C) and the other was fermented at an optimum temperature (25°C). Samples were taken at different stages of wine fermentation. Each sample was plated on YEPD–agar and lysine–agar medium (Oxoid). DNA from each mixture was isolated as described above and quantified by QPCR to enumerate Hanseniaspora and Saccharomyces populations. These data were also complemented with the enumeration of total yeast population by QPCR, as described in a previous work (Hierro et al., 2006b).

Quantitative PCR assays

PCR amplification was performed in 25 μL final volume containing 5 μL of DNA, 200 nm of each respective primer and 12.5 μL of SybrGreen Master Mix (Applied Biosystems). All amplifications were carried out in optical-grade 96-well plates on an ABI Prism 5700 Sequence Detection System (Applied Biosystems) with an initial step at 95°C for 10 min, followed by 40 cycles of 95°C for 15 s, 60°C for 1 min and 72°C for 30 s. The CT was determined automatically by the instrument. All samples were analysed in triplicate. The coefficients of efficiency (E) were calculated using the formula E=(10−1/slope)−1 (Higuchi et al., 1993).

Results

Primer design, specificity and sensitivity of QPCR

The primers for the quantification of Saccharomyces and Hanseniaspora were designed from the 5.8S ribosomal gene and the ITS2. The forward primer was common for both groups of species and was homologous to a conserved region of the 5.8S ribosomal gene. Therefore, the specificity was determined by the reverse primer, complementary to specific sequences of Saccharomyces spp. and Hanseniaspora spp., respectively.

Primer specificity was determined by amplifying DNA extracted from the yeasts and bacteria listed in Table 1 with the designed primers. The Saccharomyces-specific primers yielded an amplicon of 175 bp with the S. cerevisiae and Saccharomyces bayanus strains tested but these primers were not able to amplify DNA from other wine yeasts, lactic acid bacteria or acetic acid bacteria. This level of specificity was also obtained for the Hanseniaspora-specific primers, which only yielded an amplicon of 121 bp for H. uvarum, H. osmophila H. guillermondii and Hanseniaspora vineae strains.

To determine the sensitivity and detection limits of the QPCR for the designed primers, DNA obtained from a S. cerevisiae 1942NT culture and an H. uvarum 1444T culture with concentrations of 106 cells mL−1 were serially diluted 10-fold. Each DNA dilution was used to construct a standard curve (Fig. 1). The assay was linear over six orders of magnitude and the detection limit was c. 10 cells mL−1 in both species. Although the CT values were higher for H. uvarum in the same cellular concentration, the values of efficiency of amplification were close to 100% in both species (105% and 108% for S. cerevisiae and H. uvarum, respectively).

1

Standard curves obtained from serially diluted Saccharomyces cerevisiae (▪) (y=−3.2018x+33.301; r2=0.9978) and Hanseniaspora uvarum (•) (y=−3.1429x+35.9; r2=0.9947) pure genomic DNA. CT values are the average of three repetitions. Bars represent SEs.

Quantification of Hanseniaspora and Saccharomyces

In the authors' opinion, wine samples must be quantified using standard curves constructed by cellular dilution rather than DNA dilution. Therefore, H. uvarum 1444T and S. cerevisiae 1942NT were used to generate species-specific standard curves. From overnight YEPD cultures, 10-fold serial dilution of these yeasts was performed and DNA from each dilution was isolated and quantified by QPCR using the corresponding species-specific primer set (Fig. 2). To check the putative inhibitory effect of the wine in the QPCR reaction, the same original cultures were incubated for 24 h in Macabeo and Cabernet Sauvignon wines. These cultures were also 10-fold serially diluted to construct standard curves from cells isolated from a complex matrix such as wine (Fig. 2). In all cases, the detection limit was c. 102 cells mL−1 and the assays were linear over five orders of magnitude. The correlation coefficients, slopes and efficiencies of the amplification for the standard curves are shown in Table 3. All values of efficiency were lower than those of the standard curves constructed by DNA dilution. This decrease in the efficiency of the QPCR reaction had to be taken as a lower efficiency of the DNA extraction process from the cell dilutions. At this point, an external reference was used to evaluate whether the efficiency of the DNA extraction was constant. Aliquots of L. kononenkoae 1967T containing 105 cells were added to each of the dilutions used to construct the standard curve. Each dilution was amplified using a Lipomyces-specific primer set and in all dilutions the expected CT value (CT=23.43±0.77, n=32) was obtained. DNA recovery was therefore uniform in all samples, which means that this DNA extraction process is both reliable and reproducible for the QPCR assay.

2

Species-specific standard curves obtained by QPCR from 10-fold serial dilution of (a) Saccharomyces cerevisiae grown in YEPD media (▪, —) and incubated in white wine (○, —) and red wine (□, —) and (b) Hanseniaspora uvarum grown in YEPD media (♦, —) and incubated in white wine (○, —) and red wine (□, —). Correlation coefficient (R2) values, slope and efficiency are shown in Table 3. Bars represent SEs.

View this table:
3

Correlation coefficient, slope and efficiency of standard curves obtained from serial dilutions of yeast cells grown in YEPD media and yeast cells incubated in white and red wine

MediaR2SlopeEfficiency(%)
S. cerevisiaeYEPD0.9968 ± 0.003−3.662 ± 0.10887.63 ± 3.49
White wine0.9641 ± 0.001−3.863 ± 0.01781.51 ± 0.46
Red wine0.9571 ± 0.005−3.934 ± 0.13979.66 ± 3.72
H. uvarumYEPD0.9771 ± 0.01−3.522 ± 0.20392.68 ± 7.6
White wine0.9922 ± 0.003−4.310 ± 0.04570.62 ± 0.95
Red wine0.9830 ± 0.001−3.986 ± 0.12378.27 ± 3.182
  • * Efficiency was estimated by the formula E=(10−1/slope)−1.

With regard to the effect of the matrix, the samples incubated in wine also showed a slight decrease in amplification efficiency. As no loss was detected of the viability of the cultures after the incubation in wine, this decrease in amplification efficiency could be explained by the presence of inhibitory compounds in the wine. Some major compounds of wine, such as polyphenols, are known to have an inhibitory effect on the PCR (Phister & Mills, 2003; Delaherche et al., 2004; Martorell et al., 2005). These authors reported that the inhibitory effect was stronger in red wines (which have a higher polyphenol concentration) than in white wines. However, this was not so in this case because a common matrix interference was detected that did not depend on the type of wine. This technique may therefore be useful to quantify Saccharomyces and Hanseniaspora populations in wine samples. The same white and red wines used for the standard curves were artificially contaminated with different mixtures of S. cerevisiae and H. uvarum ranging from 103 to 107 cells mL−1. After 24 h of incubation, samples were analysed by QPCR. The enumeration of Saccharomyces and Hanseniaspora populations from these mixed cultures was calculated using the standard curves generated from wine. The method's quantification effectiveness was tested by correlating the cell number concentration estimated by plate count and QPCR (Fig. 3). Each mixed culture was plated on two different media: the nonselective YEPD–agar medium and the selective lysine–agar medium, which does not support the growth of Saccharomyces. Therefore, the CFU mL−1 counted on lysine–agar medium are equivalent to Hanseniaspora colonies and the Saccharomyces CFU mL−1 are obtained by subtracting the CFU mL−1 counted on lysine–agar from the CFU mL−1 counted on YEPD. The results showed that the numbers obtained by QPCR did not significantly differ from those measured by plating.

Fig.3

Relationship between the results of yeast quantification by QPCR and plating of artificially contaminated white wine samples (□) and red wine samples (▴). CT values are averages of results from three replicates. Bars represent SEs.

Monitoring wine fermentations

The evolution of the Saccharomyces and Hanseniaspora populations during alcoholic fermentation was also analysed. Several authors have suggested that some non-Saccharomyces species, e.g. those of Hanseniaspora, have a better chance of growing at a low temperature than Saccharomyces (Sharf & Margalith, 1983; Heard & Fleet, 1988). To do so, the same grape must fermented at a low temperature (13°C) and an optimum temperature (25°C) was monitored. These temperatures of fermentation were routinely used to compare the effect of a low temperature in yeast metabolic activity (Novo et al., 2003; Torija et al., 2003; Beltran et al., 2006). Both fermentations proceeded spontaneously because no yeast inoculum was added. Samples from must, alcoholic fermentations and wines were plated on YEPD and lysine–agar media (Table 4) and quantified by QPCR (Table 5). The latter technique was used to enumerate the Saccharomyces and Hanseniaspora populations with the primers designed in this study but also to enumerate the total yeast population using the primers designed in a previous study (Hierro et al., 2006b).

View this table:
4

Enumeration by plating of yeast in samples from the same Macabeo must fermented at low (13°C) and optimum (25°C) temperature

YEPD-agarLysine-agar
13°C25°C13°C25°C
Must1.3 ± 0.1 × 1051.3 ± 0.4 × 105
Clear-must4.0 ± 1.3 × 1054.0 ± 1.1 × 1053.6 ± 3.5 × 1053.6 ± 2.1 × 105
IF1.5 ± 0.7 × 1072.9 ± 1.3 × 1078.5 ± 0.7 × 1059.3 ± 3.1 × 105
MF1.4 ± 0.7 × 1071.3 ± 0.6 × 1072.7 ± 0.2 × 1051.3 ± 1.1 × 105
FF9.5 ± 0.1 × 1069.5 ± 0.2 × 10605.8 ± 1.0 × 104
Wine4.9 ± 0.1 × 1061.9 ± 0.2 × 10600
  • * Must is the same for both fermentations (density 1095 g L−1).

  • Clear-must is must after settling and racking.

  • IF=initial fermentation (density of 1070 g L−1).

  • § MF=middle fermentation (density 1040 g L−1).

  • FF=final fermentation (density 1000 g L−1).

  • Wine <2 g L−1 of sugars.

View this table:
5

Quantification by QPCR of yeast in samples from the same Macabeo must fermented at low (13°C) and optimum (25°C) temperature

Total-YeastsSaccharomycesHanseniaspora
13°C25°C13°C25°C13°C25°C
Must9.5 ± 1.5 × 1063.9 ± 0.2 × 1031.2 ± 0.5 × 105
Clear-must3.7 ± 0.4 × 1066.7 ± 0.9 × 1066.7 ± 0.8 × 1053.3 ± 0.2 × 1066.2 ± 0.7 × 1056.6 ± 0.3 × 105
IF1.8 ± 0.1 × 1071.6 ± 0.2 × 1079.8 ± 0.5 × 1061.5 ± 0.2 × 1071.0 ± 0.7 × 1054.1 ± 0.9 × 104
MF1.1 ± 0.1 × 1071.1 ± 0.1 × 1076.7 ± 0.7 × 1061.2 ± 0.2 × 1078.9 ± 2.0 × 1044.0 ± 0.5 × 104
FF7.6 ± 1.0 × 1066.3 ± 0.1 × 1065.3 ± 0.9 × 1063.0 ± 0.2 × 1065.4 ± 0.8 × 1043.8 ± 0.7 × 104
Wine1.15 ± 0.2 × 1064.21 ± 0.6 × 1061.16 ± 0.2 × 1064.98 ± 1.7 × 1061.50 ± 0.3 × 1047.71 ± 2.0 × 103
  • * Must is the same for both fermentations (density 1095 g L−1).

  • Clear-must is must after settling and racking.

  • IF=initial fermentation (density of 1070 g L−1).

  • § MF=middle fermentation (density 1040 g L−1).

  • FF=final fermentation (density 1000 g L−1).

  • Wine<2 g L−1 of sugars.

A clear advantage of QPCR is that it enables microorganisms present at low population sizes to be enumerated (Table 5). This is the case of Saccharomyces, which represented approximately one-thousandth of the total population in the must sample. In this sample, the Hanseniaspora population was two orders of magnitude higher than the Saccharomyces population but this only represented c. 1% of the total population. Other non-Saccharomyces species are therefore more abundant in the must sample. The clear-must sample is the must after sulphitation, natural settling for 24 h and racking in the fermenting vats. This sample showed a decrease in the total yeast counts, perhaps because dead cells or unspecific DNA, which were amplified by the universal yeast primers, were removed during settling. However, sulphitation and settling led to a large increase in the Saccharomyces population and a slight increase in the Hanseniaspora population. As expected, the fermentation process led to increases in yeast populations that decreased in the last phase of fermentation and in the wine. This decrease is mostly a consequence of the toxic effects of the ethanol, together with the deposition of the yeasts in the bottom of the vat. This increase in yeast populations should be accounted for by the increase in the Saccharomyces population, which showed the same trend during alcoholic fermentation. The peak for the Saccharomyces population led to a decrease in the Hanseniaspora population, although a low concentration of Hanseniaspora was detected throughout fermentation. At this point, it should be stressed that as the QPCR technique does not discriminate viable cells from dead cells, it is a good idea to compare the QPCR and plating counts.

The total yeast counts of the must and clear-must samples were much higher with QPCR than with plating (YEPD–agar). This may be due to a large presence of viable but nonculturable populations in the must or due to the overestimation of yeasts by the amplification of DNA from dead yeast cells. Both of these arguments may also explain the inability of the Hanseniaspora population, detected by QPCR in the final fermentation and wine samples, to grow in lysine–agar medium. The inability of the assay to discriminate dead and living cells is indeed its major constraint.

Regarding the effect of temperature on yeast populations, although not very important differences were detected, a low fermentation temperature led to a lower proportion of Saccharomyces in favour of the non-Saccharomyces population. This was confirmed by the larger Hanseniaspora populations in the low-temperature samples. However, this higher number in the Hanseniaspora population at a low temperature was not confirmed by plating.

Discussion

Rapid and sensitive methods for detecting and enumerating yeasts are needed in the wine industry to enable winemakers to make decisions to control and avoid spoilage of their products. QPCR is a promising technique for quantifying microorganisms associated with food. This technique was recently used to detect and enumerate the total number of yeasts in wine samples (Hierro et al., 2006b). In the present study, Saccharomyces- and Hanseniaspora-specific primers were designed to enumerate these important species in the wine-making processes. The primers were designed from the ITS2 and the 5.8S rRNA gene. The forward primer, selected from the 5.8S rRNA gene, was common to both species. The reverse primers were homologous to the ribosomal intergenic spacer region (ITS2). These intergenic spacers of the ribosomal genes are first transcribed in a 37S ribosomal precursor RNA (including the 18S, 5.8S and 28S rRNA genes and the two intergenic spacers ITS1 and ITS2) and then processed and eliminated from the mature rRNAs. This means that the ITS regions are less evolutionarily conserved than the rRNA-coding genes and that they are therefore valuable regions for finding interspecific differences (Guillamón et al., 1998).

Comparison of the specificity, sensitivity and efficiency of these specific primers and of the universal yeast primers designed in a previous study revealed some similarities and some differences. The specific primers amplified for the intended group of species and none of the PCR assays was positive for any other wine microorganisms of those listed in Table 1. QPCR analysis with both Saccharomyces- and Hanseniaspora-specific primers efficiently enumerated cells at concentrations as low as 10 cells mL−1 when the standard curve was constructed by DNA dilution and at concentrations as low as 102 cells mL−1 when the standard curve was constructed from cells incubated in wine. This sensitivity is higher than that for the universal yeast primers in which the detection limit was 103 cells mL−1 for wine samples (Hierro et al., 2006b). One similarity with a previous study is that the amplification efficiency was lower in the samples from wine. This matrix interference justified the construction of standard curves from cells incubated in wine. In this study, the efficiency of the cell lysis and DNA extraction was also tested by adding the same cell concentration of the yeast L. kononenkoae to the various samples analysed. The similar CT values obtained for this external reference showed that the DNA extraction method was good. All of this enabled a good correlation between the predicted number of cells per millilitre, as determined by QPCR, and the number of CFU per millilitre, as determined by plating, which proves that the assay is reproducible and highly robust.

In this study, this sensitive and independent-culture method has been used to monitor the evolution of the major fermentative species, Saccharomyces spp., and that of the major non-Saccharomyces species, Hanseniaspora spp., during industrial alcoholic fermentations. Non-Saccharomyces yeasts may affect wine fermentations directly, by producing off-flavours, and indirectly, by modulating the growth or metabolism of the dominant Saccharomyces population (Mills et al., 2002). Although five Hanseniaspora species have been isolated from wine (H. uvarum, H. guilliermondii, Hanseniaspora occidentalis, H. osmophila and H. vineae) (Capece et al., 2005), H. uvarum (and its anamorph Kloeckera apiculata) has been widely reported as a major non-Saccharomyces yeast during the initial stages of wine fermentation (Pretorius, 2000). Some strains of this apiculate yeast have shown positive oenological properties and it has been suggested that they can be used in grape-must fermentations to enhance the aroma and flavour profiles of wines (Romano et al., 2003).

Most of these ecological studies were based on the isolation and identification of colonies plated on general or selective solid media. The limitations of plating are that it is difficult to isolate minor species against major species and that populations subjected to strong stress conditions may go into a viable but nonculturable state that prevents their detection (Millet & Lonvaud-Funel, 2000). In fact, the current application of direct molecular methods for ecological analysis has revealed the tremendous diversity of microorganisms in various habitats that were not previously detected by plating methods (Pace, 1997) even in the oenological environment. Mills (2002) used RT-PCR-denaturing gradient gel electrophoresis analysis to detect an active H. osmophila population of over 104 cells mL−1 throughout fermentation that had gone unnoticed by CFU analysis.

In the present study, the correlation between quantification by plating and quantification by QPCR in alcoholic fermentation samples was good but with some divergences. The total yeast counts of the must and clear-must samples were much higher with QPCR than with plating (YEPD-agar). A comparison of Hanseniaspora quantification by QPCR and non-Saccharomyces quantification in lysine–agar showed that Hanseniaspora represented practically 100% of the non-Saccharomyces strains in the must and clear-must samples, but this relative importance decreased in the fermentation samples. No CFU was detected in samples of the final fermentation or wine in the lysine–agar medium, whereas a Hanseniaspora population of c. 104 cells mL−1 was quantified by QPCR. In a previous study (Hierro et al., 2006b), the reverse transcription of the 26S rRNA gene was used as a template for the QPCR reaction as a better indicator of cell viability than the detection of DNA (Sheridan et al., 1998). However, this strategy cannot be used with the primers designed in this study. The reverse specific primers for both species were complementary to sequences of the ITS2 and this region is rapidly processed and eliminated from the mature rRNA. It is therefore impossible to obtain a reverse transcript or cDNA of this region for further amplification. Currently, a promising and easy-to-use alternative method to RNA-based quantification is being assayed that uses the DNA-intercalating dye ethidium monoazide bromide (Nocker & Camper, 2006).

It should be stressed that the Saccharomyces population could be quantified from the must to the wine. This population has so far been difficult to quantify by plating in the first phases of fermentation because of the major presence of the non-Saccharomyces strains. The results also showed that Saccharomyces strains were the most competitive ones in the process, increasing their population by four orders of magnitude in a few days. The toxicity of ethanol and other metabolites was also revealed by a decrease in this population in the later phases of the process. The increase in the Saccharomyces population also matched a large decrease in the Hanseniaspora population, as has been widely reported (Nissen & Arneborg, 2003; Nissen et al., 2003; Arneborg et al., 2005; Pérez-Nevado et al., 2006).

Several authors have suggested that some non-Saccharomyces species, e.g. those of Hanseniaspora, have a better chance of growing at a low temperature than Saccharomyces (Sharf & Margalith, 1983; Heard & Fleet, 1988). The present results support this hypothesis because the counts of Saccharomyces were always higher at 25°C, while the counts of Hanseniaspora were higher at 13°C. However, other non-Saccharomyces species may also have been actively growing in the low-temperature fermentation because the decrease in the Saccharomyces population was not accounted for by the increase in the Hanseniaspora population. Low-temperature fermentation has been associated with an enhancement of aroma in the final wines. It has recently been shown that low temperatures modify the metabolism of Saccharomyces, thus increasing the synthesis of some aroma compounds (Beltran et al., 2006). However, this aroma enhancement may also be explained by a major contribution of the non-Saccharomyces species throughout wine fermentation.

In conclusion, QPCR is a fast, direct (without culture), sensitive and reliable technique for quantifying wine yeasts. This method has so far been applied for the enumeration of total yeasts (Hierro et al., 2006b) and individual yeasts (Phister & Mills, 2003; Martorell et al., 2005; Rawsthorne & Phister, 2006). The design of new primers for other important wine yeast species will enable yeast diversity to be monitored during industrial wine fermentation and determine the main spoilage yeasts in wine. An important drawback of quantification using DNA as the template, however, is that live and dead cells are not differentiated. Therefore, to overcome this problem alternative strategies are needed in future studies.

Acknowledgements

This work was supported by grants AGL2004-02307/ALI and AGL2004-07494-C02-02/ALI from the Spanish Government. The authors thank the Language Service of the Rovira i Virgili University for revising the manuscript.

References

View Abstract