Evaluation of the inherent capacity of ... - Wiley Online Library

194

Release of glycosidic aroma precursors

Australian Journal of Grape and Wine Research 21, 194–199, 2015

Evaluation of the inherent capacity of commercial yeast strains to release glycosidic aroma precursors from Muscat grape must A. BISOTTO1, A. JULIEN1, P. RIGOU2, R. SCHNEIDER3 and J.M. SALMON2,4 1

Lallemand SAS, Blagnac, France Sciences Pour l’Œnologie, Institut National de la Recherche Agronomique, Montpellier, France 3 Unité Mixte Technologique Qualinnov, Institut Français de la Vigne et du Vin, Gruissan, France 4 Unité Expérimentale de Pech Rouge (UE0999), Unité Mixte Technologique Qualinnov, Institut National de la Recherche Agronomique, Gruissan, France Corresponding author: Dr Jean-Michel Salmon, email [email protected] 2

Abstract Background and Aims: The aim of this work was to establish a general method for evaluating the intrinsic activity of yeasts towards the release of volatiles terpenes from glycosidic aroma precursors. Methods and Results: The enzymatic capacity of yeast cells towards glycosidic aroma precursors was assessed on permeabilised yeast cells. In this way, the pores created into the cell membrane allowed the precursors to freely enter the cells. Conclusions: The comparison between the release of terpenes from glycosidic aroma precursors by entire viable cells during fermentation and by permeabilised cells shows that the permeation of glycosidic aroma precursors through the yeast cell membrane is the main bottleneck for such release. Significance of the Study: This work highlights the fact that Saccharomyces species can exhibit β-glycosidase activity as high as non-Saccharomyces species. This important result points out future prospects for the improvement of Saccharomyces species by classical genetics. Keywords: aroma precursor, terpene, wine, yeast

Introduction During wine fermentation, yeasts are responsible for the transformation of some aroma precursors into aromatic compounds. It is now well established that certain monoterpenols of grapes, for example linalool, geraniol, nerol, citronelol, α-terpineol and linalool oxide, which are linked to diglycosides, such as 6-O-α-L-rhamnopyranosyl-, 6-O-α-L-rhamnofuranosyl- and 6-O-β-D-apiofuranosyl-β-D-glucosides, contribute significantly to the flavour of wine (Günata et al. 1985, Williams et al. 1989, Schneider et al. 2001). The enzymatic hydrolysis of these compounds requires a sequential reaction: first, an α-Lrhamnosidase or a β-D-apiofuranosidase cleaves the (1→6)glycosidic linkage, and then, the flavour compounds are liberated from the monoglucosides by the action of a βglucosidase (Günata et al. 1988, 1993). This cleavage can also occur during wine ageing, under mild acidic conditions. Unlike acid hydrolysis, enzymatic hydrolysis is efficient and does not result in modification of the aromatic composition of the bound fraction (Günata et al. 1990). Yeast glucosidases, however, exhibit limited activity on monoterpenyl glycosides during winemaking, and a large fraction of the aromatic precursors remains unprocessed (Delcroix et al. 1994, Rosi et al. 1994, Ugliano et al. 2006). Moreover, Gil et al. (2005) have shown that induced overproduction of an endogenous exoglucanase in a Saccharomyces cerevisiae strain led to an increase in the release of glycoside-related volatile compounds in wine, suggesting the involvement of other enzymatic activities in the hydrolysis of glycosides by yeast. Furthermore, the work of Ugliano et al. (2006) indicates that production and/or activity of enzymes doi: 10.1111/ajgw.12127 © 2015 Australian Society of Viticulture and Oenology Inc.

specific for apioside substrates might be limited in Saccharomyces yeast during fermentation, at least in the three strains tested by these authors. The terpene profile of Muscat wines fermented by Saccharomyces species and hybrid yeasts was also investigated (Gamero et al. 2011). These authors did not find any relationship between β-d-glucosidase activity and the terpene profile in Muscat wines fermented with Saccharomyces species and hybrids. The release or formation of aroma compounds from precursors was found to be strongly linked to the hybrid used, and, for example, the triple hybrid S. cerevisiae × S. bayanus × S. kudriavzevii in particular and secondarily the hybrid S. cerevisiae × S. bayanus were highly efficient in the production of most varietal terpenols (Gamero et al. 2011). Altogether, these results suggest that there are two main limiting factors for a complete release of monoterpenols from grape glycosidic precursors by yeasts. The first bottleneck deals with the transport of the precursors into the yeast cells. Aroma glycosidic precursors are large molecules, which cannot travel through the plasma membrane by simple diffusion in intact viable cells. Moreover, Darriet et al. (1988) reported that the β-d-glucosidase activity is located in the periplasmic space of yeast cell, presumably easily liberated during yeast autolysis. In addition, it has also been found that lees from different yeast strains may have a slightly different ability to release volatile compounds derived from glycosidic precursors (Loscos et al. 2009). To our knowledge, however, no scientific data are available on the specific transport of such molecules in yeasts. The second point deals with the activity of the enzymes themselves.

Bisotto et al.

Previous work has established by working on entire yeast cells that only non-Saccharomyces species exhibited β-glycosidase activity (Mendes Ferreira et al. 2001). In contrast, other work has revealed that entire Saccharomyces species cells could exhibit a low β-glycosidase activity (Delcroix et al. 1994, Zoecklein et al. 1997, Ugliano et al. 2006). Therefore it is possible that some yeast strains that exhibit efficient enzymatic β-glycosidase activity may remain unable to take up the glycosidic precursors as substrates. The aim of this work is to establish a general method for evaluating the intrinsic activity of yeasts towards the release of glycosidic aroma precursors. For this purpose, we use the technique of cell permeabilisation (Salmon 1984) to create pores into the cell membrane allowing the precursors to freely enter the cells, where the enzymatic release of aglycone terpenes by yeast enzymatic activity can take place. In this way, the enzymatic capacity of the yeast cells can be easily assessed.

Material and methods Grape glycosidic precursor purification and analysis Preparation of a glycosidic fraction from Muscat must. Juice was extracted by crushing mature Muscat of Frontignan grapes (17°Brix) harvested from the experimental vineyard of INRA-Pech Rouge (Gruissan, France) with a blender. After centrifugation (8000 × g), Muscat grape juice (200 mL) was eluted through a XAD-2 column (Sigma-Aldrich Chimie, Lyon, France), followed by washing with water (100 mL), then pentane/dichloromethane (2/1 v/v, 100 mL) to remove the free fraction of terpenes. The bound glycosidic fraction was recovered by elution with 100 mL methanol (Rosi et al. 1995) and stored at −18°C.

Enzymatic hydrolysis of the Muscat glycosidic fraction. The glycosidic fraction was dried under vacuum, and the residue was then solubilised in 2 mL of phosphate citrate buffer (sodium hydrogen phosphate 0.2 mol, citric acid 0.1 mol, pH 5.0). An aliquot (200 μL) of an enzymatic preparation (AR2000 at 70 mg/mL, DSM Food Specialties, Heerlen, the Netherlands) in citrate phosphate buffer was added. After mixing, the reaction occurred at 40°C for 16 h.

Permeabilised

cell assay. Permeabilised yeast cells (4.5 × 108) were added to the glycosidic fraction previously dried under vacuum and solubilised in 2-mL citrate phosphate buffer. After mixing, the reaction occurred at 40°C or at room temperature for 1, 4, 16 or 24 h. A blank experiment was realised by replacing the permeabilised cells by the same volume of imidazole buffer (75 mmol, pH 7.5). Extraction of aglycone terpenes. After enzymatic hydrolysis, or after incubation with permeabilised yeast cells, the volatile fraction was extracted by five times 2 mL of pentane/ dichloromethane (2/1 v/v) azeotrope. The organic extract was then dried on anhydrous sodium sulfate. 4-Nonanol (6 µg) was added as an internal standard. The extract was then concentrated to about 400 μL by partial rectification at 35°C using a Dufton’s spiral column. The extract was maintained at −20°C until analysis by GC/MS.

Analysis of the terpenic aglycones by GC/MS. The aglycone extract was analysed with a Hewlett-Packard (HP) 5890 Series II GC system coupled to a HP 5989 A MS. The samples were injected in splitless mode (injector port tempera© 2015 Australian Society of Viticulture and Oenology Inc.


195

ture 245°C; purge on time 0.5-min) onto a DB-Wax column [30 m × 0.25 mm id, 0.25 μm film thickness (Agilent Technologies, Santa Clara, CA, USA)]. Compounds were separated using helium carrier gas at 1 mL/min. The temperature program began with an isotherm at 60°C for 3 min. The temperature of the oven was then raised by 3°C/min to 245°C and held for 10-min. The transfer line was held at 250°C, and compounds were detected with the source held at 150°C by ionisation by electronic impact generated at 70°C. Full scan mass spectra were recorded between 29 and 350 m/z. Data were acquired and treated with the HP 5989 B.05.02 MS Chemstation. The terpenes identified were semi-quantified using 4-nonanol as an internal standard. Limits of detection and of quantification of the method for the terpenes considered are given in Table 1.

Yeast strains All tested yeast strains were commercially available active dry yeasts (ADY) for winemaking (Lallemand, Montreal, QC, Canada): five S. cerevisiae strains (Lalvin QA23, Lalvin ICV K1M, Vitilevure DV10, Uvaferm CEG, Lalvin EC-1118); a natural hybrid of S. cerevisiae and S. uvarum (Lalvin S6U), and two non-Saccharomyces enological strains: Biodiva (Torulaspora delbrueckii) and Flavia (Metschnikowia pulcherrima).

Growth conditions The medium (YEP-20) for yeast cell growth contained yeast extract 10 g/L, peptone 20 g/L, glucose 20 g/L (pH adjusted to 3.5). Yeasts were cultured in 100 mL Erlenmeyer flasks containing 20 mL of growth medium. The flasks were directly inoculated with active dry yeasts according to the manufacturer’s recommendations. Yeast cultures were incubated overnight at 28°C with agitation (180 rpm). The β-glucosidase activity of entire viable yeast cells was evaluated in a specific synthetic culture medium. This medium (MS-50) contained 1.7 g/L yeast nitrogen base (YNB with amino acids, Becton Dickinson and Company, Franklin Lakes, NJ, USA), 5 g/L ammonium sulfate, 50 g/L glucose, 3 g/L citric

Table 1. Terpene content of the glycosidic fraction from Muscat must, which was obtained by GC-MS analysis of the volatile fraction after hydrolysis by the AR2000 enzymatic preparation. Detected compounds

Linalool Hotrienol α-Terpineol Nerol Geraniol Linalool hydrate Z-8-Hydroxylinalool trans-Pyranic linalool oxide 3,6-Terpendiol 3,7 Terpendiol 3,8-Terpendiol

4-Nonanol equivalents (μg)

LOD (μg)

LOQ (μg)

6.70 2.95 1.21 3.75 7.10 1.95 2.90 1.28 1.45 13.40 2.15

0.01 0.04 0.03 0.006 0.015 0.06 0.04 0.01 0.02 0.04 0.02

0.04 0.1 0.1 0.02 0.05 0.2 0.1 0.03 0.06 0.1 0.07

LOD (limits of detection) and LOQ (limits of quantification) of the method were calculated using the equations: LOD = 3*Hmax*R and LOQ = 10*Hmax*R, where Hmax is the maximum height of the noise and R is the response factor.

196


acid and 3 g/L malic acid (pH adjusted to 3.5 with concentrated KOH). This synthetic medium was supplemented with the solution of glycosidic precursors [dried under vacuum and redissolved in citrate phosphate buffer (sodium hydrogen phosphate 0.2 mol, citric acid 0.1 mol, pH 5.0)]. Ten millilitres of this synthetic medium were inoculated with 9 × 108 yeast cells and incubated in 50 mL-Erlenmeyer flasks for 36 h at 28°C. The synthetic medium was incubated without yeast inoculation as a blank.

Cell permeabilisation Cells were permeabilised according to the method described by Salmon (1984). Ten millilitres of culture were filtered through a Whatman filter GF/C type, and washed with 6 mL of ice-cold water. The cells retained on the filter were then re-suspended in 1 mL of imidazole buffer (75 mmol), KCl 0.1 mol, MgCl2 100 mmol (pH 7.5 using HCl). Fifty microlitres reduced glutathione (0.3 mol), 10 μL Triton X-100 (10%) and 50 μL of toluene/ethanol (1/4 v/v) were then added to the suspension, which was vigorously shaken by machine for 5 min. After mixing, the suspension was filtered, and the cells were washed with 6 mL of ice-cold water. The permeabilised cells were re-suspended in 3 mL of imidazole buffer. This suspension was kept on ice, and could be used for enzymatic determination within 6 h (Salmon 1984).

Enzyme assays β-Glucosidase activity was determined according to the methodology of Delcroix et al. (1994). One mL of permeabilised cells was added to 1 mL of ρ-nitrophenyl-β-D-glucopyranoside (8 mmol) in sodium acetate buffer (100 mmol, pH 5). After incubation at 40°C, the reaction was stopped by addition of 6 mL of Na2CO3 (1 mol). Absorbance was read at 400 nm, and activity was expressed as picokatals for 108 cells (pkat/108 cells).

Alcohol dehydrogenase activity. Permeabilised cells (50 to 200 μL) were added to 1 mL of glycine/NaOH buffer (75 mmol, pH 9), 100 μL NAD+ (27.7 mmol), 50 μL reduced glutathione (0.3 mol) and 50 μL pure ethanol. The cells were incubated at 28°C, and absorbance at 340 nm was recorded continuously for 10 min. Activity was expressed as nanokatals for 108 cells (nkat/ 108 cells). Analytical methods Cell enumeration. Yeast culture samples were first diluted (1000 to 2500 times) with Isoton II (Beckman-Coulter, Margency, France). After sonication (35 s, 10 W), cells were counted with a Coulter Z2 electronic counter (Beckman Coulter, Fullerton, CA, USA), fitted with a 100-μm aperture probe.

Results and discussion Efficiency of the cell permeabilisation technique The efficiency of permeabilisation was assessed by measuring the alcohol dehydrogenase (ADH) activity of all the yeast strains tested after permeabilisation. All the permeabilised cells, except DV10 and Biodiva strains, exhibited a detectable ADH activity (Table 2) under the conditions of the experiment. The measurement of ADH activity in the permeabilised cells confirmed that the method used for cell permeabilisation was appropriate, except for the DV10 and Biodiva strains. Resistance to the permeabilisation procedure was already observed in previous


Table 2. Alcohol dehydrogenase activity measured in permeabilised cells of eight strains of commercially available active dry yeasts for winemaking. ADH activity (nkat/108 cells)

Strain

EC-1118 K1M QA23 Flavia CEG S6U DV10 Biodiva

Mean

Standard deviation

10.02 8.85 6.11 5.08 2.02 2.05 0.31 0.19

0.47 0.33 0.42 0.52 0.65 0.53 0.23 0.25

Mean and standard deviation of four experiments (nkat/108 cells). ADH, alcohol dehydrogenase; Biodiva, Torulaspora delbrueckii; EC-1118, K1M, QA23, CEG and DV10, Saccharomyces cerevisiae; Flavia, Metschnikowia pulcherrima; SU6, natural hybrid of S. cerevisiae.

Table 3. β-Glucosidase activity measured in the permeabilised cells of eight strains of commercially available active dry yeasts for winemaking. β-Glucosidase activity (pkat/108 cells)

Strain

Flavia K1M QA23 EC-1118 CEG S6U DV10 Biodiva

Mean

Standard deviation

69.85 21.90 22.05 20.02 19.95 20.08 10.05 6.98

0.75 0.45 0.39 0.59 0.55 0.62 0.25 0.15

Mean and standard deviation of three experiments (pkat 10−8 cells). Substrate was ρ-nitrophenyl-β-D-glucopyranoside. Biodiva, Torulaspora delbrueckii; EC-1118, K1M, QA23, CEG and DV10, Saccharomyces cerevisiae; Flavia, Metschnikowia pulcherrima; SU6, natural hybrid of S. cerevisiae.

work on wine yeast strains, and was mainly attributed to a specific conformation of the cell membrane (Salmon 1984, 1986). This could be the case in the present study for the T. delbrueckii Biodiva strain.

β-Glucosidase activity in the permeabilised cells

β-Glucosidase activity was measured with the artificial substrate, ρ-nitrophenyl-β-D-glucopyranoside (Table 3). The results obtained are consistent with the previously observed low level of permeabilisation of the DV10 and Biodiva strains. The other Saccharomyces strains tested exhibited quite low β-glucosidase activity of about 20 pkat/108 cells, similar to that encountered previously (Delcroix et al. 1994). In contrast, the Flavia strain (M. pulcherrima) exhibited a greater β-glucosidase activity of about 70 pkat/108 cells. Although previous work with entire yeast cells showed that only non-Saccharomyces species exhibited β-glycosidase activity (Mendes Ferreira et al. 2001), we have demonstrated by using permeabilised cells that Saccharomyces species also exhibit β-glycosidase activity. © 2015 Australian Society of Viticulture and Oenology Inc.

Bisotto et al.


Efficiency of permeabilised cells to release terpenes from a glycosidic Muscat extract

Since some β-glucosidase activity was found in most of the permeabilised cells tested, the efficiency of permeabilised cells to release aglycone terpenes from the glycosidic fraction of a Muscat must was then assayed. The total aromatic potential of the glycosidic fraction was first evaluated by analysis of the volatile fraction obtained after hydrolysis by the AR2000 enzymatic preparation (Table 1). Second, permeabilised yeast cells were incubated for 20 h in the presence of the same glycosidic fraction at either ambient temperature or at 40°C. The increase in temperature of incubation did not change either the profile or the quantity of terpenes released but affected only the observed release rates (data not shown). Only five of the six strains tested hydrolysed the glycosidic fraction and liberated terpenes

197

(Table 4). The permeabilised cells of strain S6U, which is a natural hybrid of S. cerevisiae and S. uvarum, were unable to release any bound terpenes under the experimental conditions tested. A great variability was observed in the capacity of the different permeabilised strains to release specific classes of bound terpenes during the experiment (Figure 1). Considering the extent and breadth of terpenes released, the permeabilised cells of Flavia appeared to possess the greatest glycoside hydrolase activity among the tested strains (Table 4 and Figure 1). Globally, the monoterpendiols, as 3,6 and 3,7 monoterpendiols, were the terpenes most released by the tested strains in their permeabilised state. Interestingly, most of these odourless molecules can be transformed into other aromatic molecules during the ageing of wine by many reactions occurring in acid medium, such as isomerisation, ring closure,

Table 4. Terpenes revealed by permeabilised yeast (after 16 h of incubation at 40°C) or entire viable yeast cells (after 36 h of fermentation at 28°C) from the glycosidic fraction of a Muscat must. Detected compounds

Yeast strains QA23

Linalool Hotrienol α-Terpineol Nerol Geraniol Linalool hydrate Z-8-Hydroxylinalool trans-Pyranic linalool oxide 3,6-Terpendiol 3,7-Terpendiol 3,8-Terpendiol

K1M

EC-1118

CEG

S6U

Flavia

PC

EC

PC

EC

PC

EC

PC

EC

PC

EC

PC

EC

4.71 25.0 ND 26.3 20.4 28.0 72.3 ND ND 26.4 ND

ND ND ND ND ND ND ND ND ND ND ND

ND 87.5 ND ND ND 69.2 41.0 ND 100.0 53.0 100.0

ND 23.0 ND ND ND ND ND ND ND 8.7 ND

ND 61.3 ND 8.2 8.9 48.2 48.2 ND 52.6 35.2 ND


ND 36.5 ND 26.6 35.0 100.0 59.0 ND 100.0 100.0 93.1




49.7 9.1 82.9 40.1 39.9 25.5 94.5 100.0 100.0 11.6 99.5


EC, entire viable cells; ND, not detectable; PC, permeabilised cells. Results are expressed as a proportion of the total aglycone terpene potential revealed by hydrolysis with the AR2000 enzymatic preparation (see Table 1).

Figure 1. Monoterpenols ( , ), monoterpenol oxides ( ) and monoterpendiols ( , ) revealed by permeabilised yeast cells ( , , ) (16 h incubation at 40°C) (plain boxes) and entire yeast cells ( , ) (36 h alcoholic fermentation at 28°C) (dashed boxes) from the glycosidic fraction of Muscat grape juice, expressed as a proportion of the total aglycone terpene potential revealed by hydrolysis with the AR2000 enzymatic preparation (see Table 1).

© 2015 Australian Society of Viticulture and Oenology Inc.

198


hydration, dehydration and oxidation (Rapp et al. 1985). The other major terpenes released were mainly monoterpenols and monoterpene oxides, with a high efficiency observed with the Flavia strain. The observed release of monoterpenol oxides was more surprising. There are two pathways that may explain their origin: first, some were already present as their glycosylated forms in the original Muscat extract as shown previously (Table 1); second, yeast metabolism could oxidise a produced terpene. As far as is known, however, there are only a few descriptions of yeast-mediated terpene bio-oxidation processes (Bicas et al. 2009), mainly by Candida tropicalis, Trichosporum cutaneum and by ale and lager yeasts. The present study does not allow a final conclusion to be drawn on the predominant pathway for their formation.

Comparison between permeabilised cells and entire viable cells In order to determine whether cell permeabilisation influenced the release of bound terpenes from the glycosidic fraction of Muscat must, non-permeabilised yeasts were tested for their ability to release aglycone terpenes on the same glycosidic extract as above, during a 36-h alcoholic fermentation. Under these fermentative conditions, yeast strains were not able to reveal the same cultivar of compounds as permeabilised cells (Table 4 and Figure 1). Only hotrienol and 3,7-monoterpendiol could be detected after fermentation by some strains (K1M, EC 1118, CEG and Flavia) and at a concentration much lower than that observed with permeabilised cells. This result challenges the previous findings of Ugliano et al. (2006) who considered that hotrienol was partially formed through acidcatalysed hydrolysis of either a glycosidic precursor or aglycons released by yeast glycosidases. The comparison between the amount of aglycone terpenes released by whole and permeabilised cells (Figure 1) clearly shows the benefit of permeabilisation. Indeed, a considerable increase in the proportion of aglycone terpenes released with permeabilised cells was observed. These results suggest that the transport of the glycosidic precursors into the cell is an important limiting factor for commercial yeast strains to reveal glycosidic aroma precursors from grape must.

Conclusion The aim of this work was to study the inherent capacity of commercial wine yeast strains to reveal glycosidic aroma precursors from Muscat grape must by using permeabilised cells. Although a few yeast strains appeared to resist the permeabilisation treatment, results particularly highlight the great potential of this technique to improve the actual enzymatic capacity of yeast strains to release terpenes from their glycosidic forms. Surprisingly, some yeast strains appeared to be able to release the free odourous monoterpenols, while other release mainly odourless monoterpendiols. The results obtained confirmed that no concomitant hydrolysis of complex disaccharide glycosides associated with enzymatic activities different from β-glucosidase (e.g. α-arabinosidase, α-rhamonosidase, β-apiosidase) occurred in the release of glucosides, as previously demonstrated by Ugliano et al. (2006). The results suggest that the transport of glycosidic precursors is a limiting factor in this aromatic release, which is of specific physiological interest and could orientate breeding programs for the construction of new interspecific wine yeasts. Traditional breeding techniques, mainly based on breeding between S. cerevisiae or closely related strains, were previously used in the development of new yeast strains with altered phenotypic characteristics for winemaking (Romano et al. 1985, Chambers et al.


2009). The generation of an interspecific hybrid, however, between a commercial S. cerevisiae wine yeast strain and S. mikatae, a species not previously associated with alcoholic fermentation, was recently described for developing novel yeast strains that bring greater complexity to wine than strains currently available to the industry (Bellon et al. 2013). A similar approach could be used to improve genetically commercial Saccharomyces wine yeast strains by complementing their inherent glycoside hydrolase enzymatic activity with the ability to transport actively glycosidic aroma precursors from the grape must. This will provide additional tools for winemakers to develop new wine styles.

Acknowledgements The authors thank Jean-Paul Lepoutre and Nicolas Bouvier for their technical assistance during aroma analysis. The stay of Alexandra Bisotto was funded by Lallemand SAS.

References Bellon, J.R., Schmid, F., Capone, D.L., Dunn, B.L. and Chambers, P.J. (2013) Introducing a new breed of wine yeast: interspecific hybridisation between a commercial Saccharomyces cerevisiae wine yeast and Saccharomyces mikatae. PLoS ONE doi: 10.1371/journal.pone.0062053. Bicas, J.L., Dionı’sio, A.P. and Pastore, G.M. (2009) Bio-oxidation of terpenes: an approach for the flavor industry. Chemical Review 109, 4518–4531. Chambers, P.J., Bellon, J.R., Schmidt, S.A., Varela, C. and Pretorius, I.S. (2009) Non-genetic engineering approaches for isolating and generating novel yeasts for industrial applications. Satyanarayana, T. and Kunze, G., eds. Yeast biotechnology: diversity and applications (Springer: New York, USA) pp. 431–455. Darriet, P., Boidron, J.N. and Dubourdieu, D. (1988) L’hydrolyse des hétérosides terpèniques du Muscat à petits grains par les enzymes périplasmiques de Saccharomyces cerevisiae. Connaissance Vigne Vin 22, 189–195. Delcroix, A., Günata, Z., Sapis, J.C., Salmon, J.M. and Bayonove, C. (1994) Glycosidase activities of three enological yeast strains during winemaking – effect on the terpenol content of muscat wine. American Journal of Enology and Viticulture 45, 291–296. Gamero, A., Hernández-Orte, P., Querol, A. and Ferreira, V. (2011) Effect of aromatic precursor addition to wine fermentations carried out with different Saccharomyces species and their hybrids. International Journal of Food Microbiology 147, 33–34. Gamero, A., Manzanares, P., Querol, A. and Belloch, C. (2011) Monoterpene alcohols release and bioconversion by Saccharomyces species and hybrids. International Journal of Food Microbiology 145, 92–97. Gil, J.V., Manzanares, P., Genovés, S., Vallés, S. and Gonzalez Candelas, L. (2005) Over-production of the major exoglucanase of Saccharomyces cerevisiae led to an increase in the aroma of wine. International Journal of Food Microbiology 103, 57–68. Günata, Z., Bayonove, C., Baumes, R. and Cordonnier, R. (1985) The aroma of grapes. Extraction and determination of free and glycosidically bound fractions of some grape aroma components. Journal of Chromatography 331, 83–90. Günata, Z., Bitteur, S., Brillouet, J.M., Bayonove, C. and Cordonnier, R. (1988) Sequential enzymic hydrolysis of potentially aromatic glycosides from grape. Carbohydrate Research 184, 139–149. Günata, Z., Bayonove, C., Tapiro, C. and Cordonnier, R. (1990) Hydrolysis of grape monoterpenyl β-glucosides by various β-glucosidases. Journal of Agricultural and Food Chemistry 38, 1232–1236. Günata, Z., Dugelay, I., Sapis, J.C., Baumes, R. and Bayonove, C. (1993) Role of the enzymes in the use of the flavour potential from grape glycosides in winemaking. Schreier, P. and Winterhalter, P., eds. Progress in flavour precursor studies (Allured Publishing Corporation: Carol Stream, IL, USA) pp. 219–234. Hernandez-Orte, P., Cersosimo, M., Loscos, N., Cacho, J., Garcia Moruno, E. and Ferreira, V. (2008) The development of varietal aroma from nonfloral grapes by yeasts of different genera. Food Chemistry 107, 1064– 1077. Loscos, N., Hernandez-Orte, P., Cacho, J. and Ferreira, V. (2007) Release and formation of varietal aroma compounds during alcoholic fermentation from nonfloral grape odorless flavor precursors fractions. Journal of Agricultural and Food Chemistry 55, 6674–6684. © 2015 Australian Society of Viticulture and Oenology Inc.

Bisotto et al. Loscos, N., Hernandez-Orte, P., Cacho, J. and Ferreira, V. (2009) Fate of grape flavor precursors during storage on yeast lees. Journal of Agricultural and Food Chemistry 57, 5468–5479. Mendes Ferreira, A., Amaco, M.C. and Mendes Faia, A. (2001) The role of non-Saccharomyces species in releasing glycosidic bound fraction of grape aroma components: a preliminary study. Journal of Applied Microbiology 91, 67–71. Rapp, A., Günter, M. and Ulmeyer, M. (1985) Uber varänderungen der aromastoffe während der flashenlagerung von weisweinen der rebsorte Riesling. Zeitschrift für Lebensmittel Untersuchung and Forschung 180, 109–116. Romano, P., Grazia Soli, M., Suzzi, G., Grazia, L. and Zambonelli, C. (1985) Improvement of a wine Saccharomyces cerevisiae strain by a breeding program. Applied and Environmental Microbiology 50, 1064–1072. Rosi, I., Vinella, M. and Domizio, P. (1994) Characterization of β-glucosidase activity in yeasts of oenological origin. Journal of Applied Bacteriology 77, 519–527. Rosi, I., Domizio, P., Vinella, M. and Salicone, M. (1995) Hydrolysis of grape glycosides by enological yeast β-glycosidases. Charalambous, G., ed. Food flavours: generation, analysis and process influence (Elsevier Science: Amsterdam, The Netherlands) pp. 1623–1635. Salmon, J.M. (1984) Application of the technique of cellular permeabilization to the study of the enzymatic activities of Saccharomyces cerevisiae in continuous alcoholic fermentation. Biotechnology Letters 6, 43–48. Salmon, J.M. (1986) Mise en évidence d’un rôle anabolique de l’acide Lmalique chez Saccharomyces cerevisiae en anaérobiose et en conditions

© 2015 Australian Society of Viticulture and Oenology Inc.


199

oenologiques. PhD thesis (Microbiology), Université des Sciences et Techniques du Languedoc, Montpellier, France. Schneider, R., Razungles, A., Augier, C. and Baumes, R. (2001) Monoterpenic and norisoprenoidic glycoconjugates of Vitis vinifera L. cv. Melon B. as precursors of odorants in Muscadet wines. Journal of Chromatography. A 936, 145–157. Ugliano, M., Bartowsky, E., MacCarthy, J., Moio, L. and Henscke, P. (2006) Hydrolysis and transformation of grape glycosidically bound volatile compounds during fermentation with three Saccharomyces yeast strains. Journal of Agricultural and Food Chemistry 54, 6322–6331. Williams, P.J., Sefton, M.A. and Wilson, B. (1989) Non volatile conjugates of secondary metabolites as precursors of varietal grape flavor components. Teranishi, R., Buttery, R. and Shahidi, R., eds. Flavor chemistry: trends and developments (American Chemical Society: Washington, DC, USA) pp. 35–48. Zoecklein, B.W., Marcy, J.E., Williams, J.M. and Jasinski, Y. (1997) Effect of native yeasts and selected strains of Saccharomyces cerevisiae on glycosyl glucose, potential volatile terpenes, and selected aglycones of white Riesling (Vitis vinifera L.) wines. Journal of Food Composition and Analysis 10, 55–65.

Manuscript received: 3 June 2014 Revised manuscript received: 15 September 2014 Accepted: 24 September 2014