Journal of Applied Microbiology ISSN 1364-5072

ORIGINAL ARTICLE

Screening and evaluation of the glucoside hydrolase

activity in Saccharomyces and Brettanomyces brewing
yeasts
L. Daenen, D. Saison, F. Sterckx, F.R. Delvaux, H. Verachtert and G. Derdelinckx
Department of Microbial and Molecular Systems, Centre for Malting and Brewing Science, Faculty of Bioscience Engineering,
Katholieke Universiteit Leuven, Leuven, Belgium

Keywords
b-glucanase, b-glucosidase, bioflavouring,
brewing yeasts, co-culture,
Dekkera ⁄ Brettanomyces, hop glycosides,
Saccharomyces.
Correspondence
L. Daenen, Centre for Malting and Brewing
Science, Department of Microbial and
Molecular Systems, Katholieke Universiteit
Leuven, Kasteelpark Arenberg 22, B-3001
Leuven, Belgium.
E-mail: luk.daenen@biw.kuleuven.be
2007 ⁄ 0396: received 13 March 2007, revised
14 July 2007 and accepted 2 August 2007
doi:10.1111/j.1365-2672.2007.03566.x
Abstract
Aims: The aim of this study was to select and examine Saccharomyces and Bret-
tanomyces brewing yeasts for hydrolase activity towards glycosidically bound
volatile compounds.
Methods and Results: A screening for glucoside hydrolase activity of 58
brewing yeasts belonging to the genera Saccharomyces and Brettanomyces was
performed. The studied Saccharomyces brewing yeasts did not show 1,4-b-glu-
cosidase activity, but a strain dependent b-glucanase activity was observed.
Some Brettanomyces species did show 1,4-b-glucosidase activity. The highest
constitutive activity was found in Brettanomyces custersii. For the most interest-
ing strains the substrate specificity was studied and their activity was evaluated
in fermentation experiments with added hop glycosides. Fermentations with
Br. custersii led to the highest release of aglycones.
Conclusions: Pronounced exo-b-glucanase activity in Saccharomyces brewing
yeasts leads to a higher release of certain aglycones. Certain Brettanomyces
brewing yeasts, however, are more interesting for hydrolysis of glycosidically
bound volatiles of hops.
Significance and Impact of the Study: The release of flavour active compounds
from hop glycosides opens perspectives for the bioflavouring and product
diversification of beverages like beer. The release can be enhanced by using Sac-
charomyces strains with high exo-b-glucanase activity. Higher activities can be
found in Brettanomyces species with b-glucosidase activity.

nonvolatile glycosides can be extracted from the raw

Introduction
An improvement of the organoleptic quality of beverages
and the development of new beverages can be attained
through bioflavouring. This technique relies on the
production and conversion of flavour compounds and
flavour precursors by biological methods (Vanderhaegen
et al. 2003). One method relies on the possibility to
enhance or modify the flavour profile through an
enzymatic hydrolysis of flavour precursors such as
glycosidically bound flavour compounds, which are
present in many plants, flowers and fruits. During the
processing of foods and beverages, the odourless and
materials, which lead to a mostly unused flavour potential
(Winterhalter and Skouroumounis 1997). The enzymatic
hydrolysis of glycosides requires in a first step the
cleavage of the inter-sugar bond in di- or triglycosides,
yielding b-d-glucosides (Gunata et al. 1988). The most
appropriate enzyme for hydrolysis of these b-d-glucosides
is 1,4-b-glucosidase (EC 3.2.1.21) (Sarry and Gunata
2004).
Regarding fermented beverages, a method for introduc-
ing this b-glucosidase activity into the medium is the use
of an appropriately selected yeast strain. As most fermen-
tation media of interest contain sugars exerting catabolite

ª 2007 The Authors

478 Journal compilation ª 2007 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 478–488
L. Daenen et al.
repression effects (glucose, sucrose and fructose) (Verstre-
pen et al. 2004), screening of yeasts for b-glucosidase
activity in the presence of these sugars could be very use-
ful. Most research on b-glucosidases in yeast has focused
on species indigenous to wine-making and high activities
were demonstrated, especially in nonSaccharomyces yeasts
(Rosi et al. 1994; Ferreira et al. 2001; Rodriguez et al.
2004; Villena et al. 2005). The majority of Saccharomyces
isolates does not show b-glucosidase activity on a natural
substrate like arbutin (Rosi et al. 1994; Spagna et al. 2002;
Rodriguez et al. 2004). However, Mateo and DiStefano
(1997) demonstrated the hydrolysis of grape glycosides by
crude extracts of Saccharomyces strains. Recently, this
activity was suggested to be related to the major
exo-b-glucanase EXG1 of Saccharomyces cerevisiae (Gil
et al. 2005). An interesting feature of this yeast exo-b-glu-
canase activity is a substrate nonspecificity, by attacking
besides laminarin also pustulan and small substrates such
as laminaribiose, p-nitrophenyl-b-d-glucoside (pNPG) and
4-methylumbelliferyl-b-d-glucoside (4-MUG) (Nebreda
et al. 1986; Suzuki et al. 2001). Moreover, this
exo-b-glucanase activity is produced constitutively,
independently of the carbon source (Olivero et al.
1985) and is first secreted to the periplasmic space and
then released into the culture medium (Cid et al.
1995). Over-production of this major exo-b-glucanase
in S. cerevisiae led to a moderate release of certain
volatiles from grape glycosides (Gil et al. 2005). There-
fore, screening for high exo-b-glucanase activity in
Saccharomyces strains could be an approach for obtain-
ing a flavour enhancing strain.
Until now, no data are available on the glucoside
hydrolase activities of Saccharomyces brewing yeasts or
other yeasts involved in brewing. The study of these activ-
ities is interesting for better understanding and improving
the taste and aroma of beers. In this regard, glycosides
which are extracted from hops during the production of
beer may become important (Goldstein et al. 1999). This
is certainly important for lambic and gueuze beers, where
glycosides from added sour cherries or raspberries
(Schwab et al. 1990; Pabst et al. 1992) are extracted dur-
ing a secondary fermentation. Lambic and gueuze are spe-
cial Belgian beers obtained by spontaneous fermentation
(Verachtert and Dawoud 1984). In contrast to wine mak-
ing, where Brettanomyces yeasts are mostly regarded as
spoilage agents, these yeasts are essential to obtain lambic
and gueuze beer. The following species were found during
fermentation: Brettanomyces bruxellensis, Brettanomyces
lambicus, Brettanomyces intermedius, Brettanomyces
custersii, Brettanomyces abstinens, Brettanomyces anomalus,
Brettanomyces naardenensis and Brettanomyces custersi-
anus. Brettanomyces is also found during the refermenta-
tion of other special Belgian ales (Verachtert and Dawoud
ª 2007 The Authors
Journal compilation ª 2007 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 478–488
Glucoside hydrolase in brewing yeasts
1984; Martens et al. 1997) and is involved in the biofla-
vouring of a particular Belgian trappist beer (Vanderhae-
gen et al. 2003).
The aim of this study was to select yeast strains related
to beer-making that show a pronounced glucoside hydro-
lase activity. Both b-glucosidase and exo-b-glucanase
activity were considered. Finally, the hydrolase activity of
the yeasts was validated in fermentation experiments,
using purified hop glycosides.
Materials and methods
Chemicals
The following products were supplied by Sigma Aldrich:
2-heptanol, 4-methylumbelliferyl-b-d-glucoside (4-MUG),
Amberlite XAD-2, silicone antifoam (30% in water), ar-
butin, bovine serum albumin, cellobiose (C), glucose (D),
para-nitrophenol (pNP), para-nitrophenyl-b-d-glucoside
(pNPG), octyl-b-d-glucoside and salicin (S).
Yeast strains
Saccharomyces brewing yeasts (9 lager and 27 ale strains
shown in Fig. 1) were obtained from the Centre for Malt-
ing and Brewing Science (CMBS collection, Leuven, Bel-
gium). A commercial wine yeast S. cerevisiae UVAFERM
228 (U228) with published b-glucosidase activity was
obtained from Danstar Ferment AG (Zug, Switzerland).
Four haploid S. cerevisiae wild type strains BY4741,
BY4742, W303a, W303a and a BY4742 deletion mutant
strain (EXG1D ⁄ YLR300WD) were obtained from Euro-
scarf (Frankfurt, Germany). Dekkera and Brettanomyces
yeasts (18 strains shown in Fig. 2) were from CMBS and
were isolated and identified from fermenting lambic by
Verachtert and co-workers (Vanoevelen et al. 1977; Spae-
pen and Verachtert 1982; Verachtert and Dawoud 1984;
Martens et al. 1997).
Screening for glucoside hydrolase activity
A qualitative and fast detection of glucoside hydrolase
activity in yeast was carried out by replica plating the
yeast onto agar plates containing different b-d-glucosidic
substrates like arbutin, salicin, cellobiose or 4-MUG. Solid
media (20 g l)1 agar), all adjusted to pH 5, consisted of:
(i) Arbutin-plates: 6Æ7 g l)1 yeast nitrogen base (YNB)
and 5 g l)1 arbutin, with or without 0Æ2 g l)1 ferric
ammonium citrate, which was added as described by Rosi
et al. (1994). In addition, the same agar plates were pre-
pared with yeast extract (10 g l)1) and peptone (20 g l)1)
in stead of YNB; (ii) Cellobiose-plates: 6Æ7 g l)1 YNB and
5 g l)1 cellobiose; (iii) Salicin-plates: 6Æ7 g l)1 YNB and
479
Glucoside hydrolase in brewing yeasts L. Daenen et al.

Figure 1 Screening of Saccharomyces yeast strains on p-nitrophenyl-b-D-glucopyranoside (pNPG). Cell-associated (left grey bars), extracellular
(middle white bars) and total enzyme (right black bars) activity are presented and expressed as mU mg)1 dry weight (DW). Yeasts were grown on
YPD pH 5. LD10–18: Saccharomyces pastorianus (nine lager yeast strains); LD20–46: Saccharomyces cerevisiae (27 ale yeast strains); BY and W: S.
cerevisiae (four haploid laboratory strains); U228: S. cerevisiae (one wine yeast strain).

20 g l)1 peptone (P), 20 g l)1 glucose (D) or cellobiose

(C)) and incubated at 25C for 24, 48 or 72 h, depend-
ing on the growth of the yeast. Then, the yeast was
inoculated at a cell concentration of 1 · 106 cells ml)1
in Erlenmeyer flasks filled to 20% of their volume with
YPD or YPC and grown at 20C on an orbital shaker at
150 rev min)1 until reaching the later exponential
growth phase.

Extracellular activity
Yeast cell suspension (1 ml) was centrifuged (4C, 3000 g,
10 min) and 0Æ2 ml supernatant was kept for the determi-
nation of extracellular activity on pNPG.

Figure 2 Screening of Dekkera and Brettanomyces yeast strains on
p-nitrophenyl-b-D-glucopyranoside (pNPG). Cell-associated (left grey
bars), extracellular (middle white bars) and total enzyme (right black
bars) activity are presented and expressed as mU mg)1 dry weight
(DW). Yeasts were grown on YPD pH 5. LD70, LD86, LD87: Dekkera
bruxellensis (3); LD78-LD83: Brettanomyces bruxellensis (6); LD71,
LD73-LD77: Brettanomyces lambicus (6); LD72: Brettanomyces cust-
ersii (1); LD84: Brettanomyces anomalus (1); LD88: D. anomala (1).
5 g l)1 salicin; (iv) 4-MUG-plates: 10 g l)1 yeast extract,
20 g l)1 peptone, 20 g l)1 glucose and 0Æ4 g l)1 4-MUG
(Hernandez et al. 2003).
Assay for glucoside hydrolase activity
Yeast growth
A loopful of yeast cells was inoculated from YPD plates
in 10 ml YPD or YPC at pH 5 (10 g l)1 yeast extract (Y),
Cell-associated activity
The remaining pellet was washed with 1 ml physiologi-
cal water (sterile, cold, 7 g l)1 NaCl). After centrifuga-
tion (4C, 3000 g, 10 min) the supernatant was
removed and 0Æ2 ml McIlvaine buffer pH 5 (0Æ1 mol l)1
citric acid–0Æ2 mol l)1 Na2HPO4) was added to the
pellet.
Enzyme assay on pNPG
For determination of the extracellular and cell-associated
activity, 0Æ2 ml pNPG solution (5 mmol l)1 pNPG in
McIlvaine buffer pH 5) was added to 0Æ2 ml supernatant
and to the pellet in 0Æ2 ml buffer. After incubation at
30C for 1 h, 0Æ8 ml carbonate buffer (pH 10Æ2;
0Æ2 mol l)1) was added and the tubes were centrifuged
(4C, 3000 g, 10 min). The increase in absorbance by the
released pNP was measured spectrophotometrically at
400 nm. Blanks were prepared for the substrate and the
different fractions.

ª 2007 The Authors

480 Journal compilation ª 2007 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 478–488
L. Daenen et al.
Yeast protein fraction
To determine the substrate specificity of the glucoside
hydrolase activity, different glucosidic substrates were
examined. To measure the released glucose, yeast protein
fractions were used in stead of whole cells or culture
medium to avoid any glucose interference. The yeast
protein fractions were obtained as described below.
After growth on YPD or YPC, 50 ml of yeast cell
suspension was centrifuged at 4C, 1811 g for 10 min.
The extracellular fraction was obtained by concentrating
the culture fluid and performing a buffer exchange with
McIlvaine buffer (0Æ1–0Æ2 mol l)1) pH 5, by ultrafiltration
using Vivaspin 20 membranes (10 000 MWCO; Sartorius,
Vilvoorde, Belgium).
All the following steps were carried out with cooling
on ice. The yeast pellet was washed twice with sterile cold
physiological water. Then 0Æ5 ml PBS lysis buffer (pH 7Æ3)
was added to an appropriate amount of yeast cells.
The lysis buffer consisted of 10 mmol l)1 Na2HPO4, 1Æ8
mmol l)1 KH2PO4, 140 mmol l)1 NaCl and 2Æ7 mmol l)1
KCl added with 150 g l)1 glycerol, 2 g l)1 Tween,
2 mmol l)1 MgCl2, 1 mmol l)1 EDTA and a protease-
inhibitor mix (Complete EDTA-free; Roche Applied Sci-
ence, Vilvoorde, Belgium). The mixture was shaken with
0Æ25 ml glass beads in a FastPrepTM instrument (FP 120,
BIO101, MP Biomedicals, Illkirch, France). Lysis was
completed by shaking for 20 s at 6 m s)1, cooling on ice
for 2 min and shaken again for 10 s. After centrifugation
(4C, 15000 g, 5 min) the supernatant was collected in a
new centrifuge microtube and centrifuged once more.
The supernatant was used as the intracellular enzyme
activity. The pellet was used as the cell insoluble fraction.
The amount of protein was determined with Bradford
(1976) reagent (Sigma Aldrich) using bovine serum albu-
min for calibration.
Enzyme assay on different substrates
After incubation with the yeast protein fraction, the
enzyme activity and the substrate specificity were deter-
mined by the amount of glucose released from the sub-
strates presented in Fig. 4. To 0Æ85 ml McIlvaine buffer
(0Æ2–0Æ4 mol l)1 pH 5), 0Æ1 ml substrate (glucosides:
25 mmol l)1; laminarin: 25 mg ml)1) and 0Æ05 ml enzyme
fraction were added. The mixture was incubated at 30C
for 1 h. Then, the microtubes were put on ice and
0Æ48 ml carbonate buffer (1 mol l)1 at pH 10Æ2) was
added. The released glucose was measured using a stan-
dardized reference technique (glucose oxidase, GOD-PAP;
Dialab, Wiener Neudorf, Austria).
Dry weight
Yeast dry weight was determined by filtering 20 ml of
yeast cell suspension over a preweighted 0Æ45lm filter
ª 2007 The Authors
Journal compilation ª 2007 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 478–488
Glucoside hydrolase in brewing yeasts
paper (Macherey-Nagel, Düren, Germany). The filter
papers with yeast were oven-dried to a constant weight at
100C.
Preparation of hop glycoside extract
A mixture of glycosidic compounds extracted from hops
was used to investigate and to validate the glucoside
hydrolase activity of the selected yeast strains. Glycosides
were extracted and partially purified as described by
Goldstein et al. (1999) with some modifications. Saaz
hops (1Æ5 kg previously extracted with liquid CO2) were
shaken with 9 l of methanol ⁄ water (4 : 1 v ⁄ v) for 3 h.
The extract was filtered and concentrated using a rotary
evaporator to 1Æ1 l. Then, the glycosidic extract was
added to a column (45 · 4 cm) packed with 200 g Am-
berlite XAD-2, previously treated according to Guyot-
Declerck et al. (2000). The effluent was collected. The
column was then washed with 1Æ5 l of water and eluted
with 0Æ75 l of methanol. The column was reconditioned
and the collected effluent was brought a second time on
the column. The column was then washed and eluted as
described above. The methanol fractions were concen-
trated to dryness on a rotary evaporator and re-sus-
pended in 740 ml McIlvaine buffer pH 5. This mixture
was shaken with 25 g PVPP and three times extracted
with 145 ml of diethyl ether to remove polyphenols and
free volatiles, respectively.
Hydrolysis of hop glycoside extract
To investigate the hydrolase activity of yeast on hop gly-
cosides during an alcoholic fermentation, a loopful of
cells was incubated into 10 ml of YPD for 24–48 h at
25C. The precultures were transferred to 50 ml YPD and
incubated on an orbital shaker for 24–48 h at 20C. The
cells were centrifuged (4C, 1811 g, 10 min), washed and
re-suspended in physiological water. Finally, 75 ml YPD
(10 g l)1 yeast extract, 20 g l)1 peptone, 100 g l)1 glu-
cose) supplemented with 1 ml of hop glycoside extract
was inoculated at 10 · 106 cells ml)1. Fermentations were
carried out for 6 days at a constant temperature of 20C.
The pure culture fermentation by Br. custersii CMBS
LD72 was carried out for 12 days.
Purge-and-trap GC-MS analysis
Analysis of the released volatile aglycones was performed
as described by Vanderhaegen et al. (2004) with some
modifications. Briefly, the method was as follows. Samples
were collected in 50 ml Falcon tubes on ice and centri-
fuged (4C, 1811 g, 10 min). Then, internal standard
2-heptanol (250 mg l)1) and 10% antifoam solution was
481
Glucoside hydrolase in brewing yeasts L. Daenen et al.

added to give a final concentration of 1 mg l)1 and

0Æ04%, respectively. Of this sample, 5 ml was transferred
into a Tekmar Dohrman 3000 (Emerson, Mason, OH,
USA) purge-and-trap concentrator unit in which the
sample was purged with helium carrier gas and trapped
on a Vocarb 3000 trap (Supelco, Bellefonte, PA, USA).
Then, the volatiles were transferred via a MFA815 cold-
trap ⁄ control unit (Thermofinnigan, San Jose, CA, USA)
to a Fisons GC 8000 gas chromatograph equipped with a
Chrompack CP-Wax-52-CB column (length 50 m; inter-
nal diameter 0Æ32 mm; film thickness 1Æ2 lm; Varian,
Palo Alto, CA, USA). The temperature program was
1 min at 50C, 3C min)1 to 150C, 15C min)1 to
240C and 5 min at 240C. Total ion mass chromato-
grams were obtained in a Fisons MD 800 quadrupole
mass spectrometer (ionization energy 70 eV; source tem-
perature 250C) (Thermo fisher Scientific, Waltham,
MA). Quantification was performed using standard refer-
ence compounds.
Data analysis
Significance differences between experimental figures were
estimated using Student’s t-test at a significance of P £
0Æ05. The values represent the means of replications;
n = 2.
Results
Screening of brewing yeasts
Methods for detecting b-glucosidase activity in yeasts are
generally based on the hydrolysis of a glucosidic substrate,
which leads to the formation of biomass or to the release
of a traceable aglycon. The b-d-glucosides which are
mostly used are arbutin, 4-MUG, pNPG, esculin, salicin
or cellobiose. These substrates with the exception of escu-
lin were used in this study. A screening for glucoside
hydrolase activity was performed on 59 yeasts of which
54 brewing yeasts belonging to the genera Saccharomyces,
Dekkera and Brettanomyces, one commercial wine yeast
S. cerevisiae U228 with b-glucosidase activity and four
haploid S. cerevisiae laboratory strains. Results for screen-
ing, presented in Table 1, show that which method is
used highly determines whether activity towards gluco-
sides is present or not. Hence, YNB agar plates containing
arbutin as sole carbon source and ferric ammonium
citrate (Rosi et al. 1994) did not show growth or colour
formation by any yeast strain. When omitting ferric
ammonium citrate from the medium, growth was
detected for some strains (Table 1). On yeast extract, pep-
tone (YP) agar plates, however, containing arbutin and
ferric ammonium citrate, growth occurred and a brown

Table 1 Qualitative detection of the glucoside hydrolase activity in brewing yeasts using screening methods based on solid media

YNB-A YP-A
YPD
Solid media (2% agar) ⁄ Fe3+ ⁄ Fe3+ YNB-C YNB-S 4-MUG

Detection by: Growth Brown color Growth Fluo-halo

No. of strains
Yeast tested Comment

Saccharomyces 9 Lager brewing ) ) ) ) ) ) + or ++

pastorianus strains
Saccharomyces 27 Ale brewing ) ) ) ) ) ) + or ++
cerevisiae strains
S. cerevisiae 1 Wine strain ++ ) + + ++ ++ ++
S. cerevisiae 4 Haploid lab strains ) ) ) ) ) ) +
Dekkera bruxellensis 3 D. bruxellensis ) ) ) ) ) ) + or ++
Brettanomyces 6 Br. bruxellensis ) ) ) ) ) ) + or ++
bruxellensis 6 Brettanomyces ) ) ) ) ) ) + or ++
lambicus§
1 Br. custersii§ +++ ) + ++ +++ +++ +++
Dekkera anomala 1 D. anomala +++ ) + ++ +++ +++ +++
Brettanomyces 1 Br. anomalus +++ ) + ++ +++ +++ +++
anomalus

), No detectable activity; +, weak activity; ++, moderate activity; +++, strong activity; YNB, yeast nitrogen base; YP, yeast extract, peptone; A,
arbutin; C, cellobiose; S, salicin; 4-MUG, 4-methylumbelliferyl-b-D-glucopyranoside.
The yeast strains described here are the same as presented in Figs 1 and 2.
The S. pastorianus name is used, as proposed by Rainieri et al. (2006), for lager brewing strains which naturally occur as Saccharomyces hybrids.
§Synonyms of Dekkera Brettanomyces species according to Barnett et al. (1990).

ª 2007 The Authors

482 Journal compilation ª 2007 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 478–488
L. Daenen et al. Glucoside hydrolase in brewing yeasts

colour was formed for some Dekkera and Brettanomyces

species and with S. cerevisiae U228. For yeasts on YPD
agar plates containing 4-MUG, a fluorescent halo
appeared for all strains. The halo around the colony
increased during growth and varied in size depending on
the strain. Yeasts growth on salicin corresponded with
growth on cellobiose. This occurred with Br. intermedius
CMBS LD85, Br. custersii CMBS LD72, Br. anomalus
CMBS LD84 and Dekkera anomala CMBS LD88, but not
with any Br. bruxellensis or Br. lambicus. Neither did any
Saccharomyces strain grow on salicin nor cellobiose, ex-
cepted for the commercial strain S. cerevisiae U228.
The glucoside hydrolase activity of the strains was fur-
ther quantitatively differentiated by activity determina-
tions on pNPG. As applications of this enzyme activity
could be interesting in fermented food products contain-
ing sugars which exert catabolite repression effects, the
presence of a constitutively activity was examined by
using yeast after growth in liquid YPD and determining
the extracellular and the cell-associated activity. All strains
showed hydrolase activity, especially S. cerevisiae CMBS Figure 3 Growth of the wild type BY4742 and the exg1D
LD40 and Br. custersii CMBS LD72 (Figs 1 and 2). (= ylr300wD) mutant strain on an YPD agar plate containing 4-methyl-
umbelliferyl-b-D-glucopyranoside (4-MUG). The fluorescent halo of the
released 4-methylumbelliferone (4-MU) under UV-radiation demon-
Exo-b-glucanase strates hydrolase activity.

As mentioned above, the major exo-b-1,3-glucanase

Table 2 Specific enzyme activity of the haploid strain BY4742 and
(EXG1) of S. cerevisiae is known to hydrolyse besides lami-
the deletion mutant exg1D on para-nitrophenyl-b-D-glucoside (pNPG)
narin, the polymer pustulan and small substrates like lami- and laminarin
naribiose, pNPG and 4-MUG (Nebreda et al. 1986; Suzuki
et al. 2001). To examine the impact of EXG1 on the results BY4742 exg1D BY4742 exg1D
obtained above with the screening on 4-MUG-plates and pNPG
with the screening on pNPG (Table 1 and Fig. 1), the wild (mU ⁄ mg DW)
type strain BY4742 and the deletion mutant EXG1D
(YLR300WD) were screened in the same way (Fig. 3). On Extracellular activity* 0Æ54 <0Æ01
Whole cells* 0Æ24 0Æ01
agar plates with 4-MUG, a fluorescent halo was observed
for the wild type strain but not for the mutant strain exg1D pNPG Laminarin
(Fig. 3). On pNPG, both extracellular and cell-associated (mU ⁄ mg protein) (mU ⁄ mg protein)
activity almost completely disappeared for the mutant
Extracellular activity 49 0Æ3 149 8Æ6
strain exg1D (Table 2). To verify this activity on a glucan Intracellular activity 0Æ2 0Æ07 0Æ3 0Æ1
substrate like laminarin, different yeast protein fractions Cell insoluble fraction 1Æ6 0Æ6 3Æ5 0Æ7
were incubated both on pNPG and on laminarin
(Table 2). The hydrolase activity clearly decreased on both DW, dry weight.
*Specific activity is determined as in Figs 1 and 2 by measurement of
substrates in case of the mutant strain exg1D.
the released pNP.
Specific activity is determined by measurement of the released pNP
Selection for further experiments or glucose after incubation of the concerning yeast protein fraction
on pNPG and laminarin, respectively.
Based on the previous results, a selection of six yeast
strains was made for further characterization of the gluco- relative higher b-glucosidase activity. The haploid strain
side hydrolase activity. The commercial wine yeast S. S. cerevisiae BY4742 and the mutant strain exg1D were
cerevisiae U228 was retained as this was the only chosen to investigate the effect of exo-1,3-b-glucanase
Saccharomyces strain demonstrating 1,4-b-glucosidase (EXG1). The strains S. cerevisiae CMBS LD 25 and CMBS
activity in this study. Brettanomyces custersii CMBS LD72 LD40 were selected for a respectively low and higher glu-
isolated from a lambic fermentation was selected for a coside hydrolase activity on pNPG (Fig. 1).

ª 2007 The Authors

Journal compilation ª 2007 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 478–488 483
Glucoside hydrolase in brewing yeasts
Figure 4 Hydrolysis of different glucosidic substrates by the intracel-
lular activity of Saccharomyces cerevisiae U228 and Brettanomyces
custersii CMBS LD72 after growth on YP medium with added glucose
(U228 YPD and CMBS LD72 YPD); S. cerevisiae U228 and Br. custersii
CMBS LD72 after growth on YP medium with added cellobiose (U228
YPC and CMBS LD72 YPC). The specific enzyme activity is expressed
as mU mg)1 protein (bovine serum albumin equivalents in order of
reproduction). ( ) U228 YPD; ( ) U228 YPC; ( ) CMBS LD72 YPD;
( ) CMBS LD72 YPC.
Characterization of selected yeast strains
For the brewing yeasts, the substrate specificity of the glu-
coside hydrolase was characterized on different glucosidic
substrates. First, the b-glucosidase activity of S. cerevisiae
U228 and Br. custersii CMBS LD72 was examined. More-
over, the constitutive or inductive production of the
activity was examined by determining the activity after
growth on YP medium with respectively glucose or cello-
biose as carbon source (Fig. 4). After growth on cello-
biose, the intracellular activity of S. cerevisiae U228 and
Br. custersii CMBS LD72 was higher on all the glucosidic
substrates than after growth on glucose. The intracellular
activity of Br. custersii CMBS LD72, after growth on glu-
cose, was higher than the activity of S. cerevisiae U228
grown on glucose. Concerning substrate specificity, espe-
cially salicin, octyl-glucoside and pNPG were hydrolysed
to a great extent by S. cerevisiae U228 and Br. custersii
CMBS LD72.
Further, the glucoside hydrolase activity of the Saccha-
romyces brewing yeasts CMBS LD25 and LD40 was exam-
ined. As the results from Table 1 and Fig. 1 indicate that
this glucoside hydrolase activity is probably because of
the exo-1,3-b-glucanase, the glucan substrate laminarin
was tested as well. The results in Fig. 5 clearly demon-
484 Journal compilation ª 2007 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 478–488
L. Daenen et al.
Figure 5 Hydrolysis of different glucosidic substrates by different
yeast protein extracts [extracellular (EC), intracellular (IC) and cell
insoluble fraction (CI)] of Saccharomyces cerevisiae CMBS LD25 and S.
cerevisiae CMBS LD40 grown on YP medium with added glucose
(CMBS LD40). Specific enzyme activity is expressed as mU mg)1 pro-
tein (bovine serum albumin equivalents). ( ) CMBS LD25 EC; ( )
CMBS LD40 EC; ( ) CMBS LD25 IC; ( ) CMBS LD40 IC; ( ) CMBS
LD25 CI; ( ) CMBS LD40 CI.
strate the higher activity on the substrates pNPG, 4-MUG
and laminarin for S. cerevisiae CMBS LD40 in compari-
son with CMBS LD25.
Finally, with respect to effects on hop glycosides, fer-
mentations were started with the selected strains and also
with a mixed culture of Br. custersii CMBS LD72 and S.
cerevisiae CMBS LD25. The latter was chosen for this
mixed culture in order to minimally interfere in glucoside
hydrolysis because of a relative lower exo-b-glucanase
activity (Fig. 5). Samples for analysis of released aglycones
were taken after completion of the fermentations and
results are presented in Fig. 6. The laboratory strain
BY4742 caused a slightly higher release of methyl salicylate
and cis-3-hexen-1-ol than the mutant strain exg1D. The
strains S. cerevisiae CMBS LD40 and S. cerevisiae U228
caused a moderate but higher release of methyl salicylate
and the aliphatic alcohols 1-octen-3-ol and cis-3-hexen-1-
ol than S. cerevisiae CMBS LD25. The S. cerevisiae strains
were not very active on the linalyl-glycosides. Most inter-
esting is the high release of aglycones in fermentations
with Br. custersii CMBS LD72, in pure culture or in co-
culture with Saccharomyces. In these fermentations, linal-
ool, methyl salicylate, 1-octen-3-ol and cis-3-hexen-1-ol
were released to the highest extent.
ª 2007 The Authors
L. Daenen et al. Glucoside hydrolase in brewing yeasts

Figure 6 Concentration (mg l)1) of released aglycones after fermentation of YPD with addition of purified hop glycosides by Saccharomyces ce-
revisiae BY4742 wt, S. cerevisiae exg1D, S. cerevisiae CMBS LD25, S. cerevisiae CMBS LD40, S. cerevisiae U228, Br. custersii CMBS LD72, mixed
culture of S. cerevisiae CMBS LD25 and Brettanomyces custersii CMBS LD72. Control without any enzyme activity was carried out at pH 5. (a) ( )
linalool, (b) ( ) methyl salicylate, (c) ( ) cis-3-hexen-1-ol and (d) ( ) 1-octen-3-ol.

charomyces or Dekkera and Brettanomyces did not grow or

Discussion
The aim of this study was to select and study yeast
strains, indigenous to the brewing environment, with
hydrolase activity towards glycosidically bound volatile
compounds. This required an easy, fast and reliable detec-
tion method. The most appropriate enzyme for the
hydrolysis of glycosidically bound volatiles is b-glucosi-
dase (EC 3.2.1.21), preferably with a broad substrate spec-
ificity. Many different methods have been used to detect
b-glucosidase activity. The method with arbutin as sole
carbon source for yeast growth and addition of ferric
ammonium citrate has been proposed as reliable for
detecting b-glucosidases (Rosi et al. 1994; Ferreira et al.
2001). In our study, the brewing yeasts belonging to Sac-
develop a brown colour with that medium. This method
apparently led to false negative results with our strains, as
S. cerevisiae U228, Br. intermedius CMBS LD85, Br. cust-
ersii CMBS LD72, Br. anomalus CMBS LD84 and D. ano-
mala CMBS LD88 clearly showed the presence of a b-
glucosidase (EC 3.2.1.21) by growing on cellobiose and
salicin. By omitting ferric ammonium citrate from the
minimal medium with arbutin, these strains showed
growth. Apparently, with our strains, the combination of
arbutin and ferric ammonium citrate at the used concen-
tration inhibited growth. The assimilation of cellobiose,
salicin and arbutin by Br. intermedius, Br. custersii,
Br. anomalus and D. anomala was already reported by
Barnett et al. (1983).

ª 2007 The Authors

Journal compilation ª 2007 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 478–488 485
Glucoside hydrolase in brewing yeasts
For Saccharomyces yeasts, the presence of a sporadically
observed b-glucosidase activity remains intriguing. In our
study, none of the Saccharomyces brewing yeasts showed
such activity. On the contrary, the commercial wine yeast
S. cerevisiae U228 could grow on cellobiose, arbutin and
salicin and consequently demonstrated the presence of a
b-glucosidase. Previously, only very few S. cerevisiae iso-
lates were found which showed activity on arbutin and
thus indicating b-glucosidase activity (Rosi et al. 1994;
Spagna et al. 2002; Rodriguez et al. 2004). This is remark-
able as no gene in the genome of the haploid strain
S. cerevisiae is known for coding a 1,4-b-glucosidase (EC
3.2.1.21) (Cherry et al. 1997; http://www.yeastgenome.org;
2 January 2007). Moreover, S. cerevisiae yeasts are not
expected to assimilate cellobiose, arbutin or salicin
according to physiological identification tests proposed by
Barnett et al. (1990). However, the b-glucosidase gene of
a S. cerevisiae strain AL41 isolated on arbutin by Spagna
et al. (2002) was recently partially sequenced by Quatrini
et al. (2006). The translated amino acid sequence con-
tained one of the conserved patterns, namely FGYGLSY,
which is typical for most yeast b-glucosidases. This pat-
tern was found in the C-terminal part of the b-gluco-
sidase(s) from Saccharomycopsis fibuligera, Candida
pelliculosa and Kluyveromyces fragilis (Rojas and Romeu
1996). Apparently only some S. cerevisiae yeasts possess a
gene coding for b-glucosidase (EC 3.2.1.21) whereas the
majority does not.
On the contrary, all S. cerevisiae yeasts synthesize
b-glucanases. According to Nebreda et al. (1986), the
exo-1,3-b-glucanase EXG1 is responsible for the greatest
part of the hydrolysis of laminarin and synthetic sub-
strates like pNPG and 4-MUG, which is confirmed in this
study. The question remains whether the capacity of only
certain S. cerevisiae strains to assimilate b-glucosides like
cellobiose, salicin or arbutin is because of a b-glucanase
with an adjusted substrate specificity or to the presence of
a b-glucosidase like enzyme. Results by Ridruejo et al.
(1989) already supported the idea that yeast exo-1,3-
b-glucanases (EC 3.2.1.58) are glucosidases that also
attack laminarin, rather than glucanases capable of attack-
ing pNPG. Suzuki et al. (2001) suggested that the yeast
Exg1p may be classified as a new type of b-glucanase or
b-glucosidase that has not been described before.
Considering the regulation of the enzymes, the b-glu-
cosidase from S. cerevisiae U228 appears to be repressed
by glucose and induced by cellobiose. An inducible b-glu-
cosidase activity in S. cerevisiae was already demonstrated
by Duerksen and Halvorson (1958). The b-glucosidase of
Br. custersii CMBS LD72 also appeared to be induced by
growth on cellobiose. Possible repression effects on the
b-glucosidase production of Br. custersii CMBS LD72 will
be investigated in further research. The exo-1,3-b-glucan-
486 Journal compilation ª 2007 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 478–488
L. Daenen et al.
ase EXG1 of the S. cerevisiae strains is constitutively regu-
lated (Olivero et al. 1985). Results from this study show a
strain-dependent activity for this exo-1,3-b-glucanase,
which was most pronounced for S. cerevisiae CMBS
LD40.
This relative higher activity of S. cerevisiae CMBS LD40
led also to a higher release of certain aglycones from hop
glycosides during fermentation. This becomes an interest-
ing property of a brewer’s yeast when considering applica-
tions for bioflavouring. Fermentation with the haploid
mutant strain exg1D, led to a lower release of certain agly-
cones than the wild type. These results are in agreement
with those of Gil et al. (2005) who demonstrated the activ-
ity of the major exoglucanase EXG1 of S. cerevisiae towards
glycosidically bound volatile compounds. The reaction is
however somewhat specific, as linalool is almost not
released. This could be because of steric hindrance of the
reaction by the tertiary alcohol (Gunata et al. 1985; Gil
et al. 2005). However, Ugliano et al. (2006) observed
a pronounced release of linalool from grape glycosides
after fermentations with S. cerevisiae and S. bayanus wine
yeasts. The hydrolysis of glycosidically bound tertiary
alcohols like linalool apparently depends on the strain, the
fermentation conditions or on both. Further investigation
is required to clarify these observations.
The most interesting and efficient glucoside splitting
occurred with the b-glucosidase of Br. custersii CMBS
LD72. This yeast was isolated from fermenting lambic
(Verachtert and Dawoud 1984). The b-glucosidase of this
Br. custersii strain showed a broad substrate specificity
towards alkyl- and aryl-b-d-glucosides. For the hydrolysis
of hop glycosides, a co-culture with S. cerevisiae was also
very efficient. As Dekkera and Brettanomyces species are
slow fermenting yeasts, the duration of an alcoholic fer-
mentation can be reduced by a mixed fermentation in
which S. cerevisiae carries out the main fermentation and
Br. custersii takes care for the release of glycosidically
bound volatiles. Brettanomyces custersii was already recog-
nized for its interesting b-glucosidase activity as an effi-
cient cellobiose and cellotriose fermenting yeast in
ethanol production (Gonde et al. 1984; Spindler et al.
1992). Brettanomyces custersii was found to be the domi-
nant yeast species during the latest stages of lambic fer-
mentation (13th–24th month) indicating a possible
adaptation to the environment, through assimilation of
cellobiose released from the wooden fermentation casks
(Verachtert and Dawoud 1984; Vanderhaegen et al. 2003).
In conclusion, Saccharomyces strains show a strain-
dependent hydrolase activity towards certain glycosidically
bound volatile compounds. This would be mainly because
of the enzyme exo-1,3-b-glucanase. Only few Saccharomy-
ces strains appear to possess real 1,4-b-glucosidase activ-
ity. A commercial wine strain showed an inducible
ª 2007 The Authors
L. Daenen et al.
b-glucosidase with weak activity during fermentation.
A more pronounced b-glucosidase activity was found in
nonSaccharomyces yeasts like Br. custersii, isolated from
fermenting lambic. Fermentation carried out with a pure
culture or a co-culture of S. cerevisiae and Br. custersii led
to the highest release of volatiles from hop glycosides.
Further research will be focused on the use of Br. custersii
to introduce new flavours in beer, either through a mixed
fermentation, maturation or refermentation, either with
or without the presence of added ingredients such as
fruits, flowers or spices containing glycosides as flavour
precursors.
Acknowledgements
This research was funded by a PhD grant from the Insti-
tute for the Promotion of Innovation through Science
and Technology in Flanders (IWT-Vlaanderen, Belgium).
The authors wish to thank L. De Cooman and K. Goiris
of the Technical University KaHo in Gent for providing
the treated Saaz spent hops.
References
Barnett, J.A., Payne, R.W. and Yarrow, D. (1983) Yeasts: Char-
acteristics and Identification, 1st edn. Cambridge: Cam-
bridge University Press.
Barnett, J.A., Payne, R.W. and Yarrow, D. (1990) Yeasts: Char-
acteristics and Identification, 2nd edn. Cambridge: Cam-
bridge University Press.
Bradford, M.M. (1976) Rapid and sensitive method for quanti-
tation of microgram quantities of protein utilizing princi-
ple of protein–dye binding. Anal Biochem 72, 248–254.
Cherry, J.M., Ball, C., Weng, S., Juvik, G., Schmidt, R., Adler,
C., Dunn, B., Dwight, S. et al. (1997) Genetic and physical
maps of Saccharomyces cerevisiae. Nature 387, 67–73.
Cid, V.J., Duran, A., Delrey, F., Snyder, M.P., Nombela, C.
and Sanchez, M. (1995) Molecular-basis of cell integrity
and morphogenesis in Saccharomyces cerevisiae. Microbiol
Rev 59, 345–386.
Duerksen, J.D. and Halvorson, H. (1958) Purification and
properties of an inducible beta-glucosidase of yeast. J Biol
Chem 233, 1113–1120.
Ferreira, A.M., Climaco, M.C. and Faia, A.M. (2001) The role
of non-saccharomyces species in releasing glycosidic bound
fraction of grape aroma components – a preliminary
study. J Appl Microbiol 91, 67–71.
Gil, J.V., Manzanares, P., Genoves, S., Valles, S. and Gonz-
alez-Candelas, L. (2005) Over-production of the major
exoglucanase of Saccharomyces cerevisiae leads to an
increase in the aroma of wine. Int J Food Microbiol 103,
57–68.
Goldstein, H., Ting, P., Schulze, W.G., Murakami, A.A., Lusk,
L.T. and Young, V.D. (1999) Methods of Making and Using
ª 2007 The Authors
Journal compilation ª 2007 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 478–488
Glucoside hydrolase in brewing yeasts
Purified Kettle Hop Flavorants. Milwaukee, WI: Miller
Brewing Company. US Patent No. 5972411.
Gonde, P., Blondin, B., Leclerc, M., Ratomahenina, R.,
Arnaud, A. and Galzy, P. (1984) Fermentation of cello-
dextrins by different yeast strains. Appl Environ Microbiol
48, 265–269.
Gunata, Y.Z., Bayonove, C.L., Baumes, R.L. and Cordonnier,
R.E. (1985) The aroma of grapes. 1. Extraction and deter-
mination of free and glycosidically bound fractions of
some grape aroma components. J Chromatogr 331, 83–90.
Gunata, Z., Bitteur, S., Brillouet, J.M., Bayonove, C. and Cor-
donnier, R. (1988) Sequential enzymic-hydrolysis of poten-
tially aromatic glycosides from grape. Carbohydr Res 184,
139–149.
Guyot-Declerck, C., Chevance, F., Lermusieau, G. and Collin,
S. (2000) Optimized extraction procedure for quantifying
norisoprenoids in honey and honey food products. J Agric
Food Chem 48, 5850–5855.
Hernandez, L.F., Espinosa, J.C., Fernandez-Gonzalez, M. and
Briones, A. (2003) Beta-glucosidase activity in a Saccharo-
myces cerevisiae wine strain. Int J Food Microbiol 80, 171–
176.
Martens, H., Iserentant, D. and Verachtert, H. (1997) Microbi-
ological aspects of a mixed yeast-bacterial fermentation in
the production of a special Belgian acidic ale. J Inst Brew
103, 85–91.
Mateo, J.J. and DiStefano, R. (1997) Description of the beta-
glucosidase activity of wine yeasts. Food Microbiol 14, 583–
591.
Nebreda, A.R., Villa, T.G., Villanueva, J.R. and Delrey, F.
(1986) Cloning of genes related to exo-beta-glucanase pro-
duction in Saccharomyces cerevisiae – characterization of
an exo-beta-glucanase structural gene. Gene 47, 245–259.
Olivero, I., Hernandez, L.M. and Larriba, G. (1985) Regulation
of beta-exoglucanase activity production by Saccharomyces
cerevisiae in batch and continuous culture. Arch Microbiol
143, 143–146.
Pabst, A., Barron, D., Semon, E. and Schreier, P. (1992)
4-oxo-beta-ionol and linalool glycosides from raspberry
fruits. Phytochemistry 31, 4187–4190.
Quatrini, P., Marineo, S. and Puglia, A.M. (2006) The Beta-
Glucosidase Encoding Gene from Yeast Strains Isolated from
Sicilian musts and Wines (DQ010949). Bethesda, MD: Cen-
ter for Biotechnology Information (NCBI), available at:
http://www.ncbi.nlm.nih.gov/, accessed October 15th 2006.
Rainieri, S., Kodama, Y., Kaneko, Y., Mikata, K., Nakao, Y.
and Ashikari, T. (2006) Pure and mixed genetic lines of
Saccharomyces bayanus and Saccharomyces pastorianus and
their contribution to the lager brewing strain genome.
Appl Environ Microbiol 72, 3968–3974.
Ridruejo, J.C., Munoz, M.D., Andaluz, E. and Larriba, G.
(1989) Inhibition of yeast exoglucanases by glucosidase
inhibitors. Biochim Biophys Acta 993, 179–185.
Rodriguez, M.E., Lopes, C.A., Broock, M., Valles, S., Ramon,
D. and Caballero, A.C. (2004) Screening and typing of
487
Glucoside hydrolase in brewing yeasts
Patagonian wine yeasts for glycosidase activities. J Appl
Microbiol 96, 84–95.
Rojas, A. and Romeu, A. (1996) A sequence analysis of the
beta-glucosidase sub-family B. FEBS Lett 378, 93–97.
Rosi, I., Vinella, M. and Domizio, P. (1994) Characterization
of beta-glucosidase activity in yeasts of oenological origin.
J Appl Bacteriol 77, 519–527.
Sarry, J.E. and Gunata, Z. (2004) Plant and microbial glycoside
hydrolases: volatile release from glycosidic aroma precur-
sors. Food Chem 87, 509–521.
Schwab, W., Scheller, G. and Schreier, P. (1990) Glycosidically
bound aroma components from sour cherry. Phytochemis-
try 29, 607–612.
Spaepen, M. and Verachtert, H. (1982) Esterase-activity in the
genus Brettanomyces. J Inst Brew 88, 11–17.
Spagna, G., Barbagallo, R.N., Palmeri, R., Restuccia, C. and
Giudici, P. (2002) Properties of endogenous beta-glucosi-
dase of a Saccharomyces cerevisiae strain isolated from
Sicilian musts and wines. Enzyme Microb Technol 31,
1030–1035.
Spindler, D.D., Wyman, C.E., Grohmann, K. and Philippidis,
G.P. (1992) Evaluation of the cellobiose-fermenting yeast
Brettanomyces custersii in the simultaneous saccharification
and fermentation of cellulose. Biotechnol Lett 14, 403–407.
Suzuki, K., Yabe, T., Maruyama, Y., Abe, K. and Nakajima, T.
(2001) Characterization of recombinant yeast exo-beta-1,3-
glucanase (Exg 1p) expressed in Escherichia coli cells. Biosci
Biotechnol Biochem 65, 1310–1314.
Ugliano, M., Bartowsky, E.J., McCarthy, J., Moio, L. and Hens-
chke, P.A. (2006) Hydrolysis and transformation of grape
488 Journal compilation ª 2007 The Society for Applied Microbiology, Journal of Applied Microbiology 104 (2008) 478–488
L. Daenen et al.
glycosidically bound volatile compounds during fermenta-
tion with three Saccharomyces yeast strains. J Agric Food
Chem 54, 6322–6331.
Vanderhaegen, B., Neven, H., Coghe, S., Verstrepen, K.J., Derde-
linckx, G. and Verachtert, H. (2003) Bioflavoring and beer
refermentation. Appl Microbiol Biotechnol 62, 140–150.
Vanderhaegen, B., Neven, H., Daenen, L., Verstrepen, K.J.,
Verachtert, H. and Derdelinckx, G. (2004) Furfuryl ethyl
ether: important aging flavor and a new marker for the
storage conditions of beer. J Agric Food Chem 52, 1661–
1668.
Vanoevelen, D., Spaepen, M., Timmermans, P. and Verachtert,
H. (1977) Microbiological aspects of spontaneous wort fer-
mentation in production of lambic and gueuze. J Inst Brew
83, 356–360.
Verachtert, H. and Dawoud, E. (1984) Microbiology of lam-
bic-type beers. J Appl Bacteriol 57, R11–R12.
Verstrepen, K.J., Iserentant, D., Malcorps, P., Derdelinckx,
G., Van Dijck, P., Winderickx, J., Pretorius, I.S., Theve-
lein, J.M. et al. (2004) Glucose and sucrose: hazardous
fast-food for industrial yeast? Trends Biotechnol 22, 531–
537.
Villena, M.A., Iranzo, J.F.U., Otero, R.R.C. and Perez, A.I.B.
(2005) Optimization of a rapid method for studying the
cellular location of beta-glucosidase activity in wine yeasts.
J Appl Microbiol 99, 558–564.
Winterhalter, P. and Skouroumounis, G.K. (1997) Glycoconju-
gated aroma compounds: occurrence, role and biotechno-
logical transformation. Adv Biochem Eng Biotechnol 55,
74–99.
ª 2007 The Authors