Previous Article | Next Article ![]()
Applied and Environmental Microbiology, December 2003, p. 7558-7562, Vol. 69, No. 12
0099-2240/03/$08.00+0 DOI: 10.1128/AEM.69.12.7558-7562.2003
Copyright © 2003, American
Society for
Microbiology. All Rights Reserved.
-L-Rhamnosidase of Enological Interest
Departamento de Biotecnología, Instituto de Agroquímica y Tecnología de Alimentos (CSIC),1 Departamento de Medicina Preventiva y Salud Pública, Bromatología, Toxicología y Medicina Legal, Facultad de Farmacia, Universitat de València, Burjassot, 46100 Valencia, Spain,4 Fungal Genomics Section, Laboratory of Microbiology, Wageningen University, NL-6703-HA,2 Fungal Genetics and Technology Consultancy, 6700-AJ Wageningen,The Netherlands3
Received 17 April 2003/ Accepted 15 September 2003
|
|
|---|
-L-rhamnosidase has been expressed in both
laboratory and industrial wine yeast strains. Wines produced in
microvinifications, conducted using a combination of the genetically
modified industrial strain expressing rhaA and another strain
expressing a ß-glucosidase, show increased content mainly of
the aromatic compound
linalool. |
|
|---|
-terpineol present
in grapes determine the varietal flavor properties of young quality
wines made from Muscat varieties (for reviews, see references
19 and
21). Geraniol and
linalool are considered to be the most important of the monoterpene
alcohols, as they are present in greater concentrations and have lower
flavor thresholds than other major wine monoterpenes. In particular,
linalool is thought to be responsible for the grapelike aroma of wines
produced from the Muscat variety. A large proportion of these compounds
are found as odorless diglycoside conjugates, such as
6-O-
-L-arabinofuranosyl-ß-D-glucopyranosides,
6-O-
-L-rhamnopyranosyl-ß-D-glucopyranosides,
and
6-O-ß-D-apiofuranosyl-ß-D-glucopyranosides.
This fraction constitutes a potential pool of aroma precursors, and its
enzymatic hydrolysis has been studied as a method to release free
monoterpenes and enhance grape juice and wine flavor
(10,
34,
35). It is now well
established that this enzymatic hydrolysis occurs in two steps
(11). During the first
step, and depending on the conjugate, the glycosidic linkage is cleaved
by either a ß-D-apiosidase, an
-L-arabinofuranosidase, or an
-L-rhamnosidase, and the corresponding
monoterpenyl-ß-D-glucosides are released. In the
second step, monoterpenes are liberated by the action of a
ß-D-glucosidase. Consequently, research interest has
been focused on those glycosidases able to release flavor compounds
from glycosides. Several glycosidases, such as
-L-arabinofuranosidases
(14,
30,
31,
36),
ß-D-apiosidases
(13),
-L-rhamnosidases
(2,
3,
15,
17,
18,
20,
22,
32,
33), and
ß-D-glucosidases
(1,
12,
20,
26,
30,
31), from different
microbial sources have been characterized for their potential
application in increasing the aroma of wine.
An alternative way
to achieve increases in the content of volatile compounds is the
inoculation of must with genetically modified wine yeasts expressing
genes coding for these enzymes
(25). Previous work in
our laboratory has reported the use of several transgenic wine yeasts
expressing a Candida molischiana
ß-D-glucosidase and an Aspergillus niger
-L-arabinofuranosidase as tools to increase the
free monoterpene content in wine
(27,
28). We have also
expressed genes encoding a Trichoderma longibrachiatum
endoglucanase (23) and
different Aspergillus nidulans xylanases
(4,
5) in industrial wine
yeast strains, which resulted in increased levels of some volatile
compounds.
Recently, we cloned two genes from Aspergillus
aculeatus encoding the
-L-rhamnosidases RhaA
and RhaB (18). In the
present work, the expression of the gene encoding the A.
aculeatus RhaA in a commercial wine yeast strain was investigated.
For this purpose, a 2.2-kb BamHI fragment carrying the
rhaA cDNA (18)
(which includes a 19-amino-acid signal sequence) was ligated between
the Saccharomyces cerevisiae glyceraldehyde-3-phosphate
dehydrogenase gene promoter (i.e., GPDp) and the
phosphoglycerokinase (i.e., PGKt) terminator and
polyadenylation signals in plasmid pG-1
(29), thereby generating
plasmid pR31. Plasmids pR31 and pG-1 (a control) were used
independently to transform the laboratory yeast strain W303-1A
(MATa ade2-1 his3-11,15
leu2-3,112 can1-100 trp1-1
ura3-1) using the LiAc-ss-DNA-polyethylene glycol
protocol described by Gietz and Woods
(7). Tryptophan
prototrophic transformants were selected on synthetic
dextrose minimal supplemented medium (0.67% yeast
nitrogen base without amino acids, 20 g of glucose per liter, 20 mg of
adenine per liter, 20 mg of uracil per liter, 20 mg of histidine per
liter, and 30 mg of leucine per liter) and analyzed for
-L-rhamnosidase production by plating on synthetic
dropout minimal-supplemented media containing 4-methylumbelliferyl
-L-rhamnopyranoside (MUR) as the substrate
(16). The hydrolysis of
MUR by the action of
-L-rhamnosidase resulted in
the release of 4-methylumbelliferone, which can be visualized under UV
illumination as fluorescent halos surrounding yeast colonies. Only
yeast cells transformed with pR31 showed the corresponding activity.
Eight randomly selected pR31 transformants and a control strain
transformed with pG-1 were grown in liquid selective medium for
18 h, and
-L-rhamnosidase activity was
determined in culture filtrates by using
p-nitrophenyl-
-L-rhamnopyranoside(pNPR) as the substrate
(18). As expected,
activity was detected only in pR31 transformants. These data confirm
that the A. aculeatus rhaA gene encodes an
-L-rhamnosidase and demonstrate that it can be
efficiently expressed in S. cerevisiae. From those
transformants exhibiting the highest levels of
-L-rhamnosidase activity, one of them, designated
YR6, was isolated for further studies. This transformant produces 130
mU of
-L-rhamnosidase activity per ml (specific
activity of 2.6 U/mg) after 72 h of growth in yeast
extract-peptone-dextrose (YPD) medium (1% yeast extract,
2% bacteriological peptone, 2% glucose).
Following
the protocol previously developed except for the use of 10 mM sodium
citrate buffer (pH 3.8)
(18), the A.
aculeatus RhaA enzyme was purified to apparent homogeneity from
transformant YR6 by using cation-exchange chromatography. The level of
recovery was 12% of that of the original
-L-rhamnosidase activity. Sodium dodecyl
sulfate-polyacrylamide gel electrophoresis of RhaA revealed a single
protein band after N deglycosylation of an apparent molecular mass of
70 kDa, a finding which was in agreement with the molecular mass of
RhaA purified from A. aculeatus
(18). Optimum pH and
stability as well as kinetic constants using pNPR as the substrate were
similar to those described for wild-type RhaA
(18). Certain factors
that may affect enzyme activity under enological conditions were also
tested (28). Ethanol at a
concentration of 15% (vol/vol) reduced RhaA activity by
55%, whereas at 20% (vol/vol) ethanol, the enzyme
retained 35% of its original activity. The presence of
SO2 at concentrations of up to 150 mg/liter, a level that
can be found at different stages during wine fermentation, did not
significantly affect enzyme activity. Glucose concentration did however
affect RhaA activity, and inhibition levels of 59, 84, and 93%
were observed for glucose concentrations of 0.15, 0.50, and 1 M,
respectively. Finally, to test the ability of RhaA to hydrolyze Muscat
must glycosides, the release of L-rhamnose was analyzed by
high-performance liquid chromatography using a Waters
Sugar-Pak I column (dimensions, 6.5 by 300 mm) at 80°C and a
refraction index detector. From our data, it is possible to conclude
that the heterologously expressed RhaA at a final concentration of 48
µg/ml in the incubation mixture is able to release 46 nmol of
L-rhamnose after 24 h of incubation at
40°C. These experiments indicate that the A. aculeatus
enzyme has enologically relevant properties.
In order to express
the A. aculeatus rhaA gene in a commercial wine yeast strain,
we isolated the
GPDp::rhaA::PGKt
expression cassette from plasmid pR31 as a 4.2-kb HindIII
fragment and ligated it into the HindIII site of plasmid
YEpCR21 (23), yielding
plasmids pR37 and pR38 (corresponding to each of the two possible
orientations with respect to the vector). The industrial wine yeast
strain T73
(24) was transformed with
plasmids YEpCR21 (control), pR37, and pR38, and positive colonies were
tested for
-L-rhamnosidase activity on YPD plates
supplemented with 1 µg of cycloheximide per ml and containing
MUR as the substrate. No activity was detected in T73 or in
a T73 strain (YR13) transformed with the vector YEpCR21.
However, transformants generated with pR37 and pR38 gave similar levels
of
-L-rhamnosidase activity, demonstrating that all
the signals necessary for expression (i.e., promoter and terminator)
are present on the 4.2-kb HindIII fragment and are independent
of orientation within the vector. From the latter, transformant YR8 was
selected and used for microvinification experiments. This transformant
secretes 600 mU of
-L-rhamnosidase per ml after
72 h of growth in YPD medium, a sixfold increase in
production with respect to strain YR6. Following previously described
protocols (9), we detected
no differences in RhaA cellular location between the strains YR6 and
YR8 (30% intracellular, 10% cell wall bound, and
60% extracellular).
Due to the fact that the enzymatic
hydrolysis of grape rhamnoglucosides occurs in two steps and that the
actions of an
-L-rhamnosidase and a
ß-D-glucosidase are needed, microvinification
experiments were carried out both singly and with a combination of
transformant YR8 expressing the A. aculeatus rhaA gene and a
T73 strain (YCB35) expressing the C.
molischiana gene bgln, which codes for a
ß-D-glucosidase
(28). Controls were done
with the untransformed T73 strain and the transformant YR13.
Microvinifications were performed in triplicate and carried out at
18°C in 375-ml glass bottles containing 300 ml of Muscat grape
must inoculated with 105 cells/ml. The progress of the
fermentation was similar in all the microvinifications (Fig.
1A). After 9 days of fermentation, the final wine products contained
approximately the same ethanol concentration: 11.8% (vol/vol) in
wines produced by T73 and 12.0% (vol/vol) in those
produced by the YR13 transformant and the combination of the
transformants YR8 and YCB35. Figure
1B and C show that
-L-rhamnosidase activity is only found in the
presence of transformant YR8 and that although
ß-D-glucosidase activity was mainly detected in the
presence of YCB35 strain, as previously reported
(28), very slight levels
of activity could be detected in all fermentations. The highest levels
of both enzyme activities occurred during the first 5 days of
fermentation. High levels of
-L-rhamnosidase
activity (470 mU/ml) were reached in the fermentation carried out by
strain YR8, whereas in the presence of the YCB35
transformant, around 90 mU of
-L-rhamnosidase
activity per ml was found. ß-D-Glucosidase activity
reached levels of approximately 15 mU/ml in the YCB35
fermentation and also in the presence of the YR8 transformant. Further
studies will be done to find out if the differences in
-L-rhamnosidase activity between single and double
fermentations are a consequence of differences in growth or enzyme
production rates. Regarding possible effects on aroma compounds,
volatiles were extracted by using C18 cartridges and
analyzed by gas chromatography-mass spectrometry
(8) (Fig.
2). Fermentations carried out by YCB35 alone showed marked
increases in trans-furanlinalool oxide and geraniol contents.
These data suggest the possible existence of monoglucosides of these
terpenes or the capability of the C. molischiana
ß-D-glucosidase to hydrolyze diglycosides with
different specificities, depending on the aglycon. Similar results have
been reported for purified C. molischiana
ß-D-glucosidase added directly to Muscat wine
(6). Although a slight
increase in linalool content was also detected, greater levels of
linalool were observed for wines cofermented by YCB35 and
the novel genetically modified yeast strain YR8 producing the A.
aculeatus
-L-rhamnosidase A. Moreover,
significant increases in
-terpineol and nerol were detected
only in wines produced by both yeast strains in combination compared to
individual T73 and YR13 fermentations. These data indicate
the biotechnological feasibility of the combination of wine yeast
strains producing
-L-rhamnosidase and
ß-D-glucosidase to increase free monoterpenol
content. To our knowledge, this is the first report on the use of a
genetically modified wine yeast strain producing an
-L-rhamnosidase of enological relevance. Problems
related to the social acceptance of genetically modified foods, mainly
in the European Union, make the use of such products an alternative for
the future but not for the present.
![]() View larger version (15K): [in a new window] |
FIG. 1. Analysis
of microvinifications carried out with the commercial wine yeast strain
T73 (), the YR13 control strain transformed with
the vector YEpCR21 ( ), the YCB35 transformant
strain expressing the C. molischiana
ß-D-glucosidase ( ), the YR8 transformant
strain expressing the A. aculeatus
-L-rhamnosidase ( ), and a combination of
the YCB35 and the YR8 strains ( ). Shown are the
kinetics of reducing sugar consumption (A) and
-L-rhamnosidase (B) and
ß-D-glucosidase (C) activities,
respectively, during the course of
fermentation.
|
![]() View larger version (32K): [in a new window] |
FIG. 2. Concentrations
expressed as mean values with Tukey's honestly significant
difference intervals at 95% confidence level of
various terpenols in the final wines produced in the microvinifications
(aroma threshold values in micrograms per liter:
trans-furanlinalool oxide and cis-furanlinalool
oxide, >6,000; linalool, 100; -terpineol and nerol,
400 to 500; geraniol, 130)
(19).
|
We thank E. Ibáñez and J. A. Tamayo for their invaluable assistance and Andrew P. MacCabe for critical reading of the manuscript.
This work was supported by a Ramón Areces grant and also by the CICYT AGL2002-01906 project (Spanish Ministry of Science and Technology-FEDER).
|
|
|---|
-L-rhamnosidase of
oenological interest. Z. Lebensm.-Unters.-Forsch.
203:522-527.
-L-rhamnosidase from Aspergillus
terreus of interest in winemaking. J. Food Sci.
66:1-6.
-L-Arabinofuranosidases from
Aspergillus terreus with potential application in enology:
induction, purification and characterization. J. Agric. Food
Chem.
45:2379-2383.[CrossRef]
-L-rhamnosidase from Aspergillus niger.FEMS Microbiol. Lett.
157:279-283.[Medline]
-L-rhamnosidase from Aspergillus
nidulans. Lett. Appl. Microbiol.
31:198-202.[CrossRef][Medline]
-L-rhamnosidases, RhaA and RhaB, from
Aspergillus aculeatus. Appl. Environ.
Microbiol.
67:2230-2234.
-L-rhamnosidase of potential oenological interest.Lett. Appl. Microbiol.
28:383-388.[CrossRef]
-L-arabinofuranosidase and
ß-D-glucopyranosidase from Aspergillus
niger on a chitosan derivative to increase the aroma of wine. Part
II. Enzyme Microb. Technol.
23:413-421.[CrossRef]
-L-arabinofuranosidase and
ß-D-glucopyranosidase from Aspergillus
niger to increase the aroma of wine. Part I. Enzyme Microb.
Technol.
22:298-304.[CrossRef]
-L-rhamnopyranosidase from
Aspergillus niger to increase the aroma of Moscato wine.Enzyme Microb. Technol.
27:522-530.[CrossRef][Medline]
-L-rhamnopyranoside from
Aspergillus niger for application in beverage technologies.Enzyme Microb. Technol.
28:427-438.[CrossRef][Medline]
-L-arabinofuranosidase
from Pichia capsulata X91. Biosci. Biotechnol.
Biochem.
64:1181-1188.[CrossRef][Medline]
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»