Determination of reduced cysteine in oenological cell wall fractions of saccharomyces cerevisiae

Article abstract of DOI:10.1021/jf904047u

Compounds containing cysteine residues, such as glutathione, can affect the redox potential of must and wine by reduction of o-quinones and hydrogen peroxide. The oenological yeast cell wall fractions contain cysteine residues in their protein structure, and they could affect both oxidative and odor properties of wine. An analytical approach based on the derivatization of cysteinyl residues with p-benzoquinone followed by reversed-phase high-performance liquid chromatography separation was developed to quantify glutathione and free and protein cysteine in 16 Saccharomyces cerevisiae strains and 12 commercial samples of yeast mannoproteins, hulls, and lysates. The chemical modifications induced by the Maillard reaction following the industrial preparation of such fractions were evaluated as well. Lysates showed the highest protein cysteine content and high contents of glutathione and free cysteine. Mannoproteins showed an intense Maillard reaction (furosine >60 mg/100 g protein), and most of the samples were able to bind thiol compounds with a potentially detrimental effect toward the thiol-related odors in wine.

Full text of DOI:10.1021/jf904047u

J. Agric. Food Chem. 2010, 58, 4565–4570 4565
DOI:10.1021/jf904047u
Determination of Reduced Cysteine in Oenological Cell Wall
Fractions of Saccharomyces cerevisiae
A
NTONIO
T
IRELLI,* DANIELA
F
RACASSETTI
,
AND
IVANO
D
E
N
ONI
ꢀ
Dipartimento di Scienze e Tecnologie Alimentari e Microbiologiche (DISTAM), Universita degli Studi di
Milano, via Celoria 2, 20133 Milano, Italy
Compounds containing cysteine residues, such as glutathione, can affect the redox potential of must
and wine by reduction of o-quinones and hydrogen peroxide. The oenological yeast cell wall
fractions contain cysteine residues in their protein structure, and they could affect both oxidative and
odor properties of wine. An analytical approach based on the derivatization of cysteinyl residues with
p-benzoquinone followed by reversed-phase high-performance liquid chromatography separation
was developed to quantify glutathione and free and protein cysteine in 16 Saccharomyces
cerevisiae strains and 12 commercial samples of yeast mannoproteins, hulls, and lysates. The
chemical modifications induced by the Maillard reaction following the industrial preparation of such
fractions were evaluated as well. Lysates showed the highest protein cysteine content and high
contents of glutathione and free cysteine. Mannoproteins showed an intense Maillard reaction
(furosine >60 mg/100 g protein), and most of the samples were able to bind thiol compounds with a
potentially detrimental effect toward the thiol-related odors in wine.
KEYWORDS: Cysteine; glutathione; mannoprotein; yeast; hull; lysate; HPLC; quinone
INTRODUCTION
show the Cys residues of yeast mannoproteins (MPs) to be
effective antioxidants if they are accessible to oxidizing mole-
cules (10). Moreover, it is well-known that white wines and spark-
ling wines develop faster nonenzymatic browning, oxidation, and
loss of odor-related thiols after yeast lees are removed (6,11). The
release of MPs from yeast lees as well as the addition of yeast
glycoproteins to wine could increase the antioxidant Cys content.
Nevertheless, little is known about the Cys content of yeast cell
wall fractions as well as about the antioxidant behavior of yeast
lysates, hulls, and MPs in wine.
The chemical modifications of the oenological yeast fractions
following their preparation procedures have not been evaluated
to date. Indeed, the technologies applied to produce oenological
yeast fractions can chemically modify the MP moiety of the yeast
cell wall. Heat treatments and drying conditions, as those adopted
to produce yeast fractions, can promote browning due to
Maillard reaction (12). The formation of R-dicarbonyl com-
pounds (12) and degradation of Cys residues (13) occur as well.
Such reactions likely modify the RPC content of the yeast frac-
tions, affecting their antioxidant properties. Despite the growing
interest in exploiting such properties of yeast cell wall fractions in
winemaking, a reliable analytical method is not currently
available to quantify RPC. The adoption of classical analytical
approaches such as the Ellman’s method (14) for the determi-
nation of glycoprotein Cys, is hindered by the dark-yellow color
produced by the MP solutions. On these bases, the present paper
describes a reliable analytical method to quantify both the
glycoprotein and the nonproteinaceous Cys content of yeast
fractions. The method was applied to assess the antioxidant
properties of commercial samples of yeast lysates, hulls, and
MPs on the basis of their Cys content. The furosine index was
Thiol compounds of grape and yeast can strongly affect the
sensorial properties of wine. Some aliphatic thiols are involved in
the varietal aroma of wine even though their perception threshold
is in the order of a few tens of nanomoles per liter (1). Barrel aging
and bottle storage can decrease the concentration of such
compounds below the perception level in wine because of their
reaction with o-quinones and hydrogen peroxide arising from
phenol oxidation (2). The reaction of quinones with the odor-
related thiols, as well as other thiol compounds, produces thiol-
substituted hydroquinones (3) or disulfide compounds lacking of
aromatic properties (4, 5). Glutathione (GSH) can preserve wine
odor since it can compete against the odor-related thiols, for
oxidation if its concentration exceeds a few micromoles per
liter (6). Cysteine (Cys) residues can also hinder the alterative
or atypical aging of white wine by preventing the formation of
Maillard reaction-related compounds such as sotolone (7) and
2,5-dimethyl-4-hydroxy-3(2H)-furanone (8). The reducing pro-
perty of GSH toward the o-quinones produced by the tyrosinase
activity in must is well-known as well (9).
The mercaptans responsible for the reduced odors in wine
represent a further group of thiol compounds, and they can be
removed by racking wine under aerating conditions or by barrel
aging on yeast lees. Mercaptans are converted to thiol-substituted
hydroquinones by reaction with the o-quinones or with the Cys
residues of the yeast cell wall to give disulfides (4). The antioxi-
dant properties of the reducing protein Cys (RPC) from yeast lees
have not been extensively investigated in wine, but some results
*To whom correspondence should be addressed. Fax: þ39 02
50316672. E-mail: antonio.tirelli@unimi.it.
© 2010 American Chemical Society
Published on Web 04/01/2010
pubs.acs.org/JAFC

4566 J. Agric. Food Chem., Vol. 58, No. 8, 2010
Tirelli et al.
applied for assessing the heat damage promoted by the techno-
logical processes adopted for the production of yeast fractions.
MATERIALS AND METHODS
Chemicals. Cys, 3-mercaptopropionic acid (3MPA), and p-benzoqui-
none (pBQ) were purchased from Fluka (Switzerland). GSH, starch, and
trifluoroacetic acid (TFA) were from Sigma-Aldrich (St. Louis, MO).
Citric acid was purchased from J. T. Baker (Phillipsburg, NJ), and the
oenological lysozyme was from Intec (Verona, Italy). High-performance
liquid chromatography (HPLC) grade methanol was purchased from
Panreac (Barcelona, Spain), and HPLC water was obtained by Milli-Q
system (Millipore Filter Corp., Bedford, MA).
Samples. The following commercial samples of yeast fractions for
oenological use supplied by six different producers as dried products with
unknown compositions were evaluated: four MPs, four hulls, four lysates,
and two extracts. Additionally, 16 commercial dried yeast starters em-
ployed in winemaking were evaluated:11 strains were from Lallemand Inc.
(Ontario, Canada), and 5 strains were from Dal Cin (Milan, Italy).
Derivatization of Cys Residues in the Samples. Samples were added
with 100 μL of 400 μM methanol solution of pBQ. After 1 min of mixing,
1 mL of 500 μM 3MPA dissolved in 0.3 M citrate buffer, pH 3.5, was added.
Determination of Cys Residues in Yeast Fractions. At least 50 mg
of insoluble sample (active dried yeast, yeast hulls, and yeast lysates) was
dispersed in 50 mM of citrate buffer, pH 5.0, to obtain a 2 mL suspension
with sample concentrations in the range of 20-100 g/L. The dispersed
sample was centrifuged at 5000g for 15 min at 15 °C by a thermostatted
Sorvall centrifuge (Thermo, Waltham, MA). The supernatants were
diluted 1-10 fold, and 2 mL was submitted to derivatization. The preci-
pitated material was resuspended in 5 mL of 50 mM citrate buffer, pH 5.0,
and submitted to the derivatization reaction. After derivatization, 1 mL of
the suspension was centrifuged at 14000g for 5 min at 25 °C by a
thermostatted benchtop centrifuge (Hettich, Tuttlingen, Germany), and
the supernatant was submitted to HPLC separation. To recover the
derivatized 3MPA potentially adsorbed on the insoluble yeast fraction,
the precipitated material was rinsed with 2 mL of 0.1% hydrochloric
ethanol as described by Ummarino et al. (15), carefully resuspended, and
centrifuged at 14000g for 5 min. The supernatant was dried under vacuum,
redissolved in 1 mL of water, and injected into HPLC.
Figure 1. (A) General scheme of the derivatization reaction of Cys thiols
with pBQ and (B) UV spectra obtained for derivatized Cys (Cys-HQ), GSH
(GSH-HQ), and 3-mercaptopropanoic acid (3MPA-HQ).
(Phenomenex, Torrence, CA). Eluting solvents were water/trifluoroacetic
acid (0.05% v/v) and methanol; the concentration of the latter increased
from 10 to 35% in 18 min during the elution gradient at 1.0 mL/min flow.
Chromatographic data were acquired and processed by Millenium
software v. 4.0 (Waters).
HPLC/Electrospray Ionization-Mass Spectrometry (ESI-MS).
For MS detection, the LCQ Deca XP spectrometer, controlled by the
Excalibur software (Thermo Finnigan, San Jose, CA), was operated in
positive ion mode. A postcolumn flow splitter was used to introduce 1:15
of the HPLC flow stream into the ESI source. The ESI interface and the
ion optics settings were as follows: spray potential, 5.0 kV; nebulization gas
(nitrogen) relative flow value, 10; capillary temperature, 275 °C; and cone
voltage, 30 V. Full-scan mass spectra were acquired scanning the range
50-800 m/z. Mass accuracy was ensured by calibration with a mixture of
caffeine, reserpine, and the tripeptide PFK (in methanol:water 1:1, 0.1%
acetic acid) infused separately.
MP and yeast extract samples were prepared by dissolving 100-200 mg
of sample in 2 mL of 50 mM citrate buffer, pH 5.0. The solution was
submitted to a derivatization reaction, and then, it was ultrafiltered using
3 kDa cutoff Microcon membranes (Millipore, Billerica, MA) at 14000g
for 100 min at 25 °C by the thermostatted benchtop centrifuge. The
permeate was submitted to HPLC separation.
The retentate was added with 1 mL of 0.1% hydrochloric ethanol and
centrifuged at 14000g for 5 min. The supernatant was dried under vacuum,
and the dried material was redissolved in 500 μL of water before the HPLC
separation. Each sample was analyzed in at least duplicate.
Precision Parameters. The repeatability of the Cys residues evalua-
tion method was assessed by submitting one sample of yeast hulls and two
samples of yeast lysates to five determinations each. The response linearity
of the method was assessed for RPC concentrations up to 180 μM by
dispersing 27-100 g/L of a yeast hull sample containing 0.18 mmol RPC/
100 g product in the 50 mM citrate buffer. The response linearity of
the method for higher RPC concentrations (210-800 μM) was attained
by analyzing dispersions (10-40 g/L) of a yeast hull sample containing
2.1 mmol RPC/100 g product.
Evaluation of Matrix Effects on the Cys Residues Determination.
To evaluate possible interferences arising from high polysaccharide
contents, one sample of yeast lysate (L2) containing GSH and protein
Cys was assayed with or without addition of 30 or 80 g/L of amylose
or caramelized amylose. The caramelization was obtained by exposing
2 g of amylose suspended in 2 mL of 0.1% HCl at 110 °C for 24 h. Inter-
ference of protein disulfides (cystine) was evaluated by submitting 5 g/L
lysozyme solution in 50 mM citrate buffer, pH 5.0, to protein Cys evalua-
tion.
HPLC Separation of Derivatized Thiol Compounds. The reversed-
phase (RP)-HLPC of the thiol-substituted hydroquinones and p-hydro-
quinone (pHQ) was performed with a Waters Alliance 2695 (Milford,
MA) equipped with a photodiode array detector Waters 2996. The sepa-
Quantification of Thiol Compounds. Cys and GSH were quantified
chromatographically by the external standard method in both the soluble
and the insoluble fractions. Standard solutions containing Cys and GSH
concentrations up to 100 μM were prepared in 50 mM citrate buffer, pH
5.0.
The concentration of RPC in the evaluated sample fraction was
calculated as follows:
½RPCꢁ ¼ ½pBQꢁ - ð½Cysꢁ þ ½GSHꢁ þ ½pHQꢁ þ ½3MPA-HQꢁÞ
where [pBQ] = concentration of pBQ used for the derivatization, [Cys] =
concentration of Cys quantified chromatographically, [GSH] = concentra-
tion of GSH quantified chromatographically, [3MPA-HQ] = concentration
of S-3-mercaptopropionyl-hydroquinone quantified chromatographically,
and [pHQ] = concentration of the underivatized hydroquinone quantified
chromatographically.
Evaluation of Protein Content and Furosine Level. The protein
content and furosine level were determined as described by Resmini et al. (16)
with the following modification: 200 mg of sample (corresponding to
about 20 mg of protein) was added with 8 mL of 8 M HCl and submitted to
acid hydrolysis at 110 °C for 23 h before solid-phase extraction (SPE).
Total Cys Content. The total amount of Cys (i.e., Cys þ 2 ꢀ cystine)
was assayed on the acid-hydrolyzed sample obtained from the furosine
determination. After purification with SPE on a C18 cartridge, the sample
was dried under vacuum and redissolved in 0.15 M sodium carbonate
buffer, pH 8.6. The buffered solution was then submitted to Cys
determination according to Krause et al. (17).
˚
ration column was a hexyl-phenyl column, 250 mm ꢀ 4.6 mm, 5 μm, 110 A

Article
J. Agric. Food Chem., Vol. 58, No. 8, 2010 4567
Figure 2. HPLC separation of Cys-HQ, GSH-HQ, 3MPA-HQ, and pHQ obtained from derivatization with pBQ of (A) standard water solutions of Cys, GSH,
and their mixture and (B) MP sample added with Cys (M), yeast lysate (L), and yeast hull (H). HPLC-ESI/MS spectra of Cys-HQ (C) and GSH-HQ (D) are
reported.
RESULTS AND DISCUSSION
(Figure 1A) is fast and stoichiometric at room temperature, and
it allows Cys and GSH to be detected spectrophotometrically as
S-cysteinyl-p-hydroquinone (Cys-HQ) and S-glutathionyl-p-
hydroquinone (GSH-HQ), respectively (Figure 1B). The thiol deri-
vatization procedure was followed by the addition of 3MPA to
remove the exceeding amount of pBQ. The latter could be reduced
to pHQ by oxidation of the thiol-substituted hydroquinones to
thiol-substituted p-quinones, which are able to bind a further thiol
molecule, so forming a dithiol-substituted hydroquinone (18).
Moreover, pBQ can readily polymerize to produce brown com-
pounds.
The thiol property to reduce quinones to thiol-substituted
hydroquinones, as it spontaneously occurs for the o-quinones
in must and wine, was adopted to assess the Cys residues in the
yeast fractions. For this purpose, the symmetric pBQ was used as
a derivatizing agent to prevent the formation of multiple hydro-
quinone derivatives and to obtain the single mono derivative for
each thiol molecule. The molecular size of pBQ is comparable to
that of the o-quinones detectable in must and wine, and it can
react with the thiol residues actually accessible in the glycoprotein
structures. The formation of thiol-substituted hydroquinones

4568 J. Agric. Food Chem., Vol. 58, No. 8, 2010
Tirelli et al.
The formation of Cys-HQ and GSH-HQ in citrate buffer
solution/suspension containing Cys and GSH (Figure 2A) or
yeast fractions (Figure 2B) was confirmed by HPLC-ESI/MS
(Figure 2 C, D). Such thiol-substituted hydroquinones, including
S-mercaptopropionyl-p-hydroquinone (3MPA-HQ), showed a
maximum absorption at 303 nm wavelength (Figure 1B). The
HPLC pattern of Figure 2A also shows the formation of both
pHQ and minor amounts of dithio-substituted hydroquinones
following the derivatization reaction. The exceeding amount of
pBQ could be effectively removed by the addition of sulfur
dioxide, but the amount of the products obtained after such a
reaction is affected by a number of factors (19). On the contrary,
the addition of 3MPA allows us to verify whether the amount of
pBQ exceeds the content of Cys residues and to assess the level of
residual pBQ after thiol derivatization. As a result of the developed
analytical approach, the amount of the RPC was calculated by the
difference between the amount of added pBQ and the amounts of
GSH, Cys, pHQ, and 3MPA-HQ chromatographically evaluated.
Because the insoluble cell wall fractions were submitted to
derivatization with pBQ after removal of the soluble Cys residues,
the amount of pBQ reacted with γ-glutamyl-Cys and cysteinyl-
glycine potentially present in yeast preparations was assumed as
negligible. Such dipeptides represent about 4% of GSH in
yeast (20); therefore, their contribution was not taken into
account also for the MP samples where the nonproteinaceous
Cys thiols were never detected. The amounts of dithiol-substi-
tuted hydroquinones were not included in the calculation since
their values were close to the limit of quantification (signal-
to-noise ratio >10). The removal of the soluble material from the
samples allows an effective RPC determination in wine yeast lees
since sulfur dioxide interferences are minimized (data not shown).
The precision parameters for the quantification of Cys, GSH,
and RPC were assessed by analyzing one sample of yeast hull and
two samples of yeast lysates (Table 1). An average value of 5.4%
can be assumed for the relative standard deviation (RSD) of Cys
thiol concentrations exceeding 10 μM.
Table 1. Analytical Values (μM) and RSDs Obtained from Five Replicated
Determinations Performed on One Sample of Yeast Hull (S4) and Two
Samples of Yeast Lysates (L2 and L4)
Cys
L2
GSH
L2
RPC
L2
S4
L4
S4
L4
92
S4
L4
43.3 0.0
43.0 0.0
43.8 0.0
44.2 0.0
43.1 0.0
43.0
1.8 13.4 10.0
2.1 15.0
173
171
165
164
186
172
266
299
311
286
280
288
6.0
246
241
254
269
248
252
9.7 101
2.9 16.3 10.4 102
2.9 13.8 11.0 103
1.9 14.6 11.3
2.3 14.6 10.5
89
97
mean
RSD (%)
The calibration curves of Cys and GSH (Figure 3A) showed
linear and similar analytical responses for concentrations up to
1.2
23.4
7.7
6.4
6.9
5.1
4.2
Figure 3. Analyticalresponseoftheproposedanalyticalmethodobtainedfor(A) standardsolutionsofCysandGSHand(B) high (solidline) andlow(dashed
line) RPC concentrations. Data of duplicated determinations are reported.

Article
J. Agric. Food Chem., Vol. 58, No. 8, 2010 4569
Table 2. Effect of Amylose or Caramelized Amylose on the Quantification of
the Cys Residues in a Sample of Yeast Lysatea
Table 3. Characterization of 16 Samples of Oenological Dry Saccharomyces
cerevisiae Strains According to Their Content of Cys Forms, Protein Content,
and Intensity of the Maillard Reaction Expressed as Furosine Levela
amount (mmol/100 g sample)
reducing Cys
mmol/100 g
amylose
caramelized amylose
added amount (g/L)
GSH
RPC
GSH
RPC
furosine
(mg/100 g
protein)
overall
Cys
(mmol/100 g)
0
30
80
0.22-0.21
0.22-0.22
0.21-0.22
1.59-1.44
1.52-1.56
1.57-1.60
0.23-0.23
0.25-0.26
0.27-0.28
1.38-1.35
1.35-1.33
1.38-1.35
protein
(g/100 g)
sample
GSH
RPC
1
11.3
11.0
10.5
12.0
12.0
10.3
12.0
12.0
11.2
13.9
11.0
9.2
2.7
0.92
0.45
0.71
0.39
0.73
0.63
0.77
0.60
0.83
0.55
0.45
0.54
0.58
0.82
0.58
0.63
0.64
0.76
0.80
0.86
0.88
0.89
0.90
0.91
0.91
0.91
0.95
0.96
0.97
0.99
1.02
1.05
1.28
0.94
7.1
a Results of duplicated determinations are reported.
2
3
4
3.6
7.6
6.8
6.9
100 μM. Because no standard material is commercially available
for RPC quantification, the range of linear response was eva-
luated in yeast hull suspensions containing RPC levels up to
800 μM. Suspensions containing RPC amounts higher than 210
μM were prepared by dispersing 10-40 g/L of a yeast hull sample
containing 2.1 mmol RPC/100 g product. To not affect the
representativeness of the sampling, amounts higher than 50 mg
were used to prepare the suspensions intended for linear res-
ponse assessement. For this reason, the linear response for low
(<180 μM) RPC concentrations was evaluated analyzing sus-
pensions (27-100 g/L) of a yeast hull sample containing 0.18
mmol RPC/100 g product. Under the adopted conditions, linear
responses were observed for the entire range tested (Figure 3B).
According to the signal-to-noise ratio, the detection limits for
Cys and GSH by means of HPLC analysis were 0.30 and 0.26 μM,
respectively; for the same compounds, the quantification limits
were 1.0 and 0.85 μM, respectively. To assess whether RPC
quantification was affected by matrix effects arising from poly-
saccharides or disulfide bonds, the quantification was performed
after the addition of increasing amounts of amylose or carame-
lized amylose to a yeast lysate. No significant difference (p <
0.05) was found for RPC and GSH quantification in the presence
of either amylose or caramelized amylose (Table 2). Similarly, no
interference was observed when 5 g/L lysozyme, a protein
containing four disulfide bonds, was added to water solutions
containing GSH or Cys.
The analytical approach was first applied to the characteriza-
tion of 16 commercial samples of oenological dry active yeasts to
evaluate their natural thiol content. Because the yeast samples
tested were aimed to both effective propagation and alcoholic
fermentation, a very low intensity of Maillard reaction was
expected. Indeed, the furosine values were lower than 8 mg/100
g protein (Table 3), and they agree with the values detected in
other unprocessed biological material (21-23), although no data
are reported in the literature for dry active yeasts. Amounts of
RPC in the range 0.76-1.28 mmol/100 g were detected in yeast
samples as well as GSH levels up to 0.92 mmol/100 g (Table 3).
GSH is a cytoplasmatic metabolite, and it likely arises from the
lysis of yeast cells during drying. None of the studied samples
contained free Cys. Because the MP fractions represent about
10% of the yeast cell wall and they are mainly linked to the outer
layer of the glucan backbone (24), oenological glycoprotein
fractions with a higher content of Cys residues can be potentially
obtained. The commercial MP samples showed levels of RPC
very different from each other. Samples M1 and M2 had the
lowest Cys levels (Table 4). Sample M2 presented the highest
furosine level, thus suggesting a strong heat damage, which
agreed with the dark brown color and the burnt odor character-
izing the sample. The RPC level of sample M4 was close to the
values found in the dry yeast samples. Overall, the RPC levels
were far from those needed to obtain effective antioxidant activity
if the usual amounts of MPs added to wine (200-500 mg/L) are
taken into account. Neither GSH nor Cys was detected in the
5
6
7
8
9
10
7.8
4.5
11
12
13
16.9
13.6
14.9
11.2
12.1
2.3
6.4
7.8
7.8
14
15
16
average
5.1
6.8
a Overall Cys refers to Cys þ 2 ꢀ cystine.
Table 4. Characterization of Commercial Yeast Cell Wall Fractions and Yeast
Extracts According to Their Contents of Free Cys, GSH, RPC, and Overall Cys
(Cys þ 2 ꢀ Cystine)a
Cys
mmol/100 g
furosine
(mg/100 g
protein)
protein
sample (g/100 g)
unrecovered free GSH RPC overall
M1
M2
M3
M4
H1
H2
H3
H4
L1
2.86
10.51
9.02
28
254
62
67
17
12
6
<0.01
<0.01
0.04
0.41
0.04
0.03
0.04
0
0
0
0
0
0
0
0
0
0.03
0
0.55
0.54
2.3
0
0
0.03
0.51
0.08
0.07
0.14
8.77
7.48
0
3.1
3.3
0
8.75
8.87
0
4.4
3.3
0
11.88
9.96
12
3
2.6
0
0.85 0.86 10.6
0.45 0.73
0.33 1.4
0
6.9
15.8
L2
17.94
16.84
14.33
21.68
23.87
5
0
0
L3
38
3
0
0.32 4.6
0.07 2.8
0.81 12
L4
0
1.3
1.1
8.1
37.4
E1
E2
154
20
0.46
0.29
0
0
0
0
0.76 25.4
a The protein content and the intensity of the Maillard reaction expressed as
furosine level are reported. The unrecovered Cys refers to the amount of analytically
missing Cys after the addition of known amounts of Cys to samples lacking in
nonproteinaceous Cys residues (M, MP; H, hull; L, lysate; and E, extract).
commercial MP samples. Surprisingly, the amount of 3MPA-HQ
increased when Cys was added to samples before derivatization.
Such a behavior occurred with most of the samples not containing
free Cys and GSH, and it was likely due to the presence of oxidi-
zed phenol aminoacids in the protein structure (10). The sensorial
properties of wines containing odor-related thiols can be detri-
mentally affected by the addition of MPs capable of binding high
amounts of thiols like sample M4. In this regard, 200-400 mg/L
of such MP could deplete up to 250-500 μM of thiols from wine.
Contrarily, such MPs could be usefully added to wine as an early
treatment for removing the reduced odors since they can link

4570 J. Agric. Food Chem., Vol. 58, No. 8, 2010
Tirelli et al.
mercaptans, so avoiding exposure of wine to oxygen or the
addition of copper sulfate. The high furosine values of MP
samples indicate a strong extent of the Maillard reaction. These
commercial MPs are capable of rapidly increasing the formation
of atypical aging-related compounds in wine. Therefore, their
addition to wine to improve colloidal and tartaric stabilities or to
modify the astringency and the viscosity can also deplete wine
odor and decrease wine shelf life.
evaluation of its antioxidant properties. Food Res. Int. 2005, 38,
395–402.
(4) Lavigne, V.; Dubourdieu, D. Mise en e
´
´
vidence et interpretation de
ꢀ
l’aptitude des lies a e
´
liminer certains thiols volatils du vin. J. Int. Sci.
Vigne Vin 1996, 30, 201–206.
(5) Vasserot, Y.; Steinmetz, V.; Jeandet, P. Study of thiol consumption
by yeast lees. Antonie van Leeuwenhoek 2003, 83, 201–207.
(6) Lavigne, V.; Dubourdieu, D. Affinamento sulle fecce e freschezza dei
vini bianchi. VigneVini 2004, 31, 58–66.
Yeast hulls having a high RPC content could protect wine
during barrel aging from the oxidative effect of micro-oxygena-
tion. The analyzed commercial samples presented RPC levels
close to the values detected for MPs as well as similar capability to
bind free Cys. The amounts of GSH and free Cys revealed in
sample H4 were likely due to GSH and Cys addition during
manufacturing to increase the reducing properties of the product.
Amounts of RPC effective as antioxidant were detected in the
yeast lysates L2 and L4. The former also contained high levels of
GSH. The amounts of GSH detected in samples L2 and L3
accounted for 0.8-1.2% of yeast dry weight as reported in the
literature (25). Nevertheless, their overall Cys content was about
twice higher than the level found in the dry yeast samples, and
likely, an enrichment with exogenous GSH occurred.
Because the cytoplasmic GSH represents about 0.5-1% of
yeast dry weight, highamounts ofGSH were expected tobe found
in yeast extracts. Moreover, in spite of the high value of total Cys
(about 30 mmol/100 g sample), no GSH was detected in the two
extract samples, and up to 0.50 mmol Cys/100 g sample was
combined after the addition of Cys to the same samples. Such
data suggest anintense oxidation ofthe yeast extracts likelydue to
the chemical/heat damage arising from the industrial production
as supported by the high furosine values observed.
(7) Blank, I.; Lin, J.; Fumeaux, R.; Welti, D. H.; Fay, L. B. Formation of
3-hydroxy-4,5-dimethyl-2(5H)-furanone (Sotolone) from 4-hydro-
xy-L-isoleucine and 3-amino-4,5-dimethyl-3,4-dihydro-2(5H)-fura-
none. J. Agric. Food Chem. 1996, 44, 1851–1856.
(8) Naim, M.; Wainish, S.; Zehavi, U.; Peleg, H.; Rouseff, R. L.; Nagy,
S. Inhibition by thiol compounds of off-flavour formation in stored
orange juice. 1. Effect of L-cysteine and N-acetyl-L-cysteine on 2,5-
dimethyl-4-hydroxy-3(2H)-furanone formation. J. Agric. Food
Chem. 1993, 41, 307–312.
(9) Singleton, V. L.; Salgues, M.; Zaya, J.; Trousdale, E. Caftaric acid
disappearance and conversion to products of enzymic oxidation in
grape must and wine. Am. J. Enol. Vitic. 1985, 36, 50–56.
(10) Jaehrig, S. C.; Rohn, S.; Kroh, L. W.; Fleischer, L. G.; Kurz, T. In
vitro antioxidant activity of (1f3),(1f6)-b-D-glucan and protein
fractions from Saccharomyces cerevisiae cell wall. J. Agric. Food
Chem. 2007, 55, 4710–4716.
(11) Feuillat, M. Yeast macromulecoles: origin, composition and enolo-
gical interest. Am. J. Enol. Vitic. 2003, 54, 211–213.
(12) Friedman, M. Food browning and its prevention: an overview.
J. Agric. Food Chem. 1996, 44, 631–652.
(13) Friedman, M. Chemistry, biochemistry, nutrition, and microbiology
of lysinoalanine, lanthionine, and histidinoalanine in food and other
proteins. J. Agric. Food Chem. 1999, 47, 1295–1319.
(14) Ellman, G. L. Tissue sulfydril groups. Arc. Biochem. Biophys. 1959,
82, 70–71.
(15) Ummarino, I.; Garcia-Moruno, E.; Di Stefano, R. Interazione
polifenoli;Scorze di lievito. Riv. Vitic. Enol. 2001, 54, 37–45.
(16) Resmini, P.; Pellegrino, L.; Battelli, G. Accurate quantification of
furosine in milk and dairy products by a direct HPLC method.
Ital. J. Food Sci. 1990, 2, 173–183.
(17) Krause, I.; Bockhardt, A.; Neckermann, H.; Henle, T.; Klostermeyer,
H. Simultaneous determination of amino acids and biogenic amines by
reversed-phase high-performance liquid chromatography of the dabsyl
derivatives. J. Chromatogr. 1995, 715, 67–79.
(18) Cilliers, J. J.; Singleton, V. L. Caffeic acid autoxidation and the
effects of thiols. J. Agric. Food Chem. 1990, 38, 1789–1796.
(19) Wedzicha, B. L. Principles, properties and reactions. In Chemistry of
Sulphur Dioxide in Foods; Elsevier Applied Science Publishers: Lon-
don, 1984; pp 78-83.
(20) Rao, Y.; Xiang, B.; Bramanti, E.; D’Ulivo, A.; Mester, Z. Deter-
miantion of thiols in yeast by HPLC coupled with LTQ-Orbitrap
mass spectrometry after derivatization with p-(hydroxymercuri)-
benzoate. J. Agric. Food Chem. 2010, 58, 1462–1468.
A wide range of oenological yeast fractions are commercially
available, and yeast hulls, lysates, and MPs effective against wine
oxidation and wine atypical aging can be potentially obtained.
The data reported in this work show that a number of such
oenological samples do not have useful reducing properties.
Moreover, the technologies applied for their production are not
suitable for preserving the RPC content of the yeast fractions. On
these bases, both odor and antioxidant properties of wine could
be potentially endangered by using most of the studied samples,
which cannot be considered profitable in winemaking.
The RPC content of yeasts suggests that more useful fractions
could be obtained to protect the odor-related thiols and to
increase the shelf life of wine. In this regard, the adoption of
specific culture media to increase GSH and RPC contents in
yeasts, the selection of yeast strains with improved MP release
properties, and the use of specific enzymatic procedures could
greatly enhance the oenological properties of yeast fractions.
(21) Ruiz, J. C.; Guerra-Hernandez, E.; Garcia-Villanova, B. Furosine is
a useful indicator in pre-baked breads. J. Sci. Food Agric. 2004, 84,
366–370.
ACKNOWLEDGMENT
(22) Morales, V.; Sanz, M. L.; Martin-Alvarez, P. J.; Corzo, N. Com-
bined use of HMF and furosine to assess fresh honey quality. J. Sci.
Food Agric. 2009, 89, 1332–1338.
(23) Cattaneo, S.; Masotti, F.; Rosi, V. Evaluation of some heat-treat-
ment indices in UHT milk marketed in Italy. Ital. J. Food Sci. 2008,
20, 553–565.
Thanks are due to Dr. Charikleia Mavrommati for her
practical support and to Dr. Maria Manara of Dal Cin Co. for
providing samples and oenological support.
LITERATURE CITED
(24) Lesage, G.; Bussey, H. Cell wall assembly in Saccharomyces cerevi-
siae. Microbiol. Mol. Biol. Rev. 2006, 70, 317–343.
(25) Rauhut, D. Usage and formation of sulphur compounds. In Biology
(1) Tominaga, T.; Beltenweck-Guyot, R.; Peyrot Des Gachons, C.;
Dubourdieu, D. Contribution of volatile thiols to the aromas of
white wines made from several Vitis vinifera grape varieties. Am. J.
Enol. Vitic. 2002, 51, 178–181.
(2) Blanchard, L.; Darriet, P.; Dubourdieu, D. Reactivity of 3-mercap-
tohexanol in red wine: Impact of oxygen, phenolic fractions, and
sulfur dioxide. Am. J. Enol. Vitic. 2004, 55, 115–120.
€
of Microorganisms on Grapes, in Must and in Wine; Konig, H.,
€
Unden, G., Frohilich, J., Eds.; Springer-Verlag: Berlin, 2009; pp
181-207.
Received for review November 18, 2009. Revised manuscript received
March 4, 2010. Accepted March 10, 2010.
(3) Bassil, D.; Makris, D. P.; Kefalas, P. Oxidation of caffeic acid in the
presence of L-cysteine: Isolation of 2-S-cysteinylcaffeic acid and