koyama 2007

Inuence of Maceration Temperature in Red Wine Vinication on Extraction

of Phenolics from Berry Skins and Seeds of Grape (Vitis vinifera)

Kazuya KOYAMA,y Nami GOTO-YAMAMOTO, and Katsumi HASHIZUME

National Research Institute of Brewing, 3-7-1 Kagamiyama, Higashihiroshima, Hiroshima 739-0046, Japan

Received November 7, 2006; Accepted January 31, 2007; Online Publication, April 7, 2007

[doi:10.1271/bbb.60628]

The extraction of phenolics from berry skins and

seeds of the grape, Vitis vinifera cv. Cabernet Sau-

vignon, during red wine maceration and the inuence of

dierent temperature conditions (cold soak and/or

heating at the end of maceration) were examined.

Phenolics contained mainly in berry skins, viz., antho-

cyanin, avonol, and epigallocatechin units within

proanthocyanidins, were extracted during the early

stage of maceration, whereas those in seeds, viz., gallic

acid, avan-3-ol monomers, and epicatechin-gallate

units within proanthocyanidins, were gradually extract-

ed. In addition to their localization, the molecular size

and composition of the proanthocyanidins possibly

inuenced the kinetics of their extraction. Cold soak

reduced the extraction of phenolics from the seeds.

Heating at the end of maceration decreased the

concentration of proanthocyanidins. Thus, modication

of the temperature condition during maceration aected

the progress of the concentration of phenolics, resulting

in an alteration of their make-up in the nished wine.

Key words: grape; red wine; phenolics; proanthocyani-

dins; extraction

Phenolic compounds contribute greatly to the sensorycharacteristics of wine, and in particular, of red wine.Anthocyanins and their derivatives are predominantpigments in red wine. Flavonols and hydroxycinnamatesare bitter, and proanthocyanidins, which are polymers ofavan-3-ol units and are also called condensed tannin,contribute bitterness and astringency.13) In addition,these compounds have recently received much attentionbecause of their potential contribution to human healthdue to their antioxidant, antimicrobial, antiviral, andanticarcinogenic characteristics.4)

Phenolic compounds are present mainly in the skinsand seeds in red grape berries.3,5) Anthocyanins andavonols are found in the skins. Hydroxycinnamates3,5)

are found in the esh and skins. On the other hand,

avan-3-ol monomers68) and the majority of gallicacids3,5) in red wine are likely to originate in the seeds.Gallic acid is thought to be released by hydrolysis andheat breakdown from certain esters or more complexmolecules, such as the epicatechin-gallate unit inproanthocyanidins, which are contained mainly in theseeds. Proanthocyanidins are extracted from both theskins and the seeds.Proanthocyanidins from the skins and seeds have been

found to be organoleptically dierent. The seed proan-thocyanins were found to be coarser than the skinproanthocyanidins.1) Proanthocyanidins are multi-com-ponent polymers, and their structural features have notbeen thoroughly studied. However, an analytical methodusing acid depolymerization with nucleophilic agentshas been developed, and it enabled us to determine thesubunit composition and mean degree of polymerization(mDP) of proanthocyanidins and to evaluate theirinuence on the extraction of proanthocyanidins duringmaceration.9,10) Analyses using this method revealedthat skin proanthocyanidins dier from seed proantho-cyanidins by the presence of prodelphinidins, theirhigher mDP, and their lower proportion of galloylatedsubunits.1,7,9,11) These dierences are expected to be thereason for their organoleptic dierences.1) Peyrot desGachons and Kennedy11) showed that extraction of skinand seed proanthocyanidins can be traced separatelyduring fermentation due to their apparent dierences.In addition to proanthocyanidins, analytical methods

for other phenolic compounds using HPLC and HPLC-MS have been improved, and the concentrations of allclasses of phenolics extracted from grape skins andseeds can be quantied.Winemaking practices inuence the extraction of

phenolic compounds. Changing the temperature duringprocessing is an eective method that inuencesextraction because temperature aects the permeabilityof the cells and membranes in grape berries. Thefollowing methods are examples in which the temper-

y To whom correspondence should be addressed. Tel: +81-82-420-0826; Fax: +81-82-420-0809; E-mail: [email protected]: mDP, mean degree of polymerization; %P, percentage of epigallocatechin units within proanthocyanidins; %G, galloylation rate

within proanthocyanidins; LMWP, low-molecular-weight proanthocyanidins; HMWP, high-molecular-weight proanthocyanidins

Biosci. Biotechnol. Biochem., 71 (4), 958965, 2007

ature during maceration is changed: fermentation underhigh temperature, cold soak, must or grape freezing,heating the must at the end of maceration, and thermo-vinication.1219)

Heating at the end of maceration and cold soak areusually applied at more moderate temperatures thanthose used for thermovinication or must freezing.Gerbaux et al.13,17) showed that higher anthocyanin con-tent, color intensity, total polyphenols, and total sensoryscore were obtained in Pinot noir wine by heating at theend of maceration than by control vinication. Heatingat about 40 C for 1 d after fermentation was preferable,as this temperature had a siginicant eect on phenolicextraction without producing the large quantity ofvolatile acids or oxidized o-avor that are oftenobserved under high temperature, for example, duringa thermovinication process at 5080 C. Gerbaux didnot observe any improvement in phenolic extraction asa result of cold soak,13) a vinication method in whichwine musts are kept at low temperature for several daysbefore alcohol fermentation. However, in another study,cold soak at 10 C for 3 d enhanced the extraction ofboth anthocyanins and skin-derived proanthocyani-dins.20) In a previous study with Cabernet Sauvignon,cold soak extracted more color than was obtained in thecontrol, but there was no signicant dierence in thetotal polyphenols between the nished wine of thecontrol and that obtained from the cold soak.15) Thesediscrepancies might be due to the dierences in theconditions of the cold soak process. Comprehensiveanalysis of the kinetics of extraction of all the winephenolics during maceration should provide importantinformation for understanding the observed eects oftemperature on the extraction of phenolics from berryskins and seeds.The objective of this study was to examine the

kinetics of the extraction of all classes of phenolics fromseeds and skins during red wine maceration, which is notthoroughly understood, and to then evaluate the eectsof cold soak and heating at the end of maceration on theextraction of phenolics in red wine.

Materials and Methods

Materials. All chromatographic solvents were HPLCgrade. ()-Epicatechin, ()-epicatechin-3-O-gallate,and ()-epigallocatechin were purchased from KuritaWater Industries (Tokyo). (+)-Catechin and gallic acidwere obtained from Sigma (St. Louis, MO). Malvidin-3-glucoside, quercetin, and caeic acid were purchasedfrom Extrasynthese (Genay, France).

Winemaking. Grapes of Vitis vinifera cv. CabernetSauvignon (vintage 2004, 18:53 0:30% (w/v) ofsugars, 0:49 0:03% (w/v) of titratable acidity astartaric acid, and pH 3:46 0:03) harvested in Yama-nashi Prefecture, Japan, were used for four batches ofvinication. A 50-kg lot of grapes was destemmed and

crushed into a stainless steel tank. Sucrose was added to23% (w/v) of the nal sugar concentration in the juice.Six hours after musts were sulted to 75mg/l, wineyeast (L2323, Lallemand, Rexdale, Canada) was inocu-lated. The caps were punched down twice a day. Thetemperature was controlled by chillers and heaters underfour dierent treatments during the 10-d macerationperiod (Fig. 1). The glucose and fructose concentrationsin the musts were checked with a D-glucose/D-fructosekit (Boehringer Mannheim, Darmstadt, Germany) toensure that fermentation had ceased. For heating at theend of maceration, the must temperature was raised to42 C after fermentation (on the 9th day) and graduallydecreased to 20 C for 1 d. After maceration, the winewas pressed in a hydraulic press.Samples from each must were taken every 2 d during

the maceration period. At the 10th day of maceration,samples were taken from the wines immediately afterthey were pressed. The samples were centrifuged at15;000 g for 15min, and the supernatants were thenltered in a glass lter (GF/D, Whatman International,Maidstone, UK) and stored at 30 C until use.

Extraction of skin and seed phenolics. Skin and seedphenolics were extracted from berries of CabernetSauvignon with a modication of the method of Downeyet al.21) After peeling and deseeding, the berry skins andseeds of the grapes were immediately frozen in liquidnitrogen and stored at 80 C. Frozen samples wereground to a ne powder using a Multi-Beads Shocker(Yasui Kikai, Osaka, Japan), and extracted with 2:1acetone/water for 24 h. The extracts were used forphloroglucinolysis after fractionation.

Spectrophotometric determinations. The color density(CD) and tint were determined by the method of Somersand Evans22) using a UVvisible spectrophotometer witha 2-mm path-length cuvette. The values were convertedto those obtained with a 10-mm light-path cuvette. Theabsorbance of the must at 520 nm was divided into threefractions, viz., polymeric pigment, copigmentation, andfree anthocyanins, by the method of Boulton et al.23)

Reversed phase HPLC analysis of monomeric phe-nolics. Wine and must samples were ltered through0.45 mm PTFE syringe-tip lters (CR13mm, Pall Gel-man Laboratory, Ann Arbor, MI) before analysis ofreversed phase HPLC by a modication of the method ofRitchey and Waterhouse.24) The equipment used forHPLC analysis was an Agilent (Wilmington, DE) 1100series with a four-solvent system and a diode-arraydetector coupled to Chemstation. Zorbax SB C18 (2:1150mm, 5 mm particle size) kept at 30 C was used witha ow rate of 0.3ml/min. The solvents used for theseparation were A 20mM ammonium formate adjust-ed to pH 2.6 with formic acid; B 20% A with 80%acetonitrile; and C 2:3M formic acid (pH 1.5). Amodied solvent gradient condition is shown in Table 1.

Extraction of Phenolics from Berry Skins and Seeds of Grape 959

Compounds were identied on the basis of their UVspectra, and the m=z value obtained with HPLC/ESI-MS(LCQ Advantage, Thermo Electron, San Jose, CA)analysis operated in negative mode. For quantication,standard curves were prepared using caeic acid forcinnamates at 315 nm, catechin for avan-3-ol mono-mers at 280 nm, quercetin for avonols at 365 nm, andmalvidin-3-glucoside for anthocyanins at 520 nm. Forgallates, the concentration of gallic acid, which was apredominant gallate, was determined at 280 nm.

Fractionation of phenolic compounds by C18 sep-pack cartridges. For proanthocyanidin analysis, wineand must samples and extracts of grape skins and seedswere fractionated by the method of Sun et al.25)

Dealcoholized (deacetonized) samples adjusted topH 7.0 were applied to preconditioned Sep-Pack car-tridges (Waters, Milford, MA) connected in series (0.9 gtC18, 0.84 g C18), followed by a washing with water.After the samples were dried with an N2 stream, avanolmonomers were eluted with diethyl ether and oligomersand polymers were recovered by the elution of meth-anol. Fractions of proanthocyanidin oligomers andpolymers were used in the analyses described below.

Acid catalysis in the presence of excess phlorogluci-nol (phloroglucinolysis). Phloroglucinolysis was per-formed under the condition of Kennedy and Jones.9) Thesamples were combined with 1.5 volumes of 0.25N HClin methanol containing 125 g/l phloroglucinol and 25 g/l ascorbic acid. After the reaction was performed at50 C for 20min, 5 volumes of 40mM aqueous sodiumacetate were added to stop it.The concentrations of avan-3-ols (terminal units)

and phloroglucinol adducts (extension units) generatedby depolymerization of proanthocyanidins were deter-mined by HPLC analysis under a modication of theconditions of Kennedy and Jones.9) Seven major peakswere identied by HPLC/ESI-MS analysis. Thesecompounds were quantied using calibration curvesmade from avan-3-ol standards ((+)-catechin, ()-epicatechin, ()-epicatechin gallate, and ()-epigallo-

catechin). The mDP, galloylation rate (%G), andpercentage of epigallocatechin units (%P) were calcu-lated as the molar ratio of extention units to terminalunits, the molar ratio of galloylated units to total units,and the molar ratio of epigallocatechin units to totalunits respectively. The average molecular mass (aMM)was estimated based on the proportional compositionand mDP of the proanthocyanidins. The sums of theconcentrations of these degradation products, fromwhich the weights of the phloroglucinol moiety weresubtracted, were converted to proanthocyanidin concen-trations in the musts.

Normal-phase HPLC. Molecular mass distribution ofproanthocyanidins was analyzed by normal-phase HPLCby a modication of the method of Kennedy andWaterhouse.26) The chromatographic system was cali-brated using cacao proanthocyanidins extracted fromcacao beans kindly provided by Dr. M. Natsume (MeijiSeika Kaisha Ltd., Saitama, Japan). Proanthocyanidinsdetected at 280 nm were divided into two groups, low-molecular-weight proanthocyanidins (LMWP) and high-molecular-weight proanthocyanidins (HMWP), on thebasis of the retention time of the dimer to the tetramer,and over that of the tetramer from the cacao beanstandard, respectively. Area values were converted toequivalents of epicatechin.

Statistical analysis. A t-test was used to evaluatesignicant dierences in the mean concentration orpercentage of phenolics in the musts and the wines aftermaceration with and without cold soak. To examine theinuence of heating at the end of maceration, the dif-ferences in the mean rate of change from the 8th d to the10th d with heat treatment and without heat treatmentwere evaluated by the same method.

Table 1. Solvent Gradient Conditions for HPLC Analysis of Mono-

meric Phenolics

Time (min) Solvent A (%) Solvent B (%) Solvent C (%)

0 100 0 0

5 100 0 0

15 96 4 0

25 92 8 0

32 0 10 90

45 0 14 86

55 0 30 70

60 0 40 60

65 0 80 20

70 0 100 0

75 100 0 0

0.980

1.000

1.020

1.040

1.060

1.080

1.100

0 2 4 61 3 5 7 8 9 10Time (d)

Spec

ific g

ravit

y

05

1015

2025

3035

4045

Tem

pera

ture

(C)

Fig. 1. Fermentation Proles of Wines under Four Temperature

Conditions.

The time course of specic gravity (solid lines with closed

symbols) and temperature (dotted lines with open symbols) in the

musts. , control; , cold soak; , heating at the end of

maceration (heat treatment); , cold soak and heating at the end

of maceration (cold soak and heat treatment).

960 K. KOYAMA et al.

Results

Inuence of temperature on fermentation and winecompositionThe start of fermentation, measured by the decrease in

specic gravity, was retarded by the cold soak, since thetemperature was maintained at about 13 C for the rst2 d (Fig. 1). Although the fermentation progress wasdierent, on the 8th day, the specic gravity ceaseddecreasing under all conditions. The end of fermentationwas also conrmed by the fact that no glucose orfructose was detected in the must. The alcohol concen-trations of the wines after maceration under fourtemperatures were almost the same. Heating at the endof maceration slightly increased the wine pH (Table 2).

Changes in monomeric phenolic concentrations inmustsThe musts and wines contained six classes of phenolic

compounds, viz., cinnamates, gallates, avonols, avan-3-ol monomers, anthocyanins, and proanthocyanidins.These phenolics, except for proanthocyanidins, weremonomeric, and direct HPLC separation was used forquantication. Changes in their concentrations in themusts under dierent temperatures are shown in Fig. 2.The concentration of anthocyanins increased rapidly

to a maximum during the early stage of maceration anddecreased gradually thereafter throughout the remainderof fermentation (Fig. 2A), which is consistent with otherstudies.14,27) The cold soak limited the initial rise in theanthocyanin concentrations. At the 4th d, the anthocya-nin concentrations in the cold-soaked musts weresignicantly lower than those in the control. However,the maximum was reached on the 6th d, and cold soakdid not reduce the concentration at maximum andthereafter. Contrary to expectations, heating the must atthe end of maceration did not increase the anthocyaninconcentration.Flavonols showed an extraction curve similar to

anthocyanins (Fig. 2B). Cold soak and heat treatmentalso aected their concentrations in a way similar to thatof anthocyanins. The maximum was reached on the 8thd in the cold-soaked musts, which was later than in thecase of anthocyanins.The concentrations of cinnamates reached a max-

imum on the 4th d under all conditions, followed by a

reduction to a level 32 to 44% of the maximum aftermaceration (Fig. 2C). This high decrease might berelated with the high susceptibility of compounds in thisclass to enzymic oxidation in the must.28)

The extraction patterns in avan-3-ol monomers andgallic acid showed clear dierences from those of theothers (Fig. 2D, E). Flavan-3-ol monomers were ex-tracted mostly during the late stage of maceration, andthe concentrations continued to increase during macer-ation (Fig. 2D). Cold soak retarded their extractionsignicantly.Similarly, the concentration of gallic acid also

increased continuously, although the extraction curveswere approximately linear (Fig. 2E). Cold soak limitedits rise, and the concentration in the wine aftermaceration was signicantly lower than that in thecontrol without cold soak.In all cases, heat treatment did not increase their

concentrations, and rather decreased their levels. Gallicacid was the only compound which did not decreaseunder heat treatment.

Extraction of proanthocyanidins%P, %G, and mDP of the proanthocyanidins isolated

from the berries used for winemaking were 34.5, 1.6,and 18.1 for the skins and 0, 16.0, and 7.8 for the seedsrespectively. These dierences in the subunit composi-tion and mDP between skins and seeds are consistentwith reports by other groups.1,7,9,11)

The total proanthocyanidin concentrations in the mustduring maceration determined by phloroglusinolysis(Fig. 3A) were similar to those determined by vanillinassay (data not shown). Concentrations increased untilthe 8th d and then decreased. Cold soak signicantlyretarded the increases. Heating at the end of macerationsignicantly decreased the concentrations from those ofthe controls without heat treatment.%P of proanthocyanidins rapidly increased during the

early stage of maceration to reach a maximum percent-age of 18.6% on the 4th d and decreased thereafterduring the remainder of maceration (Fig. 3C). On theother hand, %G progressively increased (Fig. 3D). Thus,the skin proanthocyanidins were extracted more rapidlythan the seed proanthocyanidins, which is consistentwith a report from Peyrot des Gachons and Kennedy.11)

These extraction proles of skin and seed proanthocya-

Table 2. Chemical Composition of the Wines under Four Dierent Temperature Conditions after Pressing

Fraction (%) of color due toTotal

Alcohol Extract Polymeric Copigme Free phenolb Gelatin

(% v/v) (% w/v) pH CDac Tintd A520 nma pigment n- tation anthocyanins A280 nm

a (mg/l) index

Control 12.2 2.95 3.65 5.745 0.55 3.950 18.76 21.01 60.23 32.6 1434 31.6

Cold soak 12.3 3.01 3.66 6.770 0.54 4.815 16.99 23.99 59.02 33.3 1394 34.2

Heat treatment 12.1 3.11 3.75 4.895 0.58 3.520 18.89 19.89 61.22 29.7 1234 28.8

Cold soak and

heat treatment12.6 3.19 3.73 5.270 0.57 3.800 17.24 21.05 61.71 28.0 1154 27.3

aAbsorbance unit. bMeasured by the Folin-Ciocalteu analytical method. cColor density, A420 nm + A520 nm.dTint, A420 nm/A520 nm.


nidins are similar to those of skin and seeds monomericphenolics in the musts. Cold soak limited the initial risein %P during the early period of maceration. At day 4,%P in the cold-soaked must was signicantly lower thanthat in the control. The date of maximum %P was laterthan that in the control. However, no clear dierence in%P was observed afterwards. Cold soak also retardedthe rise in %G signicantly during maceration. Heattreatment decreased %G beyond that of the controlwithout heat treatment.mDP in the must proanthocyanidins increased initially

toward stabilization. In the late stage of maceration, itdecreased gradually (Fig. 3B). Cold soak retarded theinitial increase in reaching the maximum, but no cleardierence was observed on the 10th d. On the otherhand, heat treatment signicantly made the decreasesharper than that of the control without heat treatment.The molecular mass distribution of proanthocyanidins

in wine was analyzed by normal-phase HPLC. As shownin Fig. 4, heat treatment decreased the concentration ofthe high polymeric fraction. Table 3 shows quantita-tively that heat treatment decreased the HMWP con-centration signicantly (Table 3), which was consistentwith the decrease in mDP (Fig. 3B) obtained fromphloroglucinolysis.

Wine color and phenol compound indices in wineThe changes in wine color (absorbance at 520 nm)

during maceration (Fig. 5) were similar to those in theconcentrations of anthocyanins (Fig. 2A).Wine color and phenolic indices are shown in

Table 2. The rate of polymeric pigment was signi-cantly lower with cold soak, but was not aected by heattreatment (Table 2).

Discussion

Extraction of skin and seed phenolicsThis study indicates that phenolics contained in berry

skins, viz., anthocyanin, avonol, and skin proanthocy-anidins, were rapidly extracted during the early stageof maceration, whereas phenolics contained mainly inseeds, viz., gallic acid, avan-3-ol monomers, and seedproanthocyanidins, were gradually extracted and notstabilized within the maceration period (Fig. 2, Fig. 3C,D). These results conrm other reports on anthocy-anin14,27) and proanthocyanidin11) extraction. In addition,these results indicate that the extraction of the variousclasses of phenolics in berry skins had similar extractionproles, and so did that in seeds. Similarly, Canals etal.29) have reported that the total phenol (absorbance at280 nm) from skins was rapidly extracted within 45 d,whereas that from seeds was extracted progressively.Thus the dierence in the tissue structure between berryskins and seeds probably causes the dierence inphenolic extraction between them.In addition to localization, the dierence in the

chemical structures of proanthocyanidins contained even

*

050

100150200250300350

0 1 2 3 4 5 6 7 8 9 10Time (d)

Anth

ocya

nin

(mg/l

)

A*

*

0

5

10

15

20

25

0 2 3 41 5 7 986 10Time (d)

Flav

onol

(mg/l

)

B*

*

05

101520253035

0 2 3 4 5 6 71 8 9 10Time (d)

Cinn

amic

acid

(mg/l

)

C

*

*

*

*

0

10

20

30

40

50

0 2 4 5 71 3 6 8 9 10Time (d)

Flav

an-3

-ols

(mg/l

)

D

*

*

*

*

*

*

*

*

0

2

4

6

8

10

0 2 4 5 71 3 6 8 9 10Time (d)

Gal

lic a

cid (m

g/l)

E

*

*

*

*

*

Fig. 2. Changes in the Concentrations of Monomeric Phenolics in the

Musts under Four Temperature Conditions during Maceration.

A, Anthocyanin. B, Flavonol. C, Cinnamic acid. D, Flavan-3-ols.

E, Gallic acid. Symbols indicate four temperatures during macer-

ation. , control; , cold soak; , heat treatment; , cold soak and

heat treatment. Musts with heat treatment are shown by dotted lines,

in contrast to those without heat treatment, which are shown by solid

lines. Indicates that the average concentration with cold soak andthat without cold soak were signicantly dierent at p < 0:05.


in the same tissue (skin or seed) might alter theirextractability. For example, %P, which represents theextraction of skin proanthocyanidins, initially increasedto reach a maximum on the 4th d (Fig. 3C), whereasmDP increased during the early stage of macerationuntil the 6th d (Fig. 3B), although mDP in berry skins

was higher than in seeds. These results indicate thatproanthocyanidins in skins with low mDP were extract-ed more rapidly than those in skins with high mDP.Similarly, the increase in %P during the early stage ofmaceration indicates that proanthocyanidins with low%P were extracted more rapidly than those with high%P among skin proanthocyanidins. Thus, chemicalproperties, such as hydrophobicity and the number ofhydroxyl residues in proanthocyanidins, likely inu-enced the kinetics of extraction.In the course of their diusion into the must,

proanthocyanidins are thought to be repeatedly trapped

*

*

*

050

100150200250300350400

0 3 6 81 42 5 7 9 10Time (d)

Proa

ntho

cyan

idin

s (m

g/l)A

*

*

*

*

*

**

*

0123456789

0 1 2 3 4 5 6 7 8 9 10Time (d)

mD

P

B*

*

*

0

5

10

15

20

0 61 72 83 4 5 9 10Time (d)

%P

C *

*

*

0

1

2

3

4

5

6

0 41 52 63 7 8 9 10Time (d)

%G

D

*

*

Fig. 3. Changes in the Composition and Mean Degree of Polymerization of Proanthocyanidins in the Musts under Four Temperature Conditions

during Maceration.

A, Concentration of total proanthocyanidins. B, Mean degree of polymerization (mDP). C, Percentage of prodelphinidin unit (%P). D,

Percentage of galloylation (%G). Symbols and lines are shown in Fig. 2. Indicates that the average concentration with cold soak and thatwithout cold soak were signicantly dierent at p < 0:05. yIndicates that the average rate of the change from the 8th d to the 10th d under heattreatment and that without heat treatment were signicantly dierent at p < 0:05.

0

10

20

30

40

50

60

70

80

90

100

0 10 30 5020 40 60Time (min)

Abso

rban

ce (m

AU)

Fig. 4. Normal-Phase HPLC Chromatograms of Proanthocyanidins

in Wine Heated at the End of Maceration (Solid Line) and the

Control (Broken Line).

The retention times for the cacao proanthocyanidin standard are

shown as the string-of-dots line in the order of the degree of

polymerization from monomer to decamer (from left to right).

Table 3. Proanthocyanidins in the Wine after Pressing

LMWPad HMWPbdHMWP/

LMWPdaMMc

Control 272.8 1078.1 3.9 1749.2

Cold soak 258.2 957.0 3.7 1708.6

Heat treatment 248.6 787.3 3.2 1443.4

Cold soak and heat

treatment211.6 693.1 3.3 1478.1

aLow-molecular-weight polymers expressed by mg epicatechin equivalents

(ECE)/L.bHigh-molecular-weight polymers expressed by mg ECE/L.cAverage molecular mass calculated from the data obtained by phloroglu-

cinolysis.dFractionated by normal-phase HPLC.Indicates that the mean concentrations in paired samples of the wines underheat treatment and their controls without heat treatment are signicantly

dierent at p < 0:05.


in and released from internal components of the cells,such as cell wall polysaccharides, soluble pectins,glycoprotein, etc., after the partial collapse of the cellmembrane structure. The hydrogen bond and hydro-phobic interactions were hypothesized to be dominant inthe interactions between proanthocyanidins and puriedcell walls or proteins as ning agents.30,31) Similarly,proanthocyanidins with higher hydrophobicity and ahigher number of hydroxyl groups, indicative of highermDP, %P, and %G, might be selectively trapped andreleased into the must later with the aid of increasedethanol concentrations. Alternatively, the dierent cel-lular localization of proanthocyanidins with dierentcompositions might explain this selectivity.8,32,33)

The decrease in the concentrations of skin phenolicsafter reaching a maximum (Fig. 2A, B, C) might be dueto the adsorption of extracted polyphenols by yeast leesand/or grape solids, degradation, oxidation, or conden-sation. In the case of anthocyanins, the condensationreaction with proanthocyanidins has been assumed tocontribute more or less to their decrease during the latestage of fermentation.27)

Inuence of cold soak on the concentration ofphenolicsCold soak retarded the initial rise in the concentra-

tions of all classes of phenolics except for cinnamates.During maceration, cold soak reduced the extraction ofavan-3-ol monomers and gallic acid, which mainlycome from the seeds, and of proanthocyanidins (Fig. 2,Fig. 3A). The retardation in the extraction was probablydue to their lower ethanol concentration during the earlystage of maceration. Canals et al.29) found a signicantinuence of the ethanol concentration on the extract-ability of anthocyanins and, in particular, of proantho-cyanidins from skins and seeds.In the case of skin phenolics, although cold soak

retarded their extraction, it did not reduce the maximumconcentrations. During the late stage of maceration, theskin monomeric phenolic concentrations in the cold-soaked musts were either similar to or rather higher thanthose in the control. Similarly, cold soak increased theextraction of proanthocyanidins from the skins anddecreased that from the seeds in other study.20) In

general, an increase in the skin phenolics is consideredto have some advantages for the organoleptic propertiesof the wine. For example, skin proanthocyanidin isconsidered to be softer and to have higher quality thanseed proanthocyanidins.2)

Cold soak increased the ratio of anthocyanin toproanthocyanidins (a 67% increase) in the wine aftermaceration, suggesting a possible increase in theproportion of the anthocyanin-proanthocyanidin adductsagainst total polymers, which must aect the quality ofthe resultant wine after long-term storage.2,20,34)

Inuence of heating at the end of maceration on theconcentration of phenolicsHeat treatment is generally assumed to damage grape

cell membranes, which results in an increased extactionof phenolic compounds.13,14,17) However, in this study,heating at the end of maceration did not increasephenolic extraction and, moreover, decreased the con-centration of proanthocyanidins and mDP (Fig. 2,Fig. 3A, B). This decrease might be due to oxidation,degradation, and/or adsorption of proanthocyanidins, inparticular, those with higher molecular weights.30,35)

Further studies are necessary to determine the reasonsfor the dierent results obtained in our study andothers.1214,16,17) Potential variables, e.g., grape variety,berry maturity, heat conditions, and/or fermentationscale, might have aected the results.

References

1) Vidal, S., Francis, L., Guyot, S., Marnet, N.,Kwiatkowski, M., Gawel, R., Cheynier, V., and Waters,E. J., The mouth-feel properties of grape and appleproanthocyanidins in a wine-like medium. J. Sci. FoodAgric., 83, 564573 (2003).

2) Cheynier, V., Flucrand, H., Brossaud, F., Asselin, C.,and Moutounet, M., Phenolic composition as related tored wine avor. In Chemistry of Wine Flavor, eds.Waterhouse, A., and Ebeler, S., American ChemicalSociety, Washington, DC, pp. 124141 (1999).

3) Cheynier, V., Polyphenols in foods are more complexthan often thought. Am. J. Clin. Nutr., 81, 223S229S(2005).

4) Jackson, R., Wine, health, and food. In Wine Science,ed. Taylor, S., Academic Press, San Diego, pp. 591607(2000).

5) Boulton, R., Singleton, V., Bisson, L., and Kunkee, R.,Viticulture for winemakers. In Principles and Practicesof Winemaking, Chapman and Hall, New York, pp. 4146 (1996).

6) Kennedy, J., Matthews, M., and Waterhouse, A., Eectof maturity and vine water status on grape skin and wineavonoids. Am. J. Enol. Vitic., 53, 268274 (2002).

7) Cortell, J. M., Halbleib, M., Gallagher, A. V., Righetti,T. L., and Kennedy, J. A., Inuence of vine vigor ongrape (Vitis vinifera L. cv. Pinot noir) and wineproanthocyanidins. J. Agric. Food Chem., 53, 57985808 (2005).

8) Sun, B. S., Pinto, T., Leandro, M. C., Ricardo-da-Silva,

*

0.01.02.03.04.05.06.07.0

0 2 4 5 6 7 8 91 3 10Time (d)

Abso

rban

ce (5

20 nm

)*

Fig. 5. Changes in Absorbance at 520 nm during Maceration.

The symbols and lines are as in Fig. 2.


J. M., and Spranger, M. I., Transfer of catechins andproanthocyanidins from solid parts of the grape clusterinto wine. Am. J. Enol. Vitic., 50, 179184 (1999).

9) Kennedy, J. A., and Jones, J. P., Analysis of proantho-cyanidin cleavage products following acid-catalysis inthe presence of excess phloroglucinol. J. Agric. FoodChem., 49, 17401746 (2001).

10) Matthews, S., Mila, I., Scalbert, A., Pollet, B., Lapierre,C., Herve du Penhoat, C. L. M., Rolando, C., andDonnelly, D. M. X., Method for estimation of proan-thocyanidins based on their acid depolymerization in thepresence of nucleophiles. J. Agric. Food Chem., 45,11951201 (1997).

11) Peyrot des Gachons, C., and Kennedy, J. A., Directmethod for determining seed and skin proanthocyanidinextraction into red wine. J. Agric. Food Chem., 51,58775881 (2003).

12) Anli, R. E., Fidan, I., and Fidan, Y., Inuence desdierents procedes de maceration sur la compositionphenolique et la qualite organoleptique de KalecikKarasi. In Oenologie 99, 6e Symposium Internationald Oenologie, ed. Lonvaud-Funel, A., Technique etDocumentation, Paris, pp. 451455 (2000).

13) Gerbaux, V., Etude de quelques conditions de cuvaisonsusceptibles daugmenter la composition polyphenoliquedes vins de pinot noir. Rev. Oenol., 69, 1518 (1993).

14) Sacchi, K. L., Bisson, L. F., and Adams, D. O., A reviewof the eect of winemaking techniques on phenolicextraction in red wines. Am. J. Enol. Vitic., 56, 197211(2005).

15) Okubo, K., Goto-Yamamoto, N., and Okazaki, N., Eectof pre-fermentation cold soak on extraction of antho-cyanin during red wine making. Nippon Jozokyoukaishi(in Japanese), 98, 193200 (2003).

16) Zimman, A., Joslin, W. S., Lyon, M. L., Meier, J., andWaterhouse, A. L., Maceration variables aectingphenolic composition in commercial-scale CabernetSauvignon winemaking trials. Am. J. Enol. Vitic., 53,9398 (2002).

17) Gerbaux, V., Briox, C., and Vincent, B., Optimisationde la maceration nale a` chaud, interet dun enzymage etdune maceration sous chapeau immerge pour la vin-ication du Pinot noir. Rev. Fr. Oenol., 201, 1621(2003).

18) Reynolds, A., Cli, M., Girard, B., and Kopp, T. G.,Inuence of fermentation temperature on compositionand sensory properties of Semillon and Shiraz wines.Am. J. Enol. Vitic., 52, 235240 (2001).

19) Plaza, E. G., Munoz, R. G., Roca, J. M. L., Cutillas, A.M., and Fernandez, J. I. F., Phenolic compounds andcolor stability of red wines: eect of skin macerationtime. Am. J. Enol. Vitic., 52, 266270 (2001).

20) Cheynier, V., Duenas-Paton, M., Salas, E., Maury, C.,Souquet, J. M., Sarni-Manchado, P., and Fulcrand, H.,Structure and properties of wine pigments and tannins.Am. J. Enol. Vitic., 57, 298305 (2006).

21) Downey, M. O., Harvey, J. S., and Robinson, S. P.,Analysis of tannins in seeds and skins of Shiraz grapesthroughout berry development. Aust. J. Grape Wine Res.,9, 1527 (2003).

22) Somers, T. C., and Evans, M. E., Spectral evaluation of

young red wines: anthocyanin equilibria, total phenolics,free and molecular SO2, chemical age. J. Sci. FoodAgric., 28, 279287 (1997).

23) Boulton, R., Neri, R., Levengood, J., and Vaadia, M.,Copigmentation of anthocyanins in Cabernet Sauvignonand Merlot wines from the Napa Valley of California. InOenologie 99, 6e Symposium International dOenolo-gie, ed. Lonvaud-Funel, A., Technique et Documenta-tion, Paris, pp. 3538 (2000).

24) Ritchey, J. G., and Waterhouse, A. L., A standard redwine: monomeric phenolic analysis of commercialCabernet Sauvignon wines. Am. J. Enol. Vitic., 50, 91100 (1999).

25) Sun, B., Leandro, C., Ricardo da Silva, J. M., andSpranger, I., Separation of grape and wine proanthocya-nidins according to their degree of polymerization.J. Agric. Food Chem., 46, 13901396 (1998).

26) Kennedy, J. A., and Waterhouse, A. L., Analysis ofpigmented high-molecular-mass grape phenolics usingion-pair, normal-phase high-performance liquid chroma-tography. J. Chromatogr. A, 866, 2534 (2000).

27) Zimman, A., and Waterhouse, A. L., Incorporation ofmalvidin-3-glucoside into high molecular weight poly-phenols during fermentation and wine aging. Am. J.Enol. Vitic., 55, 139146 (2004).

28) Singleton, V. L., Salgues, M., Zaya, J., and Trousdale,E., Caftaric acid disappearance and conversion toproducts of enzymic oxidation in grape must and wine.Am. J. Enol. Vitic., 36, 5056 (1985).

29) Canals, R., Llaudy, C., Valls, J., Canals, J. M., andZamora, F., Inuence of ethanol concentration on theextraction of color and phenolic compounds from theskin and seeds of Tempranillo grapes at dierent stagesof ripening. J. Agric. Food Chem., 53, 40194025(2005).

30) Renard, C. M. G. C., Baron, A., Guyot, S., and Drilleau,J.-F., Interactions between apple cell walls and nativeapple polyphenols: quantication and some consequen-ces. Int. J. Biol. Mol., 29, 115125 (2001).

31) Manchado, P. S., Deleris, A., Avallone, S., Cheynier, V.,and Moutounet, M., Analysis and characterization ofwine condensed tannins precipitated by proteins used asning agents in enology. Am. J. Enol. Vitic., 50, 8186(1999).

32) Geny, L., Saucier, C., Bracco, S., Daviaud, F., andGlories, Y., Composition and cellular localization oftannins in grape seeds during maturation. J. Agric. FoodChem., 51, 80518054 (2003).

33) Amrani Joutei, K., Glories, Y., and Mercier, M., Local-isation des tanins dans la pellicule de baie de raisin.Vitis, 33, 133138 (1994).

34) Vidal, S., Cartalade, D., Souquet, J. M., Fulcrand, H.,and Cheynier, V., Changes in proanthocyanidin chainlength in winelike model solutions. J. Agric. FoodChem., 50, 22612266 (2002).

35) Mazauric, J., and Salmon, J., Interactions between yeastlees and wine polyphenols during simulation of wineaging. I. Analysis of remnant polyphenolic compoundsin the resulting wines. J. Agric. Food Chem., 53, 56475653 (2005).


koyama 2007

Documents