Remarkable Proanthocyanidin Adsorption ... - ACS Publications

Article pubs.acs.org/JAFC

Remarkable Proanthocyanidin Adsorption Properties of Monastrell Pomace Cell Wall Material Highlight Its Potential Use as an Alternative Fining Agent in Red Wine Production Ana Belén Bautista-Ortín, Yolanda Ruiz-García, Fátima Marín, Noelia Molero, Rafael Apolinar-Valiente, and Encarna Gómez-Plaza* Departamento de Tecnologı ́a de Alimentos, Nutrición y Bromatologı ́a, Facultad de Veterinaria, Universidad de Murcia, Campus de Espinardo, 30071 Murcia, Spain S Supporting Information *

ABSTRACT: The existence of interactions between the polysaccharides of vegetal cell walls and proanthocyanins makes this cell wall material an interesting option for its use as a fining agent to reduce the level of proanthocyanins in wines. Pomace wastes from the winery are widely available and a source of cell wall material, and the identification of varieties whose pomace cell walls present high proanthocyanin binding capacity and of processing methods that could enhance their adsorption properties could be of great interest. This study compared the proanthocyanin adsorption properties of pomace cell wall material from three different grape varieties (Monastrell, Cabernet Sauvignon, and Syrah), and the results were compared with those obtained using fresh grape cell walls. Also, the effect of the vinification method has been studied. Analysis of the proanthocyanidins in the solution after reaction with the cell wall material, using phloroglucinolysis and size exclusion chromatography, provided quantitative and qualitative information on the adsorbed and nonadsorbed compounds. A highlight of this study was the observation that Monastrell pomace cell wall material showed a strong affinity for proanthocyanidins, with values similar to that obtained for fresh grapes cell walls, and a preferential binding of high molecular mass proanthocyanidins, so these pomace cell walls could be used in wines to reduce astringency. The use of maceration enzymes during vinification had little effect on the retention capacity of the pomace cell walls obtained from this vinification, although an increase in the retention of low molecular mass proanthocyanidins was observed, and this might have implications for wine sensory properties. KEYWORDS: proanthocyanidins, tannins, cell wall material, grapes, pomace, phloroglucinolysis, size exclusion chromatography

■

INTRODUCTION The properties and characteristics of grape skin and flesh cell wall material (CWM) are closely related to some enological events and some observations concerning phenolic compounds in grapes and wines. In this way, previous studies have demonstrated that the composition and quantity of skin cell walls may determine the mechanical resistance, texture, and ease with which berries are processed, as well as anthocyanin extractability during fermentative maceration.1,2 Cell walls (CWs) are composed of polysaccharides, phenolic compounds, and proteins and are stabilized by ionic and covalent linkages. Among the polysaccharides, the presence of hemicellulose, cellulose, and pectins is important, and the last two account for 30−40% of the polysaccharide components of the cell wall.3 Cell wall polysaccharides contain hydroxyl groups as well as aromatic and glycosidic oxygen atoms that have the ability to form hydrogen bonds and hydrophobic interactions with some molecules,4,5 including proanthocyanidins.6 The existence of CWM−proanthocyanidin interactions may explain some experimental observations such as the decrease in grape proanthocyanidin (PA) extractability as ripening progresses,7,8 which can be explained as a retention of PAs in the cell wall material,9−11 and the reduced extractability of PAs during fermentative maceration,12 which may be related not only to these PAs being retained in the solid parts but also to a substantial © 2014 American Chemical Society

proportion of them being adsorbed by the CWM in suspension in the must and finally precipitated during settling. Moreover, some authors have stated that there are differences in the affinity of PAs for the different polysaccharides present in CWs, their affinity for covalently linked pectin being higher than that for xyloglucan and cellulose.5 This observation may indicate that differences in the CW and PA composition among varieties may be the cause of the varietal differences in PA extractability observed in other experiments.12 Another enological practice that may be affected by these interactions is the use of enological tannins. The addition of exogenous commercial tannins to increase their levels in wine is a common practice for several reasons: they may help stabilize color, increase body, mask green characters, etc.13 Depending on the final objective, they are used in the fermentative or postfermentative phase of winemaking. Their addition during fermentation may result in a large part of these exogenous tannins being eliminated by their interaction with the skin and pulp cell walls present in the must after crushing of the grapes, as reported by Bautista-Ortı ́n et al.14 Another point of interest of the capacity of CWs to interact with PAs is their possible use as fining agents. Some red wines Received: Revised: Accepted: Published: 620

July 30, 2014 December 14, 2014 December 20, 2014 December 20, 2014 DOI: 10.1021/jf503659y J. Agric. Food Chem. 2015, 63, 620−633

Article

Journal of Agricultural and Food Chemistry

initiated with Levuline Gala (commercial dry Saccharomyces cerevisiae yeast, OenoFrance, Bordeaux, France) inoculated at 10 g/hL. After 10 days of maceration, the wine was pressed at 150 kPa in a 75 L tank membrane press (Hidro 80L, Ausavil, Spain), and the pomace was recovered. The skins from pomace were separated from seeds and other solid parts, and they represented the pomace from the control vinification. Other vinification was carried out with the addition of a commercial enzyme preparation (3 g/100 kg, Enozym Vintage, Agrovin, Spain) just after crushing. The fermentation, pressing conditions, and pomace recovery were as described for the control vinification. Cell Wall Material. Purified CWM was extracted from V. vinifera L. cv. Monastrell, Cabernet Sauvignon, and Syrah fresh skins and from the pomace obtained after 10 days of fermentative maceration. Cell walls were isolated according to the method of De Vries et al.,21 as adapted by Apolinar-Valiente et al.22 Briefly, the grape skin tissue or the pomace skin tissue (10 g) was suspended in 15 mL of boiling water for 5 min and then homogenized. One part of the homogenized material was mixed with two parts of 96% ethanol and extracted for 30 min at 40 °C. The raw alcohol-insoluble solids were separated by filtration on paper and extracted again with fresh 70% ethanol for 30 min at 40 °C. The washing treatment with fresh 70% ethanol was repeated several times until no sugars remained in the 70% ethanol phase. Then, the alcohol-insoluble solids (AIS) were washed twice with 96% ethanol and once with acetone and finally dried overnight under an air stream at 20 °C. The composition of the cell walls can be found in the Supporting Information (Figures S1−S4). Proanthocyanidins Used in the Interaction Studies. Two purified PA fractions from ripened grape skin (sPA) and seed (sdPA) were selected for the experiment as well as two commercial samples, a skin-derived (T1) and a seed-derived (T2) enological tannin. The purity of the commercial tannins was estimated spectrophotometrically following the method of Riberau-Gayon et al.23 after acid hydrolysis of the samples. For the purified fractions, V. vinifera L. cv. Monastrell grape berries were used as the source material for grape skin and seed proanthocyanidins. A random sample (1 kg) was collected at technological maturity. Grape berries were kept at −20 °C until processed. For the proanthocyanidin isolation and purification, the skins and seeds were separated from the berry mesocarp and rinsed with distilled−deionized water, and the intact tissues were extracted in covered Erlenmeyer flasks with 2:1 acetone/water at room temperature for 24 h. To minimize proanthocyanidin oxidation, solutions were sparged with nitrogen and the extraction was carried out in the dark. Following extraction, the extract was concentrated under reduced pressure at 35 °C to remove acetone and then lyophilized to a dry powder. The crude proanthocyanidins were purified using Toyopearl TSK HW 40-F size exclusion media (Sigma-Aldrich). The column (400 × 26 mm) was equilibrated with 1:1 MeOH/water containing 0.1% (v/v) trifluoroacetic acid. The proanthocyanidin powder was dissolved in a minimum amount of this mobile phase and then applied to the column. The column was then rinsed with 5 column volumes of the mobile phase to remove carbohydrate and low-molecular-weight flavan-3-ol monomer material. The proanthocyanidins were then eluted with 3 column volumes of 2:1 acetone/water containing 0.1% (v/v) trifluoroacetic acid. The eluent was concentrated under reduced pressure at 35 °C to remove acetone and then lyophilized to a dry powder. For fractionation, the method proposed by Kennedy and Taylor24 was used. Grape skin proanthocyanidin samples (2 g each) were dissolved in 50 mL of MeOH/water/formic acid (49.8:49.8:2.4) and quantitatively applied to a Sephadex LH-20 column (400 × 26 mm, Sigma-Aldrich) equilibrated with MeOH/water/formic acid (49.8:49.8:2.4). The column was then washed sequentially using different solvent systems and volumes. Eluted proanthocyanidin fractions were collected into test tubes. The fractions were rotary evaporated, lyophilized to dryness, and stored at −80 °C. One purified fraction from seed and another from skin were chosen for the experiments, on the basis of their quantitative availability and a mean degree of polymerization closer to that of red Monastrell wines. Binding Reactions between Tannins and Cell Wall Material. Skin or pomace CWM was weighed into 3 mL tubes. CWM samples

present high levels of astringency and bitterness, due to an excessive extraction of tannins from grapes or an excessive addition of commercial enological tannins,15 and may need a fining agent to reduce the concentrations of tannins in wines. Traditionally, proteins have been used for fining purposes, due to their high affinity for PAs. Most of the proteins commonly used are of animal origin, such casein, albumin, and gelatin.16 Lately, those of vegetal origin (Arabic gum, pea protein) are becoming more popular, mainly due to the allergenic properties of some of the animal-derived proteins and the legal obligation to indicate their presence on the label (EU Directive 2007/68/EC November 27, 2007), a fact that may reduce their attraction for consumers. The ability of CWM to bind PAs makes them an interesting object of study for their possible use as new fining agents. Such studies were initiated by Bindon and Smith17 and Guerrero et al.,18 who studied the fining action of insoluble fibers prepared from fresh apple and different grape sources and compared them with commercial proteins, reporting that grape fibers may be considered as alternative fining agents for red wines. It is clear that fiber or purified CWM derived from fresh grapes would not be of commercial interest because grapes are a valuable product, but products derived from the pomace obtained after fermentation and devatting might be an interesting way of adding value to this byproduct, as also stated by Bindon and Smith.17 However, compositional differences among different varieties, the possible compositional changes that flesh, but especially skin CWM may suffer during the fermentation process, and how these changes affect their binding capacity need to be considered. In fact, previous studies have shown that the fermentation process may alter the polysaccharide composition of the CWM, as may do the different treatments used during fermentative maceration as it is the use of maceration enzymes.19,20 The use of maceration enzymes decreased the uronic acids and total sugar content of the cell wall due to their action on the pectin matrix20 and that may change the level of CW−PA interactions. In this study, the ability of purified CWM derived from pomace (from three different varieties and from two different vinification methods) to bind different types of PAs was studied, searching for a source of pomace material with a high adsorption capacity and clear selectivity for high molecular mass proanthocyanidins that could be used to reduce wine astringency.

■

MATERIALS AND METHODS

Chemicals. Chromatographic solvents were of high-performance liquid chromatography (HPLC) grade, and chemicals were of analytical reagent grade. Acetonitrile, acetone, methanol, ethanol, formic acid, and trifluoroacetic acid were from Merck (Darmstadt, Germany). The phloroglucinol reagent was sourced from Sigma-Aldrich (St. Louis, MO, USA). Sodium acetate was sourced from J. T. Baker (Deventer, The Netherlands). The standards (+)-catechin, (−)-epicatechin, (−)-epicatechin gallate, and (−)-epigallocatechin were obtained from Extrasynthese (Genay, France). Instrumentation. The HPLC apparatus was a Waters 2695 (Waters, Milford, MA, USA) equipped with a Syrahstem autosampler and a Waters 2996 photodiode array detector. Grapes and Pomace. Vitis vinifera L. cv. Monastrell, Cabernet Sauvignon, and Syrah grapes, cultivated in Murcia (Spain), were sampled at 23.8, 24.8, and 25.6 °Brix, respectively, and with a titratable acidity ranging from 2.9 to 3.3 g/L, expressed as tartaric acid. Grapes from each of the three cultivars (90 kg) were destemmed and crushed. The crushed grapes were macerated in 100 L stainless steel tanks. Potassium bisulfite (8 g/100 kg grapes) was added. Fermentation was carried out with temperature control (25 °C); fermentations were 621

DOI: 10.1021/jf503659y J. Agric. Food Chem. 2015, 63, 620−633


■

were then combined with the four different proanthocyanidins: those contained in two different commercial enological tannins (a seedderived and a skin-derived grape tannin) and the two purified skin- and seed-derived proanthocyanidins. They were dissolved in a model solution (12% ethanol at pH 3.6 adjusted with trifluoroacetic acid) to a concentration of 2 g/L. The reaction volume was 2.5 mL, and the CWM was 13 mg/mL. The different samples were shaken at 300 rpm in an orbital shaker at room temperature for 90 min. Each experiment (each tannin solution with each of the three different CWM, isolated from fresh grape skins and from the pomace of the two different vinifications) was made in duplicate and for each variety. For each reaction, a blank without CWM and a blank without tannin were also included, the latter to monitor any possible desorption of CWM-bound tannin. After the binding reaction, samples were centrifuged at 13000 rpm, and the supernatant was transferred to a new tube. Samples were then dried under vacuum at 35 °C. Recovered tannin was reconstituted in 250 μL of methanol and then analyzed by phloroglucinolysis and size exclusion chromatography (SEC). Analysis of Proanthocyanidins Using the Phloroglucinolysis Reagent. The proanthocyanidin content of the control sample (2 g/L) or those remaining in solution after the interaction with the CWs was analyzed using the phloroglucinol reagent, according to the method described by Kennedy and Jones25 with some modifications described below. A solution of 0.2 M HCl in methanol, containing 100 g/L phloroglucinol and 20 g/L ascorbic acid, was prepared (phloroglucinolysis reagent). The methanolic extract was left to react with the phloroglucinolysis reagent (1:1) in a water bath for 20 min at 50 °C and then combined with 2 volumes of 200 mM aqueous sodium acetate to stop the reaction. HPLC analysis followed the conditions described by Busse-Valverde et al.26 Samples (10 μL injection volume) were injected on an Atlantis dC18 column (250 × 4.6 mm, 5 μm packing) protected with a guard column of the same material (20 mm × 4.6 mm, 5 μm packing) (Waters). The elution conditions were as follows: 0.8 mL/min flow rate; oven temperature, 30 °C; solvent A, water/formic acid (98:2, v/v); and solvent B, acetonitrile/solvent A (80:20 v/v). Elution began with 0% B for 5 min, linear gradient from 0 to 10% B in 30 min, and gradient from 10 to 20% in 30 min, followed by washing and re-equilibration of the column. Proanthocyanidin cleavage products were estimated using their response factors at 280 nm relative to (+)-catechin, which was used as the quantitative standard. These analyses allowed determination of the recovery by mass of the total proanthocyanidin content, the apparent mean degree of polymerization (mDP), and the percentage of each constitutive unit. The mDP was calculated as the sum of all subunits (flavan-3-ol monomer and phloroglucinol adducts, in moles) divided by the sum of all flavan-3-ol monomers (in moles). The percentage of conversion was calculated gravimetrically from the percentage of conversion of tannins to known proanthocyanin subunits. To determine the naturally occurring proanthocyanidin monomers (catechin and epicatechin), the methanolic extract was analyzed without reaction with the phloroglucinolysis reagent. Analysis of Proanthocyanidins by Size Exclusion Chromatography. An adaptation of the method described by Kennedy and Taylor24 was used for SEC. The method used two PLgel (300 × 7.5 mm, 5 μm, 500 (effective molecular mass range of up to 4000 using polystyrene standards) by 100 Å (effective molecular mass range of 500−30000 using polystyrene standards) columns connected in series and protected by a guard column containing the same material (50 × 7.5 mm, 5 μm), all purchased from Polymer Laboratories (Amherst, MA, USA). The amount of sample injected was 40 μg. The isocratic method used a mobile phase consisting of N,N-dimethylformamide containing 1% glacial acetic acid, 5% water, and 0.15 M lithium chloride. The flow rate was maintained at 1 mL/min with a column temperature of 60 °C, and elution was monitored at 280 nm. The SEC peak area results were calibrated to the gravimetric value (A280/mg gravimetric) for each tannin type. Statistical Analysis. All of the statistical analyses were made using Statgraphics Centurion XVI (Statpoint Technologies Inc., Warranton, VA, USA).

Article

RESULTS AND DISCUSSION

Proanthocyanidins Used in the Experiment. Because it has been suggested that the composition, molecular size, conformational mobility, and shape influence the interactions between tannins and CWM,4,5 PA samples with different characteristics were used for this study. We have worked with two commercial samples, a skin-derived (T1) and a seed-derived (T2) commercial tannin, and the results were compared with a purified PA fraction from ripened grape skin (sPA) or seed (sdPA). The phloroglucinolysis assay showed that the lowest conversion was observed in T1 and the highest in seed PA, with a conversion of 72% (Table 1). These results were expected because higher conversions are always found when working with seed tannins compared with skin.7 It is well-known that grape PAs slightly differ in their composition according to their origin in the berry. Seed proanthocyanidins contain greater levels of flavan-3-ols esterified to gallic acid, they generally have a lower degree of polymerization (mDP) than those from skins, and no trihydroxylation of the B ring has been observed.25,27 However, and contrary to expectations, the mDP of the skin-derived commercial enological tannin was very low considering its origin (skin tannin sample), with values very similar to those obtained for seed-derived commercial sample. Higher mDPs were observed for the purified samples, the highest being found in sPA. The maximum galloylation percentage was observed in the seed-derived tannins, especially in the purified samples, and lower values were found in the skin-derived samples, the lowest values in T1. With regard to the subunit composition, commercial samples differed from purified samples mainly in the lower quantities of terminal units and higher percentage of extension epicatechin. Comparison of both purified samples showed the main differences were in the percentages of epicatechin gallate, both terminal and extension subunits (as was expected, given the higher percentage of galloylation) and the absence of epigallocatechin subunits in the seed samples. The composition of T1, labeled as skin tannin, showing a low mDP, relatively high galloylation percentage, and low percentage of epigallocatechin, may indicate that this sample is really a mix of skin and seed tannins. The tannin samples and purified fractions were also analyzed using SEC. Given the relatively low conversion percentage of the phloroglucinolysis study, the use of SEC also allows the fraction that was not depolymerized by the phloroglucinolysis assay to be studied. This technique provides complementary information concerning the adsorption rate and molecular mass distribution of the tannins. The profiles are shown in Figure 1, and differences were observed among the different samples. For the study of the molecular mass distribution, and unlike other studies that reported mass distribution information as a cumulative (% volume elution) molecular mass, in the current study the mass distribution is reported as elution time and the SEC chromatograms were divided in three areas representing the high (those eluting from minute 10 to 11.8, F1 (see Table 2)), medium (from 11.8 to 13.6 min, F2), and low molecular mass PAs (from 13.6 to 15.0 min, F3). The distribution of the eluted area showed that sPA and sdPA had the largest percentages of areas eluting at the first elution time range (67.8 and 66.9%, respectively), indicating a higher percentage of high molecular mass PAs, whereas in T1 and T2 the area in this fraction of elution time was lower 622


Article

Abbreviations: T1, commercial enological tannin from grape skins; T2, commercial enological tannin from grape seeds; sPA, purified grape skin proanthocyanidin; sdPA, purified grape seed proanthocyanidin; mDP, mean degree of polymerization; %G, percentage of galloylation; extC, percentage of extension (+)-catechin; extEC, percentage of extension (−)-epicatechin; extECG, percentage of extension (−)-epicatechin gallate; tC, percentage of terminal (+)-catechin; tEC, percentage of terminal (−)-epicatechin; tECG, percentage of terminal (−)-epicatechin gallate.

Figure 1. Size exclusion chromatogram of the different tannins used in the experiment.

Table 2. Total Area Observed in the SEC Analysis for Each Tannin and Percentage Eluted from 10 to 11.8 min (F1, Corresponding to High Molecular Mass Proanthocyanidins), from 11.8 to 13.6 min (F2, Corresponding to Medium Molecular Mass Proanthocyanidins), and from 13.6 min to 15.00 min (F3, Corresponding to Low Molecular Mass Proanthocyanidins)a tannin

concentration (mg/L)

F1

F2

F3

T1 T2 sPA sdPA

1208b 1594c 2000 2000

37.5 36.2 67.8 66.9

48.2 49.3 28.8 31.5

12.5 13.0 1.7 0.7

a

Abbreviations: T1, commercial enological tannin from grape skins; T2, commercial enological tannin from grape seeds; sPA, purified grape skin proanthocyanidin; sdPA, purified grape seed proanthocyanidin. bThis concentration corresponds to the real tannin content of a 2 g of the commercial sample T1, determined by acid hydrolysis. cThis concentration corresponds to the real tannin content of a 2 g of the commercial sample T2, determined by acid hydrolysis.

(37.5 and 36.2%), which is in agreement with the differences in mDP. At the range of time between 11.8 and 13.6 min, corresponding to medium molecular mass PAs, the values were, this time, higher for T1 and T2 (48.2 and 49.3%) than for sPA (28.8%). Again, a substantial variation was observed at elution time from 13.6 to 15 min (low molecular mass PAs), the relative area being very low for sPA and sdPAs (1.7 and 0.7%) and higher for T1 and T2 (12.5 and 13.0%). Effect of the Addition of the Different Cell Wall Materials to Model Solutions Containing the Different Proanthocyanidin Samples. The proanthocyanidin samples (commercial tannins and purified fractions) were combined with the purified skin CWM (gCW) from Monastrell, Cabernet Sauvignon, and Syrah grapes and their corresponding CWM from the pomace obtained after a control vinification (cmCW) or a vinification where a maceration enzyme was used (emCW). The different CWs were suspended in model solutions to determine the effect of the interaction in terms of PA concentration, composition, and molecular mass distribution. The effect of CWM additions on PAs was determined by observing the changes in the PAs that remained in solution compared with a control (Tables 3− 5 and Figures 2−5).

a

3.12 ± 0.00 4.86 ± 0.03 2.78 ± 0.02 12.27 ± 0.04 6.33 ± 0.01 0.00 16.46 ± 0.02 0.00 47.89 ± 0.18 46.57 ± 0.12 67.53 ± 0.31 62.79 ± 0.04 5.16 ± 0.04 9.66 ± 0.02 1.47 ± 0.01 7.36 ± 0.01 0.74 ± 0.00 2.47 ± 0.00 0.00 4.74 ± 0.03 19.33 ± 0.08 16.18 ± 0.04 5.54 ± 0.33 6.56 ± 0.08 17.42 ± 0.13 20.27 ± 0.14 6.23 ± 0.03 6.28 ± 0.11 38.60 ± 0.26 61.37 ± 0.25 57.19 ± 5.53 72.74 ± 0.95 771.91 ± 5.24 1227.48 ± 5.03 1143.72 ± 110.57 1454.78 ± 19.07 T1 T2 sPA sdPA

2.67 ± 0.02 2.57 ± 0.01 8.50 ± 0.26 5.69 ± 0.1

3.86 ± 0.00 7.33 ± 0.03 2.78 ± 0.02 17.01 ± 0.07

extEC extC tECG tEC tC %G mDP % conversion total tannins (mg/L) sample

Table 1. Composition of the Enological Tannins Used in the Experimenta

extEGC

extECG


623


36.9 ± 2.3 38.2 ± 1.8 35.8 ± 0.7

37.6 ± 0.8 34.6 ± 1.0 31.6 ± 0.4

56.2 ± 2.0 60.8 ± 2.1 52.4 ± 1.5

42.6 ± 1.1 27.9 ± 4.1 34.1 ± 0.3

771.91 ± 5.24 487.01 ± 20.46 476.60 ± 35.04 495.43 ± 12.75

1227.48 ± 5.03 765.41 ± 11.72 802.07 ± 58.76 838.85 ± 7.51

1143.72 ± 110.57 500.94 ± 39.30 448.28 ± 41.89 544.37 ± 12.39

1454.78 ± 19.07 834.06 ± 110.23 1049.19 ± 108.70 958.05 ± 7.90

T1 T1 + gCWM T1 + cmCWM T1 + emCWM


sPA sPA + gCWM sPA + cmCWM sPA + emCWM

sdPA sdPA + gCWM sdPA + cmCWM sdPA + emCWM

5.69 ± 0.1 5.80 ± 0.08 5.72 ± 0.03 5.91 ± 0.04

8.50 ± 0.26 9.01 ± 0.10 8.91 ± 0.19 9.20 ± 0.20

2.57 ± 0.01 2.30 ± 0.02 2.36 ± 0.00 2.40 ± 0.02

2.67 ± 0.02 2.51 ± 0.00 2.51 ± 0.03 2.51 ± 0.00

mDP

17.01 ± 0.07 16.47 ± 0.03 16.58 ± 0.03 16.56 ± 0.01

2.78 ± 0.02 2.77 ± 0.03 2.70 ± 0.07 2.72 ± 0.00

7.33 ± 0.03 5.84 ± 0.14 7.62 ± 0.02 7.68 ± 0.09

3.86 ± 0.00 3.56 ± 0.03 3.43 ± 0.03 3.36 ± 0.01

%G

6.28 ± 0.11 7.00 ± 0.06 6.98 ± 0.23 6.77 ± 0.05

6.23 ± 0.03 7.61 ± 0.03 7.91 ± 0.06 7.41 ± 0.17

20.27 ± 0.14 22.67 ± 0.02 22.60 ± 0.07 22.33 ± 0.11

17.42 ± 0.13 18.46 ± 0.11 18.45 ± 0.45 18.58 ± 0.03

tC

6.56 ± 0.08 5.09 ± 0.25 5.56 ± 0.17 5.26 ± 0.03

5.54 ± 0.33 3.49 ± 0.16 3.31 ± 0.18 3.47 ± 0.07

16.18 ± 0.04 18.59 ± 0.33 17.57 ± 0.04 17.25 ± 0.15

19.33 ± 0.08 20.87 ± 0.19 20.89 ± 0.02 20.85 ± 0.10

tEC

4.74 ± 0.03 5.17 ± 0.05 4.94 ± 0.03 4.87 ± 0.02

0.00 0.00 0.00 0.00

2.47 ± 0.00 2.16 ± 0.02 2.12 ± 0.00 2.06 ± 0.01

0.74 ± 0.00 0.58 ± 0.02 0.58 ± 0.01 0.46 ± 0.01

tECG

7.36 ± 0.01 7.80 ± 0.05 7.66 ± 0.04 7.70 ± 0.01

1.47 ± 0.01 1.64 ± 0.02 1.74 ± 0.03 1.72 ± 0.01

9.66 ± 0.02 9.49 ± 0.12 9.38 ± 0.00 9.41 ± 0.00

5.16 ± 0.04 4.95 ± 0.02 5.17 ± 0.03 5.16 ± 0.01

extC

62.79 ± 0.04 63.63 ± 0.17 63.21 ± 0.1 63.70 ± 0.12

67.53 ± 0.31 68.62 ± 0.16 68.16 ± 0.50 68.08 ± 0.04

46.57 ± 0.12 43.41 ± 0.36 42.83 ± 0.00 43.34 ± 0.16

47.89 ± 0.18 45.99 ± 0.00 46.28 ± 0.42 45.91 ± 0.03

extEC

0.00 0.00 0.00 0.00

16.46 ± 0.02 15.86 ± 0.33 16.19 ± 0.35 16.61 ± 0.20

0.00 0.00 0.00 0.00

6.33 ± 0.01 6.17 ± 0.07 5.79 ± 0.02 6.14 ± 0.03

extEGC

12.27 ± 0.04 11.31 ± 0.01 11.65 ± 0.06 11.68 ± 0.01

2.78 ± 0.02 2.77 ± 0.03 2.70 ± 0.07 2.72 ± 0.00

4.86 ± 0.03 3.68 ± 0.11 5.50 ± 0.02 5.63 ± 0.10

3.12 ± 0.00 2.98 ± 0.01 2.85 ± 0.04 2.90 ± 0.02

extECG

Abbreviations: T1, commercial enological tannin from grape skins; T2, commercial enological tannin from grape seeds; sPA, purified grape skin proanthocyanidin; sdPA, purified grape seed proanthocyanidin; gCWM, CWM from the skin of fresh grapes; cmCWM, cell wall material from the pomace obtained after a control vinification; emCWM, cell wall material from the pomace obtained after a vinification where a maceration enzyme was used; mDP, medium degree of polymerization; %G, percentage of galloylation; extC, percentage of extension (+)-catechin; extEC, percentage of extension (−)-epicatechin; extECG, percentage of extension (−)-epicatechin gallate; tC, percentage of terminal (+)-catechin; tEC, percentage of terminal (−)-epicatechin; tECG, percentage of terminal (−)-epicatechin gallate.

a

% reaction

tannins (mg/L)

sample

Table 3. Effect of the Addition of the Different Monastrell CWMs on the Reduction in Tannin Concentration and on the Composition of the PAs Remaining in Solutiona

Journal of Agricultural and Food Chemistry Article

624


52.3 ± 0.9 27.5 ± 0.8 34.1 ± 1.0

45.5 ± 2.6 28.7 ± 2.0 35.3 ± 2.3

52.7 ± 0.7 47.4 ± 0.7 47.5 ± 0.5

37.8 ± 3.5 27.1 ± 1.8 29.4 ± 1.4

771.91 ± 5.24 367.97 ± 4.21 559.47 ± 10.94 508.37 ± 8.30

1227.48 ± 5.03 668.21 ± 100.11 875.46 ± 53.86 793.85 ± 90.17

1143.72 ± 110.57 540.17 ± 3.08 601.88 ± 2.40 600.07 ± 2.00

1454.78 ± 19.07 904.68 ± 106.25 1060.66 ± 60.45 1027.36 ± 20.15





5.69 ± 0.1 5.84 ± 0.04 5.86 ± 0.12 5.87 ± 0.04

8.50 ± 0.26 11.46 ± 0.44 11.80 ± 0.06 11.57 ± 0.16

2.57 ± 0.01 2.41 ± 0.00 2.43 ± 0.01 2.44 ± 0.01

2.67 ± 0.02 2.62 ± 0.04 2.58 ± 0.02 2.56 ± 0.01

mDP

17.01 ± 0.07 15.65 ± 0.06 15.64 ± 0.20 15.55 ± 0.02

2.78 ± 0.02 3.00 ± 0.03 3.13 ± 0.27 3.30 ± 0.11

7.33 ± 0.03 6.44 ± 0.02 8.22 ± 0.00 8.19 ± 0.05

3.86 ± 0.00 4.00 ± 0.01 3.58 ± 0.02 3.68 ± 0.03

%G

6.28 ± 0.11 6.96 ± 0.18 6.75 ± 0.10 6.84 ± 0.14

6.23 ± 0.03 7.10 ± 0.30 6.90 ± 0.10 6.99 ± 0.08

20.27 ± 0.14 21.58 ± 0.08 21.96 ± 0.18 21.86 ± 0.10

17.42 ± 0.13 17.59 ± 0.27 18.14 ± 0.32 18.02 ± 0.00

tC

6.56 ± 0.08 5.38 ± 0.27 5.68 ± 0.24 5.57 ± 0.20

5.54 ± 0.33 1.63 ± 0.03 1.57 ± 0.06 1.65 ± 0.04

16.18 ± 0.04 17.75 ± 0.08 17.11 ± 0.02 16.98 ± 0.02

19.33 ± 0.08 19.99 ± 0.26 20.13 ± 0.09 20.54 ± 0.08

tEC

4.74 ± 0.03 4.78 ± 0.04 4.64 ± 0.01 4.61 ± 0.02

0.00 0.00 0.00 0.00

2.47 ± 0.00 2.17 ± 0.02 2.15 ± 0.00 2.13 ± 0.01

0.74 ± 0.00 0.59 ± 0.02 0.49 ± 0.00 0.47 ± 0.00

tECG

7.36 ± 0.01 7.78 ± 0.01 7.74 ± 0.02 7.75 ± 0.02

1.47 ± 0.01 1.60 ± 0.04 1.69 ± 0.08 1.64 ± 0.03

9.66 ± 0.01 9.58 ± 0.01 9.11 ± 0.03 9.16 ± 0.02

5.16 ± 0.04 4.95 ± 0.06 5.25 ± 0.02 5.27 ± 0.02

extC

62.79 ± 0.04 64.23 ± 0.15 64.19 ± 0.13 64.29 ± 0.15

67.53 ± 0.31 67.27 ± 0.99 69.87 ± 0.63 69.51 ± 0.15

46.57 ± 0.12 44.65 ± 0.02 43.59 ± 0.14 43.80 ± 0.15

47.89 ± 0.18 46.29 ± 0.55 46.72 ± 0.27 46.25 ± 0.11

extEC

0.00 0.00 0.00 0.00

16.46 ± 0.02 19.40 ± 1.39 16.84 ± 0.48 16.91 ± 0.17

0.00 0.00 0.00 0.00

6.33 ± 0.01 7.17 ± 0.09 6.18 ± 0.04 6.24 ± 0.02

extEGC

12.27 ± 0.04 10.87 ± 0.02 10.99 ± 0.21 10.93 ± 0.03

2.78 ± 0.02 3.00 ± 0.03 3.13 ± 0.27 3.30 ± 0.11

4.86 ± 0.01 4.27 ± 0.01 6.07 ± 0.01 6.06 ± 0.04

3.12 ± 0.00 3.41 ± 0.01 3.09 ± 0.02 3.21 ± 0.02

extECG


a

% reaction

tannins (mg/L)

sample

Table 4. Effect of the Addition of the Different Cabernet Sauvignon CWMs on the Reduction in Tannin Concentration and on the Composition of the PAs Remaining in Solutiona


625


35.84 ± 0.9 30.5 ± 0.4 35.6 ± 1.1

27.5 ± 1.2 26.6 ± 1.9 32.0 ± 1.0

45.9 ± 2.2 48.1 ± 0.5 49.9 ± 1.1

29.7 ± 0.4 28.0 ± 1.2 30.5 ± 1.9

771.91 ± 5.24 459.25 ± 25.38 536.51 ± 16.16 497.29 ± 19.40

1227.48 ± 5.03 890.28 ± 17.84 900.28 ± 57.65 834.90 ± 17.37

1143.72 ± 110.57 619.17 ± 62.29 593.21 ± 1.85 572.83 ± 16.47

1454.78 ± 19.07 1023.09 ± 1.24 1047.50 ± 14.10 1010.84 ± 24.46





5.69 ± 0.1 5.85 ± 0.02 5.68 ± 0.01 5.76 ± 0.01

8.50 ± 0.26 8.79 ± 0.46 11.36 ± 0.03 10.96 ± 0.00

2.57 ± 0.01 2.39 ± 0.00 2.45 ± 0.00 2.43 ± 0.01

2.67 ± 0.02 2.60 ± 0.00 2.56 ± 0.02 2.53 ± 0.00

mDP

17.01 ± 0.07 15.47 ± 0.22 15.63 ± 0.07 15.73 ± 0.04

2.78 ± 0.02 3.14 ± 0.16 2.84 ± 0.04 2.87 ± 0.01

7.33 ± 0.03 6.58 ± 0.02 8.14 ± 0.01 7.82 ± 0.02

3.86 ± 0.00 3.92 ± 0.35 3.51 ± 0.07 3.49 ± 0.00

%G

6.28 ± 0.11 6.83 ± 0.15 6.86 ± 0.14 6.88 ± 0.08

6.23 ± 0.03 7.35 ± 0.16 7.22 ± 0.03 7.54 ± 0.04

20.27 ± 0.14 22.05 ± 0.04 21.77 ± 0.02 21.96 ± 0.09

17.42 ± 0.13 18.06 ± 0.18 18.33 ± 0.03 18.50 ± 0.10

tC

6.56 ± 0.08 5.58 ± 0.06 6.04 ± 0.16 5.79 ± 0.07

5.54 ± 0.33 4.04 ± 0.43 1.58 ± 0.01 1.58 ± 0.04

16.18 ± 0.04 17.54 ± 0.07 16.93 ± 0.03 17.08 ± 0.01

19.33 ± 0.08 19.87 ± 0.11 20.19 ± 0.19 20.59 ± 0.15

tEC

4.74 ± 0.03 4.69 ± 0.04 4.70 ± 0.02 4.68 ± 0.01

0.00 0.00 0.00 0.00

2.47 ± 0.001 2.21 ± 0.04 2.11 ± 0.01 2.10 ± 0.01

0.74 ± 0.00 0.56 ± 0.00 0.47 ± 0.01 0.47 ± 0.01

tECG

7.36 ± 0.01 7.78 ± 0.02 7.70 ± 0.00 7.70 ± 0.02

1.47 ± 0.01 1.71 ± 0.05 1.76 ± 0.01 1.75 ± 0.02

9.66 ± 0.02 9.54 ± 0.07 9.33 ± 0.05 9.40 ± 0.06

5.16 ± 0.04 5.16 ± 0.04 5.17 ± 0.06 5.19 ± 0.04

extC

62.79 ± 0.04 64.35 ± 0.12 63.77 ± 0.50 63.91 ± 0.51

67.53 ± 0.31 68.15 ± 0.44 70.28 ± 0.04 69.91 ± 0.02

46.57 ± 0.12 44.28 ± 0.06 43.83 ± 0.10 43.73 ± 0.12

47.89 ± 0.18 46.36 ± 0.21 46.69 ± 0.30 46.20 ± 0.00

extEC

0.00 0.00 0.00 0.00

16.46 ± 0.02 15.61 ± 0.06 16.32 ± 0.02 16.34 ± 0.05

0.00 0.00 0.00 0.00

6.33 ± 0.01 6.63 ± 0.04 6.11 ± 0.09 6.03 ± 0.01

extEGC

12.27 ± 0.04 10.78 ± 0.18 10.93 ± 0.04 11.04 ± 0.05

2.78 ± 0.02 3.14 ± 0.16 2.84 ± 0.04 2.87 ± 0.01

4.86 ± 0.03 4.37 ± 0.07 6.03 ± 0.05 5.72 ± 0.03

3.12 ± 0.00 3.36 ± 0.35 3.04 ± 0.06 3.02 ± 0.01

extECG


a

% reaction

tannins (mg/L)

sample

Table 5. Effect of the Addition of the Different Syrah CWMs on the Reduction in Tannin Concentration and on the Composition of the PAs Remaining in Solutiona


626


Article


Figure 2. Comparison of the size exclusion chromatography of the commercial enological tannin T1 in solution and that for this remaining after the addition of different CWMs (gCWM, CWM from the skin of fresh grapes; cmCWM, cell wall material from the pomace obtained after a control vinification; emCWM, cell wall material from the pomace obtained after a vinification where a maceration enzyme was used).

Figure 3. Comparison of the size exclusion chromatography of the T2 in solution and that for this remaining after the addition of different CWMs (gCWM, CWM from the skin of fresh grapes; cmCWM, cell wall material from the pomace obtained after a control vinification; emCWM, cell wall material from the pomace obtained after a vinification where a maceration enzyme was used).

The results of the phloroglucinolysis analysis showed that the addition of the different CWMs to the suspension containing the tannins produced, in all cases, a decrease in the quantity of tannins in the solution; that is, all of them could act as fining agents, reducing tannins in the wine. For Monastrell samples, this decrease ranged from 27.9% for the solution containing sdPA and cmCWM to 60.8% for sPA and cmCWM. The differences in the adsorption capacity of Monastrell fresh skin CWs compared with those of the pomace were not very large, being only slightly lower for cmCW with T1, T2, and sdPA. The data shown in the Supporting Information indicate that the percentual carbohydrate composition did not significantly change between fresh skin CWs and pomace CWs. In Monastrell CW samples, only the use of a maceration enzyme promoted a small decrease in the uronic acid proportion of the pomace CWs, the other sugars barely changing. Guerrero et al.18 also reported that pomace CWs presented a chemical and polysaccharide composition quite similar to that of fresh grapes, although they seemed to present larger dry volume

per unit mass and an increased expansion in hydroalcoholic solution. The potential contribution of an expanded surface area of the hydrated fiber, and therefore greater surface interaction, might well contribute to the high affinity of pomace CWs for PAs. The adsorption capacity of Monastrell emCWs was always similar to that of fresh skin CWs, even when pectins (measured as uronic acids) were present at lower proportion. Le Bourvellec et al.28 stated that the elimination of pectins decreased the apparent affinity values of cell walls and PA. However, the increase in cell wall porosity after pectin elimination may enhance the encapsulation of PA within a more open network and explain the very similar behaviors of gCW and emCWs. Maximum adsorption was observed for purified skin PA samples, with lower values being observed for purified seed PA, the adsorption values of which were similar to those of the commercial samples, which, in turn, showed very similar behavior. Bindon et al.,29 working with solutions containing different 627


Article


Figure 4. Comparison of the size exclusion chromatography of the purified skin tannin (sPA) in solution and that for this remaining after the addition of different CWMs (gCWM, CWM from the skin of fresh grapes; cmCWM, cell wall material from the pomace obtained after a control vinification; emCWM, cell wall material from the pomace obtained after a vinification where a maceration enzyme was used).

Figure 5. Comparison of the size exclusion chromatography of the purified seed tannin (sdPA) in solution and that for this remaining after the addition of different CWMs (gCWM, CWM from the skin of fresh grapes; cmCWM, cell wall material from the pomace obtained after a control vinification; emCWM, cell wall material from the pomace obtained after a vinification where a maceration enzyme was used).

percentages of skin and seed PAs, found that the total PA amount bound to flesh CWM gradually declined as the proportion of seed PA in the solution increased (from 57% removed for a solution containing 100% skin PA to 47% removed for a solution containing 100% seed PA), although, contrary to our results, they did not find any pattern of PA mass decrease when using skin CWs. The differences in adsorption between skin and seed PA are probably related to their different compositions. Haslam6 reported that interactions between PAs and CWM increase as mDP increases (because each additional monomer increases the number of reactive sites, allowing the tannin molecule to bind simultaneously to more than one site in the cell wall), whereas the presence of epigallocatechin subunits in the skin tannin composition may also contribute to the differences because epigallocatechin has an additional hydroxyl group on ring B that provides another site for hydrogen bonding to occur.5

The percentage of galloylation (%G) slightly decreased in all of the nonadsorbed tannins (except in the case of the interaction of the pomace CWM with T2, where a small increase was observed). The mDP also tended to decrease in the fraction of nonadsorbed tannins, except for the samples where purified seed PA was used, when almost no changes were observed, and in the samples where skin PA was used, when the mDP of the remaining tannins slightly increased. In the case of purified samples, where large quantities of high molecular mass tannins are present, saturation of the available binding sites on the CWM surface might occur, whereas smaller PA molecules might still penetrate the interior of the CWM structure, therefore increasing the mDP of the remaining tannins, as also stated by Bindon et al.11 628


Article


Figure 6. Percent changes in the main parameters when Monastrell CWs are used (T1, commercial enological tannin from grape skins; T2, commercial enological tannin from grape seeds; sPA, purified grape skin proanthocyanidin; sdPA, purified grape seed proanthocyanidin; gCWM, CWM from the skin of fresh grapes; cmCWM, cell wall material from the pomace obtained after a control vinification; emCWM, cell wall material from the pomace obtained after a vinification where a maceration enzyme was used; mDP, medium degree of polymerization; %G, percentage of galloylation; F1, F2, and F3, percentage of area of the size exclusion chromatogram corresponding to high (from 10 to 11.8 min), medium (from 11.8 to 13.6 min), and low (from 13.6 to 15 min) molecular mass PAs).

The terminal and extension subunits changed very little; a decrease in terminal epicatechin was always observed in sPA, concomitant with the increase in mDP. When Cabernet Sauvignon CWM (Table 4) was used, the results showed that gCWM, in general, retained similar or larger quantities of tannins than Monastrell gCWM, which can be attributed to this variety presenting the highest content in pectins in their CWs, but this retention significantly decreased when the pomace CWM of this variety was used. The losses in sugars and pectins observed when the fresh skin CWs and the pomace CWs of this variety are compared might explain these differences. Also, another key structural difference that could reduce PA adsorption to Cabernet Sauvignon pomace cell walls could be, as suggested by Bindon et al.,10 the endogenously higher concentrations of insoluble PA and lignin and lower cell wallbound protein than in fresh skin cell walls. These differences may confer reduced flexibility and porosity to the pomace cell walls. In the case of T1, the retained quantities decreased by almost 50% compared with those retained in gCWM. The decrease was less pronounced when sPA was used, observing, in general, a higher adsorption of tannins when skin PAs (both commercial and purified) were used, similarly to that described for Monastrell. The behaviors of mDP and galloylation were exactly the same as those observed in Monastrell CWM, the observed increases in mDP in the nonadsorbed sPA being even higher. In the case of Syrah, the retention of PAs on gCWM was the lowest of the three varieties studied, perhaps because the CW pectin content was the lowest and the cellulose content the highest (see Supporting Information, Figure S3). However, the adsorption capacities of gCWM and pomace CWM were almost the same, as in Monastrell, only a small decrease being observed (although much lower than that observed in Cabernet Sauvignon pomace CWM). The effect on mDP and the percentage of

galloylation of the remaining PAs in solution was similar to that previously described for the other varieties, as was the higher affinity for skin PA samples. These results suggest that, regarding binding capacity, important varietal differences exist and Monastrell pomace CWs could be the more interesting for reducing tannins and astringency in wines than pomace from other varieties. The results of SEC analysis, comparing the PA profiles before and after the addition of CWM, are shown in Figures 2−5, whereas a comparison of the percentages of retention measured in the phloroglucinolysis analysis and SEC, including the retention of high, medium, and low molecular mass PAs, is shown in Figures 6−8. As can be seen, the results were quite coincident with the phloroglucinolysis analysis, although the retention measured by SEC was greater than that observed by phloroglucinolysis for all of the varieties and all of the different PAs. This was probably due to the presence in the samples of some type of structures, such as oxidized tannins, that are not depolymerized by the phloroglucinol reagent but are measured by the SEC analysis.30 This chromatographic analysis also showed that the intermediate (those eluting in SEC between 11.8 and 13.6 min) and especially the high molecular mass tannin fractions (those eluting from 10 to 11.8 min) were preferentially removed in all cases. From a study of each individual tannin, the results for T1 (Figures 2 and 6) indicate that the highest retention was reported for Monastrell CWs. For Syrah (Figures 2 and 7) and Cabernet Sauvignon (Figures 2 and 8) CWs, the quantities of adsorbed tannin were lower when the pomace CWs were in the solution than with that of fresh grapes CWs and lower than that reported for Monastrell. With regard to T2 (Figure 3), the behavior of Monastrell CWs was very similar to that reported for T1, the CWM of this variety 629


Article


Figure 7. Percent changes in the main parameters when Syrah CWs are used (T1, commercial enological tannin from grape skins; T2, commercial enological tannin from grape seeds; sPA, purified grape skin proanthocyanidin; sdPA, purified grape seed proanthocyanidin; gCWM, CWM from the skin of fresh grapes; cmCWM, cell wall material from the pomace obtained after a control vinification; emCWM, cell wall material from the pomace obtained after a vinification where a maceration enzyme was used; mDP, medium degree of polymerization; %G, percentage of galloylation; F1, F2, and F3, percentage of area of the size exclusion chromatogram corresponding to high (from 10 to 11.8 min), medium (from 11.8 to 13.6 min), and low (from 13.6 to 15 min) molecular mass PAs).

Figure 8. Percent changes in the main parameters when Cabernet Sauvignon CWs are used (T1, commercial enological tannin from grape skins; T2, commercial enological tannin from grape seeds; sPA, purified grape skin proanthocyanidin; sdPA, purified grape seed proanthocyanidin; gCWM, CWM from the skin of fresh grapes; cmCWM, cell wall material from the pomace obtained after a control vinification; emCWM, cell wall material from the pomace obtained after a vinification where a maceration enzyme was used; mDP, medium degree of polymerization; %G, percentage of galloylation; F1, F2, and F3, percentage of area of the size exclusion chromatogram corresponding to high (from 10 to 11.8 min), medium (from 11.8 to 13.6 min), and low (from 13.6 to 15 min) molecular mass PAs).

Cabernet Sauvignon, the differences between gCWs and emCWs were almost nonexistent, whereas adsorption was lower in cmCWs.

showing the highest adsorption compared with the results obtained in Cabernet Sauvignon and Syrah, the pomace CWs retaining even more tannins than fresh skin CW. For Syrah and 630


Article


proanthocyanidins, which cannot be measured by the phloroglucinolysis analysis. Finally, a discriminant algorithm was used to identify the specific variables potentially useful as predictors of discrimination between the three varieties and between the origins of the CWs. The discriminant analysis was based on a backward step algorithm (Table 6 and Figure 9). For the discrimination of

In the purified samples, sPA (Figure 4) was highly retained by Monastrell CWs, reaching 67 and 65% in cmCW and emCWs. Also, high values of total adsorption were observed for Syrah and Cabernet Sauvignon CWs, the values being higher than that obtained with commercial tannins and, now, differences in the adsorption capacity between gCW and pomace CW were very small, especially in Syrah. In the SEC profiles for sdPA (Figure 5), and contrary to the results observed in the phloroglucinolysis analysis, large differences were observed between varieties. The retention in Monastrell reached 64% for emCW, whereas for Cabernet Sauvignon the maximum value was only 38%, the use of pomace CWs leading to a retention of only 27 and 29%. Syrah showed intermediate values. It is clear that the SEC profiles of T1 and T2 and their behaviors were quite similar with regard to the retention properties (similar characteristics, especially the mDP, were observed in both commercial tannins, despite their different origins), whereas for purified samples, sPA and sdPA, behaved clearly differently, the higher adsorption being observed for sPA, as also found by Bindon et al.29 for flesh CWs. The results obtained by SEC are quite similar to the results observed with the phloroglucinolysis analysis and confirm that CWs from Monastrell pomace could be a really interesting material for wine tannin fining, being very effective in binding both nonpurified commercial tannins and native skin and seed tannins. With regard to the molecular mass distribution of the retained tannins, the highest retention was always observed in the high molecular mass PAs, especially with Monastrell CWs, for which values of >70% were obtained for sPA + cmCW and T2 + cmCW. Bindon et al.29 also found that for a 100% skin PA, the greatest PA-CW adsorption occurred in the higher molecular mass range, with less material removed by CWM in the lower molecular mass range. In general, emCW retained, percentually, lower quantities of high molecular mass PAs than gCW, perhaps due to a more linear structure, with slightly more affinity for low molecular mass tannins. Bindon and Smith17 also observed that lower molecular PAs were preferentially bound when the Cabernet Sauvignon fibers were used. These observations and differences can be much more clearly observed using a multivariate analysis of variance, which allows the different affinity properties of CWs (measured as the percentage of retained PAs calculated by phloroglucinolysis or SEC analysis) to be observed separately, taking into account their varietal or technological (fresh or processed skins) origin (Supporting Information, Figures S5 and S6). Monastrell CWs generally retained the highest quantities of PAs, especially when observed by SEC. It could be hypothesized that Monastrell CWs, due to their composition, have a higher capacity to bind oxidized tannins, which would explain the differences in percentage of retention observed between the phloroglucinolysis analysis and SEC. This would also explain why differences between varieties are much higher when the binding capacity is measured by SEC. The same argument regarding varieties could be applied to the type of CW. Considering all of the data and for all of the varieties, gCWM showed a high PA affinity, in both phloroglucinolysis and SEC analyses. On the other hand, a large difference was observed in the binding capacity of emCWs, depending on the type of analysis. They were seen to retain a high quantity of PAs when the nonadsorbed PAs in solution were analyzed by SEC and much lower quantities when they were determined by the phloroglucinolysis analysis, which suggests that emCWs present a high affinity for oxidized and other nondepolymerized

Table 6. Discriminant Functions Used To Predict to Which Group the Observations Belong standardized coefficient

function 1

function 2

Variety SEC area mDP CW Origin mDP phloroglucinolysis area % of retention at 90% elution time

1.24 0.94

−0.22 0.84

1.03 0.75 −0.84

0.30 1.02 0.79

varieties, the analysis was conducted with an F-to-enter = 2. This strategy resulted in a considerable reduction in the dimensionality of the information, because it led to the selection of very few variables that can be considered as the most important for the differentiation of the samples and which could indicate whether the various types of CWs could really be classified as different. When the variables were used to classify the CWs according to their varietal origin, a 72% correct classification was obtained. CWs from the Monastrell variety were well classified, only one sample being misclassified. CWs from Cabernet Sauvignon and Syrah showed slightly poorer separation, which is coincident with the experimental results, Monastrell CWs showing higher affinity for PAs than the CWs from Cabernet Sauvignon or those from Syrah grapes. The backward discriminant analysis allowed this separation to be reached with only two variables, the percentage of PA adsorbed to the CWs, measured as the percentage of decrease in the SEC area, and the mDP of the nonadsorbed tannins. The classification obtained according to the technological origin of CW (from fresh skins or pomace from control or enzyme vinifications) was not very good (as low as 63%), indicating that the differences in their adsorption properties were not very different. The mDP of the nonadsorbed tannins, the percentage of adsorbed tannins measured by the area of phloroglucinolysis analysis, and the retention of the low molecular mass tannins were the variables used to discriminate the samples. In conclusion, this study has confirmed that pomace CWs, a byproduct of the enology industry, present a high affinity for PAs and has demonstrated that this was especially remarkable in those of Monastrell grapes, for which pomace CWs present a very high adsorption capacity of all types of PA structures but especially for those of high molecular mass, and so their use in wines could reduce astringency. However, pomace cell walls present a slightly higher affinity for low molecular mass PAs than CWs from fresh skin grapes, and this might have implications for wine sensory properties such as astringency and bitterness and must be studied in greater depth. Moreover, and given that Guerrero et al.18 observed that, in general, grape fibers not only reduced PAs but also reduced anthocyanins, total phenolics, and wine color density, to confirm the potential use of Monastrell pomace cell walls as an alternative fining agent to proteins, their role in the retention of other phenolic and volatile compounds in 631


Article


Figure 9. Plot of discriminant functions for (A) variety and (B) type of cell wall. (6) Haslam, E. Molecular recognition. Phenols and polyphenols. In Practical Polyphenolics. From Structure to Molecular Recognition and Physiological Action; Cambridge University Press: Cambridge, UK, 1998. (7) Kennedy, J. A.; Matthews, M. A.; Waterhouse, A. L. Changes in grape seed polyphenols during fruit ripening. Phytochemistry 2000, 55, 77−85. (8) Bautista-Ortin, A. B.; Rodríguez-Rodríguez, P.; Gil-Muñoz, R.; Jímenez-Pascual, E.; Busse-Valverde, N.; Martínez-Cutillas, A.; LópezRoca, J. M.; Gómez-Plaza, E. Influence of berry ripeness on concentration, qualitative composition and extractability of grape seed tannins. Aust. J. Grape Wine Res. 2012, 18, 123−130. (9) Bindon, K.; Kennedy, J. Ripening-induced changes in grape skin proanthocyanidins modify their interaction with cell walls. J. Agric. Food Chem. 2011, 59, 2696−2707. (10) Bindon, K.; Bacic, A.; Kennedy, J. Tissue-specific and developmental modifications of grape cell walls influence the adsorption of proanthocyanidins. J. Agric. Food Chem. 2012, 60, 9249−9260. (11) Bindon, K.; Madani, S.; Pendleton, P.; Smith, P.; Kennedy, J. Factors affecting skin tannin extractability in ripening grapes. J. Agric. Food Chem. 2014, 62, 1130−1141. (12) Busse-Valverde, N.; Bautista-Ortín, A. B.; Gómez-Plaza, E.; Fernández-Fernández, J. I.; Gil-Muñoz, R. Influence of skin maceration time on the proanthocyanidin content of red wines. Eur. Food Res. Technol. 2012, 235, 1117−1123. (13) Hill, M.; Kaine, G. The Market Potential of Tannin Related Innovations in the Wine Industry; Victorian Department of Primary Industries: Australia, 2007; Tatura pp. 41. (14) Bautista-Ortin, A. B.; Cano-Lechuga, M.; Ruiz-García, Y.; GómezPlaza, E. Interactions between grape skin cell wall material and commercial enological tannins. Practical implications. Food Chem. 2014, 152, 558−565. (15) Bautista-Ortín, A. B.; Martínez-Cutillas, A.; Ros-García, J. M.; López-Roca, J. M.; Gómez-Plaza, E. Improving colour extraction and stability in red wines: the use of maceration enzymes and enological tannins. Int. J. Food Sci. Technol. 2005, 40, 1−12. (16) Boulton, R.; Singleton, V.; Bisson, L.; Kunkee, R. Principles and Practices of Winemaking; Chapman & Hall: New York, 1996. (17) Bindon, K.; Smith, P. Comparison of the affinity and selectivity of insoluble fibres and commercial proteins for wine proanthocyanidins. Food Chem. 2013, 136, 917−928.

wine, the effect on wine chromatic characteristics, and some technological aspects, such as the amount of lees formed during their used, need to be study.

■

ASSOCIATED CONTENT

S Supporting Information *

Figures S1−S5. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*(E.G.-P.) E-mail: [email protected]. Phone: 34 868887323. Fax: 34 868884147. Funding

This work was made possible by financial assistance of the Ministerio de Economı ́a y Competitividad and FEDER funds, Project AGL2012-39845-C02-01. Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Ortega-Regules, A.; Romero-Cascales, I.; Ros-García, J. M.; LópezRoca, J. M.; Gómez-Plaza, E. A first approach towars the relationship between grape skin cell-wall composition and anthocyanin extractability. Anal. Chim. Acta 2006, 563, 26−32. (2) Rolle, L.; Torchio, F.; Ferrandino, A.; Guidoni, S. Influence of wine-grape skin hardness on the kinetics of anthocyanin extraction. Int. J. Food Prop. 2012, 15, 249−261. (3) Nunan, K. J.; Sims, C. A.; Bacic, A.; Robinson, S. P.; Fincher, G. B. Isolation and characterization of cell walls from the mesocarp of mature grape berries (Vitis vinifera). Planta 1997, 203, 93−100. (4) McManus, J.; Davis, K.; Beart, J.; Gaffney, S.; Lilley, T.; Haslam, E. Polyphenols interactions. I. Introduction, some observations on the reversible complexation of polyphenols with proteins and polysaccharides. J. Chem. Soc., Perkin Trans. 2 1985, 1429−1438. (5) Le Bourvellec, C.; Guyot, S.; Renard, C. M. Noncovalent interaction between procyanidins and apple cell wall material. Part 1. Effect of some environmental parameters. Biochim. Biophys. Acta 2004, 1672, 192−202. 632


Article

Journal of Agricultural and Food Chemistry (18) Guerrero, R.; Smith, P.; Bindon, K. Application of insoluble fibers in the fining of wine phenolics. J. Agric. Food Chem. 2013, 61, 4424− 4432. (19) Apolinar-Valiente, R. Pared celular de uva y polisacáridos de vinos de ́ enzimáticas y de frio. distinta procedencia, elaborados mediante tecnologias Ph.D. thesis, University of Murcia, 2012. (20) Romero-Cascales, I.; Ros-García, J. M.; López-Roca, J. M.; Gómez-Plaza, E. The effect of a commercial pectolytic enzyme on grape skin cell wall degradation and colour evolution during the maceration process. Food Chem. 2012, 130, 626−631. (21) de Vries, J. A.; Voragen, A. G. J.; Rombouts, F. M.; Pilnik, W. Estraction and purification of pectins from alcohol insoluble solids from ripe and unripe apples. Carbohydr. Polym. 1981, 1, 117−127. (22) Apolinar-Valiente, R.; Romero-Cascales, I.; López-Roca, J. M.; Gómez-Plaza, E.; Ros-García, J. M. Application and comparation of four selected procedures for the isolation of cell wall material from the skin of grapes cv. Monastrell. Anal. Chim. Acta 2010, 660, 206−210. (23) Ribéreau-Gayon, P.; Glories, Y.; Maujean, A.; Dubourdieu, D. Traité d’Oenologie. 2. Chimie du Vin. Stabilisation et Traitements; Dunod: Paris, France, 1998. (24) Kennedy, J. A.; Taylor, A. W. Analysis of proanthocyanidins by high-performance gel permeation chromatography. J. Chromatogr. 2003, 995, 99−107. (25) Kennedy, J. A.; Jones, G. P. Analysis of proanthocyanidin cleavage products following acid-catalysis in the presence of excess phloroglucinol. J. Agric. Food Chem. 2001, 49, 1740−1746. (26) Busse-Valverde, N.; Gómez-Plaza, E.; López-Roca, J. M.; GilMuñoz, R.; Fernández-Fernández, J. I.; Bautista-Ortín, A. B. Effect of different enological practices on skin and seed proanthocyanidins in three varietal wines. J. Agric. Food Chem. 2010, 58, 11333−11339. (27) Moutounet, M.; Rigaud, J.; Souquet, J. M.; Cheynier, V. Caracterisation structurale des tanins de las baie de raisin. Quelques exemples de l’incidence du cepage, du terroir et du monde de conduite de la vigne (1). Bull. O.I.V. 1996, 433−443. (28) Le Bourvellec, C.; Watrelot, A.; Ginies, C.; Imberty, A.; Renard, C. M. Impact of processing on the noncovalent interactions between procyanidin and apple cell wall. J. Agric. Food Chem. 2012, 60, 9484− 9494. (29) Bindon, K.; Smith, P.; Kennedy, J. Interaction between grapederived proanthocyanidins and cell wall material. 1. Effect on proanthocyanidin composition and molecular mass. J. Agric. Food Chem. 2010, 58, 2520−2528. (30) Poncet-Legrand, C.; Cabane, B.; Bautista-Ortin, A. B.; Carrillo, S.; Fulcrand, H.; Pérez, J.; Vernhet, A. Tannin oxidation: intra- versus intermolecular reactions. Biomacromolecules 2010, 11, 2376−2386.

633


Remarkable Proanthocyanidin Adsorption ... - ACS Publications

Recommend Documents