Impact of Oak Wood Barrel Tannin Potential and Toasting on White

Article Cite This: J. Agric. Food Chem. XXXX, XXX, XXX−XXX

pubs.acs.org/JAFC

Impact of Oak Wood Barrel Tannin Potential and Toasting on White Wine Antioxidant Stability Maria Nikolantonaki,*,† Samar Daoud,† Laurence Noret,† Christian Coelho,† Marie-Laure Badet-Murat,‡ Philippe Schmitt-Kopplin,§,∥ and Reǵ is D. Gougeon†

Downloaded via BUFFALO STATE on July 17, 2019 at 06:22:08 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

UMR PAM Université de Bourgogne/Agro Sup Dijon, Institut Universitaire de la Vigne et du Vin, Jules Guyot, 21000 Dijon, France ‡ Œnologie by MLM, 25 rue Aurel Chazeau, 33160 Saint-Médard-en-Jalles, France § Research Unit Analytical BioGeoChemistry, Helmholtz Zentrum Muenchen, 85764 Neuherberg, Germany ∥ Technische Universität München, Analytical Food Chemistry, Alte Akademie 10, 85354 Freising, Germany S Supporting Information *

ABSTRACT: Wines aged in oak wood barrels with various uniform tannin contents (which were classified according to their total ellagitannins contents as predicted by Near Infrared Spectroscopy on the untoasted wood) and different toasting levels (high precision toasting by radiation) were distinguished according to their overall abilities to resist against oxidation. Wine trials were carried out on two different vintages (2015, 2016) and three grape varieties (Sauvignon blanc, Sémillon, Chardonnay). Regardless of the vintage and the wine matrix, a relationship was established between wine oxidative stability (based on EPR spin trapping methodology) and oak barrel tannin potential. The extraction kinetic of ellagitannins by wines appeared linear during barrel aging and achieved its maximum at six or eight months, in a grape variety dependent manner. Oak wood barrel tannin potentials and toastings had no effect on wine glutathione and polyphenols contents. However, wines aged in new barrels with both low and medium tannin potentials, preserved at the end of aging and important number of S−N containing compounds, which was in addition to the known ellagitanins, revealed wines better antioxidant stability. KEYWORDS: chardonnay wine, sauvignon wine, ellagitannins, glutathione, phenolic compounds, sulfur compounds, oxidation, radical chemistry, EPR

1. INTRODUCTION The signature of great dry white wines, besides their organoleptic complexity, is their ability to improve with age. Since oxidative instability was observed, winemakers have taken greater precautions to avoid this premature aging. The management of barrel aging, an intrinsic step in the production of premium quality wines, plays a major role in the issue. Originally, oak wood barrels were used as a mode of transport and aging which allowed micro-oxygenation and enrichment of phenolic and volatile compounds that complexify wine sensory perception and increase antioxidant stability.1 Wood contact during winemaking improves organoleptic quality in terms of complex ellagitanins and quercotriterpenosides extraction as well as volatiles constituents coming generally from ethanolysis and thermolysis of lignin and hemicelluloses.2−4 Several studies have emphasized the importance of barrel aging for red wine color stabilization and mouth feel, related to bitterness and astringency,5−7 but there is a lack of knowledge concerning its impact on white wine ageability. Oak wood has an antioxidant capacity that influences the oxidation−reduction potential of wines and therefore its oxidative stability. The antioxidant capacity of oak wood is dependent on the ellagitannin content and the toasting process.8,9 Alañon et al. (2011) on the basis of phenolic compounds profiling showed a strong correlation between wood extract antioxidant capacity and ellagitannin concentration. Hence, hydrolyzable tannins as high molecular weight © XXXX American Chemical Society

polyphenols are among the known compounds mainly responsible for the antioxidant capacity of oak wood extracts.10 This antioxidant capacity is due to the physicochemical properties of ellagitannins and their structure.11 Indeed, ellagitanins are composed of several benzyl aromatic rings with hydroxyl groups, which gives them an ability to readily yield a hydrogen atom and the ability to support the unpaired electron when compared to low molecular weight phenolic compounds.9,11 Despite many analytical methods being available to assess wood extract antioxidant capacity in vitro, the estimation of young white wine antioxidant stability contributed by oak wood is a difficult task because of the matrix effect, lees aging, oxygen intake during barrel, aging and other oenological practices (i.e., sulfur dioxide addition). Therefore, in our recent study, we propose a fine classification of wines according to their resistance against oxidation via a novel tool based on EPR spin trapping methodology.12 After this methodology, the monitoring of POBN-1-hydroxyethyl spin adduct formation in wines initiated via the Fenton reaction allows the assessment of wine resistance against oxidation after modelization of Imax POBN‑1‑HER and rPOBN‑1‑HER analytical parameters. Low Received: Revised: Accepted: Published: A

January 23, 2019 May 22, 2019 July 5, 2019 July 5, 2019 DOI: 10.1021/acs.jafc.9b00517 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article

Journal of Agricultural and Food Chemistry

Figure 1. Experimental set up of white wines (A, B, C, D, E, F) from different vintages and appellations, aged in oak barrels. Toasting: blanche = 150 °C/1 h; ivoire = 160−170 °C/1 h 30. LTP and MTP/Tannin Potential corresponding to different ellagitannin content in untoasted wood: low or LTP from 2000 to 4000, medium or MTP from 4001 to 6000 of ellagic acid equivalent/g of dry wood. Control: one year old barrel. ellagitannin content was performed by near-infrared spectroscopy (NIRS, AOTF, Brimrose, U.S.A.). This device has the advantage of providing an extremely rapid spectral scanning, while providing a perfect resolution and excellent spectral reproducibility. The classification from the NIRS method was used to sort the untoasted staves into two groups of tannin potential (TP), directly related to the ellagitannin content in untoasted wood: Low or LTP from 2000 to 4000 μg, Medium or MTP from 4001 to 6000 μg of ellagic acid equivalent per g of dry wood. 2.3.2. Wood Toasting Methodology. Toasting was made after a 4 min automated steam bending, which allows us to obtain a neutral white barrel inside. Toasting, controlled by a computer, was performed using radiant heat rather than direct contact with flame. The toasting pot was fed regularly with fuel (100% oak pellets) by an auger. The barrel, placed on a turntable, rotates around a double cone that covers the fire and channels the heat source, during the entire toasting phase. An infrared sensor, performing measurements on the internal surface of the shell, provides temperature control, with a heating accuracy of ±3 °C. In our study, we used LTP and MTP barrels paired with two toasting levels: blanche (150 °C/1 h) and ivoire (160 to 170 °C in the two steps of 30 min each). 2.4. Wine Resistance against Oxidation Evaluated by EPR. The estimation of wine resistance against oxidation was performed by electronic paramagnetic resonance (EPR) after free radical initiation of oxidation as recently described by Nikolantonaki et al. (2019). Wine samples were asperged with CO2 to remove residual free SO2 before analysis. For the EPR analysis, three different solutions were prepared in amber vials. FeSΟ4·7Η2Ο (50 μM) and H2O2 (954 μΜ) were used as sources of hydroxyl radicals, and 30 mM POBN solution was used as a spin trap. Fenton reaction was initiated in an amber vial, and sample were stirred and quickly transferred to an EPR capillary for analysis at room temperature (298 K). EPR measurements were performed using an ER300 EPR spectrometer. The parameters used for the experiments were as follows: modulation frequency 100 kHz, modulation amplitude 0.9 G, time constant 10.24 ms, conversion time 2.56 ms, microwave power 10 mW, and receiver gain 104. Serial 2 min EPR acquisitions were performed. The intensity of the EPR signals was determined by the WINESR software program. 2.5. UPLC-DAD Analysis of Total Ellagitanins. The concentration of total ellagitanins after acid hydrolysis (2 h at 100 °C, 2 N HCl in MeOH) was determined according to the protocol proposed by Michel et al. (2016) with some modifications. A 50 mL sample of white wine was evaporated under reduced pressure at a temperature of approximately 60 °C and redissolved in 20 mL of methanol. Then, 4 mL of this mixture was placed in three hydrolysis tubes with 1 mL of HCl (37%) and kept at 100 °C for 2 h. At the same time, two control tubes containing 1 mL each of this mixture were kept at room temperature. Before hydrolysis, 100 or 25 μL of methanol with 2.5 g L−1 4-chloro-1-naphthol as an internal standard was added to the hydrolysis and control tubes. After hydrolysis, the samples were filtered (0.45 μm, PTFE) prior to UPLC−UV analysis. The quantity

Imax POBN‑1‑HER and rPOBN‑1‑HER values suggest that the wine matrix is able to quench a substantial amount of radicals and avoid deleterious oxidative reactions. The aims of the present study were 3-fold: The first aim was to measure wine resistance against oxidation when they are aged in barrels classified according to their tannin potentials and toasting levels. The second aim was to study the extraction kinetics of oak wood ellagitanins in white wines and their impact on white wine phenolic and glutathione content. The third aim was to characterize the wood chemical space related to white wine resistance against oxidation by Fourier transform ion cyclotron resonance ultra high-resolution mass spectrometry (FT-ICR-MS).

2. MATERIALS AND METHODS 2.1. Chemicals. All solvents were HPLC grade from SigmaAldrich (France). Glutathione, tartaric acid, iron sulfate heptahydrate (FeSO4·7H2O), hydrogen peroxide, trifluoroacetic acid, 4-pyridil-1oxide-N-tert-butylnitrone (POBN), ellagic acid, trans-caftaric acid, gentisic acid, trans-caffeic acid, trans-ferulic acid, salicylic acid, transcoumaric acid, gallic acid, protocatechuic acid, hydroxybenzoic̈ acid, hydroxytyrosol, chlorogenic acid, tyrosol, 4-chloro-1-naphtol, catechin, and epicatechin were purchased from Sigma-Aldrich (France). Proanthocyanidin B1 and proanthocyanidin B2 were purchased from Extrasynthese (Lyon, France). 2.2. Wine Trials. Wine trials were conducted in the Bordeaux (South France) and Burgundy (North France) regions in 2015 and 2016 using manually harvested Sauvignon blanc (B, E, F), Semillion (D), and Chardonnay (A, C) grapes. Grapes were pneumatically pressed, and juices were cold settled. Clarified juices were afterward transferred to oak barrels for alcoholic fermentation with selected dry yeasts at cellar temperature (17−22 °C) (Figure 1). Malolactic fermentation was conducted only in chardonnay trials (A and C). Wines were sulphited (30 mg L−1 free SO2) at the end of the alcoholic fermentation (AF) for B, E, F, and D wines and at the end of malolactic fermentation (MLF) for A and C. Wines pH, titratable acidity, volatile acidity, alcohol by volume, and malic and lactic contents were determined using an FTIR analyzer at different stages of the vinification process (Table S1). All trials were conducted in triplicates, and sampling for GSH, total ellagitannins, and phenolic compounds analysis was performed every two months during eight months of barrel aging. EPR analysis was performed after AF (T0) and at the end of barrel aging (Tf). 2.3. Oak Barrel Characteristics. 2.3.1. Barrel Classification According to Oak Wood Tannin Potential. The commercial barrels (225 L for Bordeaux wines and 228 L for wines from Burgundy) were made from woods coming from French forests (90%) and neighboring countries (10%, mainly Germany) that were naturally seasoned for 30 months. Wood classification according to its B

DOI: 10.1021/acs.jafc.9b00517 J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Article


Figure 2. Wine classification (A, B, C) during aging in barrels with different tannin potential (LTP and MTP/Tannin Potential corresponding to different ellagitannin content in untoasted wood: low or LTP from 2000 to 4000, medium or MTP from 4001 to 6000 of ellagic acid equivalent/g of dry wood) according to the maximal intensity (Imax POBN‑1‑HER) and kinetic curve gradient rPOBN‑1‑HER of formation of radicals POBN-1-HER measured by EPR after chemical initiation by Fenton reaction at room temperature. Analysis carried out just after alcoholic fermentation (T0: open circle) and end of oak barrel aging (Tf: solid circle). Control: one year old barrel. The samples discription is defined in Figure 1. *Significant differences between Tannin Potential levels for each time point (Tukey’s test p < 0.05). epicatechin; 0.01−10 mg L−1 for proanthocyanidin B1; and 0.01−10 mg L−1 for proanthocyanidin B2. The regression equations were “y = 18387x” for trans-caftaric acid, “y = 11261x” for gentisic acid, “y = 41750x” for trans-caffeic acid, “y = 46696x” for trans-coutaric acid, “y = 54214x” for 2-S-glutathionylcaftaric acid (GRP), “y = 52859x” for trans-ferulic acid, “y = 38733x” for salicylic acid, “y = 55228x” for trans-coumaric acid, “y = 20759x” for gallic acid, “y = 31275x” for protocatechuic acid, “y = 43762x” for hydroxybenzoic̈ acid, “y = 626890x” for hydroxytyrosol, “y = 1628813x” for tyrosol, “y = 1222412x” for catechin, “y = 1261079x” for epicatechin, “y = 561579x” for proanthocyanidin B1, and “y = 574124x” for proanthocyanidin B2. Trans-coutaric acid and GRP concentrations were expressed in trans-caftaric acid equivalents. 2.7. UPLC-MS-QToF Analysis of Reduced Glutathione. Glutathione was quantified in wine samples after derivatization with 4-methyl-1,2-benzoquinone according to the protocol proposed by Nikolantonaki et al. (2014). A 23 mM solution of 4-methyl-1,2benzoquinone in acetronitrile was prepared after activated periodate resin induced the oxidation of 4-methyl-catechol. A 50 μL portion of 4-methyl-1,2-benzoquinone was added in 1 mL each of the wine samples, and the mixtures were stirred for 1 min at room temperature. At 10 min, sulfur dioxide was added to quench the reaction. Samples were analyzed within 24 h in an UPLC-MS-QToF system. The separation was done with a ultrahigh-pressure liquid chromatograph (Dionex Ultimate 3000, ThermoFischer) coupled to a MaXis plus MQ ESI-Q-TOF mass spectrometer (Bruker, Bremen, Germany). Derivatized glutathione was separated in reversed phase liquid chromatography (RP-LC) by injecting 5 μL in an Acquity UPLC BEH C18 1.7 μm column 100 × 2.1 mm (Waters, Guyancourt, France). Elution was performed at 40 °C by (A) acidified water (0.1% (v/v) of formic acid) and (B) acetonitrile (0.1% (v/v) of formic acid) with the following gradient: 0−1.10 min 5% (v/v) of B and 95% (v/v) of B at 6.40 min. The flow rate was set at 400 μL min−1. Solvent and analytes were ionized with an electrospray (Nebulizer pressure = 2 bar and nitrogen dry gas flow = 10 L min−1). Ion transfer was done with an end plate offset at 500 V and transfer capillary voltage at 4.500 V. A

of ellagic acid released by ellagitannins during acid hydrolysis was determined by UPLC-DAD analysis using a Waters UPLC H-Class system (Waters, Milford, MA, U.S.A.) formed by an autosampler, a quaternary pump, a column manager, and a diode array detector controlled by Empower 2. Under optimized conditions, the column oven was thermostated at 35 °C, and the sample system was thermostated at 6 °C. A 5 μL portion of the sample was loaded to a reverse phase Raptor ARC C18 (150 × 2.1 mm, 2.7 μm, Restek, Lisses, France) column and eluted by a gradient of (A) H2O/ methanol/trifluoroacetic acid (95:5:0.28%) and (B) methanol (100%). The elution gradient was 15−48% of B in 5 min, 48% of B for 5 min, 48 to 60% in 2 min, and 60 to 100% in 4 min. The diode array detector was set at 252 nm for ellagic acid and at 295 nm for the internal standard. All the analyses were performed in triplicate, and the peak area was measured. Quantification was achieved using regression curves in the following ranges: 0.1−50 mg L−1 for the ellagic acid (y = 39320x + 2074.9). 2.6. UPLC-DAD Analysis of Phenolic Compounds. The concentrations of the wine phenolics, trans-caftaric acid, gentisic acid, trans-caffeic acid, trans-coutaric acid, 2-S-glutathionylcaftaric acid (GRP), trans-ferulic acid, salicylic acid, trans-coumaric acid, gallic acid, protocatechuic acid, hydroxybenzoic̈ acid, hydroxytyrosol, tyrosol, catechin, epicatechin, proanthocyanidin B1, and proanthocyanidin B2, were determined according to the analytical method described by Nikolantonaki et al. (2019). All the analyses were performed in triplicate, and the peak area was measured. Quantification was achieved using regression curves in the following ranges: 0.05−50 mg L−1 for the trans-caftaric acid; 0.1−150 mg L−1 for gentisic acid; 0.01−50 mg L−1 for the trans-caffeic acid; 0.1−50 mg L−1 for the trans-coutaric acid; 0.05−150 mg L−1 for the 2-Sglutathionylcaftaric acid (GRP); 0.05−50 mg L−1 for the trans-ferulic acid; 0.01−50 mg L−1 for the salicylic acid; 0.05−50 mg L−1 for the trans-coumaric acid; 0.1−50 mg L−1 for the gallic acid; 0.1−50 mg L −1 for the protocatechuic acid; 0.1−50 mg L −1 for the hydroxybenzoic̈ acid; 0.1−150 mg L−1 for hydroxytyrosol; 0.1−150 mg L−1 for tyrosol; 0.01−50 mg L−1 for catechin; 0.01−50 mg L−1 for C


Article


Figure 3. Wine classification (E, F, G) during aging in barrels with different toasting levels (blanche (Bl) = 150 °C/1 h; ivoire (Iv) = 160−170 °C/ 1 h 30) according to the maximal intensity (Imax POBN‑1‑HER) and kinetic curve gradient rPOBN‑1‑HER of formation of radicals POBN-1-HER measured by EPR after chemical initiation by Fenton reaction at room temperature. Analysis carried out just after alcoholic fermentation (T0: empty symbol) and end of oak barrel aging (Tf: full symbol). Control: one year old barrel. The samples discription is defined in Figure 1. The absence of asterisk (*) indicates no significant differences between toasting levels for each time point (Tukey’s test p < 0.05). divert valve was used to inject four times diluted ESI-L low concentration tuning mix (Agilent, Les Ulis, France) at the beginning of each run, allowing a recalibration of each spectrum. The mass spectrometer was calibrated with undiluted tuning mix before batch analysis in enhanced quadratic mode, with less than 0.5 ppm errors after calibration. Analyses were done in mass ranges: between 100 to 1.500 m/z in positive ionization mode. Quality control was used to guarantee the stability of the LC-MS system before each sample run. Quantification of derivatized glutathione was performed using [(Q + GSH −2H)+H]+, C17H23N3O8S = 430.1279 ion. The calibration curve was determined using a chardonnay wine spiked with diluted solutions containing final concentrations ranging from 0.01 to 40 mg L−1. 2.8. FT-ICR-MS Analysis and Data Processing. FT-ICR-MS sample preparation, analysis, and data processing were carried out as described by Roullier-Gall et al. (2017). 2.9. Statistical Analysis. Statistical data analysis was performed using the analysis of variance (ANOVA) of Statistica V.7 software (Statsoft Inc., Tulsa, OK). Tukey’s HSD (p < 0.05) test was used as a comparison test when samples were significantly different after ANOVA (p < 0.05) for chemical analysis. Hierarchical Cluster Analysis (HCA) were obtained with Perseus 1.5.1.6 (http://www. perseus-framework.org, Max Planck Institute of Biochemistry, Germany).15 The clustering was performed using a Pearson’s correlation.

value of the maximal intensity (Imax POBN‑1‑HER) were chosen as representative values to distinguish the different wines. On the basis of our analytic approach, wines with low Imax POBN‑1‑HER and rPOBN‑1‑HER values are considered to be more stable against oxidation.12−14 Figure 2 illustrates the impact of oak wood tannin potential (TP) on wine antioxidant stability during barrel aging as probed by ESR spectroscopy. The wine distribution according to its Imax and rPOBN‑HER values enabled us to demonstrate a relationship between barrel TP and wine oxidative stability. Independently of the wine matrix, the MTP modalities showed better stability at the end of barrel aging than the LTP modalities after 8 months of barrel aging in new barrels, which had similar characteristics to the control modalities (wines elaborated in one year old barrels). This phenomenom, validated for three different wine matrices (A, B, and C), highlights the fact that the amount of ellagitanins extracted represents an important component of the total wine antioxidant metabolome, accounting for a significant gain on wine resistance against oxidation. Figure 3 illustrates the impact of barrel toasting on wine antioxidant stability during barrel aging as probed by ESR spectroscopy. Within each wine matrix (D, E, and F), a global increase was observed in its antioxidant stability after several months of barrel aging. This set of experiments investigated wines aged in LTP barrels, associated with two types of light toasts: blanche (150 °C/1 h) and ivoire (160−170 °C/1h30). At the end of aging, no significant difference was found between these two toasting profiles. 3.2. Impact of Oak Wood Tannin Potential and Toasting Levels on Wine Ellagitannin Content. The evolution kinetics of the overall hydrolyzable tannin content for the different modalities are shown in Figure 4. The measurements were carried out starting from the end of the AF

3. RESULTS AND DISCUSSION 3.1. Impact of Oak Wood Tannin Potential and Toasting Levels on Wine Resistance against Oxidation. EPR was used to discern resistance to oxidation of the white wines aged in barrels with different tannin potentials and toasting levels. The method for analyzing wines oxidative stability is based on assessment of the kinetics of 1hydroxyethyl radical formation, after free radical initiation in the wine, which is then captured by a POBN paramagnetic probe.12 The kinetic curve’s gradient (rPOBN‑1‑HER) and the D


Article


Figure 4. (A) Total ellagitannin concentration (mg eq ellagic acid/L) of white wines (A to C) aged in barrels with different tannin potential (low (LTP); medium (MTP)) and (B) different toasting levels (150 °C (blanche); 160 °C (ivoire)) for 8 months. The samples discription is defined in Figure 1. Values with different letters are significantly different (Tukey’s test p < 0.05). Bars represent standard deviation (n = 3).

MTP modalities, respectively; a rapid increase was then observed after 4 months (9.07 and 11.4 mg L−1 ellagic acid equivalents for LTP and MTP, respectively) and after 6 months of aging (10.59 and 13.85 mg L−1 ellagic acid equivalents for LTP and MTP, respectively). These results show that the time necessary to reach the maximum ellagitannin concentration in the wine is related to the wine matrix and not the oak TP. For wine B, ellagitannin extraction peaked at 6 months of aging (14.28 mg L−1 ellagic acid equivalents for LTP, 14.30 mg L−1 ellagic acid equivalents for MTP), which was followed by a decrease after 8 months; whereas ellagitannin extraction constantly increased through-

up until 8 months of aging. A relationship was demonstrated, independent of the matrix and the oak’s tannin content, between the TP of the oak wood and the concentration of total ellagitannin in the wine (Figure 4A). The higher the TP of the barrel, the greater the ellagitannin content in the wine. As previously described by Watrelot et al.16 about red wine, a very sharp increase is noted in the first 3 months, which implies significantly higher extraction kinetics at the beginning of barrel aging. It is also important to note that very little ellagitannins are extracted during AF. After 2 months of barrel aging, the average ellagitannin concentration in the wines was 4.2 and 4.6 mg L−1 ellagic acid equivalents for the LTP and E


Article


Table 1. Glutathione (mg/L) Concentrations of White Wines (A to F) Aged in Barrels with Different Ellagitannin Content (Low (LTP) and Medium (MTP) and Toasting Levels (150 °C (Blanche) and 160 °C (Ivoire)) for 8 Months GSH Aging time (months) A_Control A_LTP A_MTP B_Control B_LTP B_MTP C_Control C_LTP C_MTP D_Control D_Iv D_Bl E_Control E_Iv E_Bl F_Control F_Iv F_Bl

0 3.88 4.48 4.16 5.36 5.50 5.21 11.45 11.02 11.67 3.77 3.35 3.40 3.81 2.93 2.85 0.46 0.50 0.48

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

2 0.10xa 0.12f 0.11f 0.14a 0.15a 0.14a 0.30a 0.29a 0.31a 0.10a 0.19a 0.19a 0.10a 0.07f 0.07f 0.01a 0.01a 0.01a

3.09 4.33 4.54 5.17 5.28 5.01 8.09 6.87 7.09 3.11 2.89 2.71 2.71 2.52 2.68 tr tr tr

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

4 0.09b 0.13g 0.14g 0.16b 0.14b 0.15b 0.25b 0.21b 0.21b 0.19b 0.08b 0.08b 0.04b 0.07b 0.05b

2.54 4.25 4.42 4.78 4.52 4.78 6.34 5.21 4.68 2.21 2.42 2.01 2.11 2.01 2.22 tr tr tr

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

6 0.06c 0.11h 0.10h 0.12c 0.12c 0.12c 0.17c 0.14c 0.12c 0.05c 0.06c 0.05c 0.15c 0.14c 0.16c

2.11 4.46 3.83 4.21 4.17 4.08 4.19 4.02 3.90 1.42 1.61 1.57 1.81 1.52 1.58 tr tr tr

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

8 0.06d 0.13i 0.11i 0.12d 0.12d 0.12d 0.12d 0.12d 0.11d 0.04d 0.04d 0.04d 0.05d 0.02d 0.06d

1.40 1.98 1.98 3.36 3.24 3.42 3.11 2.24 1.61 1.01 0.83 0.82 1.44 1.46 1.32 tr tr tr

± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.04e 0.05j 0.05j 0.10e 0.09e 0.10e 0.09e 0.06e 0.04e 0.03d 0.02d 0.01d 0.06e 0.03e 0.04e

x

Results are the mean of technical triplicates; values with different letters are significantly different (Tukey’s test, p < 0.05).

with a light toast has no influence on the glutathione concentration after AF. Afterward, during barrel aging, the GSH concentration decreases progressively, depending on the TP and the type of wine. It should be noted that for wine A, the new barrel had a significant positive effect on GSH preservation, whereas for wines B and C, this was not the case. 3.4. Impact of Oak Wood Tannin Potential and Toasting Levels on Wine Phenolic Content. Phenolic compounds that originated from grapes (flavonoids and nonflavonoids) and alcoholic fermentation products (i.e., tyrosol) both have good antioxidant properties, while the catechol group (1,2-dihydroxybenzene) reacts very readily with oxidants in the form of free radical reactive oxygen species to form a very stable radical anion, the semiquinone radical. The compounds with catechol or 1,4-dihydroquinone functionality are especially easy to oxidize because the resulting phenoxyl radical can be stabilized by the adjacent oxygen anion. It is stable enough that it does not attract hydrogens from other substances and will persist long enough to react with another semiquinone radical, resulting in a disproportionation reaction which yields a quinone and a phenol, quenching two radicals in total. Wine is rich in substances with the catechol groups, and these compounds impart an antioxidant activity to wine that acts as a natural preservative.24 Table S2 shows the phenolic content of all wine samples and is defined as total polyphenols for specific chemical families: phenolic acids (trans-caftaric acid, gentisic acid, trans-caffeic acid, trans-coutaric acid, transferulic acid, salicylic acid, trans-coumaric acid, gallic acid, protocatechuic acid, hydroxybenzoic̈ acid, and cinnamic acids), flavan-3-ols (catechin, epicatechin, proanthocyanidin B1, proanthocyanidin B2), glutathionylcaftaric acid (GRP), and tyrosol (hydroxytyrosol and tyrosol). There was a wide range of concentrations for the total phenol content in wines related to the vintage, variety, and grape origin.25,26 Thus, at T0, the ranges of total phenolic contents in wine were from 29.74 to 49.23 mg L−1 for the phenolics acids, 0.61 to 2.81 for the flavan-3-ols, 5.24 to 20.81 mg L−1 for the GRP, and 8.00 to 27.45 mg L−1 for the tyrosol concentrations relevant to white

out aging for wines A and C. Various physicochemical parameters, such as pH, alcohol content (ABV), and temperature, can modify the ellagitannin extraction rate in wine.17,18 In our experimental conditions, wines A, B, and C showed similar pHs and ABVs and were aged in cellars with mild temperatures (15−18 °C). Thus, we can hypothesize that the greater rate of ellagitannin consumption in wine B after 6 months of aging could be linked to its higher oxidation resistance, as measured by the EPR method. Trials combining the LTP with the light toasts (blanche and ivoire) did not demonstrate an impact on the ellagitannin extraction kinetics (Figure 4B). However, the wine matrix effect on the extraction kinetics was confirmed for this series of trials for three distinct wines (D, E, and F). 3.3. Impact of Oak Wood Tannin Potential and Toasting Levels on Wine Glutathione Content. The evolution in glutathione content, a known antioxidant,19,20 according to oak tannin potential, was also monitored throughout barrel aging (Table 1). GSH levels are dependent on wine fermentative conditions,21 and for that reason, it should be mentioned that no significant differences in the fermentation kinetics during AF were observed. The average levels of glutathione in all wines varied significantly between samples. The higher GSH levels were recorded at T0 in C wines (11.38 mg L−1), and the lowest levels were recorded in F wines (0.48 mg L−1). These concentrations were similar and in some cases slightly higher than those reported in literature for white wine made from Sauvignon blanc, Semillon, and Chardonnay grapes.22,23 In all wines, the GSH concentration decreased gradually during barrel aging and an average loss of more than 50% on GSH level was observed at 8 months. Generally, the magnitude of the influence that the tannin potential had on GSH levels was not significantly more important than that of the toasting level during barrel aging. At the end of AF, GSH concentration was identical for the different modalities of each type of wine. Identical results were obtained for wines aged in LTP barrels coupled with blanche and ivoire toasts. We can thus deduce that a LTP combined F


Article


Figure 5. (A) Hierarchical cluster analysis of the normalized data of all m/z ions and their intensities observed in wines aged in barrels with different ellagitannin content (low (LTP) and medium (MTP)) at 8 months. (B and C) Van Krevelen diagrams depict the most representative characteristic molecular formulas related to tannin potential on wines A and C compositions at 8 months of barrel aging. Van Krevelen plots were colored according to molecular classes, i.e., CHO (gray), CHON (orange), CHOS (green), CHONS (red). The bubble area depicts the relative mass peak intensity within the respective subcluster. (D) Histograms indicating the number of molecular formulas presented in the Van Krevelen diagrams.

wines. The levels of phenolics showed constancy during barrel aging independent of the wine matrix, the tannin potential, and the toasting effect. 3.5. Impact of Oak Wood Tannin Potential on Wine Metabolome. At this stage, the targeted analytical approach seems limited to provide a molecular explanation of the variability of wine oxidative stability during barrel aging. In order to evaluate the impact of barrel tannin potential on the chemical composition of wines with distinct resistances against oxidation (A and C), we performed a non targeted metabolomics approach with ultrahigh resolution mass spectrometry (FT-ICR-MS) (Figures S1−S3). Multivariate statistics were finally applied to analyze complex sets to unveil many independent variables (mass signals) that are highly correlated to each other. According to HCA (Figure 5A), two distinct groups could be discriminated. The first group encloses wines A and C aged in barrels with different ellagitannin contents (low (LTP) and medium (MTP)) at 8 months of aging, while the second represents the controls. Up to 230 elemental formulas were correlated with oak wood TP and particularly associated with metabolites below 650 amu and with the respective prevailing domination of CHONS (45%), CHO (27%), CHON (16%), and CHOS (12%) containing compounds (Figure 5B,C). The TP wood metabolome was correlated with compounds in the van Krevelen area corresponding to essentially the chemical space of amino acids, peptides, and phenolic compounds. Among the 232 TP distinctive molecular markers (Figure 5D), 46 could be identified putatively with the help of known databases (KEGG, HMDB, LipidMap, and our in house wine and plant database) (Table S3). The accuracy of the annotated masses varied between 0.01 and 0.6 ppm. The listed compounds included prominent low-molecular-weight ellagitannins, amino com-

pounds, peptides, and sulfonated compounds already identified in wine or other food matrices. As an example, the ion m/z 337.0200 was identified as hexahydroxydiphenic acid that is a component of some ellagitannins.28 Sulfur containing compounds including interesting sulfonated forms of ferulic acid (m/z = 273.0007 C11H10O7S) and γ-glutamyl-S-(1-propenyl) cysteine (m/z = 305.0810 C11H11O6N2S), and an important number of N−S containing small peptides that could confer antioxidant properties to wines during barrel aging were also putatively13,14,27 identified.29,30 This unsupervised analysis revealed a strong impact of barrel tannin potential to wine chemical composition during aging, correlated to wine resistance against oxidation and associated with an involved chemical diversity that encompasses the polyphenolic chemical space of wood. Our studies clearly demonstrated a positive correlation between wine oxidative stability and barrel aging, independent of the grape variety. The oak tannin potential has a significant effect on wine oxidative stability, with strong interactions with the wine matrix. The specific analyses demonstrate a high level of homogeneity in the tested barrels and a concordance between the total ellagitannin content in the wines and the barrel classification. Untargeted ultrahigh resolution mass spectrometry analyses provided evidence that barrel aging in oak wood, with a high tannin potential, preserves the native antioxidant compounds in wines. We will continue our research in order to better understand the wood compounds that are conducive to better oxidative stability in wines.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jafc.9b00517. G


Article


■

wood used in wine ageing and the correlation with polyphenol composition. Food Chem. 2011, 128, 997−1002. (10) Quideau, S.; Feldman, K. S. Ellagitannin Chemistry. Chem. Rev. 1996, 96, 475−504. (11) Vivas, N.; Laguerre, M.; Pianet de Boissel, I.; Vivas de Gaulejac, N.; Nonier, M.-F. Conformational Interpretation of Vescalagin and Castalagin Physicochemical Properties. J. Agric. Food Chem. 2004, 52, 2073−2078. (12) Nikolantonaki, M.; Coelho, C.; Noret, L.; Zerbib, M.; Vileno, B.; Champion, D.; Gougeon, R. D. Measurement of white wines resistance against oxidation by Electron Paramagnetic Resonance spectroscopy. Food Chem. 2019, 270, 156−161. (13) Nikolantonaki, M.; Magiatis, P.; Waterhouse, A. L. Measuring protection of aromatic wine thiols from oxidation by competitive reactions vs wine preservatives with ortho-quinones. Food Chem. 2014, 163, 61−67. (14) Roullier-Gall, C.; Hemmler, D.; Gonsior, M.; Li, Y.; Nikolantonaki, M.; Aron, A.; Coelho, C.; Gougeon, R. D.; SchmittKopplin, P. Sulfites and the wine metabolome. Food Chem. 2017, 237, 106−113. (15) Tyanova, S.; Temu, T.; Sinitcyn, P.; Carlson, A.; Hein, M. Y.; Geiger, T.; Mann, M.; Cox, J. The Perseus computational platform for comprehensive analysis of (prote)omics data. Nat. Methods 2016, 13, 731. (16) Watrelot, A. A.; Badet-Murat, M.-L.; Waterhouse, A. L. Oak barrel tannin and toasting temperature: Effects on red wine condensed tannin chemistry. LWT 2018, 91, 330−338. (17) Jordäo, A. M.; Ricardo-Da-Silva, J. M.; Laureano, O. Extraction of some ellagic tannins and ellagic acid from oak wood chips (Quercus pyrenaica L.) in model wine solutions: Effect of time, pH, temperature and alcoholic content. S. Afr. J. Enol. Vitic. 2005, 26, 83− 89. (18) Puech, J.-L.; Mertz, C.; Michon, V.; Le Guernevé, C.; Doco, T.; Hervé du Penhoat, C. Evolution of Castalagin and Vescalagin in Ethanol Solutions. Identification of New Derivatives. J. Agric. Food Chem. 1999, 47, 2060−2066. (19) Nikolantonaki, M.; Julien, P.; Coelho, C.; Roullier-Gall, C.; Ballester, J.; Schmitt-Kopplin, P.; Gougeon, R. D. Impact of Glutathione on Wines Oxidative Stability: A Combined Sensory and Metabolomic Study. Front. Chem. 2018, 6, 182 DOI: 10.3389/ fchem.2018.00182. (20) Sonni, F.; Clark, A. C.; Prenzler, P. D.; Riponi, C.; Scollary, G. R. Antioxidant Action of Glutathione and the Ascorbic Acid/ Glutathione Pair in a Model White Wine. J. Agric. Food Chem. 2011, 59, 3940−3949. (21) Kritzinger, E. C.; Bauer, F. F.; du Toit, W. J. Role of Glutathione in Winemaking: A Review. J. Agric. Food Chem. 2013, 61, 269−277. (22) Fracassetti, D.; Lawrence, N.; Tredoux, A. G. J.; Tirelli, A.; Nieuwoudt, H. H.; Du Toit, W. J. Quantification of glutathione, catechin and caffeic acid in grape juice and wine by a novel ultraperformance liquid chromatography method. Food Chem. 2011, 128, 1136−1142. (23) Fuleki, T.; Ricardo da Silva, J. M. Effects of cultivar and processing method on the contents of catechins and procyanidins in grape juice. J. Agric. Food Chem. 2003, 51, 640−646. (24) Waterhouse, A. L. Wine Phenolics. Ann. N. Y. Acad. Sci. 2002, 957, 21−36. (25) Singleton, V. L.; Trousdale, E. White wine phenolics: Varietal and processing differences as shown by HPLC. Am. J. Enol. Vitic. 1983, 34, 27−34. (26) Chamkha, M.; Cathala, B.; Cheynier, V.; Douillard, R. Phenolic composition of champagnes from Chardonnay and Pinot noir vintages. J. Agric. Food Chem. 2003, 51, 3179−3184. (27) Romanet, R.; Coelho, C.; Liu, Y.; Bahut, F.; Ballester, J.; Nikolantonaki, M.; Gougeon, D. R. The Antioxidant Potential of White Wines Relies on the Chemistry of Sulfur-Containing Compounds: An Optimized DPPH Assay. Molecules. 2019, 27, (7).

Wines basic enological parameters, concentrations of white wines phenolic compounds (mg/l) aged in barrels with different ellagitannin content, discriminant m/z ions according to oak wood tannin potential, MS spectra of molecular wood markers that decrease with the tannin potential (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Maria Nikolantonaki: 0000-0002-5576-9373 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Vicard generation 7, the Regional Council of Bourgogne − Franche Comté, the “Fonds Européen de Developpement Régional (FEDER)”. The authors thank Nadine Gublin (Domaine Jacques Prieur, Meursaut, Burgundy), Sylvain Pabion (Château de Marsannay, Marsannay, Burgundy), Yann Laudeho (Château Smith Haut Lafitte, Pessac-Léognan, Bordeaux) and all the Bordeaux wineries which participated in this research for implementing and monitoring these trials. We also thank Dr. Chloé RoullierGall for help in the acquisition of the FTICR-MS analysis.

■

REFERENCES

(1) Larrauri, J. A.; Sánchez-Moreno, C.; Rupérez, P.; Saura-Calixto, F. Free Radical Scavenging Capacity in the Aging of Selected Red Spanish Wines. J. Agric. Food Chem. 1999, 47, 1603−1606. (2) Prida, A.; Chatonnet, P. Impact of Oak-Derived Compounds on the Olfactory Perception of Barrel-Aged Wines. Am. J. Enol. Vitic. 2010, 61, 408−413. (3) Marchal, A.; Waffo-Téguo, P.; Génin, E.; Mérillon, J.-M.; Dubourdieu, D. Identification of New Natural Sweet Compounds in Wine Using Centrifugal Partition Chromatography-Gustatometry and Fourier Transform Mass Spectrometry. Anal. Chem. 2011, 83, 9629− 9637. (4) Michel, J.; Albertin, W.; Jourdes, M.; Le Floch, A.; Giordanengo, T.; Mourey, N.; Teissedre, P.-L. Variations in oxygen and ellagitannins, and organoleptic properties of red wine aged in French oak barrels classified by a near infrared system. Food Chem. 2016, 204, 381−390. (5) Michel, J.; Jourdes, M.; Silva, M. A.; Giordanengo, T.; Mourey, N.; Teissedre, P.-L. Impact of Concentration of Ellagitannins in Oak Wood on Their Levels and Organoleptic Influence in Red Wine. J. Agric. Food Chem. 2011, 59, 5677−5683. (6) Glabasnia, A.; Hofmann, T. Sensory-Directed Identification of Taste-Active Ellagitannins in American (Quercus alba L.) and European Oak Wood (Quercus robur L.) and Quantitative Analysis in Bourbon Whiskey and Oak-Matured Red Wines. J. Agric. Food Chem. 2006, 54, 3380−3390. (7) Chassaing, S.; Lefeuvre, D.; Jacquet, R.; Jourdes, M.; Ducasse, L.; Galland, S.; Grelard, A.; Saucier, C.; Teissedre, P.-L.; Dangles, O.; Quideau, S. Physicochemical Studies of New Anthocyano-Ellagitannin Hybrid Pigments: About the Origin of the Influence of Oak CGlycosidic Ellagitannins on Wine Color. Eur. J. Org. Chem. 2010, 2010, 55−63. (8) Jordão, A. M.; Correia, A. C.; DelCampo, R.; González SanJosé, M. L. Antioxidant capacity, scavenger activity, and ellagitannins content from commercial oak pieces used in winemaking. Eur. Food Res. Technol. 2012, 235, 817−825. (9) Alañoń , M. E.; Castro-Vázquez, L.; Díaz-Maroto, M. C.; Gordon, M. H.; Pérez-Coello, M. S. A study of the antioxidant capacity of oak H


Article

Journal of Agricultural and Food Chemistry (28) Vrhovsek, U.; Palchetti, A.; Reniero, F.; Guillou, C.; Masuero, D.; Mattivi, F. Concentration and Mean Degree of Polymerization of Rubus Ellagitannins Evaluated by Optimized Acid Methanolysis. J. Agric. Food Chem. 2006, 54, 4469−4475. (29) Kaneda, H.; Osawa, T.; Kawakishi, S.; Munekata, M.; Koshino, S. Contribution of carbonyl-bisulfite adducts to beer stability. J. Agric. Food Chem. 1994, 42, 2428−2432. (30) Saito, K.; Jin, D.-H.; Ogawa, T.; Muramoto, K.; Hatakeyama, E.; Yasuhara, T.; Nokihara, K. Antioxidative Properties of Tripeptide Libraries Prepared by the Combinatorial Chemistry. J. Agric. Food Chem. 2003, 51, 3668−3674.

I


Impact of Oak Wood Barrel Tannin Potential and Toasting on White

Recommend Documents