Effect of enzyme additions on the oligosaccharide composition of Monastrell red wines from four different wine-growing origins in Spain

Food Chemistry 156 (2014) 151–159

Contents lists available at ScienceDirect

Food Chemistry journal homepage: www.elsevier.com/locate/foodchem

Effect of enzyme additions on the oligosaccharide composition of Monastrell red wines from four different wine-growing origins in Spain Rafael Apolinar-Valiente a,⇑, Pascale Williams b, Gérard Mazerolles b, Inmaculada Romero-Cascales a, Encarna Gómez-Plaza a, José María López-Roca a, José María Ros-García a, Thierry Doco b a b

Departamento de Tecnología de Alimentos, Nutrición y Bromatología, Facultad de Veterinaria, Universidad de Murcia, 30100 Murcia, Spain INRA, Joint Research Unit 1083 Sciences for Enology, 2 Place Viala, 24 F-34060 Montpellier, France

a r t i c l e

i n f o

Article history: Received 10 October 2013 Received in revised form 12 December 2013 Accepted 23 January 2014 Available online 6 February 2014 Keywords: Wine Oligosaccharides Terroir Pectic enzymes Monastrell

a b s t r a c t The release of oligosaccharides during winemaking depends on the grape skin cell wall degradation, which can be facilitated by the use of enzymes. Oligosaccharide quantities and composition in wine could be influenced by the ‘‘terroir’’ effect. Monastrell wine was elaborated from grapes from four different ‘‘terroirs’’ (Cañada Judío, Albatana, Chaparral-Bullas and Montealegre). Monastrell wines were also treated with b-galactosidase enzyme addition and commercial enzyme addition. The results showed significant differences in the Monastrell wine oligosaccharide fractions, according to the geographical origin of grapes. A higher quantity of oligosaccharides was found for three out of four terroirs studied when commercial enzymes were added. The use of commercial enzyme modified the Arabinose/Galactose and the Rhamnose/Galacturonic acid ratios in Cañada Judío and Albatana terroirs wines, and it modified the (Arabinose + Galactose)/Rhamnose ratio in Cañada Judío, Albatana and Chaparral-Bullas terroirs wines. Therefore, the ‘‘terroir’’ impacts the effect of commercial enzyme treatment on wine oligosaccharide composition. Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction Polyphenols, polysaccharides and proteins are the main macromolecules of wines. They have been thoroughly studied, due to their importance for wine technological and sensory properties. In contrast, oligosaccharides have only recently been shown to occur in wine and much less studied. These natural molecules can be found in important medicinal, food, agricultural applications, play a role in plant defence responses (Darvill & Albersheim, 1984; Elleuch et al., 2011; Qiang, YongLie, & QianBing, 2009) and behave as dietary fibres and prebiotics (Gibson & Roberfroid, 1995; Hopkins & Macfarlane, 2003; Macfarlane, Steed, & Macfarlane, 2008). Concerning their significance for wine quality, some oligosaccharides have physicochemical properties such as chelation of cations (Cescutti & Rizzo, 2001). Oligosaccharide fractions from

⇑ Corresponding author. Tel.: +34 868 887662; fax: +34 868 884147. E-mail addresses: [email protected] (R. Apolinar-Valiente), williams@ supagro.inra.fr (P. Williams), [email protected] (G. Mazerolles), [email protected] (I. Romero-Cascales), [email protected] (E. Gómez-Plaza), jmlroca @um.es (J.M. López-Roca), [email protected] (J.M. Ros-García), thierry.doco@supagro. inra.fr (T. Doco). http://dx.doi.org/10.1016/j.foodchem.2014.01.093 0308-8146/Ó 2014 Elsevier Ltd. All rights reserved.

red wines were first isolated and characterised by Ducasse, Williams, Meudec, Cheynier, and Doco (2010). The Carignan and Merlot wines investigated in this study contained rather large concentrations (approximately 300 mg/L) of oligosaccharides structurally related to plant cell wall polysaccharides (Ducasse et al., 2010). The oligosaccharide structures detected by MS spectrometry include short chains of galacturonic acid with a degree of polymerization between 2 and 6, and as a short chain of rhamnogalacturonan-oligomers constituted by the repeats of rhamnose and galacturonic acid, arising from smooth regions or hairy regions of pectin, respectively, but also 4-OMe-oligo-glucuronoxylan oligosaccharides produced from hemicellulose (Ducasse et al., 2010). Recently, similar structures have been isolated from other wines by Bordiga et al (2012). These authors have isolated and characterised forty-five complex free oligosaccharides in red and white wines, Grignolino and Chardonnay, respectively. The concentrations were around 100 mg/L in both wines, and the oligosaccharides corresponded to hexose-oligosaccharides, xyloglucans, and arabinogalactans, that may be the natural byproduct of the degradation of cell wall polysaccharides. In red wine winemaking the use of enzymes is now expanding into various more targeted applications beyond the classic press

152

R. Apolinar-Valiente et al. / Food Chemistry 156 (2014) 151–159

yield and clarification purposes. The main use of commercial pectinolytic enzyme preparations is to enhance the extraction of anthocyanin pigments and other desirable phenolics into the juice or red wine during maceration. The permeability of the grape skin cell walls to polyphenols can be increased by enzymes, that can help the partial hydrolysis of cell wall polysaccharides. Enzyme treatments have been shown to modify the wine polysaccharide composition (Ayestaran, Guadalupe, & León, 2004; Doco, Williams, & Cheynier, 2007; Guadalupe, Palacios, & Ayestaran, 2007). They produce an increase of RG-II and a decrease of PRAGs in red wine, along with a particular modification of AGPs, with loss of their terminal arabinose residues (Doco et al., 2007). Studies on the effect of enzymes on wine oligosaccharides are scarce (Bordiga et al., 2012; Ducasse et al., 2010; Ducasse et al., 2011). Ducasse et al. (2011) observed differences in the total oligosaccharide concentration between Merlot wines treated with enzymes; in most cases, enzyme treated wines contained lower amounts of oligosaccharides than the control. Qualitative differences were also found in oligosaccharide composition between enzyme-treated and control Merlot wines. A terroir can be defined as a grouping of homogeneous environmental units, or natural terroir units, based on the typicality of the products obtained (Laville, 1993). This word is particularly associated with the local production of wine (Barham, 2003), and implies a link between the wine and the area of this production. For example, the ‘‘terroir’’ characteristics could influence on phenol profiles in wines from different wine-growing regions, as several authors detected (Li, Pan, Jin, Mu, & Duan, 2011; Rastija, Srecˇnik, & Medic´-Šaric´, 2009). The soil and climate are the main elements of the French notion of ‘‘terroir’’, but the concept also includes human factors that may affect production in different ways (Morlat, 2005). Monastrell, also known internationally as the French name of Mourvedre, is the main wine grape cultivar in Southern Spain but, there is no information about the oligosaccharide composition of Monastrell wines. An influence of terroir on wine polysaccharide composition has been demonstrated (Apolinar-Valiente et al., 2013). This factor could also influence in oligosaccharide quantities and composition in wine depending on the cell wall ‘‘degrability’’, although it has not been studied. Moreover, the interaction between this effect and that of enzyme treatments which could facilitate the grape skin cell wall degradation has not been investigated either. The aim of this work was to study the effect of different winemaking techniques (classical winemaking, addition of b-galactosidase and commercial enzyme, separately) on the oligosaccharide fractions in Monastrell wines from four different ‘‘terroirs’’ (Cañada Judío, Albatana, Chaparral-Bullas and Montealegre).

2. Materials and methods 2.1. Samples Grapes from Vitis vinifera cv. Monastrell, cultivated near Murcia (S. E. Spain), were harvested at commercial maturity over vintage 2008 from four different origins (Cañada Judío, Albatana, Chaparral-Bullas and Montealegre). The geographical information about vineyard plots is: Cañada Judío (1°210 37.0200 O Longitude; 38°330 15.8400 N Latitude; 450 m. Altitude); Albatana (1°270 49.7800 O Longitude; 38°320 28.5600 N Latitude; 693 m. Altitude); Chaparral-Bullas (1°400 59.0600 ’’O Longitude; 38°020 38.2400 N Latitude; 432 m. Altitude); and Montealegre (1°170 42.8200 O Longitude; 38°460 39.8500 N Latitude; 771 m. Altitude). Cañada Judío terroir plot is close to Jumilla village, and is composed by dolomites, loams, limestone and sandstorm. Albatana vineyard is closer to Albatana village, and its terroir is composed

by gravel, conglomerates, sand and slime. Evaporites, vulcanites, sandstones, clay and limestones can be found in Chaparral-Bullas terroir, near Cehegín village. Montealegre vineyard is closer to Montealegre del Castillo village, and sand, clay, gravel, mud and gypsum form its terroir. Climate information of different terroir plots was obtained in weather stations between September 2007 and October 2008. Supplementary Table 1 shows climatic parameters: monthly mean temperature, monthly pluviometry, monthly maximum and minimum mean temperatures, monthly mean relative humidity, monthly maximum and minimum relative humidity, monthly mean wind speed and monthly maximum wind speed. Used instrumental equipment was: HMP45AC thermohigrometer (Vaisala, Helsinki, Finland), 05103-5 wind anemometer (Young Company, Michigan, USA), and different pluviometer models: PCP-214 (Geónica, Madrid, Spain), 4.4031.30.006 (Thies-CLima, Göttingen, Germany) and ARG-100 (Campbell Scientific Ltd., Loughborough, UK. The maturity control of grapes from four studied terroirs (Cañada Judío, Albatana, Chaparral-Bullas and Montealegre) was carried out in triplicate. Berry sampling was done weekly from veraison to harvest, on 50 vines per treatment. Groups of five to six berries from different parts of the cluster and from different clusters on the same vine were sampled randomly. Berry samples (ca. 300 g), collected from all vines of the same treatment, were placed in plastic bags and stored in ice during the transport from the field to the laboratory, where enological analysis were determined. Grape analysis involved the traditional flesh measurements (°Brix, pH and total acidity) and total anthocyanins content. Total soluble solids (°Brix) were measured using a digital refractometer (Atago RX-5000; Atago Co., Ltd, Tokyo, Japan). Titratable acidity and pH were measured using an automatic titrator (Metrohm, Herisau, Switzerland) with 0.1 N NaOH. The anthocyanin content of the solution (total anthocyanins) was chemically assayed by measuring the absorbance of the samples at 520 nm at pH 3.6. The physical–chemical characteristics of the grape berry samples from Cañada Judío, Albatana, Chaparral-Bullas and Montealegre were (respectively) as follows: weight of a hundred grape berries: 191.5, 190.7, 198.4 and 197.5 g; °Brix: 24.8, 24.8, 25.2 and 20.3; pH: 3.8, 3.8, 4.1 and 3.5; total anthocyanins content: 1161, 1245, 799 and 943 mg/L; % skin: 17, 18, 14 and 14.

2.2. Preparation of samples 2.2.1. Control trials Three 90 kg lots of Monastrell grapes from four different terroirs (Cañada Judío, Albatana, Chaparral-Bullas and Montealegre) were destemmed and crushed, using a crusher/destemmer unit (Gamma 30, Zambelli Enotech, Italy) and distributed into 100 L stainless steel tanks to yield triplicate control lots named JUCO, ALCO, BUCO and MTCO. At the same time, sodium metabisulfite (8 g/100 kg grape) was added. This basic winemaking process was followed in all the wines.

2.2.2. Commercial enzyme addition trials The same process as in the control was followed in triplicate except that a commercial enzyme was added to the tanks (5 g/100 kg) and the resulting wines were named JUCE, ALCE, BUCE and MTCE. The company (Agrovin Company, Alcázar de San Juan, Spain) which produces the commercial enzyme (Enozym Vintage) provided the following information on the enzyme: polygalacturonase activity, 546.6 IU/g; pectinesterase activity, 7.3 IU/g; pectin lyase activity, 2.8 IU/g; and b-glucanase activity, 179.6 IU/g.

R. Apolinar-Valiente et al. / Food Chemistry 156 (2014) 151–159

2.2.3. Galactosidase enzyme addition trials a and b-galactosidase (Agrovin Company, Alcázar de San Juan, Spain) was added in the same way into tanks (1 g/100 kg) in triplicate from each terroir, and they were named (JUGE, ALGE, BUGE and MTGE). 2.2.4. Fermentation All fermentations were started by inoculating commercial dry yeast (Laffort, Servian, France) at 10 g/hL and were carried out in 100 L stainless steel tanks equipped with temperature control (25 °C) enabling to regulate fermentation kinetics. During the fermentative pomace contact period (ten days in all vinifications), the cap was punched down twice a day and the temperature and must density were recorded. Each lot was fermented to completion, and when alcoholic fermentation was finished (controlled by sugar analysis), the musts were pressed at 1.5 bars in a 75 L tank membrane press. Free-run juice and press wines of each trial were combined and transferred to 50 L tanks. One month later, the wines were racked. When spontaneous malolactic fermentation was finished (one month later in all samples), the wines were racked again and supplied with 25 mg/L sulphur dioxide. The wines were not clarified or filtered, but cold stabilized (3 °C) for 1 month, bottled and stored in the experimental wine cellar at 18° C until analysis. 2.3. Enological analysis Concentration of ethanol, pH values, total and volatile acidities, and chromatic characteristics of wines were determined according to the official methods of the European Union (Analyse des moûts et des vins, 1990). 2.4. Isolation of oligosaccharide fractions The oligosaccharide fractions were isolated as previously described.8 The wines (5 ml), were partially depigmented by decolourization onto a column of MN Polyamide SC6 (5  1 cm) previously equilibrated with 1 M NaCl. Wine oligosaccharides not retained on the polyamide column were eluted by 2 bed volumes of 1 M NaCl (Brillouet, Moutounet, & Escudier, 1989). High-resolution size exclusion chromatography was performed by loading 2 mL of the previously concentrated fraction using rotary evaporator (Buchi, Switzerland) on a Superdex 30-HR column (60  1.6 cm, Pharmacia, Sweden) with a precolumn (0.6  4 cm), equilibrated at 1 mL/min in 30 mM ammonium formiate pH 5.6. The elution of polysaccharides was followed with an Erma-ERC 7512 (Erma, Japan) refractive index detector combined with a Waters Baseline 810-software. The Monastrell oligosaccharide fraction was collected according to elution time between 60 and 93 min, freezedried, re-dissolved in water, and freeze-dried again four times to remove completely the ammonium salts. 2.5. Sugar composition as trimethylsilyl derivatives The neutral and acidic sugar composition was determined after solvolysis with anhydrous MeOH containing 0.5 M HCl (80 °C, 16 h), by GC of their per-O-trimethylsilylated methyl glycoside derivatives. The TMS derivatives were separated on 2 DB-1 capillary columns (30 m  0.32 mm i.d., 0.25 lm film) coupled to a single injector inlet through a two-holed ferrule, with H2 as the carrier gas on a Hewlett–Packard Model 6890 gas chromatograph. The outlet of one column was directly connected to a flame ionisation detector and the second column via a desactived fused-silica column (0.25 m  0.11 lm i.d.) was connected to a mass detector (HP 5973 mass selective detector). The chromatograph was operated with temperature programming (120–145 °C at 1 C/min,

153

145–180 °C at 0.9 °C/min and 180–230 °C at 50 C/min) (Doco, O’Neill, & Pellerin, 2001). 2.6. ESI mass spectrometry Wine oligosaccharide samples (50 lg) in 1:1 MeOH–water (5 lL) were injected directly into an AccuTOF (AccuTOF™ JMS-T100LC, Jeol, Japan) mass spectrometer equipped with an ESI source and a time-of-flight (TOF) mass analyser operated in the negative ion mode. The source voltage was set at 2000 V, the orifice voltage at 45 V, the desolvating chamber temperature at 250 °C, the orifice temperature at 80 °C, with the mass range being from 200 to 4000 Da. ESI-TOF spectra were obtained and extracted as ASCII files. 2.7. Statistical data treatment Average values, standard deviation and statistical significance were calculated and performed with the package Statgraphics Plus 5.1. An extension of Principal Components Analysis (PCA), called Anova-SCA (ASCA) was used to relate the variability of spectra collected on the wine oligosaccharide fractions. For every factor of variability, this technique suggests forming from the initial results, a data set constituted by the average spectra of its different levels. In this way, the percentage of variability which can be attributed to the variation of the average spectra can be calculated. Then, PCA is performed on each of these files, the individual results being projected onto the principal components obtained, to return the variability around the mean points. Before PCA, spectra underwent a pre-treatment to obtain a homogeneous data matrix; each spectrum was then normalised by dividing each intensity recorded by the sum of all the intensities (TIC). Matlab 7.3 software (The Mathworks Inc., MA, USA) was used for ASCA analysis using an ‘‘home made routine’’. 3. Results and discussion 3.1. Enological analysis Values of enological parameters for Monastrell wines from different terroirs (Table 1), indicate the good course of the winemaking processes. Alcohol, pH, total and volatile acidities, total polyphenols and colour indexes and hue give information on wine quality. Alcohol, pH and total acidity data were in agreement with the values ranges of normal dry young wines, although differences between terroirs were observed. Alcohol level was lower in wine from Montealegre terroir, than in wines from the other three terroirs. However, Montealegre showed lower pH values (3.37), and higher values of total acidity (6.3 g/L tartaric acid) than Chaparral-Bullas wine (3.54, 5.8 g/L tartaric acid,). pH values of Montealegre, Cañada Judío and Albatana wines were similar (3.31). Concerning phenolic parameters, Total Polyphenols Index was similar in Montealegre and Chaparral-Bullas wine but much lower than in the wines from the other terroirs. The colour intensity of Chaparral-Bullas wines was slightly lower than that of Montealegre wines while that of the other wines was much higher. The hue values were highest in the ChaparralBullas wines. Optimum maturity is difficult to define because there is not one but several maturity levels according to different elements (sugar content, acidity, berry size, phenolic content and taste) (Meléndez et al., 2013). The g parameters are highly dependent upon the grapevine genotype and its environment (Jackson & Lombard, 1993). Montealegre terroir showed specific properties, as higher

154

R. Apolinar-Valiente et al. / Food Chemistry 156 (2014) 151–159

Table 1 Enological data of four different terroirs Monastrell wines. Terroir/Treatment

Experimental nomenclature

Alcohola

pH

TAb

VAc

TPId

CIe

Huef

CAÑADA JUDIO Control Commercial enzyme addition Galactosidase enzyme addition

JUCO JUCE JUGE

14.6 ± 0.2a 14.3 ± 0.2a 14.4 ± 0.2a

3.31 ± 0.05a 3.34 ± 0.02a 3.32 ± 0.02a

6.8 ± 0.1a 6.9 ± 0.1a 7.1 ± 0.1a

0.23 ± 0.02a 0.22 ± 0.01a 0.24 ± 0.03a

56.6 ± 3.2a 55.1 ± 3.0a 57.6 ± 0.7a

16.2 ± 0.6a 15.9 ± 0.8a 17.1 ± 0.7a

0.42 ± 0.00a 0.42 ± 0.01a 0.41 ± 0.00a

ALBATANA Control Commercial enzyme addition Galactosidase enzyme addition

ALCO ALCE ALGE

14.4 ± 0.4a 14.5 ± 0.1a 14.5 ± 0.3a

3.29 ± 0.02a 3.28 ± 0.04a 3.34 ± 0.07a

6.8 ± 0.1a 6.8 ± 0.2a 6.6 ± 0.4a

0.25 ± 0.04a 0.27 ± 0.04a 0.26 ± 0.02a

54.4 ± 1.7a 56.1 ± 0.2a 56.6 ± 1.1a

14.7 ± 0.3a 15.8 ± 0.6a 14.7 ± 1.4a

0.46 ± 0.01a 0.46 ± 0.01a 0.48 ± 0.03a

BULLAS Control Commercial enzyme addition Galactosidase enzyme addition

BUCO BUCE BUGE

14.2 ± 0.4a 14.3 ± 0.2a 14.2 ± 0.2a

3.54 ± 0.08a 3.58 ± 0.08a 3.55 ± 0.03a

5.8 ± 0.2a 5.7 ± 0.3a 5.8 ± 0.2a

0.23 ± 0.04a 0.25 ± 0.02a 0.27 ± 0.02a

34.1 ± 1.0a 35.0 ± 1.4a 35.2 ± 1.9a

8.6 ± 0.2a 9.0 ± 1.1a 8.8 ± 0.1a

0.62 ± 0.05a 0.64 ± 0.03a 0.64 ± 0.06a

MONTEALEGRE Control Commercial enzyme addition Galactosidase enzyme addition

MTCO MTCE MTGE

11.9 ± 0.1a 12.2 ± 0.3a 12.3 ± 0.5a

3.37 ± 0.03a 3.41 ± 0.03a 3.42 ± 0.04a

6.3 ± 0.1a 6.2 ± 0.1a 6.2 ± 0.1a

0.13 ± 0.00a 0.14 ± 0.00a 0.14 ± 0.00a

36.4 ± 2.1a 35.2 ± 1.1a 36.5 ± 3.4a

10.0 ± 0.5a 9.3 ± 0.1a 9.3 ± 0.6a

0.47 ± 0.01a 0.49 ± 0.01a 0.49 ± 0.01a

Different letters within the same terroir column represent significant differences according to an LSD test (p < 0.05). a Voluminal alcoholometric title % vol. b,c Total and volatile acidities g L1 tartaric acid. d Total polyphenols index at 280 nm. e Colour intensity (Abs 420 nm + Abs 520 nm + Abs 620 nm). f Abs 420 nm/Abs 520 nm.

altitude (771 m above sea levels), lower monthly mean maximum and minimum temperatures (°C) or higher monthly mean wind speed (m/s) (Supplementary Table 1). These characteristics together with other aspects that confirm the multi-factorial and complex terroir concept, could contribute to obtain well matured grapes (as shown by the pH and total acidity values) with lower sugars content from Montealegre terroir in comparison with grapes from Cañada Judío, Albatana and Chaparral-Bullas. Differences in the anthocyanin content were unrelated to sugar and acidity levels, indicating that terroir affects the level of anthocyanin compounds in grape. No treatment had a marked effect on the alcohol level, pH, and total and volatile acidities respect the control samples. The enzyme treatments did not have any effect on these parameters, regardless of the terroir. 3.2. Effect of terroir on Monastrell wine oligoccharide fractions: quantification and characterization Fig. 1 gives the molecular weight distributions of polysaccharides and oligosaccharides of Monastrell control wines from the

Fig. 1. Purification by high-resolution size-exclusion chromatography of oligosaccharide fractions isolated on Superdex 30-HR column from Cañada Judío (JU), Albatana (AL), Chaparral-Bullas (BU) and Montealegre (MT) red wines. (Relative Refractive Index versus Retention Time (Minutes).)

four terroirs. The fraction eluted on Superdex 30-HR column between 60 and 93 min (Bordiga et al., 2012; Ducasse et al., 2010) contained a complex mixture of small sugars and has been collected as the oligosaccharide fraction. The first peak in the range 41–50 min corresponded to the polysaccharide fraction of highest mass and rich in PRAGs (Polysaccharides Rich in Arabinose and Galactose) and mannoproteins (Doco, Quellec, Moutounet, & Pellerin, 1999; Ducasse et al., 2010). The second peak eluted between 49 and 60 min corresponded to the fraction containing mainly RG-II (Doco et al., 1999; Ducasse et al., 2010). Evident differences between the profiles of the control samples from the four studied locations were found (Fig. 1). The oligosaccharidic fractions in the Cañada Judío and Montealegre profiles were higher than oligosaccharide population in Albatana and Chaparral-Bullas profiles. Table 2 shows the glycosyl residue composition of the wines. They contained most of the sugars known to take part in the composition of wine carbohydrates (Belleville, Williams, & Brillouet, 1993; Doco & Brillouet, 1993; Pellerin, Vidal, Williams, & Brillouet, 1995; Vidal, Williams, Doco, Moutounet, & Pellerin, 2003; Waters, Pellerin, & Brillouet, 1994). Oligosaccharides of all wines included sugars such as rhamnose, arabinose, galactose, xylose and galacturonic and glucuronic acids coming from the pecto-cellulosic cell walls of grape berries but also mannose and glucose released from yeast and/or bacteria polysaccharides. Montealegre wines showed significantly higher quantities of rhamnose, arabinose and galactose in comparison with the other terroir wines. On the other hand, higher galacturonic acid amount was detected in Cañada Judío compared to the other terroir wines. These identified sugars confirmed the presence of mannan-, arabinogalactan-, homogalacturonan- and rhamnogalacturonanlike structures in all oligosaccharide fractions. As observed earlier by several authors (Carpita & Gibeaut, 1993; Doco, Vuchot, Cheynier, & Moutounet, 2003; Ducasse et al., 2010, and Ducasse et al., 2011) the presence of xylose, glucuronic and 4-OMe glucuronic acid residues indicated that traces of hemicelluloses might be solubilized from grape berry cell walls and recovered as oligosaccharide structures in wines. The amount of oligosaccharide isolated fraction of wines from different terroir is shown in Fig. 2. It can be observed that total

155

c

a

b

Average of measurements from three wine elaborations and standard deviation. Rha, Rhamnose; Fuc, Fucose; Ara, Arabinose; Xyl, Xylose; Man, Mannose; Gal, Galactose; Glc, Glucose; Gal A, Galacturonic acid; Glc A, Glucuronic acid; 4-OMeGlc A, 4-O methyl Glucuronic acid. Different letters within the same terroir column represent significant differences according to an LSD test (p < 0.05).

3.40 ± 0.18a 3.40 ± 0.43a 1.07 ± 0.24a 0.74 ± 0.33a 1.72 ± 0.53a 0.98 ± 0.41a MONTEALEGRE MTCO 34.0 ± 15.0a MTGE 28.3 ± 10.3a

3.9 ± 1.4a 3.2 ± 1.6a

73.6 ± 36.7a 46.4 ± 21.8a

18.4 ± 7.9a 14.8 ± 5.2a

32.5 ± 7.9a 32.5 ± 5.2a

40.7 ± 11.0a 46.7 ± 5.3a

37.1 ± 9.4a 43.2 ± 10.0a

31.3 ± 12.6a 39.5 ± 12.3a

7.5 ± 1.7a 6.9 ± 0.9a

1.3 ± 0.5a 1.1 ± 0.5a

7.7 ± 0.5a 7.5 ± 2.0a

7.54 ± 0.42b 4.29 ± 0.34a 6.78 ± 0.50b 0.10 ± 0.02a 0.15 ± 0.07a 0.17 ± 0.08a 1.18 ± 0.52a 1.24 ± 0.24a 1.27 ± 0.74a 2.5 ± 2.6a 6.1 ± 1.2a 2.8 ± 2.2a 5.6 ± 2.1a 18.4 ± 3.5b 5.7 ± 3.1a BULLAS BUCO BUCE BUGE

22.9 ± 13.0a 43.2 ± 8.5a 21.2 ± 16.6a

14.5 ± 5.1a 16.3 ± 1.7a 14.3 ± 5.4a

33.3 ± 5.1a 42.7 ± 1.7a 33.5 ± 5.4a

19.1 ± 3.0a 34.9 ± 2.3b 16.7 ± 4.4a

33.3 ± 7.9a 52.8 ± 9.7a 34.3 ± 9.1a

56.7 ± 9.3b 136.8 ± 39.5c 41.9 ± 37.2a

7.1 ± 6.4b 19.2 ± 13.7c 2.9 ± 0.5a

1.3 ± 0.9a 0.8 ± 0.8a 1.6 ± 0.8a

6.9 ± 0.5a 5.9 ± 0.9a 7.3 ± 0.8a

3.62 ± 0.37b 2.67 ± 0.37a 0.28 ± 0.05b 0.17 ± 0.04a 0.84 ± 0.36a 1.54 ± 0.25b 2.4 ± 1.4a 9.1 ± 4.6a ALBATANA ALCO 11.7 ± 4.4a ALCE 33.6 ± 0.0b

19.1 ± 9.5a 54.9 ± 11.1b

15.5 ± 6.0a 18.7 ± 2.3a

31.2 ± 5.5a 59.2 ± 9.7b

22.4 ± 1.4a 35.2 ± 1.4b

37.5 ± 9.0a 69.9 ± 10.2b

42.0 ± 9.9a 202.7 ± 37.0b

4.0 ± 0.2a 5.9 ± 2.3a

2.0 ± 1.2a 1.1 ± 0.5a

6.8 ± 0.7a 7.5 ± 0.9a

4.25 ± 0.44b 2.97 ± 0.46a 4.70 ± 0.26b 0.13 ± 0.03b 0.06 ± 0.00a 0.13 ± 0.03b 0.77 ± 0.18a 1.26 ± 0.32b 0.61 ± 0.05a 8.8 ± 1.4a 2.7 ± 4.6a 4.5 ± 1.7a 2.6 ± 0.9a 3.5 ± 1.1a 1.3 ± 0.5a CAÑADA JUCO JUCE JUGE

JUDIO 11.8 ± 3.4b 17.1 ± 1.9c 4.9 ± 2.0a

21.7 ± 7.0a,b 28.8 ± 11.1b 8.8 ± 3.5a

14.7 ± 3.8a 6.9 ± 2.9a 8.1 ± 3.4a

37.0 ± 7.4b 41.1 ± 4.6b 19.6 ± 7.4a

27.6 ± 3.2b 22.4 ± 2.8b 14.3 ± 5.2a

28.0 ± 7.8a,b 38.4 ± 5.5b 19.1 ± 6.7a

89.7 ± 30.9b 264.0 ± 30.5c 36.1 ± 11.9a

7.5 ± 3.8a 2.4 ± 1.4a 2.0 ± 0.7a

1.7 ± 0.7a 0.3 ± 0.4a 2.1 ± 3.7a

Rha/Gal Aa,b Ara/Gala,b Glca,b Gala,b Mana,b Xyla,b Araa,b Fuca,b Rhaa,b

Table 2 Glycosyl Composition (mg/L) and characteristic ratios of Oligosaccharides from different Monastrell terroirs wines.

Gal Aa,b

Glc Aa,b

Xylitola,b

4-OMeGlc Aa,b

(Ara + Gal)/Rhaa,b

R. Apolinar-Valiente et al. / Food Chemistry 156 (2014) 151–159

Fig. 2. Total oligosaccharide concentration (mg/L) in Cañada Judío (JU), Albatana (AL), Bullas (BU) and Montealegre (MT) wines. (CO: Control; CE: Commercial enzyme; GE: Galactosidase enzyme.)

oligosaccharide concentrations were higher, although not statistically significant, in the Cañada Judío and Montealegre terroirs wines (251 and 288 mg/L, respectively) in comparison with Albatana (195 mg/L) and Chaparral-Bullas (203 mg/L) terroirs wines. Therefore, chemical quantitative analysis corroborated obtained profiles. Several characteristic ratios have been calculated from oligosaccharide sugar composition: Arabinose/Galactose, Rhamnose/ Galacturonic acid, and (Arabinose + Galactose)/Rhamnose (Table 2). The Arabinose/Galactose ratio is characteristic of the wine Polysaccharides Rich in Arabinose and Galactose (PRAGs) (Bordiga et al., 2012; Doco, Williams, Pauly, O’Neill, & Pellerin, 2003; Ducasse et al., 2010; Vidal et al., 2003) and it can be used to characterise the wine oligosaccharides (Ducasse et al. 2010 and Ducasse et al. 2011). The obtained Arabinose/Galactose ratio for oligosaccharide fractions exhibited small differences for Cañada Judío (0.77), Albatana (0.84) and Chaparral-Bullas (1.18) wines. However, Montealegre wines showed a higher Arabinose/Galactose ratio (1.72), which suggests a greater release of arabinose or oligosaccharides rich in arabinose arising from the pectic framework into this wine (Vidal et al., 2003). Previously, Apolinar-Valiente et al. (2013) also found that polysaccharides of Monastrell wines from four different terroirs showed differences concerning this ratio. The Ara/Gal ratio obtained for the oligosaccharide fraction of the control wine for the four different wines were much lower than the ratios obtained previously for Carignan and Merlot wines, 2.2 and 2.8 (Ducasse et al., 2010). This result could suggest that the oligosaccharides present in Monastrell red wines are less rich in arabinose residues. The Rhamnose/Galacturonic acid ratio reflects the relative richness of the wine in homogalacturonan oligosaccharides (from smooth regions of pectins) versus rhamnogalacturonan oligosaccharides (from hairy regions of pectins) (Arnous & Meyer, 2009). The low value calculated for this ratio in Cañada Judío (0.13), Albatana (0.28) and Chaparral-Bullas (0.10) oligosaccharides showed an homogalacturonan structure predominance, whereas the ratio close to 1 for the Montealegre oligosaccharides (1.07) indicated a majority of rhamnogalacturonan structure likely organised with a repeat unit of [?2)-a-L-Rhap-(1?4)-a-D-GalpA-(1?]. Both types of structures have been previously described in wine oligosaccharides (Ducasse et al., 2010 and Ducasse et al., 2011). As one might presume that most of the arabinose and galactose are associated with pectin hairy regions, the relative importance of the neutral side-chains to the rhamnogalacturonan bakcbone was deduced from the ratio of (Arabinose + Galactose) to Rhamnose. This ratio was considerably lower in Cañada Judío (4.25), Albatana (3.62) and Montealegre (3.40) oligosaccharides in comparison with

156

R. Apolinar-Valiente et al. / Food Chemistry 156 (2014) 151–159

Chaparral-Bullas (7.54) oligosaccharides. It could indicate that the rhamnogalacturonan oligomers present in Chaparral-Bullas wine carry more neutral lateral chains. In this work, all the winemaking parameters were similar, except terroir characteristics. Therefore, all obtained results could suggest a terroir influence on wine oligosaccharide concentration, composition and structure, even with the same grape variety. 3.3. ESI mass spectrometry: analysis of oligosaccharide fractions from four different terroirs Monastrell wines The data set corresponding to whole ESI-TOF mass spectra of oligosaccharides isolated from Cañada Judío (e), Albatana (N), Chaparral-Bullas (d) and Montealegre (j) terroirs Monastrell wines, after normalisation, was treated by an extension of principal component analysis (PCA), called Anova-SCA (ASCA) (Fig. 3). Mass spectrometry cannot be considered a quantitative method because oligosaccharide species may exhibit different desorption capacities according to their structure. However, the MS profiles (i.e., relative intensities of mass signals) obtained were highly reproducible for each sample and very different between samples, thus allowing comparison among samples. The variability which can be attributed to the variation of the different terroir average spectra was found equal to 26.2%. The wine projections on the two first axis obtained by applying PCA on the mean spectra of the different terroirs were presented in Fig. 3A. The wine projections on the first axis show clear separation of Montealegre wines (j) from Cañada Judío (e), Albatana (N) and Chaparral-Bullas (d) wines. The wine projections on the second axis show clear separation of Albatana wines (N) from Cañada Judío (e), Montealegre (j) and Chaparral-Bullas (d) wines. The predominant ions observed in the first axis mass spectra (Fig. 3B) were ion at m/z 545, 559, 567, 605 and 735 for Cañada Judío, Albatana and Chaparral-Bullas oligosaccharides, and ion at m/z 661 for Montealegre oligosaccharides. As detailed by Ducasse et al. (2010), the ion at 545 m/z corresponds to [GalA]3, the ion at m/z 559 corresponds to [GalA]3-CH3 coming from the homogalacturonan backbones of the pectins,

the ion at 567 m/z corresponds to [GalA]3-Na and the ion at m/z 605 corresponds to 4-OMe-glucuronic acid (4-OMe-GlcA), two xylose residues (Xyl) linked in 1?4, and a xylitol residue in nonreducing position, coming this last oligosaccharide from the hemicellulosic cell wall structures. Although hemicelluloses are not identified in soluble polysaccharides of the wine, these fragments of 4-OMe-oligo-glucuronoxylan have already been reported in oligomer fractions (Ducasse et al., 2010). Finally, the ion at m/z 735 corresponds to [GalA]4-CH3. Concerning Montealegre oligosaccharides, the ion at m/z 661 corresponds to the repetition of the basic unit [(14)-a-D-GalAp-(1?2)-a-L-Rhap] two times, and the ion at m/z 983 corresponds to [Rha-GalA]3. The predominant ions observed in the second axis mass spectra (Fig. 3C) were ions at m/z 559 and 735 for Albatana oligosaccharides, and ions at m/z 567, 605 and 661 for Cañada Judío, Chaparral-Bullas and Montealegre oligosaccharides, which were previously described. Therefore, this analysis has revealed that the oligosaccharide composition can be influenced by terroir effect. 3.4. Effect of winemaking treatments on Monastrell wine oligoccharide fractions: quantification and characterization Table 2 shows the glycosyl residue composition analysis of oligosaccharides in different terroir wines (commercial enzyme addition in the case of Cañada Judío, Albatana and Chaparral-Bullas terroir wines, and b-galactosidase addition in the case of Cañada Judío, Chaparral-Bullas and Montealegre terroir wines). There were statistically higher amounts of rhamnose and galacturonic acid in Cañada Judío wine oligosaccharides, of rhamnose, arabinose, mannose, galactose, glucose and galacturonic acid in Albatana wine oligosaccharides, and of rhamnose, galactose, galacturonic and glucuronic acids in Chaparral-Bullas wine oligosaccharides, when commercial enzyme is added in comparison to the control samples. Concerning galactosidase enzyme addition, Table 2 shows lower concentrations of galacturonic and glucuronic acids in ChaparralBullas wine oligosaccharides, and of rhamnose, mannose, galactose and galacturonic acid in Cañada Judío wine oligosaccharides, than in the corresponding control wines.

Fig. 3. Analysis in principal components of oligosaccharide mass spectra of Cañada Judío (e), Albatana (N), Chaparral-Bullas (d) and Montealegre (j) terroirs red wines: (A) wine projections on axis formed by the first two principal components (PC 26.2%).; (B) contribution of the variables to principal component 1 and (C) principal component 2.

R. Apolinar-Valiente et al. / Food Chemistry 156 (2014) 151–159

Chemical quantitative analysis corroborated SEC profiles, showing higher total oligosaccharide concentrations when commercial enzyme was added in Cañada Judío (428 mg/L), Albatana (498 mg/L) and Chaparral-Bullas (377 mg/L) terroirs wines in comparison with control samples (Fig. 2). In the case of total oligosaccharide concentration from Cañada Judío wine, a decreasing trend was observed when galactosidase was employed (121 mg/L) in comparison with control wine, although this behaviour is not statistically significant. Taken together, these results indicate an obvious effect of commercial enzyme addition, which caused a 2-fold increase of the oligosaccharide amount in all treated wines, whereas the galactosidase enzyme influence over this parameter was reported just in one of treated wines (in Cañada Judío wine). Ducasse et al. (2011) also detected differences in oligosaccharide concentration between enzyme-treated and control wines, depending on the activities of the enzyme preparation used. At the same time, these results could denote an interaction between enzymatic treatments and grape origin, what would indicate that the effect of the enzyme addition technique could be valid only in grapes from certain terroirs. Table 2 shows the Arabinose/Galactose, Rhamnose/Galacturonic acid and (Arabinose + Galactose)/Rhamnose ratios calculated for each wine. The Arabinose/Galactose ratio was significantly higher in commercial enzyme treated wines from Cañada Judío (1.26) and Albatana (1.54) terroirs, in comparison with their control samples (0.77 and 0.84, respectively). The increase of this ratio in the oligosaccharide fractions suggests a release of arabinose or oligosaccharides rich in arabinose arising from the pectic framework (Carpita & Gibeaut, 1993). No significant difference was observed between galactosidase treated wines and control wines. These different behaviours depending on the grape origin in relation with this ratio could be explained by possible differences in pectin composition in grapes cell wall skin. The Rhamnose/Galacturonic acid ratio presents significantly lower values in commercial enzyme treated wines in Cañada Judío terroir (0.06) and Albatana terroir (0.17) compared to control wines (0.13 and 0.28, respectively). The lower values calculated

157

for this ratio indicates an increase of homogalacturonan-oligosaccharide predominance over rhamnogalacturonan-oligosaccharide when commercial enzyme was added in these two terroir wines. No difference was observed in Rhamnose/Galacturonic acid ratio between galactosidase treated wines and control wines in any of the studied terroirs. The ratio of (Arabinose + Galactose) to Rhamnose showed significantly lower values in wines from Cañada Judío (2.97), Albatana (2.67) and Chaparral-Bullas (4.29) treated with commercial enzyme, compared to control wines (4.25, 3.62 and 7.54, respectively). It could suggest that rhamnogalacturonan oligomers released in Cañada Judío, Albatana and Chaparral-Bullas wines carried less neutral lateral chains when commercial enzyme was added. Concerning galactosidase treated wines, there were no (Arabinose + Galactose)/Rhamnose ratio differences in comparison to the control wines in any studied terroir. Therefore, these results take together suggest a notable terroir influence over the action of commercial enzyme treatment. On the other hand, it seems there is no terroir effect on the galactosidase enzyme action. 3.5. ESI mass spectrometry: analysis of oligosaccharide fractions from four different terroirs Monastrell wines elaborated with enzymatic treatments The data set corresponding to whole ESI-TOF mass spectra of oligosaccharides isolated from control wines (h) and wines elaborated with commercial (N) and galactosidase enzyme (d), after normalisation, was treated by an extension of principal component analysis (PCA), called Anova-SCA (ASCA) (Fig. 4). The variability which can be attributed to the variation of the different enzymatic treatment average spectra was found equal to 28.9%. The wine projections on the two axis obtained by applying PCA on the mean spectra of the different enzymatic treatment were presented in Fig. 4A. The wine projection on first axis (Fig. 4A) showed evident differences among commercial enzyme treated wines (N)

Fig. 4. Analysis in principal components of oligosaccharide mass spectra of Monastrell terroirs red wines: control wines (h), commercial enzyme treated wines (N) and galactosidase treated wines (d): (A) wine projections on axis formed by the first two principal components (PC 28.9%); (B) contribution of the variables to principal component 1 and (C) principal component 2. (h: Control; N: Commercial enzyme; d: Galactosidase enzyme.)

158

R. Apolinar-Valiente et al. / Food Chemistry 156 (2014) 151–159

oligosaccharides and the group of galactosidase treated (d) and control (h) wines oligosaccharides. The predominant ion associated with the first axis (Fig. 4B) was the ion at m/z 605 for galactosidase treated and control wines oligosaccharides, and ions at m/z 499, 545, 567, 661 and 735 for oligosaccharides from wines treated with the commercial enzyme. The ion at m/z 605 characteristic of control and galactosidase treated wines corresponds to [4-OMe-GlcA-[Xyl]2-Xylitol], that was described earlier in oligosaccharide fractions from Merlot and Carignan wines (Ducasse et al., 2010). This type of structure corresponding to 4-OMe-oligo-glucuronoxylans was released in all wines and does not seem to be specific of an enzymatic treatment (Ducasse et al., 2010). So it seems that galactosidase enzyme treatment had no evident effect on released oligosaccharides composition, what is coherent with quantitative obtained results (Table 2). Concerning commercial enzyme treated wines oligosaccharides, the ion at m/z = 499 corresponds to Rha-[GalA-CH3]-Rha and the ion at m/z 661 corresponds to the repetition of the basic unit [(1?4)-a-D-GalAp-(1?2)-a-L-Rhap] two times. This represents rhamnogalacturonic zones (hairy regions), associated with oligosaccharides of enzyme-treated wines, matching up with the results obtained by Ducasse et al. (2010). These results confirm that commercial enzymatic preparation induced a greater release of pectin hairy regions (rhamnogalacturonan-like structures carrying neutral lateral chains). The other two ions in the negative part of the first axis were the ion at m/z 545 corresponding to the trigalacturonic acid, the ion at m/z 567 corresponding to [GalA]3-Na and the ion at m/z 735 corresponding to (GalA)4-CH3 (Ducasse et al., 2010). These ions represent the structures of homogalacturonans (smooth regions) resulting from pectinolytic activity of the commercial enzyme. Therefore, our results confirm a considerable effect of commercial enzyme addition on released oligosaccharide composition in comparison to the control wines. On the other hand, the wine projections on second axis (Fig. 4A) did not seem indicates important differences among wine samples. In conclusion, the analysis over 2008 vintage wines has revealed that the oligosacharide composition could be influenced by terroir effect. In the same way, our results confirm a considerable effect of commercial enzyme addition on released oligosacharide composition and amount in comparison to the control wines. On the other hand, galactosidase enzyme addition had no obvious effect on oligosaccharide release respect control Monastrell wines. Taken together, our results showed that grape origin impacts the effect of enzyme treatments on wine oligosaccharide composition. Obviously further studies should be carried out in order to a better understanding of the oligosaccharide content and composition in wines. Acknowledgements This work was made possible by financial assistance of the Ministerio de Ciencia e Innovación of Spain, Project AGL2006-11019C02-01/ALI. Author R. Apolinar-Valiente is the holder of a FPI fellowship from the Government of Spain. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.foodchem.201 4.01.093. References Analyse des moûts et des vins. (1990). Réglement (CEE) 2676/90 de la commission du 17 septembre 1990 déterminant des methods d´analyse communautaires

applicables dans le secteur du vin. Official Journal of the European, Communities, 35–49, 81–83; 84–87; 121–122; 167–176. Apolinar-Valiente, R., Williams, P., Romero-Cascales, I., Gomez-Plaza, E., Lopez-Roca, J. M., Ros-Garcia, J. M., et al. (2013). Polysaccharide composition of Monastrell red wines from four different spanish terroirs: Effect of wine-making techniques. Journal of Agricultural and Food Chemistry, 61, 2538–2547. Arnous, A., & Meyer, A. S. (2009). Quantitative prediction of cell wall polysaccharide composition in grape (Vitis vinifera L.) and apple (Malus domestica) skins from acid hydrolysis monosaccharide profiles. Journal of Agricultural and Food Chemistry, 57(9), 3611–3619. Ayestaran, B., Guadalupe, Z., & León, D. (2004). Quantification of major grape polysaccharides (Tempranillo v.) released by maceration enzymes during the fermentation process. Analytica Chimica Acta, 513, 29–39. Barham, E. (2003). Translating terroir: the global challenge of French AOC labelling. Journal of Rural Studies, 19, 127–138. Belleville, M. P., Williams, P., & Brillouet, J. M. (1993). A linear arabinan from a red wine. Phytochemistry, 33, 227–229. Bordiga, M., Travaglia, F., Meyrand, M., German, J. B., Lebrilla, C. B., Coïsson, J. D., et al. (2012). Identification and characterization of complex bioactive oligosaccharides in white and red wine by a combination of mass spectrometry and gas chromatography. Journal of Agricultural and Food Chemistry, 60, 3700–3707. Brillouet, J. M., Moutounet, M., & Escudier, J. L. (1989). Fate of yeast and grape pectic polysaccharides of a young red wine in the cross-flow microfiltration process. Vitis, 28, 49–63. Carpita, N. C., & Gibeaut, D. (1993). Structural models of primary cell walls in flowering plants: Consistency of molecular structure with the physical properties of the walls during growth. Plant Journal, 3, 1–30. Cescutti, P., & Rizzo, R. (2001). Divalent cation interactions with oligogalacturonides. Journal of Agricultural and Food Chemistry, 49, 3262–3267. Darvill, A. G., & Albersheim, P. (1984). Phytoalexins and their elicitors-a defense against microbial infection in plants. Annual Review of Plant Biology, 155, 507–516. Doco, T., & Brillouet, J. M. (1993). Isolation and characterization of a rhamnogalacturonan II from red wine. Carbohydrate Research, 243, 333–343. Doco, T., O’Neill, M. A., & Pellerin, P. (2001). Determination of the neutral and acidic glycosyl residue compositions of plant cell polysaccharides by GC-EI-MS analysis of the trimethylsilyl methyl glucoside derivatives. Carbohydrate Polymers, 46, 249–259. Doco, T., Quellec, N., Moutounet, M., & Pellerin, P. (1999). Polysaccharide patterns during the aging of Carignan noir red wines. American Journal of Enology and Viticulture, 50(1), 28–32. Doco, T., Vuchot, P., Cheynier, V., & Moutounet, M. (2003). Structural modification of arabinogalactan-proteins during aging of red wines on lees. American Journal of Enology and Viticulture, 54, 150–157. Doco, T., Williams, P., & Cheynier, V. (2007). Effect of flash release and pectinolytic enzyme treatments on wine polysaccharide composition. Journal of Agricultural and Food Chemistry, 55(16), 6643–6649. Doco, T., Williams, P., Pauly, M., O’Neill, M. A., & Pellerin, P. (2003). Polysaccharides from grape berry cell walls. Part II: Structural characterization of the xyloglucan polysaccharides. Carbohydrate Polymers, 53, 253–261. Ducasse, M. A., Williams, P., Canal-Llauberes, R. M., Mazerolles, G., Cheynier, V., & Doco, T. (2011). Effect of macerating enzymes on the oligosaccharide profiles of Merlot red wines. Journal of Agricultural and Food Chemistry, 59(12), 6558–6567. Ducasse, M. A., Williams, P., Meudec, E., Cheynier, V., & Doco, T. (2010). Isolation of Carignan and Melot red wine oligosaccharides and their characterization by ESIMS. Carbohydrate Polymers, 79, 747–754. Elleuch, M., Bedigian, D., Roiseux, O., Besbes, S., Blecker, C., & Attia, H. (2011). Dietary fibre and fibre-rich by-products of food processing: Characterisation, technological functionality and commercial applications: A review. Food Chemistry, 124, 411–421. Gibson, G. R., & Roberfroid, M. B. (1995). Dietary modulation of the human colonic microbiota: Introducing the concept of prebiotics. Journal of Nutrition, 125, 1401–1412. Guadalupe, Z., Palacios, A., & Ayestaran, B. (2007). Maceration enzymes and mannoproteins: A possible strategy to increase colloidal stability and color extraction in red wines. Journal of Agricultural and Food Chemistry, 55(12), 4854–4862. Hopkins, M. J., & Macfarlane, G. T. (2003). Nondigestible oligosaccharides enhance bacterial colonization resistance against Clostridium difficile in vitro. Applied and Environmental Microbiology, 69, 1920–1927. Jackson, D., & Lombard, P. B. (1993). Environmental and management practices affecting grape composition and wine quality: A review. American Journal of Enology and Viticulture, 44, 409–430. Laville, P. (1993). Unités de terroir naturel et terroir. Une distinction nécessaire pour redonner plus de cohérence au système d´appellation d´origine. Bulletin OIV. 745/746, 227–251. Li, Z., Pan, Q., Jin, Z., Mu, L., & Duan, C. (2011). Comparison on phenolic compounds in Vitis vinifera cv. Cabernet Sauvignon wine from five wine-growing regions in China. Food Chemistry, 125, 77–83. Macfarlane, G. T., Steed, H., & Macfarlane, S. (2008). Bacterial metabolism and health-related effects of galacto-oligosaccharides and other prebiotics. Journal of Applied Microbiology, 104, 305–344. Meléndez, E., Ortiz, M. C., Sarabia, L. A., Íñiguez, M., & Puras, P. (2013). Modelling phenolic and technological maturities of grapes by means of the multivariate relation between organoleptic and physicochemical properties. Analytica Chimica Acta, 761, 53–61.

R. Apolinar-Valiente et al. / Food Chemistry 156 (2014) 151–159 Morlat, R. (2005). Les principales approches et méthodes d´étude des terroirs viticoles. Etude et valorisation. Collection Avenir Ed., Paris. Pellerin, P., Vidal, S., Williams, P., & Brillouet, J. M. (1995). Characterization of five type II arabinogalactan-protein complexes from red wine with increasing uronic acid content. Carbohydrate Research, 190, 183–197. Qiang, X., YongLie, C., & QianBing, W. (2009). Health benefit application of functional oligosaccharides. Carbohydrate Polymers, 77(3), 435–441.

159

Rastija, V., Srecˇnik, G., & Medic´-Šaric´, M. (2009). Polyphenolic composition of Croatian wines with different geographical origins. Food Chemistry, 115, 54–60. Vidal, S., Williams, P., Doco, T., Moutounet, M., & Pellerin, P. (2003). The polysaccharides of red wine: Total fractionation and characterisation. Carbohydrate Polymers, 54, 439–447. Waters, E. J., Pellerin, P., & Brillouet, J. M. (1994). A Saccharomyces mannoprotein that protects wine from protein haze. Carbohydrate Polymers, 23, 185–191.