Colorimetric characteristics of the phenolic fractions obtained from Tempranillo and Graciano wines through the use of different instrumental techniques

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ “This is an Accepted Manuscript of an article published by Elsevier in Analytica Chimica Acta on 30 June 2012, available at: https://doi.org/10.1016/j.aca.2012.03.020 .” 1 1 TITLE: Colorimetric characteristics of the phenolic fractions obtained 2 from Tempranillo and Graciano wines through the use of different 3 instrumental techniques. 4 AUTHORS: Matilde García-Marinoa, M. Luisa Escudero-Gileteb, M. Teresa 5 Escribano-Bailóna, M. Lourdes González-Miretb, Julián C. Rivas-Gonzaloa, Francisco J. 6 Herediab* 7 Address: 8 aGrupo de Investigación en Polifenoles, Unidad de Nutrición y Bromatología, Facultad 9 de Farmacia, Universidad de Salamanca, Campus Miguel de Unamuno, E37007 10 Salamanca, Spain 11 bFood Colour and Quality Lab., Dept. Nutrition and Food Science. Universidad de 12 Sevilla. Facultad de Farmacia, E41012, Sevilla, Spain 13 14 *CORRESPONDING AUTHOR: 15 Francisco J. Heredia 16 Food Colour & Quality Lab., Dept. Nutrition & Food Science. Facultad de Farmacia. 17 Universidad de Sevilla. 41012-Sevilla, Spain 18 Tel.: +34 954556495 Fax: +34 954557017 19 e-mail: here[email protected] 20 2 21 Abstract 22 The aim of the present work was to determine whether there is any relationship between 23 measured by transmission and reflection (in the latter case, with and without contact 24 with the sample). We also evaluated which methodology used would offer a better 25 interpretation of the results in visual terms. For this purpose, different colorimetric 26 techniques such as transmission spectrophotometry, diffuse reflectance 27 spectrophotometry and spectroradiometry were applied. The samples consisted of 28 increasing dilutions (0, 20, 40, 60, 80 and 100%) of the phenolic fractions obtained from 29 4 wines: Tempranillo (T) and Graciano (G) monovarietal wines, and 80:20 mixtures: M 30 (wine elaborated by blending grapes) and W (a blend of the T and G wines) (9 fractions 31 per wine). Fractionation was performed using gel permeation chromatography with a 32 Toyopearl HW-40S column, and the dilutions of the fractions were performed with 33 synthetic wine (pH=3.6). The spectroradiometric measurements permitted the 34 differences due to the dilution effect on the fractions to be established more clearly than 35 with the results obtained using diffuse reflectance and spectrophotometry. Thus, this 36 technology is very suitable for use in comparative interpretations by the human eye. In 37 turn, we assessed the changes in colour due to the effect of dilution on the fractions, 38 observing that the effect of dilution led to an increase in the values of lightness (L*), 39 while the chroma values (C*ab) followed the opposite trend, in agreement with its role 40 as a variable related to chromatic intensity or vividness of the sample. In contrast, hue 41 (hab) did not seem to be affected by dilution of the fractions, in consonance with the 42 qualitative nature of this parameter. 43 Keywords 44 Wine colour; Anthocyanins; Tempranillo; Graciano; Colorimetric techniques; CIELAB 3 45 1. Introduction 46 In the field of oenology, the visual analysis of wine colour is included in the tasting; this 47 is also known as organoleptic examination, which consists of appreciation by sight, 48 smell and taste of the qualities of a wine. Sampling or tasting involves subjecting the 49 wine to our sensory skills with a view to becoming familiar with it and determining its 50 sensory characteristics, eventually appreciating it (or not) [1]. In red wines, colour 51 represents the first sensory characteristic perceived by the taster. Moreover, strong 52 correlations have been found between the colour and overall quality of wines [2,3], 53 although colour provides not only information about possible defects or the type or state 54 of the evolution of the wine, but also has a strong influence on its acceptability [4] and 55 price [5]. 56 The initial violet-red colour of young red wines is the net result of all the monomer, 57 oligomer and polymer anthocyanins extracted from the grape skins, together with their 58 copigmented forms and intensity and hue, dependent upon factors such as the nature and 59 concentration of the individual anthocyanins and their degree of degradation, 60 temperature, pH, the nature of the solvent, the presence of SO2, oxygen, enzymes, 61 copigments, sugars, etc.[6-12]. However, during the ageing process of wines this colour 62 evolves to reddish-orange hues, mainly due to the progressive structural changes 63 undergone by the anthocyanins. These changes occur through different mechanisms 64 [7,13-18]. 65 In the winery, the parameters traditionally used to described the variation in colour of 66 the anthocyanin solutions have mainly been the changes in max in the visible part of the 67 spectrum as a measurement of variations in hue, together with changes in absorbance 68 for the variation in colour intensity [19-20]. Both indices are easy to calculate and 69 interpret and are those most frequently used in winery [21]. Nevertheless, Gonnet [22] 4 70 reported that an adequate description of variations in the colour of anthocyanin 71 solutions, originated -for exampleby pH, require the following: (a) that the spectral 72 variations considered should be those affecting the whole spectral curve, not only its 73 visible max; (b) that it would be appropriate to use the three colour attributes (hue, 74 saturation and lightness) for its description, and (c) that these should refer to the 75 conditions of the observer and of the light source. 76 The CIE has proposed different systems for colour representation in an attempt to find 77 one that will reflect the visual sensation perceived by observers in an appropriate way. 78 When the determination of a colour is carried out it is necessary to determine the 79 position of the observer, the light source, and the interval of data acquisition (for 80 example, every 3 nm). 81 The colour of objects can be expressed through the colour coordinates of the different 82 colour spaces (CIEXYZ, CIELAB, CIELUV, etc.). 83 The CIELAB space is a Cartesian coordinate system defined by three colorimetric 84 coordinates L*, a*, and b*, where L* represents lightness. This can only take values 85 between 0 and 100, 0 corresponding to a black sample and 100 to a white one. The 86 colorimetric coordinates a* and b* form a perpendicular plane with lightness. The a* 87 coordinate defines the deviation from the achromatic point corresponding to lightness; 88 towards the red if a*>0 and towards the green when a*<0. Accordingly, it is called the 89 red-green component. Likewise, the b* coordinate defines the deviation towards the 90 yellow if b*>0, and towards the blue if b*<0, thus being referred to as the yellow-blue 91 component. Chroma (C*ab) is the distance between the lightness axis and the stimulus in 92 question. The farther away it is from the L* axis, the more colour there is. It has a value 93 of 0 for achromatic stimuli and, in general, it does not surpass 150, although it may 94 reach higher values for monochromatic stimuli. hab is hue: this is the angle of the 5 95 circumference and lies between 0º and 360º, and for achromatic stimuli (a*=b*=0) it is 96 an undefined magnitude. Although the CIELAB space is currently the one most used, 97 both CIELUV and CIELAB are recommended by the CIE, despite certain anomalies 98 and limitations that may discourage its use in some cases. The CIELAB space is also 99 adapted as a UNE norm, in which colorimetric magnitudes that can be considered to be 100 a response of standard observers to a light stimulus are defined [23]. 101 Many authors have addressed the colour of pure anthocyanins solutions by using the 102 CIELAB colour space parameters [22,24-29]. Generally, in these studies the colour of 103 wines has been measured by transmission spectrophotometry. However, recently other 104 methodologies based on reflectance measurements have been applied to characterize the 105 colour of orange juices and honeys [30,31]. These techniques aim at obtaining the 106 radiometric measurement of the spectral distribution of a source of radiation (primary or 107 secondary), with the same components as the spectrophotometer except for the light 108 source, which in this case is external to the instrument. Accordingly there is influence 109 from external factors. Regarding colour, this device serves to determine the distribution 110 of the spectral radiant energy from any source so that its colour coordinates can be 111 calculated from that distribution. 112 The aim of this work was to study the influence of the dilution effect on the colour of 113 the fractions and to evaluate the relationship between the colour of the phenolic 114 fractions of red wines containing pigments, measured by transmission and reflection 115 techniques, and to determine which of these techniques allows a better interpretation of 116 the colour to be obtained. 6 117 2. Material and methods 118 2.1. Winemaking and samples 119 Three wines were elaborated separately from red grapes Vitis vinifera L. in Bodegas 120 Roda S.A. (La Rioja, Spain): T from the Tempranillo variety, G from the Graciano 121 variety, and M from a 80:20 mixture of Tempranillo and Graciano grapes. A fourth 122 wine W was elaborated by blending (80:20 v/v) the T and G wines after post- 123 fermentative maceration. 124 2.2. Sample fractionation 125 After three months of ageing in a barrel, 180 mL aliquots of each wine sample (T, G, M 126 and W wines) were collected and fractioned with a Toyopearl HW-40(s) gel column 127 (Tosoh, Japan) [32]. The elution solvents were ethanol/H2O (80:20 v/v) and 128 methanol/H2O (80:20 v/v). The different coloured bands formed during elution as well 129 as the bleaching eluates were collected separately. Thus, nine fractions were obtained, 130 dependent upon the change in colour produced inside the chromatographic column, each 131 considered as a different family of pigments according to the major compounds present 132 (Fig 1.). All fractions were acidified to pH=1 in order to reverse the existing bisulphite- 133 anthocyanin adducts, concentrated under vacuum, re-dissolved in water, and freeze- 134 dried. Solutions of the freeze-dried fractions were prepared to have similar contents as 135 in the wines. Thus, depending on the fraction different amounts (mg) were dissolved in 136 5 mL of synthetic wine (pH 3.6, 0.2M). 137 2.3. HPLC-DAD-MS analysis 138 The solutions of the fractions were acidified with 0.1N HCl (Panreac® Barcelona, 139 Spain) and injected into the chromatographic system after filtration through a 0.45 m 140 Millex® syringe-driven filter unit (Millipore Corporation). 7 141 HPLC-DAD analysis was performed with a Hewlett-Packard 1100 series liquid 142 chromatograph. The LC system was connected to the probe of the mass spectrometer 143 via the UV cell outlet. The mass analyses were performed using a FinniganTM LCQ ion 144 trap detector (Thermoquest, San Jose, CA, USA) equipped with an API source, using an 145 electrospray ionisation (ESI) interface. The HPLC-DAD-MS analysis of red pigments 146 was carried out in accordance with García-Marino et al. [33]. 147 2.4. Quantification 148 For quantitative analyses, calibration curves were obtained using standards of 149 anthocyanin 3-O-glucosides (delphinidin 3-O-glucoside, cyanidin 3-O-glucoside, 150 petunidin 3-O-glucoside, peonidin 3-O-glucoside and malvidin 3-O-glucoside). 151 Anthocyanins were purchased from Polyphenols Labs., Sandnes, Norway. 152 All pigments were quantified from the areas of their chromatographic peaks at 520 nm, 153 and the results were expressed in mgL-1 of wine. The total content of the different 154 groups of pigments studied was calculated from the sum of the individual 155 concentrations obtained for each individual compound. 156 2.5. Colorimetric measurements 157 Prior to spectrophotometric analysis, the fractions were filtered through Millipore-AP20 158 filters (Millipore Corporation, Bedford, MA, USA). Plastic cuvettes (475×350×10 mm) 159 were used for the measurements. 160 Transmission measurements were made with a UV/Visible HP8452 (Hewlett-Packard, 161 Palo Alto, CA, USA) spectrophotometer diode-array. The whole visible spectra were 162 recorded (380-780 nm -1964 10º standard observer and CIE D65 163 standard illuminant (corresponding to day light) were taken as references to calculate 8 164 e 165 (CIE, 2004), by applying CromaLab software [34]. 166 The reflectance measurements were performed spectroradiometrically, with the 167 spectroradiometer connected to a TOP 100 telescopic optical probe (Instrument 168 Systems, Munich, Germany) and a Tamron SP 23A zoom (Tamron USA, Inc., 169 Commack, NY, USA), and with diffuse reflectance spectrophotometry, with a ISP80 170 integration sphere (Instrument Systems, Munich, Germany), both coupled to the CAS 171 140B (Instrument Systems, Munich, Germany). Samples were measured against white 172 backing (pressed barium sulphate) and the whole visible reflectance spectra were 173 recorded (380- =2 nm). In this case, the CIELAB parameters were calculated 174 using IS-Specwin v.1.8.1.6 (Instrument Systems, Munich, Germany) software. 175 2.6. Dilution assays pigment fractions 176 Dilutions of fractions were assayed in order to study the influence of each dilution on 177 the colour of the fractions and to generate a greater number of samples to allow the 178 different colorimetric techniques to be compared. To accomplish this, increasing 179 volumes of the target fraction were obtained (9 fractions obtained from the fractionation 180 of each of the 4 wines studied) and were diluted in synthetic wine, pH 3.6, finally 181 obtaining mixtures with different percentages (0, 20, 40, 60, 80 and 100%), as shown in 182 Fig 2. The total number of fraction samples was 216; i.e., 54 per fractionated wine. 183 The colour changes due to the dilution effect were evaluated by the three techniques 184 described above. 185 2.6. Statistical analyses 186 Data are presented as means ± standard deviations (±S.D.) of three experiments 187 performed in triplicate. Significant differences were determined by one-way analysis of 15 [12] G. Mazza, E. Miniati, in: G. Mazza, E. Miniati (Eds.), Anthocyanins in Fruit, Vegetables and Grains. CRC Press, Boca Raton, Florida, 1993, pp. 1-199. [13] E. Haslam, Phytochemistry 16 (1980) 1625. [14] J.C. Rivas-Gonzalo, S. Bravo-Haro, C. Santos Buelga, J. Agic. Food Chem. 43 (1995) 1444. [15] M.T. Escribano-Bailón, O. Dangles, R. Brouillard, Phytochemistry 41 (1996) 1583. [16] C. Santos-Buelga, E.M. Francia-Aricha, S. De Pascual Teresa, J.C.Rivas-Gonzalo, Eur. Food Res. Technol. 209 (1999) 411. [17] M. Monagas, P.J. Martín-Álvarez, B. Bartolomé, C. Gómez-Cordovés, Eur. Food Res. Technol. 222 (2006) 702. [18] M. Monagas, B. Bartolomé, in: M.V. Moreno-Arribas, M.C. Polo, (Eds.), Wine chemistry and biochemistry, Springer, New York, 2009, pp. 439-462. [19] Y. Glories, Connaiss. Vigne Vin 18 (1984) 195. [20] P. Sudraud, Ann. Technol. Agric. 7 (1958) 203. [21] A.I. Negueruela, J.F. Echávarri, M.M. Pérez, Am. J. Enol. Viticult. 46 (1995) 353. [22] J.F. Gonnet, Food Chem. 63 (1998) 409. [23] CIE, 1978. Recommendations on Uniform Color Spaces, Color Difference Equations, Psychometric Color Terms. En Publicaciones CIE. Commission CIE, 15.2. [24] J. Bakker, C.F. Timberlake, J. Agr. Food Chem. 45 (1997) 35. [25] F.J. Heredia, E.M. Francia-Aricha, J.C. Rivas-Gonzalo, I.M. Vicario, C. Santos- Buelga, Food Chem. 63 (1998) 491. 16 [26] J.F. Gonnet, Food Chem. 66 (1999) 387. [27] J.F. Gonnet, Food Chem. 75, (2001) 473. [28] M.M. Giusti, L.E. Rodríguez-Saona, R.E. Wrolstad, J. Agr. Food Chem. 47 (1999) 4631. [29] F.C. Stintzing, A.S. Stintzing, R. Carle, B. Frei, R.E. Wrolstad, J. Agr. Food Chem. 50 (2002) 6172. [30] A. Terrab, D. Hernanz, F.J. Heredia, J. Agr. Food Chem. 52 (2004) 3441. [31] A.J. Meléndez-Martínez, G. Britton, I.M. Vicario, F.J. Heredia, Food Chem. 101, (2007) 1145. [32] C. Alcalde-Eón, E. Boido, F. Carrau, E. Dellacassa, J.C. Rivas-Gonzalo, Am. J. Enol. Viticult. 57 (2006) 449. [33] M. García-Marino, J.M. Hernández-Hierro, J.C. Rivas-Gonzalo, M.T. Escribano- Bailón, Anal. Chim. Acta, 660 (2010) 134. [34] F.J. Heredia, C. Álvarez, M.L. González-Miret, A. Ramirez, CromaLab®, análisis de color. Registro General de la Propiedad Intelectual SE-1052-04, Sevilla, Spain, 2004. 17 Figure captions Fig. 1. Example of the colour of the redissolution pigment fractions of red wine. Fig. 2. Scheme of the preparation of different dilutions (0, 20, 40, 60, 80 y 100%) for assays of the dilution effect on the colour of the fractions. Fig. 3. Values of L* (1-left), C*ab (2-centre) and hab (3-right) obtained using spectrophotometry of the different dilutions derived from the 9 fractions (F1-F9) of the T (A), G (B), M (C) and W (D) wines. Fig. 4. Values of L* (1-left), C*ab (2-centre) and hab (3-right) obtained by diffuse reflectance of the different dilutions derived from the 9 fractions (F1-F9) of the T (A), G (B), M (C) and W (D) wines. Fig. 5. Values of L* (1-left), C*ab (2-centre) and hab (3-right) obtained using spectroradiometry of the different dilutions derived from the 9 fractions (F1-F9) of the T (A), G (B), M (C) and W (D) wines. Fig. 6. Localisation area of the wine fractions on the diagram (a*,b*) measured by spectrophotometry (A), diffuse reflectance (B) and spectroradiometry (C). F1 F2 F3 F4 F5 F6 F7 F8 F9 F1 F2 F3 F4 F5 F6 F7 F8 F9 F1 F2 F3 F4 F5 F6 F7 F8 F9 F1 F2 F3 F4 F5 F6 F7 F8 F9 F1 F2 F3 F4 F5 F6 F7 F8 F9 F1 F2 F3 F4 F5 F6 F7 F8 F9 Fig. 1. 0 mL 0.2 mL 0.4 mL 0.6 mL 0.8 mL 1 mL Problem Fraction Synthetic Wine pH 3.6 2 mL 1.8 mL 1.6 mL 1.4 mL 1.2 mL 1 mL 1 mL 1.2 mL 1.4 mL 1.6 mL 1.8 mL 0 mL 0.2 mL 0.4 mL 0.6 mL 0.8 mL 1 mL Problem Fraction Synthetic Wine pH 3.6 2 mL 1.8 mL 1.6 mL 1.4 mL 1.2 mL 1 mL Fig. 2. 20 40 60 80 100 20 40 60 80 100 20 40 60 80 100 Tempranillo (T) A.1. A.2. A.3. 100 60 90 50 80 40 70 60 30 80 60 40 20 0 F1 F2 F3 F4 50 40 30 20 40 60 80 100 % 20 10 0 20 40 60 80 100 % -20 20 -40 -60 -80 40 60 80 100 F5 F6 F7 F8 % F9 Graciano (G) B.1. B.2. B.3. 100 90 80 70 60 50 40 30 20 40 60 80 100 % 60 50 40 30 20 10 0 20 40 60 80 100 % 80 60 F1 40 F2 20 F3 0 F4 -20 F5 -40 F6 -60 F7 -80 F8 % F9 Blend of grapes (M) C.1. C.2. C.3. 100 90 80 70 60 50 40 30 20 40 60 80 100 % 60 50 40 30 20 10 0 20 40 60 80 100 % 80 60 F1 40 F2 20 F3 0 F4 -20 F5 -40 F6 -60 F7 -80 F8 % F9 Blend of wines (W) D.1. D.2. D.3. 100 90 80 70 60 50 40 30 Fig. 3. 20 40 60 80 100 % 60 50 40 30 20 10 0 20 40 60 80 100 % 80 60 F1 40 F2 20 F3 0 F4 -20 F5 -40 F6 -60 F7 -80 F8 % F9 20 40 60 80 100 20 40 60 80 100 20 40 60 80 100 60 40 20 0 20 40 60 20 40 60 80 100 Tempranillo (T) A.1. A.2. A.3. 100 90 80 70 60 50 40 30 60 50 40 30 20 10 0 20 40 60 80 100 % 20 40 60 80 100 % 80 F1 F2 F3 F4 - F5 - F6 - F7 -80 F8 % F9 Graciano (G) B.1. B.2. B.3. 100 90 80 70 60 50 40 30 20 40 60 80 100 % 60 50 40 30 20 10 0 20 40 60 80 100 % 80 60 F1 40 F2 20 F3 0 F4 -20 F5 -40 F6 -60 F7 -80 F8 % F9 Blend of grapes (M) C.1. C.2. C.3. 100 90 80 70 60 50 40 30 20 40 60 80 100 % 60 50 40 30 20 10 0 20 40 60 80 100 % 80 60 F1 40 F2 20 F3 0 F4 -20 F5 -40 F6 -60 F7 -80 F8 % F9 Blend of wines (W) D.1. D.2. D.3. 100 90 80 70 60 50 40 30 Fig. 4. 20 40 60 80 100 % 60 50 40 30 20 10 0 20 40 60 80 100 % 80 60 F1 40 F2 20 F3 0 F4 -20 F5 -40 F6 -60 F7 -80 F8 % F9 20 40 60 80 100 20 40 60 80 100 20 40 60 80 100 20 40 60 80 100 Tempranillo (T) A.1. A.2. A.3. 100 90 80 70 60 50 40 30 20 40 60 80 100 % 60 50 40 30 20 10 0 20 40 60 80 100 % 80 60 F1 40 F2 20 F3 0 F4 -20 F5 -40 F6 -60 F7 -80 F8 % F9 Graciano (G) B.1. B.2. B.3. 100 90 80 70 60 80 50 60 F1 40 40 F2 30 20 F3 60 50 40 30 20 40 60 80 100 % 20 10 0 20 40 60 80 100 % hab 0 F4 -20 F5 -40 F6 -60 F7 -80 F8 % F9 Blend of grapes (M) C.1. C.2. C.3. 100 90 80 70 60 50 40 30 20 40 60 80 100 % 60 50 40 30 20 10 0 20 40 60 80 100 % 80 60 F1 40 F2 20 F3 0 F4 -20 F5 -40 F6 -60 F7 -80 F8 % F9 Blend of wines (W) D.1. D.2. D.3. 100 90 80 70 60 50 40 30 Fig. 5. 20 40 60 80 100 % 60 50 40 30 20 10 0 20 40 60 80 100 % 80 60 F1 40 F2 20 F3 0 F4 -20 F5 -40 F6 -60 F7 -80 F8 % F9 A Spectrophotometry diagram (a*,b*) 20 15 10 5 0 0 5 10 15 20 a* B 60 50 40 30 20 10 0 0 10 20 30 40 50 60 a* C 60 50 40 30 20 10 0 Fig. 6. 0 10 20 30 40 50 60 a* 90° 80° 70° 60° 50° 40° 1 2 3 4 5 6 7 8 9 30° 20° 10° 0 Diffuse reflectance diagram (a*,b*) 90° 80° 70° 60° 50° 40° 1 2 3 4 5 6 7 8 9 30° 20° 10° 0 Spectrorradiometry diagram (a*,b*) 90° 80° 70° 60° 50° 40° 1 2 3 4 5 6 7 8 9 30° 20° 10° 0