comprehensive colorimetric study of anthocyanic copigmentation

8/13/2019 Comprehensive Colorimetric Study of Anthocyanic Copigmentation

1/10

Comprehensive Colorimetric Study of Anthocyanic Copigmentationin Model Solutions. Effects of pH and Molar RatioBelen Gordillo, Francisco J. Rodrguez-Pulido, M. Luisa Escudero-Gilete, M. Lourdes Gonza lez-Miret,and Francisco J. Heredia*

Food Colour and Quality Laboratory, Department of Nutrition and Food Science, Faculty of Pharmacy, University of Sevilla, 41012Sevilla, Spain

ABSTRACT: New colorimetric variables have been defined in the uniform CIELAB color space to assess the quantitative andqualitative color changes induced by copigmentation and their incidence on visual perception. The copigmentation process wasassayed in model solutions between malvidin 3-glucoside and three phenolic compounds (catechin, epicatechin, and caffeic acid)as a function of the pH and the pigment/copigment molar ratio. Along the pH variation, the greatest magnitude ofcopigmentation was obtained at pH 3.0, being significantly higher with epicatechin and caffeic acid. At high acidic pH, the maincontribution of copigmentation to the total color was qualitative, whereas between pH 2.0 and 4.0, the main colorimetric

contribution was quantitative. The contribution of epicatechin and caffeic acid to the color changes was more marked for thequantitative characteristics. On contrast, particularly at higher pH values, the qualitative contribution was more important incatechin copigmented solutions. Increasing copigment concentration induced perceptible color changes at molar ratios higherthan 1:2, consisting in a bluish and darkening effect of the anthocyanin solutions. Among the different CIELAB attributes, huedifference was the best correlated parameter with the increase of copigment concentration, proving the relevance of thisphysicochemical phenomenon on the qualitative changes of anthocyanin color.

KEYWORDS: anthocyanin, color, copigmentation, Tristimulus Colorimetry

INTRODUCTIONAnthocyanins are the pigments accounting for the brilliant red,purple, and blue colors in many fruits, vegetables, and derivedfood products such as fruits juices, jams, and red wines.1

It is well-known that the stability of anthocyanic pigments isgreatly limited because their resonating structure confersintrinsic instability against several physicalchemical condi-tions. Evaluation of the factors affecting the stability ofanthocyanins indicates that pH is the most important extrinsicfactor of anthocyanin degradation. Depending on the pH of themedium, the red-colored flavylium cation coexists as anequilibrium mixture with other forms of anthocyanins: theblue-purple quinonoidal bases, the colorless hemiacetal B, andthe pale yellow chalcones. Therefore, the same anthocyaninsolution may show different colors.2

However, the chemical and colorimetric stability ofanthocyanins can be improved by associations with othersubstances. The copigmentation phenomenon, among others,

represents one of the most complex and efficient mechanismsof anthocyanin chromophore stabilization in nature and foodsystems.3 In food science, this phenomenon is considered to bea relevant interaction because obtaining stable and attractivecolors is a major focus for quality control purposes.4 Especially,in winemaking, it is assumed that it plays a key role in the colorevolution and stability of young red wines.5

Copigmentation reactions consist of noncovalent interac-tions between anthocyanins among themselves (self-associa-tion), between the central anthocyanin chromophore andaromatic acyl residues covalently linked to their glycosylmoieties (intramolecular copigmentation), or through inter-molecular interaction with a wide variety of colorless organic

compounds named copigments or copigmentation cofactors(basically other phenolic compounds, but also amino acids,sugars, organic acids, polysaccharides, etc.).47

From a molecular point of view, the anthocyanincopigmentcomplexes adopt a typical sandwich configuration (stacking) via hydrophobic interaction. This structural con-formation protects the red-colored flavylium cation against thenucleophilic attack of water, peroxide and sulfur dioxidebleaching, and pH changes, reducing the formation of theother colorless species in the anthocyanin equilibrium (hemi-acetal and chalcone).6

In addition, copigmentation not only confers greater stabilityto anthocyanins but also induces color variations. This kind ofmolecular association is responsible for the typical changes inthe spectral properties of the chromophore group, that is, anincrease of the absorptivity and frequently a shift of the visiblemaxtoward greater wavelengths. Consequently, copigmenta-tion produces both quantitative and qualitative color changes in

anthocyanin solutions. In this sense, the measurement andevaluation of these colorimetric changes is of great interest tothe food industry because color is one of the main sensoryparameters for the quality of foods influencing customerselection.7

The contribution of the copigmentation phenomenon incolor has been widely studied using spectrophotometricmethods, in both model and food systems that contain

Received: November 11, 2011Revised: February 29, 2012

Accepted: February 29, 2012Published: February 29, 2012

Article

pubs.acs.org/JAFC

2012 American Chemical Society 2896 dx.doi.org/10.1021/jf2046202| J. Agric. Food Chem. 2012, 60, 28962905
http://localhost/var/www/apps/conversion/tmp/scratch_3/pubs.acs.org/JAFChttp://localhost/var/www/apps/conversion/tmp/scratch_3/pubs.acs.org/JAFC


2/10

individual anthocyanins and added copigments or in red winesin which anthocyanins naturally co-occur with other phenoliccompounds.815 The evaluation of the changes in the visiblespectrum of anthocyanins, or specifically in the max(520 nm),has permitted the determination of the influence of numerousfactors on the effectiveness of the copigmentation including theconcentrations of pigments and cofactors, their chemicalstructures, the cofactor/pigment molar ratio, the pH of the

medium, etc. In the same way, several studies in red wines haveshown that the magnitude of copigmentation and its evolutionduring winemaking is extremely variable according toviticultural, agronomical, or enological practices, accountingfor approximately 2550% of the total color of young redwines.5,1620

In any case, although spectral methods have beendemonstrated to be valid, simple, and quick tools toquantitative estimations, it is generally accepted that theyprovide limited precision and accuracy for color specifications.The lower precision and accuracy achievable can be explainedbecause an adequate description of the color variations requires(i) that spectral variations considered should be those affectingthe entire spectral curve, not only its visible ma x, and (ii) theuse of atleast three colorimetric attributes: hue, saturation, andlightness.21 Moreover, the specific changes at the ma xare alsooften interpreted incorrectly in colorimetric terms. It has beendescribed that the hyperchromic and bathocromic effects makeanthocyanin solutions appear bluer and with more intensecolor.3,22,23 However, the variation in a single wavelength doesnot explain the complete behavior of the color due to thisphenomenon.

For these reasons, to advance the knowledge of the globalcolorimetric role of copigmentation it is necessary to considerboth quantitative and qualitative color changes. In this context,Tristimulus Colorimetry, which is based on transmittancevalues of the whole spectra, represents a useful methodoly thatwidely improves the objective analysis of color.

Thus, through Tristimulus Colorimetry, in this study isperformed a precise colorimetric interpretation of thecopigmentation phenomenon using the uniform 1976-(L*a*b*) color space (CIELAB), which has been recom-mended by the Commission Internationale de lEclairage (CIE)as a more appropriate tool for color specification in mostindustrial applications. For this purpose, diverse colorimetricvariables based on both the rectangular (L*, a*, b*) andcylindrical (L*, C*ab,hab) color coordinates have been definedto assess the quantitative and qualitative color implicationsassociated to copigmentation and their incidence on visualperception. These colorimetric variables have been applied andcompared with the most used simplified method in the

evaluation of the copigmentation effects in model solutions.For the model solution assay, the pigment chosen wasmalvidin 3-glucoside, one of the s ix most commonanthocyanins in nature, fruits, and vegetables. The flavanolscatechin, epicatechin, and caffeic acid were also selectedbecause they are some of the main polyphenolic copigmentsdescribed, especially in red wines.

MATERIALS AND METHODS

Standards and Copigmented Models Solutions.The pigmentmalvidin 3-glucoside (Mv 3-gl) was isolated in the laboratory fromskins ofVitis viniferared grapes of the Tempranillo variety. Extraction

was made with acidic methanol (methanol/HCl 1 N; 95:5 v/v), andthe extracts were purified by semipressure liquid chromatography

using a reversed-phase column, as described by Heredia et al.24 Thecopigments (+)-catechin, ()-epicatechin, and caffeic acid werepurchased from Sigma Chemical Co. (St. Louis, MO).

All of the model solutions were prepared in a wine-like mediumcontaining 5 g/L tartaric acid in 12% ethanol with ionic strengthadjusted to 0.2 M by the addition of sodium chloride.

To evaluate the effect of the pH on the copigmentationphenomenon, three copigmented solutions of Mv 3-gl/(+)-catechin(MC), Mv 3-gl/()-epicatechin (ME), and Mv 3-gl/caffeic acid (MF),

as well as a reference solution (Mv 3-gl), were prepared in wine-likemedium at different pH values: 1.0, 2.0, 3.0, 4.0, and 5.0. The referencesolution contained 200 mg/L (0.41 mM) Mv 3-gl. Copigmentedsolutions contained the same anthocyanin concentration and thecorresponding copigment using a pigment/copigment molar ratio of1:5.

The effect of the copigment concentration was also assessed. Twocopigmented solutions of Mv 3-gl/(+)-catechin (MC) and Mv 3-gl/()-epicatechin (ME) and a reference solution were prepared in thesame wine-like medium adjusted to pH 3.60. Copigmented solutionscontained the same anthocyanin concentration (0.41 mM) and thecorresponding copigments to give the required pigment/copigmentmolar ratios: 1:1, 1:2, 1:5, and 1:7.

All of the solutions were prepared in triplicate and equilibrated toreach the equilibrium for 2 h and stored closed in darkness at 25 C,

after which their absorption spectra were recorded.Color Analysis. The absorption spectra (380770 nm) of all themodel solutions were recorded at constant intervals (= 2 nm) witha Hewlett- Packard UVvis HP8452 spectrophotometer (Palo Alto,CA), using 2 mm path length glass cells and distilled water as areference. The CIELAB parameters (L*, a*, b*, C*ab, and hab) weredetermined by using CromaLab software,25 following the recom-mendations of the Commission International de L Eclariage:26 the 10Standard Observer and Standard Illuminant D65.

The L* value is the vertical axis and defines the lightness, theproperty according to which each color can be considered asequivalent to a member of the gray scale, between black and white,taking values within the range of 0100, respectively. The a*and b*

values represent the chromaticity scalar coordinates, which in turnrepresent opponent redgreen and blueyellow scales.

From L*, a

*, and b

*, other parameters are defined, such as hue(hab) and chroma (C*ab). Hue angle (hab) is the attribute according to

which colors have been traditionally defined as red, green, etc. On theother hand, chroma (C*ab) is the attribute that allows each hue to bedetermined by its degree of difference in comparison to a gray color

with the same lightness. Moreover, these colorimetric parameters canbe distinguished as quantitative or qualitative color attributes as theyindicate quantitative (L*and C*ab), or qualitative (hab) contributionsto color.

Color difference, which is very important to evaluate relationshipsbetween visual and numerical analyses,27 was determined by means ofthe CIE76 color difference parameter (E*ab). It was calculated as theEuclidean distance between two points in three-dimensional spacedefined byL*, a*, and b*: E*ab= ((L*)

2 + (a*)2 + (b*)2)1/2

Colorimetric Analysis of Copigmentation in the CIELABColor Space. The colorimetric effect of copigmentation was

evaluated by comparing the color of the pure anthocyanin solutionsand the color of the same solutions containing different copigmentconcentrations.

We consider a new variable E, which expresses the total colorof asolution as the color difference between the corresponding L*,a*, andb* values with respect to distilled water (L* = 100, a* = 0, b* = 0).Therefore, the total color of the copigmented and noncopigmentedsolutions was expressed as EC and E0, respectively. From EC and E0,the percentage of the anthocyanin color solutions that is due tocopigmentation was calculated as the following equation:

= E E ECCI (( )/ ) 100C 0 0

The absolute color variation induced by copigmentation wasassessed as the CIELAB color difference formula (E*ab) applied

Journal of Agricultural and Food Chemistry Article

dx.doi.org/10.1021/jf2046202| J. Agric. Food Chem. 2012, 60, 289629052897


3/10

between the color of the copigmented and noncopigmentedanthocyanin solutions, as follows:

*

= * * + * * + * *

E

L L a a b b(( ) ( ) ( ) )

ab(C 0)

C 02

C 02

C 02 1/2

In the same way, the absolute lightness, chroma, and hue differences(L*, C*ab, and hab) were used to asses the trend of the colorchanges induced by copigmentation. Specifically, habis the difference

between two hues, in sexagesimal degrees.In addition, the relative contribution of the three color attributesthat make up the total CIELAB color difference was also calculated tocompare the copigmentation effect according to different factors.Thus, the weight of each color attribute was calculated as

= * L L E% (( ) /( ) ) 100ab2 2

= * C C E%% (( ) /( ) ) 100ab2 2

= * H H E% (( ) /( ) ) 100ab2 2

Hbeing deduced from E*ab, L, and Cvalues as follows:

= * + H E L C(( ) (( ) ( ) ))ab2 2 2 1/2

Therefore, H, L, and Care scalar magnitudes in CIELAB units.Statistical Analysis. For the statistical treatment of the data,

Statistica v. 8.0 software28 was used.

RESULTS AND DISCUSSION

Effect of pH and the Chemical Structure of Copig-ment on Copigmentation.A color analysis of the Mv 3-glcopigmented with different phenolic compounds was per-formed in the CIELAB space at different pH values (from 1 to5 units) to evaluate the influence of the copigmentationreaction on the anthocyanin equilibria. Figure 1 depicts thelocation of the pure Mv 3-gl solution and its correspondingcopigmentation with catechin (MC), epicatechin (ME), and

caffeic acid (MF) within the (a*b*) diagram and lightnessvalues (L*) according to pH value.

As previously reported by Heredia,24 pH variations provokeimportant changes in both quantitative and qualitativepsychophysical components of the anthocyanin color. As thepH increases, the total color (E) of the pure Mv 3-gl solutionstrongly decreased from 59.51 to 5.12 units, the progressivecolor degradation being especially remarkable at pH values >3.0(Table 1). However, the addition of the different copigmentsincreased significantly the total color of the pure Mv 3-glsolution, confirming the protective effect of the copigmentationphenomenon against color degradation.

As can be seen in Figure1, between pH 1.0 and 3.0, the colorof the pure Mv 3-gl solution underwent a slight decrease toblue hues and became less vivid and lighter. This colorevolution reflects the kinetic and thermodynamic competitionbetween the flavylium cation and the other colorless species inthe anthocyanin equilibria, that is, hemiacetals, chalcones, andthe blue anionic quinonoidal bases.2 However, at this pH range,a notable displacement of all copigmented solutions withrespect to the pure Mv 3-gl solution was produced in theCIELAB space, which tended progressively toward 350 color

area (

10). Therefore, whereas the pH effect diminishes thevalues ofa* and increases the values ofL*, which agrees withthe increase of the discoloration (evolution to achromaticcolors), copigmentation mainly diminishes the values ofL*andb*. As a consequence, pure Mv 3-gl solutions exhibited thedarkest and more vivid bluish color when they were in thepresence of the copigments, revealing the positive influence ofthe copigmentation phenomenon on the anthocyanin equili-brium and, thus, on its color. Specifically, at pH 3.0, the totalcolor (E) increased significantly from 29.1 (Mv 3-gl) to 31.8,32.7, and 32.8 units, respectively.

On the contrary, according to their location in the (a*b*)colorimetric diagram, at pH values >3 units, both copigmentedand noncopigmented solutions appear to be more grouped in

Figure 1.Change in the location of the pure Mv 3-gl and copigmented solutions (MC, ME, and MF) within (left panel) the ( a*b*) diagram and(right panel) lightness values (L*) as a function of the pH value.




4/10

the same color area nearer the coordinates origin, reflecting thatthe effect of the copigmentation was less relevant, being almostimperceptible at pH 5.0. At this pH range, all of theanthocyanin solutions became more and more achromatic(a*andb*tended toward zero) and clearer (L*increases) andhad a lower color intensity, which means a progressive loss ofcolor.

Figure 2 shows the change in the magnitude of copigmentation for the copigmented solutions evaluated byTristimulus Colorimetry as the ratio ((EC E0)/E0) 100.The results showed that copigmentation occurred over theentire pH range we have studied for all of the phenolics used ascopigments. However, the magnitude of the copigmentationand its colorimetric effect was strongly pH-dependent andinfluenced by the nature of the copigment used. The greatestmagnitude of copigmentation was obtained at pH 3.0, beingsignificantly higher with epicatechin and caffeic acid. They bothincreased respectively the total color of Mv 3-gl by 13.7 and13.8%, whereas catechin, which was the less effectivecopigment, reached an increase of only 9.3%. This agreeswith other studies that have shown that among the different

flavan-3-ol copigments, the more planar molecules such as()-epicatechin or with electron-donor substituents such ascinnamic acids can better stackwithanthocyanins, resulting inhigher copigmentation effect.11,2931 The slightly lower valuesof the copigmentation magnitude obtained in this study withrespect to those reported in the literature could be due to boththe lower pigment/copigment molar ratio used and the

chemical nature of the tested copigments. It has beenestablished that the differences in the number, size, or spatiallocation of the substituents make monomeric flavan-3-ols, suchas catechin, epicatechin, or simple phenolic acids, exhibitweaker copigmentation than more planar phenolic compoundssuch as flavonols.32

As shown in Table1, the CIELAB differences (E*ab, L*,C*ab, hab) between Mv 3-gl solutions with and withoutcopigments were also calculated. Along the pH variation, thehighest color differences (E*ab) were produced from pH 1.0to 3.0, confirming the colorimetric stabilization of the flavyliumion by copigmentation at lower rather than higher pH values.Specifically, the highest color differences were reached at pH3.0, being 3.1, 3.7, and from 3.4 units in MC, ME, and MF

Table 1. Mean Values of the A520nmand the Total Color (E) Obtained for the Pure Anthocyanic Solution and Its RespectiveCopigmentation at Each pH Value (MV 3-gl: 200 mg/L, Molar Ratio 1:5) as well as the CIELAB Differences (E*ab, L*,C*ab, hab) between Mv 3-gl Solutions with and without Copigments

pigment pigment/copigment

pH Mv 3-gl MC ME MF

A520nm 1 0.86 0.001 a 0.84 0.001 b 0.84 0.008 b 0.83 0.004 b

2 0.73 0.002 a 0.78 0.03 a 0.73 0.002 a 0.77 0.03 a

3 0.32 0.001 a 0.34 0.001 b 0.36 0.004 c 0.37 0.003 c4 0.10 0.003a 0.10 0.003 a 0.11 0.001 b 0.12 0.005b

5 0.06 0.001 a 0.05 0.005a 0.06 0.001 a 0.05 0.002 a

E 1 59.51 0.98 a 60.10 0.23 b 60.77 0.26 c 60.79 0.16 c

2 54.37 1.67 a 57.51 1.77 b 55.83 0.17 b 57.88 1.42 b

3 29.13 0.98 a 31.85 0.17 b 32.68 0.34 c 32.79 0.20 c

4 9.47 1.67 a 10.22 0.27 b 10.82 0.01 c 11.06 0.38c

5 5.12 0.60 a 4.93 0.38 a 5.40 0.12 a 5.52 0.47 a

E*ab 1 1.73 0.06 a 2.88 0.04 b 3.39 0.13 c

2 2.92 0.48 a 2.29 0.12 a 3.78 1.06 a

3 3.06 0.17 a 3.69 0.37 b 3.87 0.24 b

4 1.29 0.08 a 1.38 0.02 a 1.59 0.27 a

5 1.01 0.14 a 0.54 0.09 a 0.71 0.14 a

L* 1 0.46 0.11 a 0.91 0.29 ab 0.95 0.10 b

2 2.64 0.23 a 1.16 0.02 a 2.27 0.82 a

3 0.89 0.01 a 1.97 0.20 b 2.14 0.13 b

4 0.03 0.20 a 0.77 0.21 b 1.03 0.40 b

5 +0.23 0.28 a 0.29 0.10 a 0.20 0.40 a

C*ab 1 +0.41 0.21 a +0.95 0.15 b +0.94 0.14 b

2 +0.47 0.20 a +1.03 0.19 a +2.77 1.17 b

3 +2.59 0.18 a +2.95 0.27 a +2.98 0.15 a

4 +0.91 0.19 a +1.09 0.01 a +1.17 0.16 a

5 +0.02 0.06 a +0.05 0.01 a +0.44 0.24 b

hab 1 1.74 0.07 a 2.74 0.20 b 3.35 0.15 c

2 1.30 0.75 a 1.93 0.07 a 1.99 0.53 a

3 2.91 0.08 a 2.12 0.29 b 2.61 0.35 ab

4 6.65 0.65 a 2.75 0.84 b 2.20 0.17 b

5 17.28 2.30 a 5.56 0.92 b 6.41 3.07 b




5/10

solutions, respectively. Therefore, although flavan-3-ols havebeen usually considered to be poorer anthocyanin copigments,at most acidic pH values, the color changes induced werealways perceptible to the human eye.33 However, at pH values>3.0 units, there was no significant difference between the coloreffect induced by the flavanols tested, all of the color differencesbeing


6/10

and hue (%H) to the total color difference for each pigment/copigment solution were also calculated. As can be seen in

Figure 3, the main contribution to the measured colordifferences E*ab at pH 1.0 was qualitative, which wasevidenced by the significantly higher contribution of hue%H (8088%) with respect to lightness %L or chroma%C (710 and 610%, respectively). These results arecoherent because at pH


7/10

With regard to the changes in the contribution of thecopigmentation to the total anthocyanin color, the addedcopigments caused significant concentration-dependent in-creases in the magnitude of the effect, as can be seen in Figure2. From 1:1 to 1:7 molar ratios, catechin induced an increase ofthe Mv 3-gl total color from 1.2 to 16.0%, whereas withepicatechin, the increase was from 3.9 to 17.9%. Notwithstand-ing, although the pattern evolution was similar between bothflavanol copigments, once again, the ability of the epicatechin toact as Mv-3gl copigment was stronger than that of catechinbecause significantly higher values for copigmentation wereachieved for most of the molar ratios tested.

As noted in the previous section of this study, thedependence between the amplitude of the color effect inducedby copigmentation was evaluated by means of the CIELABcolor differences (E*ab) (Table 2). As expected, theprogressive formation of copigmentation complexes wasconfirmed by a successive increase in the color differencesbetween anthocyanin and flavanol solutions when copigmentconcentrations were increasingly added. For all of theconcentration levels tested, epicatechin caused always signifi-cant increases (p < 0.05) in the color differences. In contrast,higher increases in the molar ratio were necessary to inducesignificant changes with catechin copigmented solutions.

Figure 4.Change in the location of the pure Mv 3-gl and copigmented solutions (MC, ME) within (a) the ( a*b*) diagram and (b) lightness values(L*) as a function of the molar ratio.




8/10

In any case, calculation of the simple regression coefficientsbetween total color differences and the copigment concen-

tration for each copigmented solution confirmed that therelationship was significantly high with regard to the copigmentused (r*values ranged from 0.95 to 0.98 units, p < 0.05). Thelowest color differences were found between pure Mv 3-glsolution and the copigmented ones at the lowest molar ratioused (1:1), taking values of 1.0 and 2.3 units with catechin andepicatechin, respectively. However, the color differencesobtained at all molar ratios higher than 1:1 were >3 units

and, hence, visually relevant. The largest color changes wereinduced with the highest copigment concentration (molar ratio1:7), being slightly more marked with epicatechin (10.79 vs9.76 units, in MC solutions). This was consistent with thehigher values also obtained regarding the total color (E) andthe magnitude of copigmentation, although the difference wasnot significant.

In relation to the quantitative and qualitative color changesthat the original Mv 3-gl solution underwent with increasingmolar ratio, the tendency was always toward higher decreases ofL*and hab, which means an increasing bluish and darkeningeffect. Simultaneously, chroma differences C*ab tendedtoward higher increases resulting in an increasing vivid effect,although this tendency was less marked than lightness and huetrends.

Univariate correlations between individual color attributesdifferences (L*, C*ab, hab) and copigment concentrationwere explored to determine the significance of these changes.Across the different copigmented solutions, the best relation-ships were obtained with simple regression for quantitativeattributes (L* C*ab) and with second-degree polynomialregression for the qualitative (hab). Calculation of thecoefficient regressions revealed that all of the relationshipswere strong and significant (p < 0.05), except for thosecorresponding to C*ab for MC solutions. Regressioncoefficients took negative signs for lightness and hue differ-ences, ranging as mean values between 0.71 and 0.95 and

between 0.93 and 0.99, respectively; these values were positivefor chroma differences. Among the different CIELAB attributes,hue differences (hab) were the best correlated parameterbecaise >95% of cases showed high quadratic regressioncoefficients (R2 = 0.99), proving the relevance of thisphysicochemical phenomenon in the qualitative changes ofanthocyanin color. A schematic representation of the huedifferences (hab) evolution as a function of molar ratio isshown in Figure5.

Again, the higher precision and accuracy of the colorimetricmethod to better evaluate the global color changes associatedwith copigmentation were manifested by comparing theA520nm,E, and E*abvalues. As can be observed in Table 2, no effect ofcopigmentation was detected when the molar ratio was

increased from 1:2 to 1:5 in MC solution, because no changeson the A520nm were obtained. However, the increase in thecatechin concentration resulted in increases of the total colorand of the total color difference.

Finally, by comparing the relative contributions of lightness(%L), chroma (%C), and hue (%H) obtained at eachmolar ratio (Figure6), it was observed that for all of the molarratios tested, the absolute color differences induced bycopigmentation were due mainly to quantitative changes(%L+ %C= 87.5%, as mean values) and to a lesser extentto the qualitative ones (%H = 12.5%), the weight of thelightness modifications being, in general, more marked than inchroma. However, increasing concentration for both copig-ments influenced especially the pattern evolution of the

Figure 5.CIELAB hue differences (hab) as a function of molar ratiofor Mv 3-gl/catechin (MC) and Mv 3-gl/epicatechin (ME) solutions.

Figure 6. Relative contribution of lightness (%L), chroma (%C),and hue (%H) to the total color difference for each pigment/copigment as a function of the molar ratio.




9/10

lightness contribution, which notably decreased from low tohigh molar ratios (%L= 67 vs 44%, respectively).

In contrast, chroma and hue contributions were in generalmore variable and did not show a clear tendency. Interestingly,at a given molar ratio value, both chroma and hue changes wereinfluenced in different ways by the type of copigment. Forexample, at a molar ratio of 1:5, both copigments inducedsimilar absolute color differences (Table 2), but catechin

induced higher changes in hue and epicatechin in chroma(%H= 30 vs 11% and %C= 30 vs 50%, respectively).

As a summary, the colorimetric interpretation of copigmen-tation based on the CIELAB color space has demonstrated tobe of practical interest because both quantitative and qualitativecolor changes can be better understood. It has beendemonstrated that pH, copigment structure, and concentrationhave significant influences on the copigmentation process,which induced different absolute and relative color changes inanthocyanin solutions. The greatest magnitude of copigmenta-tion and color effect was obtained at pH 3.0, being significantlyhigher with epicatechin and caffeic acid. At high acidic pHvalues (1.0) the main contribution to color differences was

qualitative, whereas between pH 2.0 and 4.0, the maincolorimetric contribution was quantitative, the chromamodifications %C being particularly more marked thanlightness %L. At pH 3.60, the greatest color effects wereinduced by epicatechin at a molar ratio of 1:7. Increasing molarratio produced always an increasing bluish, vivid, and darkeningeffect on anthocyanin solutions, affecting especially thelightness contribution %L. On the other hand, epicatechinand caffeic acid produced similar effects on the total color ofthe Mv 3-gl, contributing more marked quantitatively changesthan qualitative changes. In contrast, particularly at the higherpH values, the qualitative contribution was more important incatechin copigmented solutions.

AUTHOR INFORMATION

Corresponding Author

*Phone: +34 954556495. Fax: +34 954557017. E-mail:[email protected].

Funding

We are indebted to the Ministry of Science and Innovation ofSpain (Project AGL2008-05569-C02-02 and Fellowship BES-2009-025429) and the Consejeria de Economia, Innovacion yCiencia, Junta de Andalucia (Project P07-AGR-02893) forfinancial support.

Notes

The authors declare no competing financial interest.

REFERENCES

(1) Delgado-Vargas, F.; Jimenez, A. R.; Paredes-Lopez, O.; Francis, F.J. Natural pigments: carotenoids, anthocyanins, and betalains characteristics, biosynthesis, processing, and stability. Crit. Rev. FoodSci. Nutr. 2000, 40, 173289.

(2) Brouillard, R. Chemical structure of anthocyanidins. InAnthocyanins as Food Colors; Markakis, P., Ed.; Academic Press: NewYork, 1982; pp 140.

(3) Cavalcanti, R. N.; Santos, D. T.; Meireles, M. A. A. Non-thermalstabilization mechanisms of anthocyanins in model and food systemsan overview. Food Res. Int. 2011, 44, 499509.

(4) Hutchings, J. B. Food Colour and Appearance; Blackie Academicand Professional: Glasgow, Scotland, 1994.

(5) Boulton, R. The copigmentation of anthocyanins and its role inthe color of red wine: a critical review. Am. J. Enol. Vitic.2001,52, 6787.

(6) Castaneda-Ovando, A.; Pacheco-Hernandez, M.; Paez-Hernandez, M.; Rodrguez, J. A.; Galan-Vidal, C. A. Chemical studiesof anthocyanins: a review. Food Chem. 2009, 113, 859871.

(7) Montes, C.; Campos, J.; Pons, A.; Heredia, F. J. Instrumentalfactors influencing absorption measurements for fluid food colordetermination.J. AOAC Int. 2004, 87, 632638.

(8) Brouillard, R.; Wigand, M. C.; Dangles, O.; Cheminat, A. pH andsolvent effects on the copigmentation reaction of malvin withpolyphenols, purine and pyrimidine derivatives. J. Chem. Soc., PerkinsTrans. 2 1991, 8, 12351241.

(9) Davies, A. J.; Mazza, G. Copigmentation of simple and acylatedanthocyanins with colorless phenolic compounds.J. Agric. Food Chem.1993, 41, 716720.

(10) Baranac, J. M.; Petranovic, N. A.; Dimitric-Markovic, J. M.Spectrophotometric study of anthocyan copigmentation reactions. 2.Malvin and the nonglycosidized flavone quercetin.J. Agric. Food Chem.1997, 45, 16941697.

(11) Eiro, M. J.; Heinonen, M. Anthocyanin color behavior andstability during storage: effect of intermolecular copigmentation. J.

Agric. Food Chem.2002, 50, 74617466.(12) Talcott, S. T.; Brenes, C. H.; Pires, D. M.; Del Pozo-Insfran, D.

Phytochemical stability and color retention of copigmented andprocessed muscadine grape juice. J. Agric. Food Chem. 2003,51, 957963.

(13) Gris, E.; Ferreira, E.; Falcao, L.; Bordignon-Luiz, M. Influence offerulic acid on stability of anthocyanins from Cabernet Sauvignongrapes in a model system and a yogurt system. Int. J. Food Sci. Technol.2007, 42, 992998.

(14) Gonza lez-Manzano, S.; Duenas, M.; Rivas-Gonzalo, J. C.;Escribano-Bailon , M. T.; Santos-Buelga, C. Studies on thecopigmentation between anthocyanins and flavan-3-ols and theirinfluence in the colour expression of red wine. Food Chem. 2009,114,649656.

(15) Lambert, S. G.; Asenstorfer, R. E.; Williamson, N. M.; Iland, P.G.; Jones, G. P. Copigmentation between malvidin-3-glucoside andsome wine constituents and its importance to colour expression in red

wine.Food Chem. 2011, 125, 106

115.(16) Hermosn-Gutierrez, I. Influence of ethanol content on the

extent of copigmentation in a Cencibel young red wine. J. Agric. FoodChem. 2003, 51, 40794083.

(17) Hermosin-Gutierrez, I.; Lorenzo, E. S.-P.; Espinosa, A. V.Phenolic composition and magnitude of copigmentation in young andshortly aged red wines made from the cultivars, Cabernet Sauvignon,Cencibel, and Syrah. Food Chem. 2005, 92, 269283.

(18) Darias-Martn, J.; Carrillo-Lopez, M.; Echavarri-Granado, J. F.;Daz-Romero, C. The magnitude of copigmentation in the colour ofaged red wines made in the Canary Islands. Eur. Food Res. Technol.2007, 224, 643648.

(19) Guadalupe, Z.; Ayestaran, B. Effect of commercial mannopro-tein addition on polysaccharide, polyphenolic, and color compositionin red wines. J. Agric. Food Chem. 2008, 56, 90229029.

(20) Gordillo, B.; Lopez-Infante, M. I.; Ramrez-Perez, P.; Gonza lez-Miret, M. L.; Heredia, F. J. Influence of prefermentative coldmaceration on the color and anthocyanic copigmentation of organicTempranillo wines elaborated in a warm climate. J. Agric. Food Chem.2010, 58, 67976803.

(21) Gonnet, J. F. Colour effects of co-pigmentation of anthocyaninsrevisited 1. A colorimetric definition using the CIELAB scale.FoodChem. 1998, 63, 409415.

(22) Asen, S.; Stewart, R. N.; Norris, K. H. Co-pigmentation ofanthocyanins in plant tissues and its effect on color. Phytochemistry1972, 11, 11391144.

(23) Dangles, O.; Brouillard, R. A spectroscopic method based on theanthocyanin copigmentation interaction and applied to the quantita-tive study of molecular complexes. J. Chem. Soc., Perkin Trans. 2 1992,247257.


mailto:[email protected]:[email protected]


10/10

(24) Heredia, F. J.; Francia-Aricha, E. M.; Rivas-Gonzalo, J. C.;Vicario, I. M.; Santos-Buelga, C. Chromatic characterization ofanthocyanins from red grapes I. pH effect. Food Chem. 1998, 63,491498.

(25) Heredia, F. J.; A lvarez, C.; Gonza lez-Miret, M. L.; Ramrez, A.CromaLab, ana lisis de color. Registro General de la PropiedadIntelectual, 2004.

(26) CIE. Technical Report Colorimetry; Commission Internationalede lEclairage Central Bureau: Vienna, Austria, 2004.

(27) Melgosa, M.; Hita, E.; Poza, A. J.; Alman, D. H.; Berns, R. S.Suprathreshold color-difference ellipsoids for surface colors. Col. Res.

Apl. 1997, 22, 148155.(28) StatSoft Inc. STATISTICA(data analysis software system), v 8;

StatSoft Inc.: Tulsa, OK, 2007.(29) Darias-Martn, J.; Martn-Luis, B.; Carrillo-Lopez, M.; Lamuela-

Raventos, R.; Daz-Romero, C.; Boulton, R. Effect of caffeic acid onthe color of red wine. J. Agric. Food Chem. 2002, 50, 20622067.

(30) Gomez-Mguez, M.; Gonzalez-Manzano, S.; Gonza lez-Miret, M.L.; Escribano-Bailon, M. T.; Heredia, F. J.; Santos-Buelga, C. Influenceof different phenolic copigments on the color of malvidin 3-glucoside.

J. Agric. Food Chem.2006,54, 54225429.(31) Cruz, L.; Bras, N. F.; Teixeira, N.; Mateus, N.; Ramos, M. J.;

Dangles, O.; De Freitas, V. Vinylcatechin dimers are much bettercopigments for anthocyanins than catechin dimer procyanidin B3. J.

Agric. Food Chem.2010, 58, 3159

3166.(32) Kunsagi-Mate , S.; Szabo , K.; Nikfardjam, M. P.; Kollar, L.

Determination of the thermodynamic parameters of the complexformation between malvidin-3-O-glucoside and polyphenols. Copig-mentation effect in red wines. J. Biochem. Biophys. Methods 2006, 69,113119.

(33) Martnez, J. A.; Melgosa, M.; Perez, M. M.; Hita, E.; Negueruela,A. I. Note. Visual and instrumental color evaluation in red wines.FoodSci. Technol. Int. 2001,7, 439444.

(34) Berke , B.; De Freitas, V. A colorimetric study of oenincopigmented by procyanidins. J. Sci. Food Agric. 2007, 87, 260265.



comprehensive colorimetric study of anthocyanic copigmentation

Documents