Introduction

The use of oenological tannins is common practice in winemaking. Although the use of tannins is authorized by the International Organisation of Vine and Wine (OIV) only for facilitating the clarification of wines and musts (OIV, 2017), it is unquestionable that they are also used for many other purposes. Indeed, several other properties have been attributed to oenological tannins, such as antioxidant activity (for protection of wines against oxidation) (Hagerman et al., 1998), direct consumption of dissolved oxygen (Pascual et al., 2017), the ability to scavenge peroxyl radicals (Magalhaes et al., 2014), the ability to chelate iron(II), prevention of oxidative damage mediated by Fenton-based reactions (Pérez et al., 2009), antioxidasic activity (antilaccase activity) (Obradovic et al., 2005), improvement of wine structure and mouthfeel (Vivas, 2001), color improvement and stabilization of red wines (Canuti et al., 2012), copigmentation effect (Neves et al., 2010), direct formation of new pigments (Versari et al., 1998), elimination of reduction odors (Vivas, 2001) and even bacteriostatic effects (Lempereur et al., 2002a).

Many commercial tannins of different botanical origins and chemical compositions are currently available on the market (Obreque-Slíer et al., 2009; Malacarne et al., 2016). The so-called oenological tannins include ellagitannins from oak or chestnut, gallotannins from gall nuts, and condensed tannins from grape seeds and skins or from other plant sources, such as tara, quebracho and mimosa (Versari et al., 2013).

The chemical structure of ellagitannins consists of an open-chain glucose esterified at positions 4 and 6 by a hexahydroxydiphenoyl unit and a nonahydroxyterphenoyl unit esterified at positions 2, 3 and 5 with a C-glycosidic bond between the carbon of the glucose and position 2 of the trihydroxyphenoyl unit (Quideau et al., 2004; Okuda et al., 2009). The structure of various ellagitannins have been described, but castalagin and vescalagin are the most abundant in oak wood, accounting for between 40% and 60% of the total by weight (Fernández de Simón et al., 1999).

Gallotannins are formed by the esterification of gallic acid with the hydroxyl group of a polyol carbohydrate such as glucose (Hagerman, 1998). Gallotannin extracts, which are also commercially known as tannic acid, are mixtures of polygalloyl glucose or polygalloylquinic acid esters with a number of galloyl moieties per molecule ranging from 2 to 12, depending on the plant source used to extract the tannic acid (Sylla et al., 2015).

Condensed tannins are polymers of flavan-3-ol units; they are also called proanthocyanidins, because they release anthocyanins by acidic cleavage in accordance with the Bate-Smith reaction (Bate-Smith, 1973). Their composition, which depends on their botanical origin, varies mainly as regards the monomer released after acidic cleavage, the mean degree of polymerization, and their levels of galloylation and ramification (Versari et al., 2013).

Grape-skin tannins are composed of procyanidins and prodelphinidins, because their acidic cleavage produces cyanidin and delphinidin. In contrast, grape-seed tannins comprise only procyanidins, because they release only cyanidin (Bate-Smith, 1973). Additionally, grape-skin proanthocyanidins have a high mean degree of polymerization and a low level of galloylation, whereas grape-seed proanthocyanidins have a lower mean degree of polymerization and a high level of galloylation (Souquet et al., 1996). Regarding tannins from non-grape sources, quebracho tannins have a high level of ramification and are profisetinidins, because their acidic cleavage produces fisetinidin, whereas mimosa tannins are prorobinetidins, because they release robinetinidin (Celzard et al., 2015).

Of the various properties attributed to oenological tannins, their ability to improve the color intensity and stability of red wines is probably one of the main reasons why they are widely used in winemaking. The color of red wines is without doubt one of their main quality parameters, because it is the first sensory attribute experienced by consumers (Parpinello et al., 2009). Indeed, poor color can be an indicator of possible wine deficiencies, such as insufficient grape phenolic maturity, infection of grapes by grey rot, and inadequate winemaking and stabilization processes. Because it evolves over time, the color of a wine also gives information about its age, its conservation status and its aging ability.

Anthocyanins are the main compounds that determine red wine color, the visible expression of which depends, among other factors, on pH. They are mainly present in the form of the flavylium cation (which is red) at very acidic pH, but its proportion decreases progressively as pH increases, and this in turn generates the appearance of the quinoidal base (which is blue) and the hemiketal form (which is colorless) (Brouillard and Dubois, 1977). The hemiketal form can in turn give rise, via the opening of its heterocyclic ring, to the chalcone form (which is slightly yellow), and this may later be degraded in the presence of oxygen. This degradation is greatly favored by light and high temperature (Furtado et al., 1993).

Wine color is also strongly determined by copigmentation, a phenomenon based on interactions between anthocyanins and either other anthocyanins or other molecules acting as copigments (Boulton, 2001). This comes about through the formation of vertical π–π stacking non-covalent complexes (van der Waals interactions) between the planar polarizable nuclei of anthocyanins and the colorless copigment (de Freitas et al., 2017). The resulting complexes adopt a sandwich-like structure that protects the flavylium cation against nucleophilic attack by water, thus avoiding formation of the colorless hemiketal (Santos-Buelga and de Freitas, 2009). Copigmentation therefore increases wine color intensity (i.e. it has a hyperchromic effect), but it can also change the color hue through bathochromic shift (Brouillard and Dangles, 1994).

Red wine contains several substances that can play a role as copigments, such as some amino acids, phenolic acids, flavonols, flavan-3-ol monomers, proanthocyanidins and even anthocyanins themselves (He et al., 2012). It has also been suggested that oenological tannins may contribute as copigments to improve the color of red wines (Lempereur et al., 2002b), although to our knowledge no direct scientific evidence for this effect has been reported.

Anthocyanins can also react with other molecules, especially flavanols (flavan-3-ol monomers and proanthocyanidins), to produce new pigments (Francia-Aricha et al., 1997; He et al., 2012). Different mechanisms have been proposed to explain the formation of these new pigments. Some of these involve the direct condensation of anthocyanins and flavanols without the participation of oxygen (Santos-Buelga et al., 1995), but the most important reactions are probably those involving oxygen (Timberlake and Bridle, 1976). During winemaking and aging, the presence of small quantities of oxygen leads to the formation of ethanal from ethanol. The ethanal can in turn react with flavanols to induce the formation of a very reactive carbocation that quickly reacts with either another flavanol molecule or an anthocyanin, thus producing ethyl-bridged flavanol–flavanol or flavanol–anthocyanin oligomers (Es-Safi et al., 1999).

Additionally, cycloaddition reactions between anthocyanins and other small molecules can produce a new family of anthocyanin-derived pigments called pyranoanthocyanins. Specifically, the reaction with pyruvic acid or ethanal generates vitisin A or B, respectively (Bakker and Timberlake, 1997). The reaction between anthocyanins and vinylphenol can also generate vinylphenol adducts (Schwarz et al., 2003), and the previously formed ethylidene-bridged compounds can dissociate and generate vinylflavanol adducts (Cheynier et al., 2006). More recently, it has been reported that anthocyanins can also react with ellagitannins to form new pigments (Chassaing et al., 2010).

It is generally accepted that copigmentation is responsible for a substantial proportion (32–43%) of the color of young red wines (Hermosín-Gutiérrez et al., 2005). It has also been suggested that copigmentation may be the first step in the formation of covalent bonds between anthocyanins and flavanols (flavan-3-ol monomers and proanthocyanidins), thus creating new anthocyanin-derived pigments either directly or through the participation of ethanal (Francia-Aricha et al., 1997; de Freitas et al., 2017). Indeed, during storage or aging, copigmentation decreases and the concentration of polymerized structures increases, with consequent changes in the wine (Brouillard and Dangles, 1994; Gutiérrez et al., 2005; Liao et al., 1992).

In previous studies, some researchers have shown the effectiveness as copigments of different pure phenolic compounds in model wine solutions. However, to our knowledge very little scientific information exists about the effectiveness as copigments of these commercial tannins, which are evidently not pure substances. As previously mentioned, the use of commercial oenological tannins is a common practice in modern winemaking, and for this reason the OIV is now reconsidering the properties of oenological tannins in order to rewrite the OIV International Oenological Codex and authorize their use for other purposes.

Considering all the above, the aim of the present research was to study the effect of diverse oenological tannins on the color of the anthocyanin malvidin-3-O-monoglucoside in a model wine solution, and to propose a new index to measure the actual effectiveness as copigments of the different commercial products.

Materials and methods

Results and discussion

1. Effects on malvidin-3-O-glucoside concentration

Figure 1 shows the concentration of malvidin-3-O-monoglucoside on days 1 and 7 in the positive control solution and the solutions containing malvidin-3-O-glucoside and different doses of (–)-epicatechin and the various oenological tannins. On day 1, the concentration of malvidin-3-O-glucoside was 50.8 ± 0.2 mg/L in all samples. After 7 days of stockage, the concentration of malvidin-3-O-monoglucoside of the control solution containing only this anthocyanin decreased significantly to 47.6 ± 0.3 mg/L. This reduction (about 6.3%) was probably due to oxidation (Furtado et al., 1993). Similar results were obtained by Jordao et al. (2008) with malvidin-3-O-glucoside in a model wine solution without the presence of any other phenolic compounds.

Figure 1 also shows how the solutions of malvidin-3-O-glucoside were affected by the presence in the media of the copigments, (–)-epicatechin and the oenological tannins, at different concentrations after 7 days. The presence of all copigments except skin tannin and gallotannin brought about a slight but significant decrease in the concentration of malvidin-3-O-glucoside, and this reduction tended to be greater at higher concentrations of copigment. However, the decrease in malvidin-3-O-glucoside concentration was small and may be related to its oxidation or transformation; it has been reported that proanthocyanidins, flavan-3-ol monomers (Francia-Aricha et al., 1997) and even ellagitannins (Jordão et al., 2008; Chassaing et al., 2010) can react with malvidin-3-O-glucoside to produce new pigments.

Figure 1. Malvidin-3-O-glucoside concentration of the model wine solution supplemented with increasing concentrations of (-)-epicatechin and diverse oenological tannins.

Save View full size Expand inline Collapse inline

All data are the mean ± SD of three replicates. Different letters indicate the existence of statistically significant differences (p < 0.05) between different doses of each copigment. CP/P ratio correspond to the quotient between the copigment and pigment concentrations expressed as g/L.

The absorption spectra of these new pigments differ from those of malvidin-3-O-glucoside, and the degree of difference depends greatly on the nature of the new pigment. Pigments created by direct unions between malvidin-3-O-glucoside and flavanols have a red hue, for example, whereas those created by unions between malvidin-3-O-glucoside and flavanols via an ethyl bridge have a purple hue and the different products of cycloaddition (pyranoanthocyanins) have an orange hue (Francia-Aricha et al., 1997; de Freitas et al., 2017). Consequently, the formation of these new pigments can change the absorption spectrum of the solution. However, the variations in malvidin-3-O-glucoside concentration brought about by the presence of the different copigments were not very important, with values ranging from 0 (low doses of skin tannin or gallotannin) to 6.6% (in the case of the highest dose of seed tannin), and a global mean value below 2%. For this reason, changes in the absorption spectrum of the different samples are attributable mainly to copigmentation phenomena, although in some cases it could be hypothesized that a small and unknown proportion may be due to the presence of very low amounts of new pigments derived from malvidin-3-O-glucoside.

2. Effects on absorption spectra and CIELab color parameters

2.1 Effects on absorption spectra

Figure 2 shows the full spectrum in the visible range (400–800 nm) on day 7 of the malvidin-3-O-glucoside control sample and solutions containing also different doses of (–)-epicatechin and the various oenological tannins. In all cases, the presence of copigments increased the height and the surface of the absorption spectrum in the visible range, and this effect was dose-dependent. These results confirm that all the oenological tannins investigated can improve the color of a malvidin-3-O-glucoside solution and probably also the color of red wines. Similar results have been reported previously (Lempereur et al., 2002b; Vignault et al., 2019).

Figure 2. Visible spectra of a model wine solution of malvidin-3-O-glucoside (50 mg/L) supplemented with increasing concentrations of (-)-epicatechin and diverse oenological tannins at seventh day of experimentation.

Save View full size Expand inline Collapse inline

CP/P ratio correspond to the quotient between the copigment and pigment concentrations expressed as g/L.

Because the absorption spectrum results give only a general idea of the effects of (–)-epicatechin and the various oenological tannins on color, the effects on the red color specifically (absorption at 520 nm, A_520nm) and the CIELab coordinates were measured to better understand the copigmentation phenomenon: A_520nm, lightness (L*) and chroma (C*) describe quantitative attributes of the color, whereas hue (h*) describes qualitative attributes. The evaluation of these parameters makes it possible to determine the hyperchromic and bathochromic effects produced by the presence of the various copigments.

Figure 3 shows the A_520nm on days 1 and 7 of the solutions of malvidin-3-O-glucoside containing the different doses of (–)-epicatechin and the various oenological tannins. As shown in each graph, the A_520nm of the solution clearly decreased between day 1 and day 7, when no copigments were present. This decrease was undoubtedly related to the reduction in malvidin-3-O-glucoside concentration shown in Figure 1.

Figure 3. Influence of the copigment/pigment ratio on the absorbance at 520 nm of a model wine solution of malvidin-3-O-glucoside (50 mg/L) supplemented with (-)-epicatechin and diverse oenological tannins.

Save View full size Expand inline Collapse inline

All data are the mean ± SD of three replicates. CP/P ratio correspond to the quotient between the copigment and pigment concentrations expressed as g/L.

The A_520nm of the malvidin-3-O-glucoside solutions increased with the presence of (–)-epicatechin and each of the oenological tannins, showing a reasonably linear correlation with copigment-to-pigment ratio (CP:P) (see Figure 3). The slopes of the regression lines obtained, along with their statistical significance, are shown in Table 1. The values of the slopes indicate the effectiveness of (–)-epicatechin and the various oenological tannins in improving the red color of the solution; the steeper the slope, the greater their effectiveness. Berké and de Freitas (2005) have reported a similar slope for (–)-epicatechin, although under different experimental conditions. As shown by the slopes obtained, ellagitannin produced the greatest hyperchromic effect on day 1, followed in decreasing order by gallotannin and (–)-epicatechin (the slopes for these two tannins were statistically similar). In contrast, the smallest hyperchromic effects were produced by seed tannin, skin tannin and quebracho tannin.

No significant differences were observed between the slopes of the regression lines for day 1 and day 7 in the case of quebracho tannin, ellagitannin and gallotannin; the lines were nearly parallel (see Figure 3D–F). However, some differences were observed in the case of (–)-epicatechin, seed tannin and skin tannin (see Figure 3A–C). Specifically, the slopes for the seed and skin tannins became significantly steeper, whereas the slope of (–)-epicatechin did the opposite. The steeper slopes of the regression lines for the seed and skin tannins on day 7, crossing and passing the regression line for day 1 at the highest CP:P ratio, suggest that the copigmentation phenomenon needs time to reach its maximal effect, or that new pigments with a greater contribution to the red color were being formed. Oligomeric and polymeric proanthocyanidins have been reported to react with malvidin-3-O-glucoside to form new red pigments (Malien-Auberg et al., 2002).

The significantly less steep slope of the regression line for (–)-epicatechin on day 7 compared to that on day 1 suggests that the maximal copigmentation effect had already been reached on the first day. There are a number of possible explanations for the decrease in the incline of the slope. On the one hand, it may be due to the formation of new pigments with a lesser effect on the intensity of the red color. According to Dueñas et al. (2006), (–)-epicatechin and malvidin-3-O-glucoside can react to form an adduct anthocyanin–flavanol, which in turn is quickly transformed into a xanthylium cation with a yellow color. On the other hand, (–)-epicatechin can also form other compounds. Indeed, it has been reported that (–)-epicatechin can form adducts in the presence of glyoxylic acid produced by oxidation of tartaric acid and later form xanthylium pigments (Gómez-Míguez et al., 2006). These new compounds are probably less effective copigments than (–)-epicatechin; however, further studies are needed to verify this.

2.2 Effects on lightness

Figure 4 shows the relation between L* and CP:P for (–)-epicatechin and each of the oenological tannins. A decrease in L* was brought about by the presence in the media of each of the copigments. This indicates that their presence increased the color intensity of the resultant solution. It should be highlighted that a decrease in L* better indicates the existence of a hyperchromic effect than an increase in A_520nm, because L* refers to all the visible range of the spectrum and not just to a single wavelength.

Figure 4. Influence of the copigment/pigment ratio on the Lightness (L*) of a model wine solution of malvidin-3-O-glucoside (50 mg/L) supplemented with (-)-epicatechin and diverse oenological tannins.

Save View full size Expand inline Collapse inline

All data are the mean ± SD of three replicates. CP/P ratio correspond to the quotient between the copigment and pigment concentrations expressed as g/L.

The value of L* on day 7 was significantly higher than on day 1, when no copigments were present. This significant increase in L* is to be expected, because the malvidin-3-O-glucoside concentration and the A_520nm on day 7 were significantly lower than on day 1.

The decrease in L* showed a linear relation with CP:P. However, in this case the regression coefficients were not as clear as in the case of A_520nm, probably because L* relates to the whole absorption spectrum, not just absorption at a single wavelength. The slopes for each copigment were negative (see Table 1), therefore the higher the absolute value, the greater the hyperchromic effect.

Table 1. Slopes of the regression lines corresponding to the red color (A520), Lightness (L*), Chroma (C*) and ΔEaB* in function of the copigment/pigment ratio after 1 or 7 days of the supplementation with (-)-epicatechin and diverse oenological tannins.

View popup Expand inline Collapse inline

Copigment	Time (days)	A520					Lightness (L*)					Chroma (C*)					ΔEab*
(-)-Epicatechin	1	0.0041	±	0.0004	C	β	-0.066	±	0.011	B	β	0.108	±	0.007	C	β	0.118	±	0.009	C	β
	7	0.0025	±	0.0003	b	α	-0.038	±	0.005	a	α	0.058	±	0.005	b	α	0.073	±	0.006	b	α
Seed tannins	1	0.0029	±	0.0003	B	α	-0.080	±	0.010	B	α	0.042	±	0.003	A	α	0.069	±	0.006	B	α
	7	0.0049	±	0.0002	c	β	-0.088	±	0.006	b	α	0.067	±	0.005	b	β	0.102	±	0.004	c	β
Skin tannins	1	0.0014	±	0.0006	A	α	-0.026	±	0.011	A	α	0.044	±	0.005	A	α	0.043	±	0.010	A	α
	7	0.0046	±	0.0004	c	β	-0.071	±	0.013	b	β	0.086	±	0.004	c	β	0.111	±	0.008	cd	β
Quebracho tannins	1	0.0016	±	0.0005	A	α	-0.021	±	0.007	A	α	0.049	±	0.002	A	β	0.057	±	0.005	AB	β
	7	0.0013	±	0.0004	a	α	-0.037	±	0.011	a	α	0.014	±	0.004	a	α	0.034	±	0.005	a	α
Ellagitannins	1	0.0054	±	0.0010	C	α	-0.095	±	0.022	B	α	0.073	±	0.005	B	β	0.100	±	0.018	C	α
	7	0.0041	±	0.0004	c	α	-0.094	±	0.009	b	α	0.054	±	0.008	b	α	0.096	±	0.009	c	α
Gallotannins	1	0.0051	±	0.0010	C	α	-0.068	±	0.015	B	α	0.081	±	0.004	B	α	0.098	±	0.014	C	α
	7	0.0047	±	0.0002	c	α	-0.082	±	0.007	b	α	0.081	±	0.005	c	α	0.115	±	0.006	d	β

All data are the mean ± SD of three replicates. Different capital letters indicate the existence of statistically significant differences (p < 0.05) between the different tannins at day 1. Lowercase letters indicate the existence of statistically significant differences (p < 0.05) between the different tannins at day 7. Different Greek letters indicate the existence of statistically significant differences (p < 0.05) between times of treatment.

On day 1, the highest absolute slope value and consequently the greatest hyperchromic effect was produced by ellagitannin (see Figure 4E), followed in decreasing order by seed tannin, gallotannin, (–)-epicatechin, skin tannin and quebracho tannin (see Figures 4A–D,F). The only significant difference was found between quebracho tannin and skin tannins with all the other copigments.

The slope for the L* of the solutions containing (–)-epicatechin was significantly less steep on day 7 than on day 1 (see Figure 4A), which suggests, as mentioned earlier, that the copigmentation effect of this compound tends to decrease over time, probably because of the formation of xanthylium cations (Dueñas et al., 2006; Gómez-Míguez et al., 2006). In contrast, the L* slope for skin tannin increased significantly in absolute value (see Figure 4C), showing a similar pattern to that observed for A_520 nm. No significant differences were found between the slopes for day 1 and day 7 for any of the other oenological tannins. Consequently, on day 7 the greatest hyperchromic effect was produced by ellagitannin, followed by seed tannin, skin tannin and gallotannin, which exert a similar effect. The least effective copigments in terms of hyperchromic effect were (–)-epicatechin and quebracho tannin.

2.3. Effects on chroma

Figure 5 shows the relation between C* and CP:P for (–)-epicatechin and each of the oenological tannins. As expected, C* values were significantly lower on day 7 than on day 1, when no copigments were present. Again, the decrease in malvidin-3-O-glucoside concentration explains this result. Generally, the graphs show a similar pattern to that obtained for A_520nm. The results confirm that (–)-epicatechin and each of the tannins exert a hyperchromic effect, because the higher the CP:P, the higher the C*. The C* values were more in line with CP:P, showing reasonably clear regression coefficients and confirming that the color of the malvidin-3-O-glucoside solution was more vivid in the presence of these copigments.

On day 1, the steepest slope was obtained with (–)-epicatechin (see Figure 5A), followed in decreasing order by gallotannin, ellagitannin, quebracho tannin, skin tannin and seed tannin (see Figure 5B–F). In contrast, on day 7 some interesting changes in the slopes were observed. Specifically, the slopes were significantly steeper in the case of seed tannin and skin tannin, and significantly less steep in the case of (–)-epicatechin, ellagitannin and quebracho tannin. No differences were observed in the case of gallotannin.

Figure 5. Influence of the copigment/pigment ratio on the Chroma (C*) of a model wine solution of malvidin-3-O-glucoside (50 mg/L) supplemented with (-)-epicatechin and diverse oenological tannins.

Save View full size Expand inline Collapse inline

All data are the mean ± SD of three replicates. CP/P ratio correspond to the quotient between the copigment and pigment concentrations expressed as g/L.

An increase in the steepness of the C* slope between days 1 and 7 suggests that the copigmentation effect had increased or that new red pigments had been formed (Malien-Auberg et al., 2002). In the case of skin tannin, the steeper slope on day 7 was probably due to an increase in the copigmentation effect, because there was no significant decrease in malvidin-3-O-glucoside concentration compared with the control solution for the same day. In contrast, the increase in incline of the C* slope on day 7 for the solutions containing seed tannin was probably due to both an increase in copigmentation and the formation of new pigments, because the malvidin-3-O-glucoside concentration was significant lower than that of the control solution on day 7.

In the case of (–)-epicatechin, ellagitannin and quebracho tannin, the decrease in the incline of the C* slope between days 1 and 7 suggests that their copigmentation effect is lower or that malvidin-3-O-glucoside had reacted to form new compounds with a different hue. Such a hypothesis seems probable, because a decrease in the malvidin-3-O-glucoside concentration was observed in the solutions containing (–)-epicatechin, ellagitannin and quebracho tannin compared with the concentration in the control solutions without these compounds. It has been reported that colorless molecules are formed in solutions containing malvidin-3-O-glucoside and oak extract, and that the ellagitannin concentration decreases over time (Jordão et al., 2008).

2.4 Effects on wavelength of maximal absorption

The wavelength of maximal absorption (λ_max) of the absorption spectra of the malvidin-3-O-glucoside solutions was determined (Table 2), with the aim of evaluating the bathochromic effect produced by the presence of the different copigments. No significant differences in λ_max (around 524 nm) between days 1 and 7 were detected in the absence of copigment. Consequently, λ_max was not displaced by the slight but significant degradation of malvidin-3-O-glucoside observed under such conditions.

Table 2. Wavelength (nm) of maximum absorbance of a model wine solution of malvidin-3-O-glucoside (50 mg/L) supplemented with increasing concentrations of (-)-epicatechin and diverse oenological tannins at 1 and 7 days of treatment.

View popup Expand inline Collapse inline

Copigment	(g/L)	1 day					7 days
Control	0.0	524.0	±	1.4	A	α	523.3	±	0.6	A	α
EC	0.1	526.7	±	1.2	AB	β	524.3	±	0.6	A	α
	0.2	528.3	±	2.5	B	β	524.5	±	0.6	A	α
	0.4	529.3	±	2.9	B	β	524.3	±	0.6	A	α
SDT	0.1	525.7	±	1.2	AB	α	524.5	±	0.5	A	α
	0.2	527.0	±	1.0	B	β	524.7	±	0.6	A	α
	0.4	528.7	±	2.1	B	β	524.7	±	0.6	A	α
SKT	0.1	528.0	±	0.0	B	α	527.0	±	1.0	B	α
	0.2	528.7	±	1.2	B	α	526.7	±	0.8	B	α
	0.4	528.7	±	1.2	B	α	531.0	±	0.1	C	β
QT	0.1	531.0	±	0.1	C	β	526.0	±	0.1	B	α
	0.2	531.0	±	0.1	C	β	526.3	±	0.6	B	α
	0.4	527.3	±	1.2	B	α	527.7	±	1.2	B	α
ET	0.1	528.0	±	0.1	B	β	526.3	±	0.6	B	α
	0.2	530.0	±	1.7	C	α	531.0	±	0.1	C	α
	0.4	531.0	±	0.1	C	α	531.0	±	0.1	C	α
GT	0.1	530.0	±	1.7	B	α	526.8	±	0.3	B	α
	0.2	528.0	±	0.1	B	α	530.3	±	0.6	C	β
	0.4	529.3	±	1.5	B	α	531.0	±	0.1	C	β

All data are the mean ± SD of three replicates. EC: (-)-epicatechin; SDT: seed tannin; SKT: skin tannin; QT: quebracho tannin; ET: ellagitannin; GT: gallotannin. Different capital letters indicate the existence of statistically significant differences (p < 0.05) between different doses of each copigment. Different Greek letters indicate the existence of statistically significant differences (p < 0.05) between times of treatment.

On day 1, the presence of (–)-epicatechin and each of the oenological tannins produced a significant bathochromic displacement. This effect seems to be dose-dependent, because generally the higher the concentration of copigment, the higher the λ_max with the exception of solutions containing skin tannin and quebracho tannin, which reached λ_max at lower concentrations. The bathochromic displacement oscillated between 1.7 and 7 nm of λ_max, being particularly high in the case of ellagitannin and quebracho tannin.

On day 7, the solutions containing skin tannin, quebracho tannin, gallotannin and ellagitannin had maintained the bathochromic displacement, although in the case of quebracho tannin the λ_max was significantly lower than on day 1. In contrast, the bathochromic effect of the solutions containing (–)-epicatechin and seed tannin had completely disappeared, because the λ_max was similar to that of the control solution of malvidin-3-O-glucoside.

The decrease in the bathochromic effect observed for (–)-epicatechin, seed tannin and, to a lesser extent, quebracho tannin may have come about for two reasons. It could have been due to a reduction in copigmentation effectiveness, or alternatively the formation of new pigments with a more red-orange hue (Malien-Auberg et al., 2002; Dueñas et al., 2006). The result may have been due to the simultaneous occurrence of both. In contrast, the maintenance of bathochromic displacement in the solutions containing skin tannin and gallotannin, together with the fact that the malvidin-3-O-glucoside concentration was similar to that in the control solution, point to the permanence of the copigmentation effect. The particular case of solutions containing ellagitannins is more complicated, because a significant decrease in malvidin-3-O-glucoside concentration was detected and it has been reported that new pigments can be formed from these compounds, which have a bluish hue (Jordão et al., 2008; Chassaing et al., 2010). In such case, maintenance of the higher λ_max could be associated with the formation of these new pigments and also to maintenance of the bathochromic effect.

2.5 Effects on hue

Figure 6 shows the relation between h* and CP:P for (–)-epicatechin and each of the oenological tannins. An increase in h* angle for the control solutions of malvidin-3-O-glucoside had come about by day 7 compared with day 1. This again confirms the degradation of malvidin-3-O-glucoside, which results in a more red-orange color. Supplementation with (–)-epicatechin and each of the tannins brought about a significant decrease in h* on day 1, therefore the color of these solutions was clearly more bluish than that of the corresponding control solution. This effect was not dose-dependent, because the decrease in h* was generally reached at the lowest CP:P. The decrease in h* was clearly maintained on day 7 in the case of solutions containing skin tannin and gallotannin (see Figure 6C,F). The other copigments also caused a decrease in h* on day 7, but to a lower extent than on day 1 (see Figure 6A,B,D,E).

Figure 6. Influence of the copigment/pigment ratio on the Hue (h*) of a model wine solution of malvidin-3-O-glucoside (50 mg/L) supplemented with (-)-epicatechin and diverse oenological tannins.

Save View full size Expand inline Collapse inline

All data are the mean ± SD of three replicates. CP/P ratio correspond to the quotient between the copigment and pigment concentrations expressed as g/L.

The solutions containing skin tannin and gallotannin were the only solutions in which no significant decrease in malvidin-3-O-glucoside concentration was detected, therefore it seems that the copigmentation effect is maintained between days 1 and 7. In contrast, the clear increase in h* values observed in the presence of the other copigments on day 7 compared with day 1 can be explained by the formation of new pigments and/or a decrease in the copigmentation effect, as discussed above.

The changes in h* values between days 1 and 7 suggest not only that skin tannin, gallotannin and, to a lesser degree, ellagitannin increase the bluish hue of the solution, but also that this effect is maintained probably because the copigmentation effect lasts longer, and in the case of ellagitannins, because new pigments with a bluish hue may be formed (Jordão et al., 2008; Chassaing et al., 2010). In contrast, (–)-epicatechin and the other oenological tannins exert a bathochromic effect initially, but this decreases over time, probably because new pigments with a red-orange hue are formed (Malien-Auberg et al., 2002; Dueñas et al., 2006) and/or because these compounds are transformed into others with a lower copigmentation effectiveness (Gómez-Míguez et al., 2006).

2.6 Effects on total color differences

To evaluate the effect of the different copigments on the intensity of color, the total color differences (the ΔEab* values) between the solutions of malvidin-3-O-glucoside with or without the different copigments and a pure white color solution were calculated. Generally, ΔEab* showed a similar trend to that of C*, although with some minor differences (Figure 7).

Figure 7. Influence of the copigment/pigment ratio on the total color differences (DEab*) of a model wine solution of malvidin-3-O-glucoside (50 mg/L) supplemented with (-)-epicatechin and diverse oenological tannins with a pure white color solution.

Save View full size Expand inline Collapse inline

All data are the mean ± SD of three replicates. CP/R ratio correspond to the quotient between the copigment and pigment concentrations expressed as g/L.

Conclusion

It can be concluded that all the oenological tannins investigated are good copigments, because they improve the color of a malvidin-3-O-glucoside solution. The improvement in color was dose-dependent for each of the tannins, because it was greater when the copigment concentration was higher. Vignault et al. (2019) obtained similar results and have shown that pH and ethanol content also have an important influence on the copigmentation effectiveness of oenological tannins. The effect on color varied depending on the botanical origin of the tannins and their chemical structure, probably because some of the tannins can form new pigments with malvidin-3-O-glucoside. Specifically, the changes in malvidin-3-O-glucoside concentration and color parameters between days 1 and 7 suggest that seed tannin, quebracho tannin and ellagitannin can form new pigments, whereas skin tannin and gallotannin do not seem to do.

Further studies, using not only synthetic media but also real wine, are required to expand our knowledge of the mechanisms by which oenological tannins can improve wine color. It seems that the improvement in color exerted by each of the oenological tannins investigated in the present study was mainly due to the copigmentation effect, because the decrease in malvidin-3-O-glucoside concentration was always very low (6.6% in the case of the maximal decrease, with a mean value below 2%). The results prove that supplementation with oenological tannins is an interesting tool for improvement of the color of red wines, and therefore these findings point to the need to include this functionality of oenological tannins in the OIV International Oenological Codex.

Finally, this study proposes a copigmentation index (see the equation) to measure the actual effectiveness of commercial tannins, which could be useful for tannin manufacturers and potential users of oenological tannins. The index is based on the total color difference (ΔEab*) between each copigment solution and a pure white color solution; it is the best parameter for determining true differences between the color of two solutions, because it reflects the Euclidian distance between two points in the CIELab space. The index also accounts for the percentage increase in ΔEab* in response to supplementation with the highest dose of tannins (400 mg/L), thereby facilitating its quantification.

Acknowledgments

We would like to thank Comisión Interministerial de Ciencia y Tecnología (CICYT) (projects AGL2014-56594-C2-1-R and AGL2014-56594-C2-2-R) and the Spanish Ministerio de Agricultura y Pesca, Alimentación y Medio Ambiente, for their financial support. We are also grateful to the OIV for the scholarship granted to Jordi Gombau.

References

• Ayala F., Echavarri J.F. and Negueruela A.I., 1997. New simplified method for measuring the color of wines. I. Red and rose wines. American Journal of Enology and Viticulture, 48, 357–363.
• Ayala F., Echavarri J.F. and Negueruela A.I., 2001.MSCVes.zip. http://www.unizar.es/negueruela/MSCV.es 2001
• Bakker J. and Timberlake C.F., 1997. Isolation, identification, and characterization of new color stable anthocyanins occurring in some red wines. Journal of Agricultural and Food Chemistry, 45, 35–43. doi:10.1021/jf960252c
• Bate-Smith E.C., 1973. Haemanalysis of tannins: The concept of relative astringency. Phytochemistry, 12, 907–912. doi: 10.1016/0031-9422(73)80701-0
• Berké B. and De Freitas V.A.P., 2005. Influence of procyanidin structures on their ability to complex with oenin. Food Chemistry, 90, 453–460. doi:10.1016/j.foodchem.2004.05.009
• Boulton R., 2001. The copigmentation of anthocyanins and its role in the color of red wine: A critical review. American Journal of Enology and Viticulture, 52, 67–87.
• Brouillard R. and Dangles O., 1994. Anthocyanins molecular interactions: The first step in the formation of new pigments during wine aging. Food Chemistry, 51, 365–371. doi:10.1016/0308-8146(94)90187-2
• Brouillard R. and Dubois J.E., 1977. Mechanism of the structural transformations of anthocyanins in aqueous media. Journal of the American Chemical Society, 99, 1359–1364. doi:10.1021/ja00447a012
• Canuti V., Puccioni S., Giovani G., Salmi M., Rosi I. and Bertuccioli M., 2012. Effect of oenotannin addition on the composition of sangiovese wines from grapes with different characteristics. American Journal of Enology and Viticulture, 63, 220-231.
• Celzard A., Szczurek A., Jana P., Fierro V., Basso M.C., Bourbigot S., Stauber M. and Pizzi A., 2015. Latest progresses in the preparation of tannin-based cellular solids. Journal of Cellular Plastics, 51, 89–102. doi:10.1177/0021955X14538273
• Chassaing S., Lefeuvre D., Jacquet R., Jourdes M., Ducasse L., Galland S., Grelard A., Saucier C., Teissedre P.L., Dangle O. and Quideau S., 2010. Physicochemical studies of new anthocyano-ellagitannin hybrid pigments: About the origin of the influence of oak C-glycosidic ellagitannins on wine color. European Journal of Organic Chemistry, 1, 55–63. doi:10.1002/ejoc.200901133
• Cheynier V., Dueñas-Paton M., Salas E., Maury C., Souquet J.M., Sarni-Manchado P. and Fulcrand H. (2006). Structure and properties of wine pigments and tannins. American Journal of Enology and Viticulture, 57, 298–305.
• De Freitas V., Fernandes A., Oliveira J., Teixeira N. and Mateus N.A., 2017. Review of the current knowledge of red wine colour. OENO One, 51. doi:10.20870/oeno-one.2017.51.1.1604
• Dueñas M., Fulcrand H. and Cheynier V., 2006. Formation of anthcyanidin – flavanol adducts in model solutions. Analytica Chimica Acta, 563, 15–25. doi:10.1016/j.aca.2005.10.062
• Es-Safi N.E., Fulcrand H., Cheynier V. and Moutounet M., 1999. Studies on the acetaldehyde induced condensation of (–)-epicatechin and malvidin-3-O-glucoside in a model solution system. Journal of Agricultural and Food Chemistry, 47, 2096–2102. doi:10.1021/jf9806309
• Fernández de Simón B., Cadahía E., Conde E. and García-Vallejo M.C., 1999. Evolution of phenolic compounds of Spanish oak wood during natural seasoning. First results. Journal of Agricultural and Food Chemistry, 47, 1687–1694. doi:10.1021/jf9805855
• Francia-Aricha E., Guerra M.T., Rivas-Gonzalo J.C. and Santos-Buelga C., 1997. New anthocyan pigments formed after condensation with flavanols. Journal of Agricultural and Food Chemistry, 45, 2262–2266. doi:10.1021/jf9609587
• Furtado P., Figueiredo P., Chaves Das Neves H. and Pina F., 1993. Photochemical and thermal degradation of anthocyanidins. Journal of Photochemistry and Photobiology A: Chemistry, 75, 113–118. doi:10.1016/1010-6030(93)80191-B
• Gil M., Kontoudakis N., González E., Esteruelas M., Fort F., Canals J.M. and Zamora F., 2012. Influence of grape maturity and maceration length on color, polyphenolic composition, and polysaccharide content of cabernet sauvignon and tempranillo wines. Journal of Agricultural and Food Chemistry, 60, 7988–8001. doi:10.1021/jf302064n
• Gómez-Míguez M., González-Manzano S., Escribano-Bailón T., Heredia F.J. and Santos-Buelga C., 2006. Influence of different phenolic copigments on the color of malvidin-3-glucoside. Journal of Agricultural and Food Chemistry, 54, 5422–5429. doi:10.1021/jf0604586
• Hermosín-Gutiérrez I., Lorenzo E.S. and Espinosa A.V., 2005. Phenolic composition and magnitude of copigmentation in young and shortly aged red wines made from the cultivars, Cabernet Sauvignon, Cencibel, and Syrah. Food Chemistry, 92, 269–283. doi:10.1016/j.foodchem.2004.07.023
• Hagerman A.E., 1998. The Tannin Handbook. Oxford, OH, USA: Miami University.
• Hagerman A.E., Riedl K.M., Jones G.A., Sovik K.N., Ritchard N.T., Hartzfield P.W. and Riechel T.L., 1998. High molecular weight plant polyphenolics (tannins) as biological antioxidants. Journal of Agricultural and Food Chemistry, 46, 1887–1892. doi: 10.1021/jf970975b
• He F., Liang N.N., Mu L., Pan Q.H., Wang J., Reeves M.J. and Duan C.Q., 2012. Anthocyanins and their variation in red wines II. Anthocyanin derived pigments and their color evolution. Molecules, 17, 1483–1519. doi:10.3390/molecules17021483
• Jordão A.M., Da Silva M.R., Laureano O., Mullen W. and Crozier A., 2008. Effect of ellagitannins, ellagic acid and volatile compounds from oak wood on the (+)-catechin, procyanidin B1 and malvidin-3-glucoside content of model wines. Australian Journal of Grape and Wine Research, 14, 260–270. doi:10.1111/j.1755-0238.2008.00029.x
• Lempereur V., Blayteyron L., Labarbe B., Saucier C., Klebek H. and Glories Y., 2002a. Groupe National de travail sur les tanins œnologiques: Premiers résultats. Revue Française d’Oenologie, 196, 23–26.
• Lempereur V., Blateyron L., Labarbe B., Saucier C., Kelebek H. and Glories Y., 2002b. Groupe national de travail sur les tannins oenologiques: Premiers re´sultats. Effets instantanés du tanisage sur la couleur. Revue Française d’Oenologie 196, 27–29.
• Liao H., Cai Y. and Haslam E., 1992. Polyphenol interactions. Anthocyanins: Copigmentation and colour changes in red wines. Journal of the Science of Food and Agriculture, 59, 299–305. doi:10.1002/jsfa.2740590305
• Magalhaes L.M., Ramos I.I., Reis S. and Segundo, M.A., 2014. Antioxidant profile of commercial oenological tannins determined by multiple chemical assays. Australian Journal of Grape and Wine Research, 20, 72–79. doi:10.1111/ajgw.12058
• Malacarne M., Nardin T., Bertoldi D., Nicolini G. and Larcher R., 2016. Verifying the botanical authenticity of commercial tannins through sugars and simple phenols profiles. Food Chemistry, 206, 274–283. doi:10.1016/j.foodchem.2016.03.038
• Malien-Auberg C., Dangles O. and Amiot M.J., 2002. Influence of procyanidins on the color stability of oenin solution. Journal of Agricultural and Food Chemistry, 59, 3299–3305. doi:10.1021/jf011392b
• Neves A.C., Spranger M.I., Zhao Y., Leandro M.C. and Sun B., 2010. Effect of addition of commercial grape seed tannins on phenolic composition, chromatic characteristics, and antioxidant activity of red wine. Journal of Agricultural and Food Chemistry, 58, 11775–11782. https:// doi.org/10.1021/jf1038037
• Obradovic D., Schulz M. and Oatey M., 2005. Addition of natural tannins to enhance the quality of red wine. Australian and New Zealand Grape Grower and Winemaker, 493, 52–54.
• Obreque-Slíer E., Pena-Neira A., Lopez-Solis R., Ramirez-Escudero C. and Zamora-Marín F., 2009. Phenolic characterization of commercial enological tannins. European Food Research andTechnology, 229, 859–866.
• [OIV] International Organisation of Wine and Vine, 2017. International Oenological Codex. http://www.oiv.int/public/medias/5369/oiv-oeno-574-2017-en.pdf
• Okuda T., Yoshida T., Hatano T. and Ito H., 2009. Ellagitannins renewed the concept of tannins. In S. Quideau (Ed.), Chemistry and Biology of Ellagitannins – An underestimated class of bioactive plant polyphenols (pp. 1–40). Singapore: World Scientific Publishing.
• Parpinello G.P., Versari A., Chinnici F. and Galassi S., 2009. Relationship among sensory descriptors, consumer preference and color parameters of Italian Novello red wines. Food Research International, 42, 1389–1395. doi:10.1016/j.foodres.2009.07.005
• Pascual O, Vignault A, Gombau J, Navarro M, Gómez-Alonso S, García-Romero E, Canals JM, Hermosín-Gutíerrez I., Teissedre P.L. and Zamora F., 2017. Oxygen consumption rates by different oenological tannins in a model wine solution. Food Chemistry, 234, 26–32. doi:10.1016/j.foodchem.2017.04.148
• Perez C.A., Wei Y.B. and Guo M.L., 2009. Iron-binding and anti-Fenton properties of baicalein and baicalin. Journal of Inorganic Biochemistry, 103, 326–332. doi:10.1016/j.jinorgbio.2008.11.003
• Quideau S., Varadinova T., Karagiozova D., Jourdes M., Pardon P., Baudry C., Genova P., Diakov T. and Petrova R., 2004. Main structural and stereochemical aspects of the antiherpetic activity of nonahydroxyterphenoyl-containing C-glycosidic ellagitannins. Chemistry & Biodiversity, 1, 247–258. doi:10.1002/cbdv.200490021
• Santos-Buelga C., Bravo-Haro S. and Rivas-Gonzalo J.C., 1995. Interactions between catechin and malvidin-3-monoglucoside in model solutions. Zeitschrift fűr Lebensmittel-Untersuchung und -Forschung, 201(3), 269–274.
• Santos-Buelga C. and De Freitas V., 2009. Influence of phenolics on wine organoleptic properties. In M.V. Moreno-Arribas and M.C. Polo (Eds.), Wine Chemistry and Biochemistry (pp. 529–570). New York, NY, USA: Springer.
• Schwarz M., Wabnitz T.C. and Winterhalter P., 2003. Pathway leading to the formation of anthocyanin vinylphenol adducts and related pigments in red wines. Journal of Agricultural and Food Chemistry, 51, 3682–3687. doi:10.1021/jf0340963
• Souquet J.M., Cheynier V., Brossaud F. and Moutounet M., 1996. Polymeric proanthocyanidins from grape skins. Phytochemistry, 43, 509–512. Doi:10.1016/0031-9422(96)00301-9
• Sylla T., Pouységu L., Da Costa G., Deffieux D., Monti J.P. and Quideau S., 2015. Gallotannins and tannic acid: First chemical syntheses and in vitro inhibitory activity on Alzheimer’s amyloid β-peptide aggregation. Angewandte Chemie International Edition, 54, 8217–8221. doi:10.1002/anie.201411606
• Timberlake C.F. and Bridle P., 1976. Interactions between anthocyanins phenolic compounds and acetaldehyde and their significance in red wines. American Journal of Enology and Viticulture, 27, 97–105. doi:10.1016/S0065-2911(08)60040-5
• Versari A., du Toit W. and Parpinello G.P., 2013. Oenological tannins: a review. Australian Journal of Grape and Wine Research, 19, 1–10. doi:10.1111/ajgw.12002
• Vignault A., Gombau J., Pascual O., Jourdes M., Moine V., Canals J.M., Zamora F. and Teissedre P.L., 2019. Copigmentation of malvidin-3-O-monoglucoside by oenological tannins: Incidence on wine model color in function of botanical origin, pH and ethanol content. Molecules, 24, 1448. doi:10.3390/molecules24081448
• Vivas N., 2001. Les tanins oenologiques, d’hier à aujourd’hui: Une révolution discrète que nous devons assimiler dans les pratiques de chais. Revue des Oenologues et des Techniques Vitivinicoles et Oenologiques, 28(98), 11–14.