Abstract
The evolution of malvidin-3-glucoside and (+)-catechin in the presence of ellagic acid and oak wood extracts and oxygen was studied in model wine over a period of 68 days. Malvidin-3-glucoside declined more rapidly in the presence of oak wood extract than alone. This difference was independent of oxygen. The presence of ellagic acid or oak wood extract had a positive effect on (+)-catechin decline. For color parameters (using the CIELAB coordinates), L* values showed a higher decrease as a consequence of oak wood presence. For a* values, a decrease was observed in solutions containing malvidin-3-glucoside and oak wood extract independent of the presence or absence of oxygen in the solutions. An increase in yellow color (b* values) was measured in all solutions containing oak wood extract.
The use of oak barrels in winemaking is thought to improve wine quality. Many constituents can be extracted from staves during aging in barrels, such as ellagitannins (Viriot et al. 1993), tannins, ferulic, vanillic, syringic, and ellagic acids, coumarins; and volatile compounds (Canas et al. 1999, Jordão et al. 2005a). For some authors, the presence of ellagic acid is regarded as a characteristic compound of barrel wine aging, and its concentration in wine may be a marker for barrel storage time (Matejícek et al. 2005).
The barrel is a porous vessel, which allows the continual flow of traces of oxygen. The oxidation in barrels is responsible for the slow evolution of wine. The slow dissolution of oxygen involves several redox reactions and leads to the formation of unstable peroxide compounds in wine. The capacity of red wine to take up oxygen has been measured (6 mL/L or 8 mL/L) (Singleton 1987). A recent study reported differences in redox potential between wood-matured ports and nonwood matured ports and indicated that there were differences in dissolved oxygen that could have affected the maturation rate of these wines (Ho et al. 2001). Thus, the pigmented polymers formed during wood maturation contributed to increases in color density, hue, and rate of browning.
Ellagic tannins have an important role in red wine aging (Singleton 1995, Vivas and Glories 1996) and may be involved in both red and white wine oxidation mechanism (Moutounet et al. 1989). During wine oxidation, these compounds rapidly react with dissolved oxygen and facilitate the hydroperoxidation of wine constituents (Vivas and Glories 1996). Oak wood ellagic tannins also affect pro-anthocyanidin condensation rate and anthocyanidin destruction (Vivas and Glories 1993).
Acetaldehyde, which derives from ethanol oxidation (Singleton 1987), is involved in the copolymerization of flavonoids and anthocyanins, increasing stable red polymer (Dallas et al. 1996, Castellari et al. 2001). A direct condensation of anthocyanin flavylium ions with (+)-catechin or tannins may also occur, improving the blue-red color (Timberlake and Bridle 1976).
One of the greatest wine transformations occurring during oak wood aging is the evolution of color. Thus, during wine-barrel aging, anthocyanins and tannins undergo several changes that may lead to color loss or stabilization. Oak wood barrels promote the formation of anthocyanin derivatives, such as vitisins, and malvidin-3-glucoside-catechin adducts (González-Sanjosé and Revilla 2001). The change from the purple red of young wines to the tawny hue observed in older red wines is attributed to the conversion of anthocyanins to new, more stable pigmented polymers. These modifications are more pronounced in wines stored in wood (Pérez-Prieto et al. 2003).
Despite the advances made in the identification of wood constituents in wine (Moutounet et al. 1989, Vivas et al. 2004) and the kinetics of their extraction (Puech et al. 1996, Jordão et al. 2005b), there have been few studies on the effect of oak constituents and oxygen on red wine phenolic compound evolution. Therefore, a model wine solution was used to evaluate the role of oak wood extract (from Quercus pyrenaica), ellagic acid, and oxygen, on malvidin-3-glucoside, (+)-catechin, and color evolution by the CIELAB method (C.I.E. 1986). The influence of oxygen on red wine phenolic compounds and color parameters, emphasizing its effect on the malvidin-3-glucoside and (+)-catechin during 68 days of storage, was also studied.
Materials and Methods
Samples.
Malvidin-3-glucoside, ellagic acid, and (+)-catechin were obtained from Polyphenols SA Laboratories (Sandnes, Norway), Fluka Biochemika (Buchs, Switzerland), and Aldrich (Paris, France), respectively. Oak wood chips used were obtained from Quercus pyrenaica (Portuguese oak wood, from Gerês region) with medium grain (3.0 to 3.5 mm) and medium toasting (20 min at 160 to 170°C on the wood surface). In order to reproduce extraction conditions similar to those in wine, the oak wood chip samples (20 g/L) were placed in 500 mL of model alcohol solution (adjusted to pH 3.5 with 2 g/L of tartaric acid and 12% alcohol content) for 15 days at 20°C (± 2) in the dark and stirred daily. This oak chip concentration was higher than commonly used in wineries, but it provided a better opportunity to monitor treatment differences. At the end of extraction, the oak extract was filtered through glass wool before use.
Model wine solutions.
Malvidin-3-glucoside, ellagic acid, and (+)-catechin were dissolved in a solution containing 12% (v/v) ethanol and 2 g/L tartaric acid adjusted to pH 3.5. Final concentrations of 50 mg/L for malvidin-3-glucoside and ellagic acid and 40 mg/L for (+)-catechin were targeted. Solutions were prepared, pipetted (10 mL of each sample in 15-mL vials), and sealed. For oxygen treatments, vials were saturated with oxygen (O2) or nitrogen (N2) and then sealed.
Fifteen different experimental mixtures were prepared in duplicate, filtered (0.45 μm), and dispensed into test tubes as follows: (1) malvidin-3-glucoside alone; (2) (+)-catechin alone; (3) oak extract; (4) malvidin-3-glucoside + (+)-catechin; (5) malvidin-3-glucoside + oak extract; (6) (+)-catechin + oak extract; (7) malvidin-3-glucoside + oak extract + (+)-catechin; (8) malvidin-3-glucoside + O2; (9) malvidin-3-glucoside + N2; (10) malvidin-3-glucoside + oak extract + O2; (11) malvidin-3-glucoside + oak extract + N2; (12) (+)-catechin + O2; (13) (+)-catechin + N2; (14) (+)-catechin + ellagic acid + O2; (15) (+)-catechin + ellagic acid + N2. All samples were kept in the dark for 68 days at controlled temperature (20°C, ± 2) and analyzed in duplicate.
Color measurement.
Spectral readings transmittance every 10 nm over the visible spectrum from 370 to 770 nm were performed with a diode-array UV-visible Unican UV4 spectrometer (Unicam, Cambridge, UK). This spectrophotometer had the appropriate software (Chroma 2.0 version; Unicam) to calculate CIELAB parameters directly. CIELAB and 10° parameters were calculated for CIE illuminant D65 standard conditions: L* (lightness), a* (from green to red), and b* (from blue to yellow). All parameters were measured in duplicate.
HPLC analysis.
A PerkinElmer system (Shelton, CT) was used for analysis of malvidin-3-glucoside, ellagic acid, and (+)-catechin evolution. It was equipped with a 410-LC pump, a solvent programmer (model 420), and a manual injector (Rheodyne 7125-A) fitted with a 20-μL loop. The column (250 x 4.6 mm, particle size 5 μm) was a C18 Li-Chrospher 100 (Merck, Darmsdadt, Germany) protected by a guard column of the same material. Detection was performed with a Konik detector coupled to a Konichrom data treatment station (Konik Instruments, Miami, FL).
Malvidin-3-glucoside and (+)-catechin HPLC were analyzed using a previously described method (Dallas et al. 1996). Ellagic acid was monitored using a previous described method (Viriot et al. 1994). All analyses were performed in duplicate.
Results and Discussion
Evolution of phenolic compounds.
During the storage period, in all model solutions a general decrease of (+)-catechin and malvidin-3-glucoside was observed (Figure 1⇓). Under our experimental conditions, the decrease in (+)-catechin was always lower in model wine solutions with oak extract than when alone or in a mixture containing malvidin-3-glucoside. After 68 days, the concentration of (+)-catechin alone and with malvidin-3-glucoside was 6.3 and 9.4 mg/L, respectively. In contrast and over the same period, 17.5 mg/L was reported for (+)-catechin solutions in the presence of oak and malvidin-3-glucoside and 16.0 mg/L for solutions containing oak. Despite the effect of malvidin-3-glucoside on (+)-catechin evolution, reported by several authors (Dallas et al. 1996, Salas et al. 2003), our results indicate that oak extract reduces the rate of (+)-catechin decline. This reduction may be due to the antioxidant activity of oak wood components, especially ellagitannins and ellagic acid (Guerra et al. 1996, Vivas and Glories 1996).
The presence of oak extract and (+)-catechin influenced the rate of malvidin-3-glucoside decline. After 68 days, the concentration of malvidin-3-glucoside in solutions containing malvidin-3-glucoside alone and in mixtures with oak was 19.0 and 5.2 mg/L, respectively; indicating that oak extract increased the rate of anthocyanin degradation.
Influence of oxygen.
During winemaking and wine maturation, oxygen plays an important role in phenolic evolution, particularly during maturation in oak wood barrels. The presence of oxygen during wine maturation is an important factor in anthocyanin condensation because the acetaldehyde produced by the oxidation of ethanol is essential in this condensation process (Wildenradt and Singleton 1974). In addition, temperature, pH, light, and enzymes influence anthocyanin and color evolution (Cabrita et al. 2000, Dallas et al. 2003, Torskangerpoll and Andersen 2005).
Influence of oxygen on malvidin-3-glucoside.
A progressive decrease in malvidin-3-glucoside occurred in all treatments (Figure 2⇓) and was more evident in the first 24 days. This decrease was greater in the presence of oxygen. However, the presence of oak extract induced the greatest malvidin-3-glucoside decrease, independent of oxygen. These results indicate that the presence of oak extract, and not simply the oxygen levels, was one important factor that determined the decrease in malvidin-3-glucoside. Most likely, in solutions containing oak extract and high oxygen values, two mechanisms for malvidin-3-glucoside degradation were operating.
Influence of oxygen and ellagic acid on (+)-catechin.
The influence of ellagic acid, a particular oak wood compound, on (+)-catechin evolution in model wine solutions was also investigated (Figure 3⇓). As expected, a progressive decrease of (+)-catechin occurred in all model wine solutions during storage. This decrease was more evident for oxygenated solutions containing (+)-catechin alone than in solutions with ellagic acid. Ellagic acid influenced (+)-catechin degradation. However, opposite findings were observed with respect to the behaviors of malvidin-3-glucoside in similar solutions. Several studies found that compounds extracted from barrel staves are much more easily oxidized than the majority of grape-based constituents in wine (Guerra et al. 1996, Vivas and Glories 1996).
Results showed a progressive decrease in ellagic acid concentration in all model wine solutions (Figure 3⇑). In general, there were differences in ellagic acid concentration in solutions containing only ellagic acid and those containing ellagic acid and (+)-catechin. The decrease was more evident in solutions containing ellagic acid and (+)-catechin than in solutions containing ellagic acid alone. Additionally, in all solutions containing ellagic acid, precipitation was not observed. Direct and indirect oxidation may explain these observations.
Monitoring of color parameters.
The evolution in color (expressed by the CIELAB coordinates) is shown in Table 1⇓. In general, values for L* (lightness) remained nearly constant during the storage period, except in solutions containing the phenolic compounds and oak extract, where a decrease was observed.
The CIELAB parameter a* (from green to red) and b* (from blue to yellow) was also measured in solutions (Table 1⇑). A decrease in a* values occurred in model wine solutions containing malvidin-3-glucoside, except for binary system containing malvidin-3-glucoside and (+)-catechin. The decrease in values was more evident for malvidin-3-glucoside solutions containing oak extract, even though these solutions showed the highest values after one day of storage. In solutions containing malvidin-3-glucoside alone, the a* values ranged from to 4.2 to 3.0, while for malvidin-3-glucoside solution containing oak extract, the a* values ranged from 7.9 to −0.03, for which the red color had completely disappeared after 68 days. Some authors have found that the rapid polymerization between anthocyanins and tannins in the presence of other derivatives, such as aldehydes (transferred from oak wood for example), increases color (Timberlake and Bridle 1976, Es-Safi et al. 2003). However, reactions with polymerized flavanols may lead to instability, precipitation, and, consequently, a decrease in color. It has been reported that numerous anthocyanin adducts are formed during red wine aging in oak wood barrels (Revilla and González-Sanjosé 2001). These new compounds presumably have an influence on the chromatic characteristics of wine.
Solutions containing oak extract, (+)-catechin with oak extract, and malvidin-3-glucoside with oak extract showed the highest b* values during storage. Oxygen also influenced the color evolution of solutions containing malvidin-3-glucoside and oak extract (Table 1⇑). Several authors reported that the presence of oxygen affects red wine maturation, especially red color evolution (Vivas and Glories 1996, Revilla and González-Sanjosé 2001). After a vigorous aeration, a decrease in red wine color intensity can occur (Ribéreau-Gayon et al. 1983).
Under our experimental conditions, the decrease in red color (a* values) was higher in oxygen-saturated solutions and in the presence of oak extract. However, these results showed that the presence of oak wood in solution had the same effect on red color as the presence of high oxygen levels, suggesting that the presence of oak wood components induce red color degradation (as a consequence of malvidin-3-glucoside decrease) similar to that produced by the oxidation process.
In general, the b* values obtained for malvidin-3-glucoside solutions with oxygen and nitrogen saturation were located between the area of blue (negative values) and yellow hues (positive values). For the b* values in mixtures containing malvidin-3-glucoside and oak extract with oxygen and nitrogen saturation, higher values were observed. These values increased during storage and were located in the yellow area, especially for solutions saturated with oxygen.
Conclusions
Results showed that the presence of oak extract increased malvidin-3-glucoside losses in model wine solutions. This decrease was independent of the presence or absence of oxygen. It appears that some oak wood components may have an effect on anthocyanin content and consequently increase the losses of the pigment. In addition, the presence of ellagic acid had an important role against (+)-catechin degradation, especially in solutions containing these two compounds in an oxygen-saturated medium. Results of the color parameters studied in solutions using the CIELAB method showed a greater decrease of L* and a* values as a consequence of oak wood presence in the solutions, independent of oxygen. This study suggests that oxygen could play an important role on color evolution during storage. Further studies to evaluate the role of oxygen and other factors on wine color change in oak barrels are needed.
Footnotes
Acknowledgments: The authors thank the company Tanoaria J.M. Gonçalves Lda (Palaçoulo, Portugal), for supplying oak wood chips. This study was sponsored by the Portuguese Science Ministry (Project POCTI/AGR/ 36168/2000).
- Received September 2005.
- Revision received November 2005.
- Revision received January 2006.
- Copyright © 2006 by the American Society for Enology and Viticulture