ANTHOCYANIN PATTERN OF SKIN EXTRACTS FROM THE BABIĆ AND PLAVAC MALI GRAPES AND ANTHOCYANIN PATTERN OF THE PRODUCED WINE

Acta Alimentaria, Vol. 38 (1), pp. 67 75 (2009) DOI: 10.1556/AAlim.2008.0029 First published online 13 January 2009 ANTHOCYANIN PATTERN OF SKIN EXTRACTS FROM THE BABIĆ AND PLAVAC MALI GRAPES AND ANTHOCYANIN PATTERN OF THE PRODUCED WINE I. BUDIĆ-LETO a *, U. VRHOVŠEK b, J. GAJDOŠ KLJUSURIĆ c and T. LOVRIĆ c a Institute for Adriatic Crops and Karst Reclamation, Put Duilova 11, HR-21000 Split. Croatia b IASMA Research Centre, Agrifood Quality Department, Via E. Mach 1, I-38010 San Michele all Adige. Italy c Faculty of Food Technology and Biotechnology, University of Zagreb, Pierottijeva 6, HR-10000 Zagreb. Croatia (Received: 17 July 2007; accepted: 28 April 2008) Anthocyanin composition and profile were analysed in skin extracts from red grapes and wines of the autochthonous Croatian cultivars of Babić and Plavac mali by HPLC-DAD-MS method. Higher relative abundance of malvidin-3-monoglucoside and lower relative abundance of peonidin-3-monoglucoside and cyanidin-3-monoglucoside were identified in wine in relation to their earlier abundance in grapes. Using principal component analysis (PCA), the wine samples showed differences between wine types, although the grape samples did not show any considerable differences concerning the anthocyanin composition. The main differences between Babić and Plavac mali wines were detected in a monoglucoside acetate (malvidin-3-monoglucoside acetate), a monoglucoside-caffeoate (malvidin-3- monoglucoside caffeoate) and a group of monoglucosides (malvidin-3-monoglucoside, delphinidin-3- monoglucoside, petunidin-3-monoglucoside and peonidin-3-monoglucoside). Keywords: grape, wine, anthocyanins, HPLC-DAD-MS, Babić, Plavac mali, PCA Anthocyanins the red, blue and violet pigments responsible for the colour of red grapes are often used for classification of cultivars and chemotaxic studies (ROGGERO et al., 1986; CALO et al., 1994; CARREÑO et al., 1997; AROZARENA et al., 2002). Anthocyanin composition of wine depends on grape anthocyanin composition, wine vinification techniques and a number of reactions occurring during wine ageing. Extraction of anthocyanins from grape into the wine depends on conditions present during the vinification process, such as duration of maceration, temperature, sulphur dioxide, presence of oxygen (DALLAS & LAUREANO, 1994; ROMERO-CASCALES et al., 2005; BUDIĆ-LETO et al., 2006), or implication for wine yeast (MONAGAS et al., 2007). * To whom correspondence should be addressed. Phone: +385 (0)21 434 420; fax: +385 (0)21 316 584; e-mail: irena@krs.hr 0139-3006/$ 20.00 2008 Akadémiai Kiadó, Budapest

68 BUDIĆ-LETO et al.: ANTHOCYANIN PATTERN OF SKIN EXTRACTS AND VINES On the other hand, anthocyanin concentration in wine varies because of its adsorption by yeasts, which occurs immediately upon completion of alcoholic fermentation, or because of condensation and polymerisation reactions, either among the anthocyanins or anthocyanins and other polyphenols (VASSEROT et al., 1997). The antioxidant activity and the phenol content shows the unevenness of different wines (GAJDOŠ KLJUSURIĆ et al., 2005; PILJAC-ZEGARAC et al., 2007). Babić and Plavac mali cultivars are very significant in the production of autochthonous high quality wines in Croatia. Plavac mali is also the most prevalent and economically most important red cultivar in Croatia. It has recently been proven that the Plavac mali cultivar is related to the American cultivar of Zinfandel, which has helped to determine that Plavac mali is its direct descendant, while Zinfandel is genetically identical to the Croatian cultivar of Kaštelanski crljenak (MALETIĆ et al., 2004). The objectives of this research are: (1) to examine the grape anthocyanin profile of the two Croatian autochthonous red grape cultivars Babić and Plavac mali, (2) to investigate the anthocyanin content of the wine produced from the mentioned grapes, (3) to identify correlations and determine possible variances between the analysed grape and wine anthocyanin profiles. 1. Materials and methods 1.1. Origin of Babić and Plavac mali grapes Three representative samples of each cultivar (9 kg) were selected at the technological stage of ripening during the 2001 harvest (Babić 105 Oe and 7.45 g l 1 of total acidity (TA) at the locations of Bucavac and Kremik and 103 Oe and 7.65 g l 1 of TA at the locations of Plošnjak, Jasenovik, Trovrh and Vadalj; and Plavac mali 99.5 Oe and 6.0 g l 1 of TA at the location of Dingač and 102 Oe and 6.0 g l 1 of TA at the location of Postup). 1.2. Winemaking For the production of wine a representative sample of 150 kg of grapes was selected from each location by random sampling. Pomace was sulphited with a 5%- solution of sulphite acid, which corresponded to the concentration of 50 mg l 1 of total sulphur(iv)-oxide and divided into three 70-litre vessels. Maceration was performed at the average temperature of 24 C during a period of six days, in three repetitions. Fermentation was spontaneous (without the addition of selected yeast). Pomace was punched every seven hours during the day and it was well mixed. Upon the completion of maceration, wine was transferred into 25-litre glass vessels. The first racking was performed in December, while the second racking was carried out in March. In May, the wine was racked into 0.75-litre glass bottles and closed with cork stoppers. The wines were analysed in August 2002.

BUDIĆ-LETO et al.: ANTHOCYANIN PATTERN OF SKIN EXTRACTS AND VINES 69 1.3. Preparation of grape skin extract Three representative samples of grapes were taken (9 kg each) out of which 200 g of representative berries were selected. Grape skin extracts were prepared by the selective method of extraction, according to MATTIVI and co-workers (2002) and analysed in August 2002. 1.4. HPLC diode array analysis Analysis of anthocyanins was performed using a Hewlett-Packard Model 1090 HPLC equipped with a Photodiode Array Ultraviolet (UV) Visible Detector and the HPChemstation software, according to the method of IACONO and co-workers (1994). Quantitative determination of compounds was carried out by the standard method with malvidin-3-glucoside (Extrasynthese, France). All anthocyanins were expressed as malvidin-3-glucosides. 1.5. HPLC-diode array detection-mass spectrometry analysis Confirmation of peak identity was performed as reported previously (PASSAMONTI et al., 2005) on Waters 2690 HPLC equipped with Waters 996 Diode Array Detector (Waters Corp., Milford, MA), Micromass ZQ ESI-MS system (Micromass, Manchester, UK), and Empower Software (Waters Corp.). 1.6. Statistical analysis Calculations were performed using the statistical application Statistica version 7.1. The LSD test yielded equal F and P-values, and the obtained P-value was then shown as P<0.001 (significance of 99.9%) and P>0.05 (no statistically significant variance). Principal Component Analysis (PCA) was used to detect the effect of the grape anthocyanin composition on the anthocyanin composition of wines. 2. Results and discussion 2.1. Anthocyanin structure of Babić and Plavac mali grape skin extracts Table 1 shows the relative abundance of various anthocyanins in grape skin extracts of the Croatian autochthonous cultivars of Babić and Plavac mali. Taking into account the total relative content of anthocyanin fractions, it can be seen that monoglucosides are the most frequent in both cultivars; they represent 86.1% in Babić and 84.7% in Plavac mali, while acylated derivatives represent 13.9% in Babić cultivar and 15.3% in Plavac mali cultivar. Malvidin-3-monoglucoside is known to be predominant in the anthocyanin structure of most Vitis vinifera cultivars (MATTIVI et al., 2006). It has been reported that the grape cultivars in which methylated anthocyanins are

70 BUDIĆ-LETO et al.: ANTHOCYANIN PATTERN OF SKIN EXTRACTS AND VINES predominant (delphinidin-3-monoglucoside, petunidin-3-monoglucoside and malvidin- 3-monoglucoside) are very pigmented and have greater potential to secure more stable wine colours (FERNANDEZ-LOPEZ et al., 1998). In Babić grape skin extracts an average of 58.3% of malvidin-3-monoglucoside was found. The ratio of other anthocyanins was: 11.1% of peonidin-3-monoglucoside, and 8.8% of petunidin-3-monoglucoside. Similar proportions were also found in Plavac mali grape skin extracts (Table 1). The total amount of anthocyanins in Babić grape skin extracts was 364.36 mg kg 1 at geographically lower locations by the sea (Bucavac and Kremik), and 547.13 mg kg 1 at geographically higher locations in the Primošten hinterland (Plošnjak, Jasenovik, Trovrh and Vadalj). The total amount of anthocyanins in Plavac mali grape skin extracts was 483.63 mg kg 1 at the Dingač location and 402.68 mg kg 1 at the Postup location. Table 1. Anthocyanin composition of the skin extract and wine of each cultivar (%) Anthocyanins Babić skin extract Babić wine Plavac mali skin extract Plavac mali wine Delphinidin-3-monoglucoside 5.87±0.58* 6.86±0.38 5.99±0.35 2.03±0.17 Cyanidin-3-monoglucoside 2.01±0.56 0.66±0.04 0.56±1.80 0.21±0.04 Petunidin-3-monoglucoside 8.81±0.44* 10.25±5.29* 7.93±0.30 6.64±0.04 Peonidin-3-monoglucoside 11.10±2.32 5.30±0.06 10.29±1.16 1.78±0.01 Malvidin-3-monoglucoside 58.25±2.66 62.19±0.82 58.74±2.45 73.06±0.20 Delphinidin-3-monoglucoside acetate 0.21±0.04 0.39±0.02* 0.19±0.02 0.51±0.27 Cyanidin-3-monoglucoside acetate nd 0.32±0.08* nd 0.21±0.05 Petunidin-3-monoglucoside acetate 0.26±0.03 0.38±0.03* 0.26±0.02 0.44±0.07 Peonidin-3-monoglucoside acetate 0.31±0.04 0.33±0.03 0.37±0.02 0.49±0.06 Malvidin-3-monoglucoside acetate 2.35±0.34 2.70±0.49* 3.55±0.29 4.96±0.16 Malvidin-3-monoglucoside-caffeoate 1.35±0.09* 0.25±0.05 1.20±0.05* 1.29±0.05 Delphinidin-3-monoglucoside p-coumarat 0.62±0.05 0.99±0.17 0.69±0.04 nd Cyanidin-3-monoglucoside p-coumarat nd 0.31±0.03 nd nd Petunidin-3-monoglucoside p-coumarat 0.78±0.01* 1.19±0.07 0.74±0.00* 0.64±0.02 Peonidin-3-monoglucoside p-coumarat 1.29±0.14 1.30±0.07 1.39±0.12 1.26±0.03 Malvidin-3-monoglucoside p-coumarat 6.88±0.57 6.61±0.44* 7.04±0.49 6.54±0.10 Total monoglucosides 86.07±0.93 85.25±0.71 84.75±1.07 83.73±0.07 Total monoglucoside acetates 3.14±0.22 4.11±0.26* 4.38±0.99 6.55±0.10 Total monoglucoside p-coumarats 9.38±0.61 9.40±0.55 9.66±0.71 8.44±0.13 Total anthocyanins (mg l 1 ) 455.75* 292.09* 443.16 187.61 * Show statistical differences (P<0.001). nd: not detected 2.2. Anthocyanin structure of the Babić and Plavac mali wines Examination of the anthocyanin structure of wine (Table 1) shows that the anthocyanin fingerprint of the wine is different from that of the grape skin extract. Several diversities were noted: cyanidin-3-monoglucoside-acetate and cyanidin-3-monoglucoside-pcoumarat, which were not present in the grape skin extracts, were detected in the wines

BUDIĆ-LETO et al.: ANTHOCYANIN PATTERN OF SKIN EXTRACTS AND VINES 71 of both cultivars. As shown in Table 1, higher abundance of malvidin-3-monoglucoside and lower abundances of peonidin-3-monoglucoside and cyanidin-3-monoglucoside were noted in the wine in comparison with the grape skin extracts. Malvidin-3- monoglucoside was the predominant anthocyanin in the Babić wine, amounting to 62.1%, and in the Plavac mali wine, with 73.1%. Comparison of the obtained data reveals that the content of malvidin-3-monoglucoside was higher by 3.8% in the Babić wine and by 14.4% in the Plavac mali wine in comparison with their content in grape skin extracts. Research of some other cultivars (REVILLA et al., 2001; ROMERO- CASCALES et al., 2005) also showed that wine contained higher proportions of malvidin- 3-monoglucoside in relation to grapes. Relative abundance of peonidin-3-monoglucoside was significantly lower in wine than in grape skin extract (Table 1). Thus, in the Plavac mali wine the low percentage of peonidin-3-monoglucoside was 1.8%, while the percentage of peonidin-3- monoglucoside previously identified in grapes was 10.3%. Considering the reactivity of anthocyanin fractions, it can be assumed that anthocyanins polymerize after extraction from grapes, which can be the cause of high reduction of the peonidin-3-monoglucoside percentage (REVILLA et al., 2001; ROMERO-CASCALES et al., 2005). The percentage of cyanidin-3-monoglucoside in wine was even lower than in grape skin extract. The obtained results could also have been caused by polymerization of cyanidin-3- monoglucoside, as revealed by some research (GARCÍA-BENEYTEZ, 2002). In order to discriminate the samples of grapes and wines, the Principal Component Analysis (PCA) was used. Comparison of the complete monitored anthocyanin patterns has shown that correlation of cultivation locations of the grape Babić (location 1: Bucavac and Kremik; location 2: Jasenovik, Plošnjak, Trovrh and Vadalj) is very high (r=0.993), as well as of the grape Plavac mali (r=0.999, for location 1: Dingač and location 2: Postup). This was an indicator that in observing the anthocyanin pattern of grape cultivation used in this study, all locations of grape Babić should be considered as one location (location 1) and for the grape Plavac mali, two locations should also be considered as one, location 2. To identify possible differences in anthocyanin patterns between grapes and wines the PCA seemed to be the best method for separation of grape skin extracts and different wines in order to determine the anthocyanin composition of grape skin extracts and wines. As presented in Fig. 1, the grapes have a similar anthocyanin composition, which can also be seen in Table 1, but after winemaking, the composition of the observed anthocyanins had changed and led to possible distinction of wines Plavac mali and Babić.

72 BUDIĆ-LETO et al.: ANTHOCYANIN PATTERN OF SKIN EXTRACTS AND VINES Fig. 1. Scores of principal component analysis using the anthocyanin composition of grape skin and wines, Babić grape skin (grape skin 1), Plavac mali grape skin (grape skin 2) To discriminate the main changes of anthocyanin composition from grape to wine, the observed anthocyanins from grapes and wines were compared using PCA (Fig. 2). In Fig. 2 the space distribution of the scores in the plane is defined by two PC functions. The differentiation of the samples using PCA considers the percentage of different anthocyanin components in the grape skin. The samples of grape skins and Babić wine were mainly differentiated based on the percentages of Cy-3-gl-pcum, Mv-3-gl, Cy-3-gl, Pn-3-gl and Mv-3-gl-ac (Fig. 2A) (for abbreviations see Fig. 2). The first PC function accounts for 90.53% of the variability and the second one for 9.38%. The higher influences on the first function are the percentages of Cy-3-gl-pcum and Mv-3-gl, whereas for the second function the variables with higher influence are the percentages of Cy-3-gl and Mv-3-gl-ac. The same differentiation as presented in Fig. 2A was performed for Plavac mali grape and wine samples (Fig. 2B). The samples of Plavac mali grape skins and wine were differentiated based on the percentage of Cy-3-ac, Mv-3-gl, Pt-3-gl, Dp-3-gl, Cy-3-glpcum, Pn-3-gl and Mv-3-gl-ac. In the comparison of the results of two wines, the results clearly indicate the percentages of Mv-3-gl, Dp-3-gl, Pt-3-gl, Pn-3-gl, Mv-3-gl-ac and Mv-3-gl-caf as differentiators of the anthocyanin composition of the examined wines.

BUDIĆ-LETO et al.: ANTHOCYANIN PATTERN OF SKIN EXTRACTS AND VINES 73 A B Fig. 2. Changes during vinification from: A) grape 1 to Babić wine and B) grape 2 to Plavac mali wine, presented using principal component analysis. Dp-3-gl: delphinidin-3-monoglucoside; Cy-3-gl: cyanidin-3- monoglucoside; Pt-3-gl: petunidin-3-monoglucoside; Pn-3-gl: peonidin-3-monoglucoside; Mv-3-gl: malvidin- 3-monoglucoside; Dp-3-gl-ac: delphinidin-3-monoglucoside acetate; Cy-3-gl-ac: cyanidin-3-monoglucoside acetate; Pt-3-gl-ac: petunidin-3-monoglucoside acetate; Pn-3-gl-ac: peonidin-3-monoglucoside acetate; Mv-3- gl-ac: malvidin-3-monoglucoside acetate; Dp-3-gl-pcum: delphinidin-3-monoglucoside-p-coumarat; Cy-3-glpcum: cyanidin-3-monoglucoside-p-coumarat; Pt-3-gl-pcum: petunidin-3-monoglucoside-p-coumarat; Pn-3- gl-pcum: peonidin-3-monoglucoside-p-coumarat; Mv-3-gl-pcum: malvidin-3-p-monoglucoside-p-coumarat; Mv-3-gl-caf: malvidin-3-monoglucoside-caffeoate

74 BUDIĆ-LETO et al.: ANTHOCYANIN PATTERN OF SKIN EXTRACTS AND VINES The process of the production of Babić and Plavac mali wines is the same, and from the PCA results, it seams that the differences arise from different polyphenolic composition of wines which leads to different polymerization and condensation reactions either among the anthocyanins or anthocyanins and other flavonoids, but verifying these assumptions should be the subject of future research. 3. Conclusions Malvidin-3-monoglucoside is the predominant anthocyanin in the grapes and wines of the Babić and Plavac mali cultivars. Anthocyanin pattern of wine are different from the anthocyanin patterns of grapes. The samples of grape skins and Babić wine were mainly differentiated based on the percentages of malvidin-3-monoglucoside, cyanidin-3- monoglucoside, malvidin-3-monoglucoside acetate and cyanidin-3-monoglucoside-pcoumarat. Samples of grape skins and Plavac mali wine were differentiated based on the percentages of malvidin-3-monoglucoside, petunidin-3-monoglucoside, delphinidin-3- monoglucoside, malvidin-3-monoglucoside acetate cyanidin-3-monoglucoside acetate and cyanidin-3-monoglucoside-p-coumarat. The results obtained using principal component analysis reveal that Babić and Plavac mali as grape cultivars are not so different considering the anthocyanin pattern, but the winemaking process leads to the conclusion that wines can be distinguished according to the same pattern. It seems that beside the anthocyanin pattern, some other grape characteristics influence the differences in wine quality. * This work was supported by the Ministry of Science, Technology and Sports of the Republic of Croatia. Anthocyanin analyses were done in the laboratory of Fulvio Mattivi (IASMA, Italy), which is gratefully acknowledged. References AROZARENA, I., AYESTARÁN, B., CANTALEJO, M., NAVARRO, M., VERA, M., ABRIL, I. & CASP, A. (2002): Anthocyanin composition of Tempranillo, Garnacha and Cabernet Sauvignon grapes from high- and lowquality vineyards over two years. Europ. Fd Res. Technol., 214, 303 309. BUDIĆ-LETO, I., LOVRIĆ, T., GAJDOŠ KLJUSURIĆ, J., PEZO, I. & VRHOVŠEK, U. (2006): Anthocyanin composition of the red wine Babić affected by maceration treatment. Europ. Fd Res. Technol., 222, 397 402. CALO, A., TOMASI, D., CRAVERO, M.C. & DI STEFANO, R. (1994): Contributo alla caratterizzazione e classificazione varietale (Vitis sp.) attraverso la determinazione degli antociani e degli acidi idrossicinnamoiltartarici della buccia a bacca rossa. (Varietal analysis and classification of the species Vitis by determination of anthocyanins and of hydroxycinnamoyl tartaric acids in the skin of the red-berry cultivars.) 3, 13 25.

BUDIĆ-LETO et al.: ANTHOCYANIN PATTERN OF SKIN EXTRACTS AND VINES 75 CARREÑO, J., ALMELA, L., MARTINEZ, A. & FERNANDEZ-LOPEZ, J.A. (1997): Chemotaxonomical classification of red table grapes based on anthocyanin profile and external colour. Lebensm.-Wiss. Technol., 30, 259 265. DALLAS, C. & LAUREANO, O. (1994): Effect of SO 2 on the extraction of individual anthocyanins and colored matter of three Portuguese grape varieties during winemaking. Vitis, 33, 41 47. FERNANDEZ-LOPEZ, J.A., ALMELA, L., MUÑOZ, J.A., HIDALGO, V. & CARREÑO, J. (1998): Dependence between colour and individual anthocyanin content in ripening grapes. Fd Res. int., 31, 667 672. GAJDOŠ KLJUSURIĆ, J., DJAKOVIĆ, S., KRUHAK, I., KOVAČEVIĆ GANIĆ, K., KOMES, D. & KURTANJEK, Ž. (2005): Application of Briggs-Rauscher reaction for measurement of antioxidant capacity of Croatian wines. Acta Alimentaria, 34, 483 492. GARCÍA-BENEYTEZ, E., REVILLA, E. & CABELLO, F. (2002): Anthocyanin pattern of several red grape. cultivars and wines made from them. Europ. Fd Res. Technol., 215, 32 37. IACONO, F., BERTAMINI, M., MATTIVI, F. & SCIENZA, A. (1994): Differential effects of canopy manipulation and shading of Vitis vinifera L. cv. Cabernet Sauvignon. I. Composition of grape berries. Wein Wissenschaft, 49, 220 225. MALETIĆ, E., PEJIĆ, I., KAROGLAN KONTIĆ, J., PILJAC, J., DANGL, G.S., VOKURKA, A., LACOMBE, T., MIROŠEVIĆ, N. & MEREDITH, C. (2004): Zinfandel, Dobričić, and Plavac mali: The Genetic Relationship among Three Cultivars of the Dalmatian Coast of Croatia. Am. J. Enol. Vitic., 55, 174 180. MATTIVI, F., PRAST, A., NICOLINI, G. & VALENTI, L. (2002): Validazione di un nuovo metodo per la misura del potenziale polifenolico delle uve rosse e discussione del suo campo di applicazione in enologia. (Validation of a new method to measure the polyphonic potential of red grapes and discussion of its use in ecology.) Riv. Viticoltura e di Enologia, 55, 55 74. MATTIVI, F., GUZZON, R., VRHOVSEK, U., STEFANINI, M. & VELASCO, R. (2006): Metabolite profiling of grape: Flavonols and anthocyanins. J. agric. Fd Chem., 54, 7692 7702. MONAGAS, M., GOMEZ-CORDOVES, C. & BARTOLOME, B. (2007): Evaluation of different Saccharomyces cerevisiae strains for red winemaking. Influence on the anthocyanin, pyranoanthocyanin and nonanthocyanin phenolic content and colour characteristics of wines. Fd Chem., 104, 814 823. PASSAMONTI, S., VRHOVSEK, U., VANZO, A. & MATTIVI, F. (2005) Fast access of some grape pigments to the brain. J. agric. Fd Chem., 53, 7029 7034. PILJAC-ZEGARAC, J., MARTINEZ, S., VALEK, L., STIPCEVIC, T. & KOVACEVIC-GANIC, K. (2007): Correlation between the phenolic content and DPPH radical scavenging activity of selected Croatian wines. Acta Alimentaria, 36, 185 193. REVILLA, E., GARCIA-BENEYTEZ, E., CABELLO, F., MARTIN-ORTEGA, G. & RYAN, J.M. (2001): Value of high-performance liquid chromatographic analysis of anthocyanins in the differentiation of red grape cultivars and red wines made from them. J. Chromat. A, 315, 53 60. ROGGERO, J.P., COEN, S. & RAGONET, B. (1986): High performance liquid chromatography survey on changes in pigment content in ripening grapes of Syrah. An approach to anthocyanin metabolism. Am. J. Enol. Vitic., ROMERO-CASCALES, I., FERNÁNDEZ-FERNÁNDEZ, J.I., LÓPEZ-ROCA, J.M. & GÓMEZ-PLAZA, E. (2005): The maceration process during winemaking extraction of anthocyanins from grape skins into wine. Europ. Fd Res. Technol., 221, 163 167. VASSEROT, Y., CAILLET, S. & MAUJEAN, A. (1997): Study of anthocyanin adsorption by yeast lees. Effect of some physicochemical parameters. Am. J. Enol. Vitic., 48, 433 437.