Food Research International
Changes in organic acids, polyphenolic and elemental composition of rosé sparkling wines treated with mannoproteins during over-lees aging
Saionara Sartor, Isabela M. Toaldo, Carolina P. Panceri, Vinícius Caliari, Aderval S. Luna, Jefferson S. de Gois, Marilde T. Bordignon-Luiz
Changes in organic acids, polyphenolic and elemental composition of rosé sparkling wines treated with mannoproteins during over-lees aging
Saionara Sartor1, Isabela M. Toaldo1, Carolina P. Panceri1,2, Vinícius Caliari3,4, Aderval S. Luna5, Jefferson S. de Gois5 and Marilde T. Bordignon-Luiz1,*
1Department of Food Science and Technology, Federal University of Santa Catarina, Admar Gonzaga Rd. 1346, 88034-001, Florianópolis, Santa Catarina, Brazil. [email protected]; [email protected]
2Federal Institute of Santa Catarina, Senadinho St., 88625-000, Urupema, Santa Catarina, Brazil. [email protected]
3Agricultural Research and Rural Extension Company of Santa Catarina, João Zardo Rd. 1660, 89560-000, Videira, Santa Catarina, Brazil. [email protected]
4University of West Santa Catarina, Paese St. 198, 89560-000, Videira, Santa Catarina, Brazil.
5Department of Analytical Chemistry, Rio de Janeiro State University, São Francisco Xavier, 524, Maracanã, 20550-900 Rio de Janeiro, RJ, Brazil. [email protected]; [email protected]
Corresponding author: Prof. Dr. Bordignon-Luiz, M. T. [email protected], (+55) 48 3721-5376.
Abstract
The effect of mannoproteins on the evolution of rosé sparkling wines during over-lees aging was investigated on the basis of the chemical characterization of polyphenols, organic acids, macro- and microelements using a combined analytical approach.
Variations on these constituents were assessed using Raman and near-infrared spectroscopy. During the biological aging, caffeic acid, catechin, gallic acid and malvidin- 3-O-glucoside were the most abundant polyphenolics in the rosé wines. The phenolic compound tyrosol, a fermentation derivative, was found at concentrations up to 98.07 mg L-1. The addition of mannoproteins significantly affected the concentrations of organic acids and individual polyphenolic compounds, particularly trans-resveratrol, quercetin, catechin, p-coumaric and hydroxybenzoic acids that showed increased concentrations over time. The positive effects of mannoproteins were mainly observed at the end of the biological aging. The mineral composition remained stable, while potassium was the most abundant mineral in all wines. The observed changes involving these constituents may offer new insights on their behavior during wine aging and on the bioactive and nutritional quality of rosé sparkling wines.
Keywords: Sparkling wines; Aging sur lie; Polyphenols; Minerals; NIR; Raman spectroscopy.
Chemical compounds studied in this article:
Malvidin-3-O-glucoside (PubChem CID:443652); Gallic acid (PubChem CID:370); Ellagic acid (PubChem CID: 5281855); Caffeic acid (PubChem CID: 689043); p- coumaric acid (PubChem CID: 637542); (-)-epicatechin (PubChem CID:72276); (+)- catechin (PubChem CID:9064); Quercetin (PubChem CID:5280343); Myricetin (PubChem CID: 5281672); trans-Resveratrol (PubChem CID:445154)
1. Introduction
Wines and sparkling wines bring out the characteristics of varietal grapes and as such are influenced by botanical, cultivation, climatic and technological factors. Many naturally occurring chemicals such as polyphenolic substances, sugars, acids and inorganic constituents are responsible for the transformations and stability of wines and are an essential aspect of wine evolution (Prakash et al., 2016). In the case of sparkling wines, a second fermentation takes place during the wine process and contributes immensely to the sensory complexity of these wines. The sparkling wines are produced from an initial base wine that undergoes a second fermentation in sealed bottles or in large containers leading to CO2 incorporation and release of secondary products of yeast fermentative metabolism (Caliari, Panceri, Rosier, & Bordignon-Luiz, 2015). During biological aging, the sparkling wines are left in contact with the yeast cells deposit in the over-lees or “sur lie” process.
The transformations due to yeast autolysis affect the body, mouthfeel and aromatic attributes of the final sparkling wine (Pérez-Magariño et al., 2015).
It is a common practice in wine industry the use of oenological agents to improve quality of sparkling wines. Many agents constituting yeast lysates, antioxidants and pectinolytic mixtures, tannins, and mannoproteins are commercially used with the aim of controlling wine evolution and can mainly affect the chemical characteristics of varietal wines (Del Barrio-Galán et al., 2012; Ghanem et al., 2017; Guadalupe, Palacios, & Ayestarán, 2007; Pérez-Magariño et al., 2015). Mannoproteins are particularly associated with a greater stability of aged wines as well as promoting technological and sensorial properties. These proteoglycans are mainly constituted of mannose, glucose and proteins and are released from yeast cell wall during the autolysis process (Guadalupe & Ayestarán, 2008).
Mannoproteins have different structures depending on their molecular weight, degree and type of glycosylation and their charges. Depending on how they are extracted, they have different tartaric or protein stabilization activities. Its addition is a common practice in winemaking allowed by the International Oenological Codex of Vine and Wine (OIV, 2018). Yeast cell wall mannoproteins play an important role in the winemaking process. When added to the wine they can promote different technological properties, such as a positive effect on foam stability and on sensorial characteristics, namely the reduction of astringency (Guadalupe, Palacios, & Ayestarán, 2007) and improvement of the aromatic profile of sparkling wines (Pérez-Magariño et al., 2015). The interaction between mannoproteins and wine phytochemicals is a subject of great interest (Guadalupe & Ayestarán, 2008) that has yet to be better elucidated regarding sparkling wines.
Apart from grapevine and varietal influence, there is now a growing interest toward modifications that lead to the compositional profile of wines and the determinant interactions that contribute to their typicality. The characterization of natural constituents has been proposed as method of authentication of grapes and wines (Sen, Ozturk, Tokatli, & Ozen, 2016; Snyder, Sweeney, Rodriguez-Saona, & Giusti, 2014). Together with the bioactive potential of polyphenols, the nutritional value of wines much depends on its multi-elemental content. The wine is a source of both essential and potentially toxic elements. Its composition is influenced by different environmental factors, including soil composition, viticulture practices and winemaking processes (Hopfer et al., 2015; Ribéreau-Gayon, Glories, Maujean, & Dubourdieu, 2006).
The use of analytical techniques for quality evaluation of wines is indispensable and well established in research studies. Liquid chromatography coupled to UV/Vis/diode array detection (LC-UV/Vis/DAD) or mass spectrometry (MS) has been ideally used for characterization of phytochemical composition of grapes and wines (Lucci, Saurina, &
Núñez, 2017). These are generally suited for geographical, variety and vintage discrimination or fermentation/aging monitoring (Bai et al, 2013; Ivanova-Petropulos et al., 2015; Sartor et al., 2017). Vibrational spectroscopy techniques, on the other hand, may provide useful information regarding changes in wine composition since the formation of new compounds can modify the spectra of the samples. Moreover, these techniques present a relatively low-cost, straightforward and non-destructive analysis approach for wine quality control (Snyder, Sweeney, Rodriguez-Saona, & Giusti, 2014).
A variety of chemical modifications are originated in the course of vinification and can influence the authenticity and quality of sparkling wines, given their second fermentation and yeast contact, along with the complexity of their constituents. Apart from the phenolic composition, the modifications of co-occurring organic and inorganic constituents can be particularly determinant of the bioactive and nutritional quality of these wines. Therefore, the aim of this work was to use chemical characterization and vibrational analysis to evaluate the effect of the addition of mannoproteins on the polyphenolic, organic acid and elemental composition of rosé sparkling wines and their chemical changes during over-lees aging.
2. Material and methods
2.1. Chemicals
Analytical standards of organic acids (tartaric, malic, lactic, succinic and citric acids) (≥ 95% purity), polyphenolic compounds (malvidin-3-O-glucoside, delphinidin-3- O-glucoside, cyanidin-3-O-glucoside, peonidin-3-O-glucoside, gallic acid, protocatechuic acid, vanilic acid, syringic acid, ellagic acid, trans-caftaric acid, caffeic acid, p-coumaric
acid, ferulic acid, myricetin, quercetin, kaempferol, (+)-catechin, (−)-epicatechin, trans- resveratrol and tyrosol, all ≥ 90% purity) and reagents 2,2-azino-bis (3- ethylbenzothiazoline-6-sulphonic acid) (ABTS) and 6-hydroxy-2,5,7,8- tetramethylchroman-2-carboxylic acid (Trolox) were obtained from Sigma-Aldrich (St. Louis, Missouri, USA). Standard solutions (1000 mg L-1) of elements (Ca, K, Mg, Fe, Mn, Zn, Sr and Cu) were obtained from Spex Certiprep Chemical (Metuchen, New Jersey, USA). Analytical grade nitric acid, hydrogen peroxide, ethanol, sodium acetate, potassium chloride, acetic acid, phosphoric acid, formic acid, and HPLC grade methanol and acetonitrile were purchased from Merck (Darmstadt, Hesse, Germany).
All solutions and sample dilutions were performed using ultrapure water obtained from an ultra-purifier system (Gehaka, Brazil). Nitric acid 14 mol L-1 was purified by distillation in a Teflon® sub-boiling system model Distill acid BSB-939-IR (Berghof, Germany), and pure H2O2 30% (w/w) was obtained from Proquimios, Brazil.
2.2. Winemaking
Grapes Vitis vinifera L. of the Merlot variety grown in the region of Videira, Santa Catarina, Brazil, vintage 2015, were used for this study. The sparkling wine production was carried out in the experimental station of the Agricultural Research and Rural Extension Company of Santa Catarina (Epagri), following the Traditional or Champenoise method. Initially, the must was obtained from the destemming, crushing and pressing of the grapes. To obtain the base wine, the must was fermented by Saccharomyces cerevisiae PB2019 yeasts (Fermol Blanc, AEB Spa, Brescia, Italy) in stainless steel tanks with controlled temperature between 15 and 17 °C, monitoring by
residual sugar and density. After the first fermentation, the base wine was cold stabilized (4 °C) and sulfur dioxide was added (30 mg L-1, Vinoaromax, AEB Spa, Brescia, Italy).
For the production of the sparkling wines by the Traditional method, inverted sugar syrup at concentration of 26 g L-1 was added to the base wine to achieve the pressure of 6 atm, together with yeasts Saccharomyces cerevisiae PB2002 (Fermol Reims Champagne, AEB Spa, Brescia, Italy) and 15 g hL-1 of bentonite solution, coadjuvant for the remuage process as clarifier (Compact gel, AEB, Brescia, Italy). Together with the tirage liquor, mannoproteins (0.30 g L-1, Oenolees®, Laffort, Bordeaux, France) abundant in a specific peptide fraction naturally released by the yeasts throughout autolysis (aging on lees) were added to the base wine prior to the second in-bottle fermentation, according to the manufacturer’s recommendation. One treatment was maintained as control (sparkling wine without the addition of mannoproteins). Thereafter, the wines were bottled and stored in the horizontal position (17 ºC) until the second fermentation was completed. Twenty-four bottles of wines of 750 mL each were bottled after adding the tirage liquor (n
= 24). Half of the bottles (n = 12) were added with mannoproteins and the other half (n = 12) corresponded to the control samples (sparkling wine without the addition of mannoproteins). Following the second fermentation, and after 3 months of contact with lees, three bottles of sparkling wine of each treatment (n = 6) were riddled, disgorged, corked and analyzed. The remaining bottles were kept in contact with the yeast lees (17
°C) and were disgorged after 6 (n = 6), 9 (n = 6) and 12 (n = 6) months of over-lees aging.
2.3. Oenological parameters
Titratable acidity, volatile acidity, alcohol content, pH, free and total SO2 were determined in the base wine and sparkling wines with 3 and 12 months of over-lees aging according to the Office International de la Vigne et du Vin (OIV, 2012).
2.4. Color parameters, antioxidant capacity and phenolic contents
The samples of rosé sparkling wines were initially degassed by magnetic stirring for 25 min and analyzed in a UV-Vis spectrophotometer (Hitachi U 2900, CA, USA) for total monomeric anthocyanins (TMA) by the differential pH method (Giusti & Wrolstad, 2001). The results were expressed as mg L-1 of malvidin-3-O-glucoside. Color intensity, color density and color tonality were determined with readings at 420, 520 and 620 nm using a 10 mm path length cuvette (Glories, 1984). The antioxidant capacity was determined by free radical-scavenging activity using the ABTS radical (Re et al., 1999). Results were expressed in Trolox equivalents (mmol TEAC L-1 wine).
2.5. HPLC analysis of polyphenolic compounds
The identification and quantification of individual polyphenolics in wine samples throughout the aging process were performed by high performance liquid chromatography with diode array detection (HPLC-DAD) (Shimadzu, Kyoto, Japan). For all polyphenolic compounds, analyte separation was achieved on a Shim-pack CLC-ODS end-capped reverse phase-column (4.6 x 250 mm, 5 µm) from Shimadzu. For the analysis, 20 μL of samples were injected into the HPLC system.
Sixteen phenolic compounds belonging to the subclasses of hydroxybenzoic acids, hydroxycinnamic acids, flavonols, flavan-3-ols, tyrosol and stilbenes were quantified in wine samples and their levels were monitored up to 12 months of over-lees aging. Sample preparation and HPLC analysis were performed according to Burin, Ferreira-Lima, Panceri, & Bordignon-Luiz (2014). In the procedure, wine samples were submitted to liquid-liquid extraction prior to injection into the chromatography system. The mobile phase A was water: acetic acid (98:2 v/v) and mobile phase B was water: acetonitrile: acetic acid (58:40:2 v/v/v). The gradient elution was as follows: 0-80 % solvent B for 55 min, 80-100 % B for 15 min and 100-0 % B for 5 min, with the flow rate set at 0.9 mL min-1 and total run time of 75 min. The detection was set at 280 mn for tyrosol, catechin and epicatechin, at 320 nm for caffeic, trans-caftaric, p-coumaric and ferulic acids, at 360 nm for myricetin, quercetin and kaempferol, and at 306 nm for trans- resveratrol.
The separation of hydroxybenzoic acids (gallic, protocatechuic, syringic, vanillic and ellagic) was carried out using a binary mobile phase consisted of ultrapure water: acetic acid (98:2, v/v) (A) and acetonitrile: solvent A (80:20, v/v) (B). The elution gradient was as follows: 0-35 % B for 35 min, 35-0 % B for 3 min, with total run time of 38 min.
The flow rate was 1.0 mL min-1 and the detection was set at 280 nm for all the compounds, with the exception of ellagic acid that was detected at 254 nm.
The 3-O-monoglucoside anthocyanins (malvidin, cyanidin, delphinidin and peonidin) were determined according to Revilla, Pérez-Magariño, González-Sanjosé, & Beltrán (1999). A gradient elution programme was used with the mobile phase consisted of water: formic acid (90:10 v/v) (A) and water: methanol: formic acid (45:45:10 v/v/v)
(B) as follows: 35-95 % solvent B for 20 min, 95-100 % solvent B for 5 min, 100-35 % solvent B for 5 min, with this maintained for 5 min for a total run time of 35 min. The flow rate was kept at 0.8 mL min-1 and the detection set at 520 nm.
The identification and quantification of all phenolic compounds was performed by comparing the analyte peaks with their respective standards solutions at concentration range of 0.01-200 mg L-1 using matrix-matched calibration.
2.6. HPLC analysis of organic acids
The same chromatograph system was used for the quantification of organic acids (tartaric, succinic, malic, citric and lactic), following the method described by Escobal et al. (1998), with modifications. For the analysis, the wine samples were diluted in ultra- pure water and 20 µL were injected into the chromatograph system. Analyte separation was performed using isocratic elution with the mobile phase consisted of ultra-pure water
1.2 % (v/v) phosphoric acid at pH 2.4. The modifications were as follows: flow rate was
0.7 mL min-1 and the total run time was 40 min. The detection was set at 212 nm for all organic acids. Quantification was performed using matrix-matched standards at concentration range of 0.003-0.5 g L-1.
2.7. Mineral profile by ICP-OES
The element determination was carried out using an ICP-OES model iCAP 6000 series (Thermo Scientific) equipped with a Meinhard® type nebulizer, and a cyclonic spray chamber. Argon with a minimum purity of 99.95 % was used as main, auxiliary, and nebulizer gas. The operational parameters for all measurements were: Radio frequency power of 1300W, auxiliary gas flow of 1.00 L min-1, and nebulizer gas flow rate of 0.45 L min-1. The monitored emission wavelengths were 396.8 nm (Ca), 324.7 nm (Cu), 259.9
nm (Fe), 766.4 nm (K), 279.5 nm (Mg), 257,6 nm (Mn), 421.5 nm (Sr), 334.5 nm (Zn),
and 361.3 nm (Sc – Internal Standard).
Sample digestion was performed in a microwave oven model DGT100 Plus (Provecto Analítica, Brazil) equipped with closed Teflon® flasks (internal volume of 80.0 mL). Microwave-assisted digestion was carried out by adding 2.0 mL of each sample directly inside the Teflon® vessels; the digestion mixture was composed of 4.0 mL of HNO3 14 mol L-1, and 1.0 mL of H2O2 which was added to the sample. The vessels were closed and submitted to the power program as follow: 250 W for 2 min, 0 W for 2 min, 250 W for 6 min, 400 W for 5 min, 600 W for 5 min, and cooling for 10 min. The samples were removed, diluted, and analyzed.
2.8. Near-infrared and Raman spectroscopic measurements
Near-infrared spectra were collected using an FT-NIR spectrometer Frontier 400 (Perkin Elmer, USA) and the spectrum acquisition was performed from 9090.91 cm-1 to 4000.00 cm-1. While the Raman spectroscopy analysis was performed using a PeakSeeker Pro-785, Agiltron, (Ocean Optics, USA). The Raman measurements were carried out from 250 to 1800 cm-1, with an excitation laser power of 100 mW (785 nm), the integration time of 60 s, and the frame of 10 s.
All analysis was carried out in each sample without dilution, the measurements were performed in five replicates for each samples and the spectrum was collected as the mean spectrum for each sample.
2.9. Statistical analysis
One-way ANOVA and multivariate statistical analyses were carried out using Statistica software version 8.0 (StatSoft Inc., Tulsa, USA). All results are expressed as the means ± standard deviation (SD). Student’s t-test was used to verify data normality and homogeneity of variances. Significant differences were assessed using the t-test and Duncan’s multiple range test (p<0.05). Data variability as influenced by aging time and the addition of mannoproteins were investigated by principal component analysis (PCA).
3. Results and discussion
3.1. Oenological parameters and levels of organic acids of the wines
Table 1 shows the classical oenological parameters and organic acids determined in the base wine and sparkling wines at the beginning of the experiment and after 12 months of over-lees aging. It was observed that the addition of mannoproteins did not alter significantly (p>0.05) the pH and titratable acidity of the sparkling wines after 3 and 12 months of aging. However, it was observed that these wines showed lower volatile acidity in comparison with the control samples (wines without mannoproteins). In addition, the wines treated with the agent showed higher concentrations of free and total SO2 and alcohol content in both samples of 3 and 12 months aging. The oenological characteristics of the base wine and sparkling wines are in accordance with established parameters for wines in international practices (OIV, 2012). With the exception of citric acid, four major organic acids present in grapes and wines were detected in the wine samples. Malic acid was the most abundant organic acid in the base wine and sparkling wines with concentrations ranging from 3.35 to 3.74 g L-1, followed by lactic and tartaric acids.
The organic acids were quantified at varying concentrations in the wine samples. The average concentrations were similar to those found in other studies with sparkling wines (Gallardo-Chacón et al., 2010; Caliari, Panceri, Rosier, & Bordignon-Luiz, 2015). These compounds contribute to the total and volatile acidities of wines (OIV, 2012). The high concentrations of malic acid found in the base wine and sparkling wines may indicate that harvesting of grapes was carried out before the technological maturation (Caliari, Panceri, Rosier, & Bordignon-Luiz, 2015). It also indicates that the malolactic fermentation was not completed, which is typical of the production of sparkling wines (Gallardo-Chacón et al., 2010). The concentrations of organic acids in the rosé wines were variably affected by the addition of the oenological agent. Levels of malic and succinic acids were significantly increased by the treatment and throughout the biological aging, while concentrations of tartaric acid were consistently decreased in wines when compared with the control samples, demonstrating the influence of mannoproteins on the concentration of these organic acids and on the sensorial quality of rosé sparkling wines. Regardless of the use of mannoproteins, changes on organic acids are generally accompanied by slight alterations of pH and were expected due to the fermentative metabolism of the yeast (Pan et al., 2011). This may explain the increase in the pH of the wines verified after 12 months of aging in this study, probably due to decreases in the concentrations of tartaric acid during aging as a consequence of precipitation of potassium bitartrate. Despite some variations, it was observed that the profile and levels of organic acids were preserved during the over-lees aging since they were individually quantified after 12 months of aging. This corroborates that these specific acids, namely malic, tartaric, lactic and succinic, were not exhaustingly metabolized neither during the second fermentation nor during aging, and thus can be accountable for the final quality of the rosé sparkling wines.
3.2. Effect of mannoproteins and evolution of wine polyphenolics during over-lees aging
The polyphenolic profile and evolution of sparkling wines are presented in Table
2. Many polyphenolics including hydroxybenzoic and hydroxycinnamic acids, flavonols, flavan-3-ols and anthocyanins were identified in the rosé wines. Caffeic acid, catechin, tyrosol, gallic acid and malvidin-3-O-glucoside were the predominant phenolics. trans- Resveratrol and tyrosol were found at high concentrations (up to 7.70 and 98.07 mg L-1, respectively), while the anthocyanin cyanidin-3-O-glucoside was not detected in the studied wines. The addition of mannoproteins affected significantly (p≤0.05) the color parameters, antioxidant activity and individual concentrations of phenolic compounds, although with a variable effect throughout the aging period.
The concentrations of caffeic and p-coumaric acids in wines treated with mannoproteins were significantly higher (p<0.05) in comparison with the control sample. For trans-caftaric acid, this effect was only observed after 12 months of aging, with treated wines showing higher levels of this acid than the control wine. Hence, the influence of mannoproteins was positively pronounced at the end of biological aging, when sparkling wines had the highest sum of hydroxycinnamic acids, respectively at 9 (30.9 mg L-1) and 12 months (27.8 mg L-1) of aging (Fig. 1). Contrarily, concentrations of caffeic and ferulic acids in these samples decreased during the over-lees aging. These changes during wine aging may be related to hydrolysis reactions of the esterified forms of these compounds to form free hydroxycinnamic acids or the participation of their free forms in polymerization reactions with anthocyanins, leading to the formation of pyranoanthocyanins (Garrido & Borges, 2013; Marquez, Serratosa, & Merida, 2013).
A total of five hydroxybenzoic acids were found in samples, with gallic acid being the most abundant at concentrations varying between 10.74 and 13.50 mg L-1 in base wine and sparkling wines, respectively. Similarly to hydroxycinnamic acids, sparkling wines added of mannoproteins showed the highest concentrations of hydroxybenzoic acids after 9 and 12 months of aging. In these wines, concentrations of protocatechuic, gallic, vanillic and ellagic acids were significantly higher in relation to their control sample. This shows the beneficial effect of mannoproteins on the content of phenolic acids, since these compounds are associated with the antioxidant potential of wines. In general, there were increases in levels of hydroxybenzoic acids in the wine samples throughout the over-lees aging. These findings are consistent with previous reports of changes in phenolic acids in aged wines, which is attributed to hydrolysis reactions during yeast autolysis, such as the formation of gallic acid resulting from the hydrolysis of tannins (Pozo-Bayón, Martínez-Rodríguez, Pueyo, & Moreno-Arribas, 2009).
Substantial changes were also verified when comparing the phenolic profile of sparkling and base wines, as it can be seen that some phenolics were not detected in all wine samples, such as kaempferol and the anthocyanins peonidin-3-O-glucoside and delphinidin-3-O-glucoside. Epicatechin was not detected in wines at the beginning of the experiment, however, was quantified at increasing concentrations at the last months of biological aging (after 9 and 12 months). The addition of mannoproteins had a variable effect on flavonols and flavan-3-ols. The concentrations of flavonols were significantly higher in treated wines in comparison with the control samples. Conversely, the levels of (+)-catechin and (-)-epicatechin were consistently decreased during aging in relation to untreated wines, with the exception of catechin after 12 months of aging (Fig. 1). These changes are probably due to copigmentation, depolymeryzation or polymerization
reactions of these compounds with other phenolics that occur during wine aging
(Ribéreau-Gayon, Glories, Maujean, & Dubourdieu, 2006). The decrease in concentrations of some polyphenolic compounds observed in mannoprotein-added samples could be attributed to the mannoproteins precipitation phenomena or formation of unstable colloids (Guadalupe, Palacios, & Ayestarán, 2007).
In regards to color quality of rosé sparkling wines, the concentrations of anthocyanins 3-O-monoglucosides and TMA showed a gradual decrease throughout the over-lees aging. Their levels in the control wines and wines added of mannoproteins showed reductions of 85 % and 86 %, respectively, at the end of 12 months of aging. On the other hand, changes in color intensity, color tonality and color density were only significant in wines after 3 and 12 months of aging. Notably, the addition of mannoproteins did not influence the concentrations of anthocyanins in rosé sparkling wines, as evidenced in other studies with sparkling wines (Pérez-Magariño et al., 2015) and red wines (Guadalupe & Ayestarán, 2008). Taken together, these results pointed out specific changes involving different types of polyphenolics occurring during over-lees aging that may affect the chemical quality of sparkling wines, and many were in fact attributed to mannoproteins. Overall, the addition of mannoproteins beneficially affected the concentrations of trans-resveratrol, quercetin, catechin, p-coumaric and hydroxybenzoic acids that showed increased concentrations over time. Their effect could be justified by the absorption or release of polyphenols and other non-phenolic substances by the yeast cell wall (Gallardo-Chacón et al., 2010), or also by hydrolysis of phenolic compounds with other small compounds released in the wine, such as yeast mannoproteins (Guadalupe, Palacios, & Ayestarán, 2007).
3.3. Mineral profile
The element composition of base wine and sparkling wines after 12 months of over-lees aging are given in Table 3. The macroelements Ca, K and Mg and the microelement Mn were quantified in all samples. Copper, Zn, Fe and Sr were not detected. Potassium was the macroelement present at the highest concentrations in all samples, whereas the sparkling wines had higher levels when compared to base wines. Calcium and Mg showed similar contents in wines. It was observed that K, Ca and Mg showed little changes during fermentation and over-lees aging as their levels in wines were quite similar throughout the experiment. These macroelements are extracted from soil by the grapevine and are present in wines due to the maceration process used for the elaboration of wines (Jos et al., 2004). They are important for multiplication and metabolism of wine yeasts (Ribéreau-Gayon, Glories, Maujean, & Dubourdieu, 2006).
Among the microelements, only Mn was detected in samples, with concentrations ranging from 1.4 to 1.7 mg L-1 in sparkling and base wines, respectively. Indeed, microelements such as Mn, Fe, Cu, Zn, Se and Co are found at deficient concentrations in grapes and its derivatives (Ribéreau-Gayon, Glories, Maujean, & Dubourdieu, 2006).
Apart from their natural occurrence in grape berries, their levels in wines can be originated from environmental contamination or processing conditions, such as the use of phytosanitary products, metallic tools or containers and use of oenological products (Hopfer et al., 2015; Ribéreau-Gayon, Glories, Maujean, & Dubourdieu, 2006). In the studied wines, levels of macro- and microelements were consistent during the over-lees aging, with no significant changes found between treated (added of mannoproteins) and untreated wines. These results demonstrate that the addition of mannoproteins did not compromise the total concentration of elements of the sparkling wines, suggesting that it did not cause precipitation or formation of insoluble substances with the studied minerals,
which remained stable in wine samples. Nevertheless, determination of elements in wines is of interest because of their nutritional and toxicological implications as well as their oenological importance regarding geographical origin and authenticity of wines (Hopfer et al., 2015; Jos et al., 2004).
3.4. Near-infrared and Raman spectra of sparkling wines
Near-infrared and Raman spectrometry measurements were performed in the base wine and in wines treated with mannoproteins in order to identify possible modifications in the chemical composition of rosé sparkling wines when in contact with lees during aging. Based on the bond vibration of molecules, those techniques provide useful information whether deterioration of molecules are taking place, in this case, additional peaks are expected to be observed in the samples when compared to the control sample (Luna et al., 2017; Santos et al., 2017). Hence, the vibrational spectra of wine samples were monitored in the base wine and after 12 months of aging (Fig. 2).
The NIR spectrum of samples (Fig. 2A) revealed absorption bands with intense peaks above 6000 cm-1, from approximately 6250 cm-1 to 7620 cm-1. The intense absorption band around 6901.31 cm-1 may be attributed to water (first overtone of OH) (Santos et al., 2017). The regions 9091-7692 and 6023-5435 cm-1 provided the best models for the calibration of acidity, total sugars, pH and density of the wines (Teixeira dos Santos et al., 2018). The signals around 7304-6301 cm-1 may also originate from the absorption of RN–H groups from proteins (Luna et al., 2017). The maximum signals observed in the region of 5405-4000 cm-1 have been assigned to absorption of many chemical bonds and groups (C=O, C–H, CH2 and CH3) originated from phenolic acids, water, organic acids and other polyphenolic compounds (Bauer et al., 2008; Martelo-Vidal
& Vázquez, 2014). Absorption bands related to sulfides added to wines were not characteristic in the NIR spectra obtained for samples. This is explained by the absorption of S–O and S=O compounds that display strong bands in the mid-infrared region, respectively from 1000 to 650 cm-1 and 1375 to 1050 cm−1 (Bauer et al., 2008).
The absence of additional peaks for the wines treated with mannoproteins and the control sample indicates that the composition of wines was not modified in terms of molecular degradation and formation of new substances. The observed intensities can be attributed to differences in the composition of wines (base and sparkling wines) and their variable levels of constituents, mainly polyphenolic compounds and organic acids, as shown by the results of chemical characterization. In previous studies, IR spectral data collected from grape derivatives such as red wines and grape juice and from the phenolic compounds cyanidin, malvidin, quercetin and catechin have shown intense peaks at ranges of 700-1660 cm-1 and of 2600-3500 cm-1 (Martelo-Vidal & Vázquez, 2014; Sen, Ozturk, Tokatli, & Ozen, 2016; Snyder, Sweeney, Rodriguez-Saona, & Giusti, 2014). According to the literature, this may be consistent with our observations. In fact, together with malvidin-3-O-glucoside and some phenolic acids, catechin was one of the most abundant phenolics in the studied wines (Table 2).
In the Raman spectrum (Fig. 2B) intense peaks were observed at approximately 840 cm-1, 1030 cm-1, 1050 cm-1 and 1440 cm-1. The intensity peak observed around 880 cm-1 is probably originated from the C-C stretching of ethanol. The bands around1250 and 450 cm-1 can be related with the H-C-C and O-C-C bendings, respectively (Santos et al., 2017; Teixeira dos Santos et al., 2018). Other peaks of weak intensity can be observed in the spectrum from 1050 to 1450 cm-1 and are presumably originated from hydroxycinnamic acids, such as caffeic, ferulic, p-coumaric, among others present in wines. Indeed, as previously reported in white wines, this phenolic family most likely
contribute to the Raman scattering around 1000 and 1600 cm-1 (Martin et al., 2015; Sen, Ozturk, Tokatli, & Ozen, 2016).
The Raman spectrum of wine samples confirmed the NIR observations that there was no formation of new species after the treatment performed in samples and after the biological aging of wines. This suggests that the mild changes observed for polyphenolic compounds and other chemical constituents found in the wine samples, justified by polymerization, depolymeryzation, chelation or copigmentation reactions, may have similarly occurred in all wines, independently of the addition of mannoproteins. Moreover, the extent of the chemical transformations that may have occurred at the molecular level during aging of the sparkling wines was not sufficiently broad to alter their vibrational spectra. This can be noted in the Raman and NIR spectra although the different absorption intensities verified for the wine samples since superposition of their corresponded spectrum is still attainable.
The interpretation of vibrational spectra of samples can be a complex task as variations may not be perceived clearly. For instance, when the NIR region is considered, the overlap of overtones and Fermi resonances may compromise the differentiation of samples (Luna et al., 2017).
3.5. Principal component analysis
The PCA was performed to explore information from the spectra better and to study which constituents had more influence on the grouping of samples. The results of PCA considering the composition of sparkling wines are presented in Fig. 3. Only wine samples at the end of the biological aging in the experiments (12 months) were included in
the analysis in order to achieve better representation of the chemical changes and final characteristics of aged sparkling wines.
It was possible to observe from the scores plot (Fig. 3A) of PCA that the sparkling wines were divided into two major components (PC1 x PC2), which accounts for 94.41 % of the total data variability. They were described by the first principal component (PC1) and accounted for the maximum variability of the data set (85.13 %), responsible for separating the samples into two groups, one being the control wines and the other the wines treated with mannoproteins, according to the variations in concentrations of minerals, organic acids and polyphenolics. The corresponding loading plots (Fig. 3B) depict the correlations between the analyzed variables. As can be drawn from the graph, the scatter plot revealed that differences due to the addition of the oenological agent in wines overshadowed differences among specific chemical constituents of the composition of control and agent-treated wines, particularly in regard to individual phenolic compounds, such as tyrosol, gallic acid, trans-caftaric, caffeic, myricetin, vanillic acid, protocatechuic acid, quercetin, p-coumaric acid, catechin, trans-resveratrol, ellagic and syringic acids, that were strongly associated with sparkling wines added of mannoproteins. Noteworthy, the majority of the studied minerals and organic acids as well as most of the polyphenolics were associated with these wines. These results corroborate that mannoproteins positively influenced the final composition of aged sparkling wines.
4. Conclusions
Some changes were observed in concentrations of phytochemicals of rosé sparkling wines obtained using the Champenoise method, with a variable influence of the time of biological aging. Levels of macro and microelements remained stable in all
samples, while maintaining the nutritional characteristics of the sparkling wines. The addition of mannoproteins positively influenced the concentrations of organic acids and phenolic compounds in wines over time and these effects were mainly observed at the end of the over-lees aging, particularly for tyrosol, trans-resveratrol, gallic acid, catechin and the hydroxycinnamic acids. NIR and Raman spectra revealed that these changes were not sufficiently pronounced to modify absorption spectra of wines. The changes observed through this multianalytical approach may offer new insights regarding the bioactive and nutritional properties of aged rosé sparkling wines.
Acknowledgements
The authors would like to thank for financial support from the Brazilian governmental agencies CAPES and CNPq. We also thank Agricultural Research and Rural Extension Company of Santa Catarina (Epagri) for providing the sparkling wine and the winemaking agent.
Conflict of interest
The authors have no conflict of interest to declare.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References
Bai, B., He, F., Yang, L., Chen, F., Reeves, M. J., & Li, J. (2013). Comparative study of phenolic compounds in Cabernet Sauvignon wines made in traditional and Ganimede fermenters. Food Chemistry, 141, 3984-3992.
Bauer, R., Nieuwoudt, H., Bauer, F. F., Kossmann, J., Koch, K. R., Esbensen, K. H. (2008). FTIR spectroscopy for grape and wine analysis. Analytical Chemistry, 80, 1371- 1379.
Burin, V. M., Ferreira-Lima, N. E., Panceri, C. P., & Bordignon-Luiz, M. T. (2014). Bioactive compounds and antioxidant activity of Vitis vinifera and Vitis labrusca grapes: Evaluation of different extraction methods. Microchemical Journal, 114, 155-163.
Caliari, V., Panceri, C. P., Rosier, J. P., & Bordignon-Luiz, M. T. (2015). Effect of the Traditional, Charmat and Asti method production on the volatile composition of Moscato Giallo sparkling wines. LWT - Food Science and Technology, 61, 393–400.
Del Barrio-Galán, R., Pérez-Magariño, S., Ortega-Heras, M., Guadalupe, Z., & Ayestarán,
B. (2012). Polysaccharide characterization of commercial dry yeast preparations and their effect on white and red wine composition. LWT - Food Science and Technology, 48, 215- 223.
Escobal, A., Iriondo, C., Laborra, C., Elejalde, E., & Gonzalez, I. (1998). Determination of acids and volatile compounds in red Txakoli wine by highperformance liquid chromatography and gas chromatography. Journal of Chromatography A, 823, 349-354.
Gallardo-Chacón, J. J., Vichi, S., Urpí, P., López-Tamames, E., & Buxaderas, S. (2010.) Antioxidant activity of lees cell surface during sparkling wine sur lie aging. International Journal of Food Microbiology, 143, 48–53.
Garrido, J., & Borges, F. (2013). Wine and grape polyphenols - A chemical perspective.
Food Research International, 54, 1844–1858.
Ghanem, C., Taillandier, P., Rizk, M., Rizk, Z., Nehme, N., Souchard J. P., & El Rayess
Y. (2017). Analysis of the impact of fining agents types, oenological tannins and mannoproteins and their concentrations on the phenolic composition of red wine. LWT - Food Science and Technology, 83, 101-109.
Giusti, M. M., & Wrolstad, R. E. (2001). Anthocyanins: characterization and measurement
with UV-visible spectroscopy. In R.E. Wrolstad (Eds.), Current protocols in Food
Analytical Chemistry. New York: John Wiley & Sons Inc.
Glories, Y. (1984). La coleur des vins rouges. Connaissance Vigne Vin, 18, 253-271.
Guadalupe, Z., & Ayestarán, B. (2008). Effect of Commercial Mannoprotein Addition on Polysaccharide, Polyphenolic, and Color Composition in Red Wines. Journal of Agricultural and Food Chemistry, 56, 9022–9029.
Guadalupe, Z., Palacios, A., & Ayestarán, B. (2007). Maceration enzymes and mannoproteins: a possible strategy to increase colloidal stability and color extraction in red wines. Journal of Agricultural and Food Chemistry, 55, 4854-4862.
Hopfer, H., Nelson, J., Collins, T. S., Heymann, H., & Ebeler, S. E. (2015). The combined impact of vineyard origin and processing winery on the elemental profile of red wines.
Food Chemistry, 172, 486–496.
Ivanova-Petropulos, V., Hermosín-Gutiérrez, I., Boros, B., Stefova, M., Stafilov, T., Vojnoski, B., Dornyei, A., & Kilár, F. (2015). Phenolic compounds and antioxidant activity of Macedonian red wines. Journal of Food Composition and Analysis, 41, 1–14.
Jos, A., Moreno, I., González, A.G., Repetto, G., & Cameán, A. M. (2004). Differentiation of sparkling wines (cava and champagne) according to their mineral content. Talanta, 63, 377–382.
Lucci, P., Saurina J., & Núñez, O. (2017). Trends in LC-MS and LC-HRMS analysis and characterization of polyphenols in food. Trends in Analytical Chemistry, 88, 1-24.
Luna, A. S., Silva, A. P., Alves, E. A., Rocha, R. B., Lima, I. C. A, & de Gois, J. S. (2017). Evaluation of chemometric methodologies for the classification of Coffea canephora cultivars via FTNIR spectroscopy and direct sample analysis. Analytical Methods, 9, 4255-4260.
Marquez, A., Serratosa, M. P., & Merida, J. (2013). Pyranoanthocyanin Derived Pigments in Wine: Structure and Formation During Winemaking. Journal of Chemistry, 2013, 1–35.
Martelo-Vidal, M. J., & Vázquez, M. (2014). Determination of polyphenolic compounds of red wines by UV-VIS-NIR spectroscopy and chemometrics tools. Food Chemistry, 158, 28-34.
Martin, C., Bruneel, J. L., Guyon, F., Médina, B., Jourdes, M., Teissedre, P. L., & Guillaume, F. (2015). Raman spectroscopy of white wines. Food Chemistry, 181, 235– 240.
OIV. Office International de la Vigne et du Vin. (2018).Codex Œnologique International.
(Edition 2012). Paris: France, (Vol. 1-2).
OIV. Office International de la Vigne et du Vin. (2012). Compendium of international methods of wine and must analysis. (Edition 2012). Rue d'Aguesseau-75008 Paris: France, (Vol. 1-2).
Pan, W., Jussier, D., Terrade, N., Yada, R. Y., & Orduña, R. M. (2011). Kinetics of sugars, organic acids and acetaldehyde during simultaneous yeast-bacterial fermentations of white wine at different pH values. Food Research International, 44, 660–666.
Pérez-Magariño, S., Martínez-Lapuente, L., Bueno-Herrera, M., Ortega-Heras, M., Guadalupe, Z., & Ayestarán, B. (2015). Use of commercial dry yeast products rich in mannoproteins for white and rosé sparkling wine elaboration. Journal of Agricultural and Food Chemistry, 63, 5670-5681.
Pozo-Bayón, M. À., Martínez-Rodríguez, A., Pueyo E., & Moreno-Arribas, M. V. (2009). Chemical and biochemical features involved in sparkling wine production: from a traditional to an improved winemaking technology. Trends in Food Science & Technology, 20, 289-299.
Prakash, S., Iturmendi, N., Grelard, A., Moine, V., & Dufourc, E. (2016). Quantitative analysis of Bordeaux red wine precipitates by solid-state NMR: Role of tartrates and polyphenols. Food Chemistry, 199, 229–237.
Re, R., Pellegrini, N., Proteggemnte, A., Pannala, A., Yang, M., & Rice-Evans, C. (1999).
Antioxidant activity applying and improved ABTS radical cation decolorization assay.
Free Radical Biology Medicine, 26, 1234-1237.
Revilla, I., Pérez-Magariño, S., González-Sanjosé, M.L., & Beltrán, S. (1999). Identification of anthocyanin derivatives in grape skin extracts and red wines by liquid chromatography with diode array and mass spectrometric detection. Journal of Chromatography A, 847, 83-90.
Ribéreau-Gayon, P., Glories, Y., Maujean, A., & Dubourdieu, D. (2006). Handbook of enology. The Chemistry of Wine Stabilization and Treatments, Vol. 2, West Sussex, UK: Wiley & Sons.
Santos, C. A. T., Páscoa, R. N. M. J., Sarraguça, M. C., Porto, P. A. L. S., Cerdeira, A. L., González-Sáiz, J. M., Pizarro, C., & Lopes, J. A. (2017). Merging vibrational
spectroscopic data for wine classification according to the geographic origin. Food Research International, 102, 504-510.
Sartor, S., Caliari, V., Malinovski, L. I., Toaldo, I. M., & Bordignon-Luiz, M. T. (2017). Bioactive profiling of polyphenolics and oenological properties of red wines from Italian grapes (Vitis vinifera L.) cultivated in a selected subtropical region. International Journal of Food Properties, 20, 1-10.
Sen, I., Ozturk, B., Tokatli, F., & Ozen, B. (2016). Combination of visible and mid infrared spectra for the prediction of chemical parameters of wines. Talanta, 161, 130-137.
Snyder, A. B., Sweeney, C. F., Rodriguez-Saona, L. E., & Giusti, M. M. (2014). Rapid authentication of concord juice concentration in a grape juice blend using Fourier- Transform infrared spectroscopy and chemometric analysis. Food Chemistry, 147, 295- 301.
Teixeira dos Santos, C. A., Páscoa, R. N. M. J., Porto, P. A. L. S., Cerdeira, A. L., González-Sáiz, J. M., Pizarro, C., & Lopes, J. A. (2018). Raman spectroscopy for wine analyses: A comparison with near and mid infrared spectroscopy. Talanta, 186, 306-314.
Figure Captions
Fig. 1. Representative chromatogram of polyphenolics in Merlot sparkling wines with and without the addition of mannoproteins after 12 months of over-lees aging. Absorbance peaks (280 nm): 1. trans-caftaric acid, 2. tyrosol, 3. (+)-catechin, 4. caffeic acid, 5. (-)-
epicatechin, 6. p-coumaric acid, 7. ferulic acid, 8. myricetin, 9. trans-resveratrol, 10. quercetin.
Fig. 2. Near-infrared (A) and Raman (B) spectra of the base wine and sparkling wines after 12 months of over-lees aging. Base wine: BW; Sparkling wine without mannoproteins: SWC; Sparkling wine added of mannoproteins: SWM.
Fig. 3. Two dimensional PCA showing (A) scores plot for wines without the addition of mannoproteins (control) and wines added of mannoproteins after 12 months of aging and
(B) loading plots of individual polyphenolics, organic acids and elemental composition of wines. Antioxidant activity (ABTS), color intensity (CI), color tonality (CT), color density (CD) and total monomeric anthocyanins (TMA). Each plot description corresponds to an individual observation in the score plot and to a variable in the loading plot of the two-first principal components.
Table 1. Classical oenological parameters and organic acids determined in the base wine and sparkling wines added of mannoproteins after 3 and 12 months of biological aging.
Aging over-lees
Base wineSparkling wines, 3 months Sparkling wines, 12 months
Control Mannoprotein Control Mannoprotein
pH 3.37±0.01 3.47±0.01a 3.46±0.01a 3.83±0.01b 3.84±0.01b
Titratable acidity 4.28±0.26 4.38±0.18b 4.68±0.58b 3.20±0.12a 3.15±0.19a
Volatile acidity 0.21±0.03 0.21±0.01a 0.17±0.01*b 0.18±0.01a 0.15±0.01*a
Free sulfur dioxide 11.20±0.80 11.20±0.01a 13.20±0.40*a 12.70±0.56b 14.60±0.25*b
Total sulfur dioxide 14.80±0.40 22.93±2.01a 36.27±3.23*a 42.01±1.90b 46.03±2.60*b
Alcohol content 11.00±0.01 12.40±0.01a 12.63±0.06*a 12.45±0.10a 12.71±0.09*b
Tartaric acid 0.95±0.01 0.94±0.01b 0.92±0.01*b 0.90±0.02a 0.85±0.01*a
Lactic acid 1.21±0.01 1.06±0.03a 1.10±0.01a 1.09±0.02a 1.11±0.01a
Malic acid 3.74±0.01 3.35±0.01a 3.43±0.01*a 3.40±0.02b 3.50±0.01*b
Succinic acid n.d. 0.20±0.03a 0.26±0.01*a 0.21±0.01b 0.29±0.02*b
Citric acid n.d. n.d. n.d. n.d. n.d.
Total organic acids 5.90 5.56 5.71 5.59 5.75
Mean values ± standard deviation (n=3). *Indicates significant difference between sparkling wines and their respective controls (Student`s t-test, p<0.05). Different letters in the same line indicate significant difference among control samples during over-lees aging (Duncan`s test, p<0.05). Different letters in the same line indicate significant difference among sparkling wines added of mannoproteins during over-lees aging (Duncan's test, p<0.05). Titratable acidity (g L-1 tartaric acid); Volatile acidity (g L-1 acetic acid); Free and total sulfur dioxide (mg L-1 sulfur dioxide); Alcohol content (%, v/v.); Organic acids (g L-1). n.d: not detected.
Table 2. Color parameters and phenolic characterization of wine samples and evolution of individual phenolic compounds (mg L-1) in rosé sparkling wines added of mannoproteins during biological aging (over-lees).
Aging over-lees
Compounds
Base wineSparkling wines, 3 monthsSparkling wines, 6 monthsSparkling wines, 9
Control Mannoproteins Control Mannoproteins Control Ma
Hydroxybenzoic acids
Gallic 10.74±0.05
13
Protocatechuic 7.27±0.03 4.09±0.04b 4.16±0.01*a 4.36±0.07c 4.27±0.01b 5.48±0.05d 5.
Vanillic 5.23±0.03 3.45±0.03a 3.97±0.01*b 3.86±0.07b 3.92±0.02a 3.95±0.02b 4.
Syringic 0.69±0.01 0.49±0.01a 0.55±0.01*a 0.61±0.02b 0.64±0.01b 0.66±0.04c 0.
Ellagic 0.34±0.01 0.57±0.01b 0.41±0.01*b 0.17±0.01a 0.23±0.01*a 3.37±0.13c 4.
Hydroxycinnamic acids
Mean values ± standard deviation (n=3). *Indicates significant difference between sparkling wines and their respective controls (Student`s t-test, p<0.05). Different letters in the same line indicate significant difference among control samples during over-lees aging (Duncan`s test, p<0.05). Different letters in the same line indicate significant difference among sparkling wines added of mannoproteins during over-lees aging (Duncan's test, p<0.05). TMA, total monomeric anthocyanins (mg L-1 malvidin-3-O-glucoside). #Radical scavenging antioxidant activity expressed as Trolox equivalents (mmol TEAC L-1). ##expressed as index. 1Value is lower than the detection limit of the method.
Table 3. Elemental composition of base wine and rosé sparkling wines treated with mannoproteins with 12 months of biological aging.
Aging over-lees
Elements (mg L-1) Base wine Sparkling wines, 12 months
Control Mannoproteins
Macroelements
Ca 61.7±1.6 63.4±2.2 69.3±10.6
K 1055.5±61.6 1152.2±56.6 1158.0±122.9
Mg 70.8±6.2 66.1±7.8 63.5±26.6
Microelements
Mn 1.7±0.2 1.4±0.2 1.4±0.5
Zn <5.31 <5.31 <5.31
Fe <1.11 <1.11 <1.11
Sr <2.41 <2.41 <2.41
Cu <1.01 <1.01 <1.01
Mean values ± standard deviation (n=3). *Indicates significant difference between sparkling wine and control (Student`s t-test, p<0.05). 1Value is lower than the detection limit of the method.
Highlights
-Changes on wine constituents were analyzed over time by multianalytical approaches.
-Aged sparkling wines added of mannoproteins had increased levels of polyphenols.
-Concentrations of organic acids were higher after Quercetin 12 months of over-lees aging.
-The mineral profile remained stable in wines aged in contact with yeast.
-There was no formation of new species in treated wines as shown by Raman spectra.