Applying Nanoparticle Tracking Analysis to Characterize the Polydispersity of Aggregates Resulting from Tannin-Polysaccharide Interactions in Wine-Like Media.

Interactions between grape seed tannin and either a mannoprotein or an arabinogalactan in model wine solutions of different ethanol concentrations were characterized with nanoparticle tracking analysis (NTA), UV-visible spectroscopy and dynamic light scattering (DLS). NTA results reflected a shift in particle size distribution due to aggregation. Furthermore, the light scattering intensity of each tracked particle measured by NTA demonstrated the presence of aggregates, even when a shift in particle size was not apparent. Mannoprotein and arabinogalactan behaved differently when combined with seed tannin. Mannoprotein formed large, highly light-scattering aggregates, while arabinogalactan exhibited only weak interactions with seed tannin. A 3% difference in alcohol concentration of the model solution (12 vs. 15% v/v) was sufficient to affect the interactions between mannoprotein and tannin when the tannin concentration was high. In summary, this study showed that NTA is a promising tool for measuring polydisperse samples of grape and wine macromolecules, and their aggregates under wine-like conditions. The implications for wine colloidal properties are discussed based on these results.

Potato Protein Fining of Phenolic Compounds in Red Wine: A Study of the Kinetics and the Impact of Wine Matrix Components and Physical Factors.

Producing wines within an acceptable range of astringency is important for quality and consumer acceptance. Astringency can be modified by fining during the winemaking process and the use of vegetable proteins (especially potato proteins) as fining agents has gained increasing interest due to consumers' requirements. The research presented was the first to investigate the effect of a potato protein dose on the kinetics of tannin and phenolic removal compared to gelatin for two unfined Cabernet Sauvignon wines. To further understand the results, the influence of the wine matrix and fining parameters (including pH, ethanol concentration, sugar concentration, temperature, and agitation) were tested according to a fractional 25-1 factorial design on one of the Cabernet Sauvignon wines using potato proteins. The results from the factorial design indicate that potato protein fining was significantly influenced by wine pH, ethanol concentration, fining temperature as well as an interaction (pH × ethanol) but not by sugar content or agitation. Insights into the steps required for the optimisation of fining were gained from the study, revealing that potato protein fining efficiency could be increased by treating wines at higher temperatures (20 °C, rather than the conventional 10-15 °C), and at both a lower pH and/or alcohol concentration.

Quantification of protein by acid hydrolysis reveals higher than expected concentrations in red wines: Implications for wine tannin concentration and colloidal stability.

Protein is reportedly negligible in most red wines, due to its loss following co-precipitation with phenolic substances. A method for protein quantification in red wine was developed which overcame analytical interference from phenolic substances, based on ethanol precipitation, followed by acid-hydrolysis and amino acid quantification. Protein concentration was surveyed in a range of red wines produced from V. vinifera and interspecific (Vitis spp) hybrids, revealing higher than expected concentrations, ranging from 23 mg/L ± 2.57 to 380 mg/L ± 16. The results showed that tannin extracted from grapes remains soluble in wine in the presence of protein even at high protein (>100 mg/L) and tannin (>500 mg/L) concentrations. As a further consequence of this, the particle size and concentration of colloids within high- and low-protein wines were similar, independent of protein or tannin concentration. Higher wine tannin concentration was also correlated with increased heat stability of wine protein.

Precipitation of Tannin-Anthocyanin Derivatives in Wine is Influenced by Acetaldehyde Concentration and Tannin Molecular Mass with Implications for the Development of Nonbleachable Pigments.

This study aimed to assess the effect of tannin molecular mass on nonbleachable pigment formation and subsequent stability under wine-like conditions. Tannin fractions of a defined molecular mass range were prepared from grape skins and seeds and reacted with malvidin-3-glucoside for 120 days in three media types: chemically defined wine media with or without acetaldehyde addition or model wine without acetaldehyde. Precipitation was observed after the reaction period and increased in response to both higher tannin molecular mass and acetaldehyde concentration. To confirm whether acetaldehyde-mediated condensation of tannin and anthocyanin modified the solubility of the nonbleachable pigments formed, HPLC-MS was used for the semiquantitative identification of vinyl derivatives and ethyl-linked adducts in soluble and precipitated materials. It was found that the proportion of vinyl derivatives and ethyl-linked anthocyanin was elevated in tannin precipitates relative to soluble pigmented material. Despite substantial losses of tannin due to precipitation, the resulting nonbleachable pigment concentration and color intensity were higher in wine media containing elevated acetaldehyde, when each tannin size category was considered independently. The results of this study indicated that the development of nonbleachable pigments from larger tannins may be limited when acetaldehyde-mediated condensation with anthocyanin predominates in wine, concomitant with precipitation.

Chemical and Sensory Impacts of Accentuated Cut Edges (ACE) Grape Must Polyphenol Extraction Technique on Shiraz Wines.

Accentuated Cut Edges (ACE) is a recently developed grape must extraction technique, which mechanically breaks grape skins into small fragments but maintains seed integrity. This study was the first to elucidate the effect of ACE on Shiraz wine's basic chemical composition, colour, phenolic compounds, polysaccharides and sensory profiles. A further aim was to investigate any potential influence provided by ACE on the pre-fermentation water addition to must. ACE did not visually affect Shiraz wine colour, but significantly enhanced the concentration of tannin and total phenolics. Wine polysaccharide concentration was mainly increased in response to the maceration time rather than the ACE technique. ACE appeared to increase the earthy/dusty flavour, possibly due to the different precursors released by the greater skin breakage. The pre-fermentation addition of the water diluted the wine aromas, flavours and astringency profiles. However, combining the ACE technique with water addition enhanced the wine textural quality by increasing the intensities of the crucial astringent wine quality sub-qualities, adhesive and graininess. Furthermore, insights into the chemical factors influencing the astringency sensations were provided in this study. This research indicates that wine producers may use ACE with pre-fermentation water dilution to reduce the wine alcohol level but maintain important textural components.

Bioavailability of red wine and grape seed proanthocyanidins in rats.

This study explored plasma levels and urinary and fecal excretion of metabolites and microbial-derived catabolites over a 24 h period following the ingestion of red wine (RWP) or grape seed (GSP) proanthocyanidin-rich extracts by rats. In total, 35 structurally-related (epi)catechin metabolites (SREMs) and 5-carbon side chain ring fission metabolites (5C-RFMs) (phenyl-γ-valerolactones and phenylvaleric acids), and 50 phenolic acid and aromatic catabolites were detected after intakes of both extracts. The consumption of the RWP extract, but not the GSP extract, led to the appearance of a ∼200 nmol L-1 peak plasma concentration of SREMs formed from flavan-3-ol monomers. In contrast, ingestion of the GSPs, but not the RWPs, resulted in a substantial increase in microbiota-derived 5-carbon side chain ring fission metabolites (5C-RFMs) in plasma. 5C-RFMs, along with low molecular weight phenolic catabolites were detected in urine after ingestion of both extracts. The GSP and RWP extracts had respective mean degrees of polymerisation 5.9 and 6.5 subunits, and the RWP extract had an upper polymer size of 21 subunits compared to 44 subunits for the GSP extract. The differences in plasma metabolite profiles might, therefore, be a consequence of this polydispersity impacting on the microbiota-mediated rates of cleavage of the proanthocyanidin subunits and their subsequent metabolism and absorption. Urinary excretion of phenolic catabolites indicated that 11% of RWPs and 7% for GSPs were subjected to microbial degradation. In all probability these figures, rather than representing the percentage of proanthocyanidins that are completely degraded, indicate partial cleavage of monomer subunits producing a much higher percentage of shortened proanthocyanidin chains. Obtaining more detailed information on the in vivo fate of proanthocyanidins is challenging because of the difficulties in analysing unabsorbed parent proanthocyanidins and their partially degraded flavan-3-ol subunit chains in feces. Further progress awaits the development of improved purification and analytical techniques for proanthocyanidins and their use in feeding studies, and in vitro fecal and bacterial incubations, with radio and/or stable isotope-labelled substrates.

Effect of Grape Seed and Skin Tannin Molecular Mass and Composition on the Rate of Reaction with Anthocyanin and Subsequent Formation of Polymeric Pigments in the Presence of Acetaldehyde.

Polymeric pigments formed via ethyl linkages between grape tannins and anthocyanins are important to the development of stable red wine color. To determine the effect of tannin structure on the stability and color properties of ethyl-linked polymeric pigments, tannin fractions with average polymerization between 4 and 43 units were prepared from grape skins and seeds and combined with malvidin-3-glucoside (M3G) in model wine containing acetaldehyde. As tannin molecular mass increased, the reaction rate with M3G increased. Compared with skin tannins of comparable molecular mass, seed tannins reacted more rapidly with M3G but were prone to precipitation. This resulted in a loss of polymeric pigments formed from seed tannins, which was greater as tannin molecular mass increased. Aggregation occurred following the reaction of seed tannin with M3G, concomitant with precipitation. The aggregation-precipitation phenomenon was not observed for skin tannin-derived pigments, indicating a greater stability in solution than those formed from seed tannins.

Oxygen exposure during red wine fermentation modifies tannin reactivity with poly-l-proline.

Red wines injected with nitrogen or oxygen during fermentation were used to identify the effect of gas exposure on tannin structure and reactivity with poly-l-proline. Tannin was purified from wine after fermentation and after three years of bottle storage. Tannin from nitrogen-treated wine had a lower percentage of galloylation and were less pigmented than tannin from oxygen-exposed wine. Self-aggregation of tannin was measured by nanoparticle tracking analysis and a larger particle size was observed for the oxidized treatment. The interaction of tannin and poly-l-proline was measured by isothermal titration calorimetry, and involved more hydrogen bonding than hydrophobic interactions in the case of nitrogen-treated wine tannin. Conversely, oxidized tannin was more hydrophobic and the association with poly-l-proline was entropy-driven due to a change of solvation. The results show meaningful changes in the structure and reactivity of tannin as a result of oxygen exposure during fermentation, which may impact astringency perception.

Commercial Saccharomyces cerevisiae Yeast Strains Significantly Impact Shiraz Tannin and Polysaccharide Composition with Implications for Wine Colour and Astringency.

To gain knowledge on the role of Saccharomyces cerevisiae yeast strains (and their hybrids) on wine sensory properties, 10 commercially available yeast strains were selected on the basis of their widespread usage and/or novel properties and used to produce Shiraz wines. Significant differences were evident post-alcoholic fermentation and after 24 months of ageing with regards to the number of wine compositional variables, in particular the concentration of tannin and polysaccharide. Strain L2323 is known for its pectinolytic activity and yielded the highest concentration of both yeast- and grape-derived polysaccharides. Wines made with the mannoprotein-producing strain Uvaferm HPS (high levels of polysaccharides) did not have elevated concentrations of yeast-derived polysaccharides, despite this observation being made for corresponding model fermentations, suggesting that mannoprotein production or retention might be limited by the wine matrix. Wine tannin concentration showed a high level of variability between strains, with L2323 having the highest, and AWRI1503 the lowest concentration. Sensory analysis of the wines after 24 months ageing revealed significant differences between the yeast strains, but only the attributes opacity (visual colour) and astringency could be predicted by partial least squares regression using the wine compositional data. Notably, the astringency attribute was associated with higher concentrations of both tannin and polysaccharide, contrary to reports in the literature which suggested that polysaccharide exerts a moderating effect on astringency. The results confirm previous reports demonstrating that the choice of yeast strain represents an opportunity to shape wine style outcomes.

Comparison of consecutive harvests versus blending treatments to produce lower alcohol wines from Cabernet Sauvignon grapes: Impact on polysaccharide and tannin content and composition.

A changing climate has led to winegrapes being harvested with increased sugar levels and at greater risk of berry shrivel. A suggested easy-to-adopt strategy to manage the associated rising wine alcohol levels is the pre-fermentative substitution of juice with either "green harvest wine" or water. Our study investigates the effects of this approach on Vitis vinifera L. cv. Cabernet Sauvignon wine quality attributes. Wines were also made from fruit collected at consecutive earlier harvest time points to produce wines comparable in alcohol to the substituted wines. Tannin concentrations and colour did not change significantly in the wines with modified alcohol content even at higher juice substitution rates. Differences in polysaccharide and tannin composition indicated variability in extraction dynamics according to substitution rate and type of blending component. In scenarios where berry shrivel is inevitable, the incorporation of water in particular offers much promise as part of a strategy to manage wine alcohol content.

Development and validation of an UHPLC-HRMS protocol for the analysis of flavan-3-ol metabolites and catabolites in urine, plasma and feces of rats fed a red wine proanthocyanidin extract.

This study developed, optimized and validated an ultra-high-performance liquid chromatography-high-resolution mass spectrometry (UHPLC-HRMS) method to identify and quantify metabolites and microbial-derived catabolites in urine, plasma and feces of rats following ingestion of 50 mg of a red wine proanthocyanidin-rich extract. The method was validated for specificity, linearity, limit of detection (LD) and quantification (LQ), intra-day and inter-day precision, recovery and matrix effects, which were determined for 34 compounds in the three biological matrices. After method validation, three parent flavan-3-ols, four 5-carbon side chain ring fission metabolites, and 27 phenolic acid and aromatic catabolites were quantified in plasma, urine and feces after red wine proanthocyanidin intake. These results establish the value of the UHPLC-HRMS protocol in obtaining a detailed picture of proanthocyanidin metabolites and their microbial-derived catabolites, along with their phase II metabolites, in biological fluids of rat, and potentially in human clinical studies designed to evaluate the bioavailability of dietary flavan-3-ols.

Influence of plant water status on the production of C13-norisoprenoid precursors in Vitis vinifera L. Cv. cabernet sauvignon grape berries.

The influence of irrigation strategy on grape berry carotenoids and C13-norisoprenoid precursors was investigated for Vitis vinifera L. cv. Cabernet Sauvignon. Two irrigation treatments were compared, one in which vines received reduced irrigation applied alternately to either side of the vine (partial rootzone drying, PRD) and a second control treatment in which water was applied to both sides of the vine. Over the two years of the experiments, PRD vines received on average 66% of the water applied to the controls. Initially, the PRD treatment did not alter midday leaf (psiL) and stem (psiS) water potential relative to the control, but decreased stomatal conductance (gs). Continued exposure to the PRD treatment resulted in treated grapevines experiencing hydraulic water deficit relative to the control treatment and induced lowered midday psiL and psiS, which was also reflected in decreased berry weight at harvest. In both irrigation treatments, the most abundant grape berry carotenoids, beta-carotene and lutein, followed the developmental pattern typical of other grape varieties, decreasing post-veraison. At certain points in time, as the fruit approached maturity, the concentration of these carotenoids was increased in fruit of PRD-treated vines relative to the controls. This effect was greater for lutein than for beta-carotene. PRD consistently caused increases in the concentration of hydrolytically released C13-norisoprenoids beta-damascenone, beta-ionone, and 1,1,6-trimethyl-1,2-dihydronaphthalene in fruit at harvest (24 degrees Brix) over two seasons. The effect of the PRD treatment on the concentration of hydrolytically released C13-norisoprenoids was greater in the second of the two seasons of the experiment and was also reflected in an increase in total C13-norisoprenoid content per berry. This suggests that the increases in the concentration of the C13-norisoprenoids in response to PRD were independent of water deficit induced changes in berry size and were not the result of an altered berry surface area to volume ratio.

Retention of Proanthocyanidin in Wine-like Solution Is Conferred by a Dynamic Interaction between Soluble and Insoluble Grape Cell Wall Components.

For better understanding of the factors that impact proanthocyanidin (PA) adsorption by insoluble cell walls or interaction with soluble cell wall-derived components, application of a commercial polygalacturonase enzyme preparation was investigated to modify grape cell wall structure. Soluble and insoluble cell wall material was isolated from the skin and mesocarp components of Vitis vinifera Shiraz grapes. It was observed that significant depolymerization of the insoluble grape cell wall occurred following enzyme application to both grape cell wall fractions, with increased solubilization of rhamnogalacturonan-enriched, low molecular weight polysaccharides. However, in the case of grape mesocarp, the solubilization of protein from cell walls (in buffer) was significant and increased only slightly by the enzyme treatment. Enzyme treatment significantly reduced the adsorption of PA by insoluble cell walls, but this effect was observed only when material solubilized from grape cell walls had been removed. The loss of PA through interaction with the soluble cell wall fraction was observed to be greater for mesocarp than skin cell walls. Subsequent experiments on the soluble mesocarp cell wall fraction confirmed a role for protein in the precipitation of PA. This identified a potential mechanism by which extracted grape PA may be lost from wine during vinification, as a precipitate with solubilized grape mesocarp proteins. Although protein was a minor component in terms of total concentration, losses of PA via precipitation with proteins were in the order of 50% of available PA. PA-induced precipitation could proceed until all protein was removed from solution and may account for the very low levels of residual protein observed in red wines. The results point to a dynamic interaction of grape insoluble and soluble components in modulating PA retention in wine.

Characterization of macromolecular complexes in red wine: Composition, molecular mass distribution and particle size.

Precipitates were prepared from two compositionally different Pinot noir wines with addition of excess ethanol, and contained primarily polysaccharide, tannin and protein. The ethanol-soluble material was further fractionated into polymeric (tannin) and monomeric phenolics. Tannin associated with precipitates was of a higher molecular mass than that remaining in ethanolic solution. Wine fractions were reconstituted at the ratios of the original wine and analyzed using nanoparticle tracking analysis. The average particle size of the tannin fraction was 75-89 nm, and increased when combined with the precipitate (≅ 200 nm). Addition of the monomeric fraction to the tannin-precipitate complex increased both the incidence and concentration of smaller particles, reducing the average particle size. The formation of aggregates occurred in all fractions and only minor differences in particle size distribution were found between wines. Differences in particle concentration between wines appear to be due to differences in the total concentration of macromolecules rather than compositional differences.

Evaluation of direct phloroglucinolysis and colorimetric depolymerization assays and their applicability for determining condensed tannins in grape marc.

To determine the optimum methods for determining condensed tannin (CT) content in grape marc, butanol-hydrochloric acid assays and phloroglucinolysis were adapted for use, applied to a range of grape marc types, and the methods compared. Porter's assay (butanol-HCl) was found to give unreliable results due to nonlinear color responses to grape skin and seed tannin concentrations, whereas the modification to include acetone (Grabber's assay) overcame this. Differences between skin and seed tannin responses highlighted the need to adequately select the correct grape tannin standard, and the formation of pH-dependent color was accounted for through acidification of blank samples. For phloroglucinolysis, the inability to remove highly bound tannins from cell wall material was highlighted, although a measure of tannins remaining post-phloroglucinolysis (Grabber's assay) showed a trend with the level of exposure to oxidative storage or processing conditions. The comparison of CT concentrations from phloroglucinolysis and Grabber's assay gave poor correlation coefficients.

Carbon allocation to the insoluble fraction, respiration and triose-phosphate cycling in the sugarcane culm.

The changes in carbon allocation to non-sucrose metabolic pathways were investigated in developing internodes of sugarcane. Radiolabelling studies were done on internode 3, 6 and 9 tissues, representing three stages of increasing maturity. Carbon partitioning into sucrose increased from 34% of total 14C uptake in internode 3, to 66% in internodes 9. In immature tissue, the protein and fibre components were the dominant competing sinks with sucrose for incoming carbon, to which 14 and 16% of carbon was allocated. Increased carbon allocation to sucrose with tissue maturity coincided with a decrease in partitioning to fibre and total respiration. Between internodes 3 and 9 carbon allocation to total respiration decreased by 9%, and to fibre by 14%. Carbon cycling between the triose- and hexose phosphate pools was evident in all internodes. More than 90% of carbon entering triose-phosphates was returned to hexose in internode 3 tissue, and this flux decreased with tissue maturity.

Selective extraction of polysaccharide affects the adsorption of proanthocyanidin by grape cell walls.

Cell wall material was isolated from two Vitis vinifera grape varieties, Cabernet Sauvignon and Shiraz, following a buffered phenol extraction method. Using sequential fractionation in chelating agent, then increasing the molarity of aqueous potassium hydroxide, polysaccharide classes were selectively extracted from cell walls to produce fractions of defined polysaccharide composition. Following the application of phloroglucinolysis and a modified HCl-butanol colourimetric assay to cell wall fractions, more than 54% of cell wall-bound proanthocyanidin was localised within the chelator-soluble (pectic) fraction. Model adsorption experiments with a purified skin proanthocyanidin confirmed that the removal of pectic polysaccharides by chelator most significantly reduced the adsorption of proanthocyanidin by cell walls. Nevertheless, cell wall hemicellulosic fractions retained a high binding capacity for proanthocyanidin, although lower than that observed when pectin was present. Following removal of hemicelluloses by fractionation, the primarily lignocellulosic residue had a significantly reduced affinity for proanthocyanidin. With the exception of lignocellulose, a greater selectivity of adsorption for higher molecular mass proanthocyanidins was observed by the respective cell wall fractions.

Low molecular weight procyanidins from grape seeds enhance the impact of 5-Fluorouracil chemotherapy on Caco-2 human colon cancer cells.

OBJECTIVE: Grape seed procyanidins (PC) are flavan-3-ol oligomers and polymers known for their biological activity in the gut. Grape seed extract (GSE) have been reported to reduce intestinal injury in a rat model of mucositis. We sought to investigate effects of purified PC fractions differing in mean degree of polymerization (mDP) combined with 5-Fluorouracil (5-FU) chemotherapy on the viability of colon cancer cells (Caco-2). DESIGN: SixPC fractions (F1-F6) were isolated from Cabernet Sauvignon seeds at two ripeness stages: pre-veraison unripe (immature) and ripe (mature), utilizing step gradient, low-pressure chromatography on a Sephadex LH-20 resin. Fractions were tested on Caco-2 cells, alone and in combination with 5-FU. Eluted fractions were characterized by phloroglucinolysis and gel permeation chromatography. Cell viability was determined by the 3-(4,5-Dimethylthiazol-2yl)-2,5-diphenyl-tetrazolium bromide) (MTT) assay. RESULTS: All isolated fractions significantly reduced Caco-2 cell viability compared to the control (P<0.05), but F2 and F3 (mDP 2-6) were the most active fractions (immature F2 = 32% mDP 2.4, F3 = 35% mDP 5.8 and mature F2 = 13% mDP 3.6 and F3 = 17% mDP 5.9; percentage of viable cells remaining) on Caco-2 cells. When combined with 5-FU, immature fractions F1-F3 enhanced the cell toxicity effects of 5-FU by 27-73% (P<0.05). Mature seed PC fractions (F1-F4) significantly enhanced the toxicity of 5-FU by 60-83% against Caco-2 cells (P<0.05). Moreover, some fractions alone were more potent at decreasing viability in Caco-2 cells (P<0.05; immature fractions = 65-68% and mature fractions = 83-87%) compared to 5-FU alone (37%). CONCLUSIONS: PCs of mDP 2-6 (immature F1-F3 and mature F1 and F4)not only enhanced the impact of 5-FU in killing Caco-2 cells, but also surpassed standard 5-FU chemotherapy as an anti-cancer agent.The bioactivity of PC is therefore attributed primarily to lower molecular weight PCs.

Factors affecting skin tannin extractability in ripening grapes.

The acetone-extractable (70% v/v) skin tannin content of Vitis vinifera L. cv. Cabernet Sauvignon grapes was found to increase during late-stage ripening. Conversely, skin tannin content determined following ethanol extraction (10, 20, and 50% v/v) did not consistently reflect this trend. The results indicated that a fraction of tannin became less extractable in aqueous ethanol during ripening. Skin cell walls were observed to become more porous during ripening, which may facilitate the sequestering of tannin as an adsorbed fraction within cell walls. For ethanol extracts, tannin molecular mass increased with advancing ripeness, even when extractable tannin content was constant, but this effect was negligible in acetone extracts. Reconstitution experiments with isolated skin tannin and cell wall material indicated that the selectivity of tannin adsorption by cell walls changed as tannin concentration increased. Tannin concentration, tannin molecular mass, and cell wall porosity are discussed as factors that may influence skin tannin extractability.

Comparison of extraction protocols to determine differences in wine-extractable tannin and anthocyanin in Vitis vinifera L. cv. Shiraz and Cabernet Sauvignon grapes.

Cabernet Sauvignon and Shiraz grapes were sourced from different regions within Australia, and microvinified with a skin contact period of 6 days. Grape samples were extracted using two protocols: a 15% v/v ethanol, 10 g/L tartaric acid extract of gently crushed berries (wine-like, WL) and a 50% v/v ethanol, pH 2 extract of grape berry homogenate. It was found that in WL extracts, grape tannin and anthocyanin concentrations were strongly related to wine tannin, anthocyanin and color density achieved during the skin contact period. No relationship was observed for grape tannin concentration analyzed in homogenate extracts and wine tannin, but a strong, positive relationship was found for anthocyanin concentration. When the data obtained from homogenate extraction was treated separately by grape variety, a stronger relationship between grape and wine tannin concentration was observed. Tannin compositional analysis in wines indicated that higher tannin concentrations were due to the extraction of tannin of higher molecular mass during fermentation, most likely from grape skins.