Characterisation of polyphenols and antioxidant potential of red and white pomace by-product extracts using subcritical water extraction

A detailed assessment of the content of high added-value compounds in grape pomace varieties was carried out following a subcritical water extraction method. High amounts of anthocyanins and tannins were recovered from fermented grape pomace at differential temperatures with high variability between t h e by-products observed. Contrary to anthocyanins, high extraction temperatures (about 200 °C) yielded higher amounts of tannins. Overall, we found that grape pomace antioxidant activity and total polyphenols, quantified by t h e Folin Ciocalteu method, were not directly related to the main polyphenol content in SWE extracts. The data obtained in our study by using laboratory-scale equipment will be useful for developing an industrial-scale SWE process. Finally, it was shown that grape pomace by-products can be considered a s an important source of polyphenols. They could therefore potentially provide a basis for the sustainable and integrated exploitation of winemaking by-products, to be used as inexpensive and readily available sources of bioactive compounds for the pharmaceutical, cosmetic and food industries. green


InTrODUCTIOn
Grapes are one of the most produced fruit in the world for human consumption, especially for wine production.About 66.4 million tonnes of grapes have been produced annually in recent years (FAOSTAT, 2012).European countries, such as France, Italy, Spain and Germany, account for more than 44 % of the production (OIV, 2013).Such high wine production results in a large amount of grape by-products, in particular grape pomace, which accounts for 17-25 % of the quantity of grape produced (Pinelo et al., 2005;Arvanitoyannis et al., 2006).Grape pomace is a source of polyphenols, oil, alcohol and tartaric acid (Jackson, 2008).The residual alcohol in grape pomace is generally extracted.The seeds found in the pomace are a main source of grape seed oil, while the grape skin -although containing high levels of polyphenols -has not yet been utilised to its full potential.Due to a high polyphenol content, pomace cannot be utilised as feedstock or crop fertiliser (Devesa-rey et al., 2011).For this reason, it is of considerable interest to examine ways of extracting polyphenols from the grape by-product.
Polyphenols are commonly extracted using an organic solvent.The method is simple, because the agents and equipment required for the process are easily obtainable (Spigno and De Faveri, 2007).However, as a result of degradation during solvent regeneration the yield is generally low.In addition, organic solvents substantially increase extraction process costs (Galanakis, 2012).Furthermore, the remaining solvent needs to be regenerated thoroughly from the product, leading to the generation of a large amount of organic solvent wastes (Yammine et al., 2014).
Alternatively, a number of studies have demonstrated the use of water at subcritical (100 °C<T<374.2°C) conditions (SWE) as an environmentally friendly and effective extraction process (Ju and Howard, 2003).Under these conditions, the dielectric constant of water (ε) changes dramatically with the change in temperature.The value of ε at 25 MPa decreases with temperature from 60 at ambient temperature to 21 at 250 °C, thus the reaction field changes from ionic reaction to radical reaction.In general, several studies h a v e d e m o n s t r a t e d t h a t water under subcritical conditions is an effective extraction solvent (Plaza and Turner, 2015), which can be applied for the extraction of several bioactive compounds, such as phenolic compounds from lemon balm (Miron et al., 2013), potato peel (Singh and Saldaña, 2011) and polysaccharides from golden oyster mushroom (Jo et al., 2012).As can be seen in  (Rajha et al., 2014).Meanwhile, the extraction of anthocyanins from red grape pomace at 110 °C was also variable, ranging from 5.93 mg/100 g DM (Ju and Howard, 2005) to 450 mg/100 g DM (Monrad et al., 2014) for different sources of by-products.The extraction recovery of other families of compounds, such as catechins and proanthocyanidins, showed that selective extractions of compounds also varied.This indicates that extraction temperature, the type of by-product utilised, and the manner in which it was treated before extraction, all have a major influence on extraction.
Thus, the aim of the present work was to characterise the phenolic compounds from the grape pomace by-products of four different cultivars of Vitis vinifera (Chardonnay, Cabernet Franc, Merlot, Dunkelfelder), in order to identify a n y r e l e v a n t properties t h e y m a y h a v e t h a t c o u l d be used as functional ingredients, and to compare them at different extraction temperatures.This investigation consisted in i) determining the total phenolic and total tannin content of the grape pomace by-products, i i ) identifying and quantifying monomeric and oligomeric (dimer and trimer) flavan-3-ol composition using High Pressure Liquid Chromatography (HPLC), and i i i ) estimating t h e i r antioxidant capacity c a r r y i n g o u t four different assays (ABTS, CUPRAC, FRAP, and ORAC).The data may contribute to the selection of suitable grape pomace for the development of antioxidant and polyphenolic rich nutraceuticals.

Chemicals
Copper

raw material
In this study, grape pomace by-products were used, which were obtained from representative red and white grape varieties (Vitis vinifera L.) cultivated in Switzerland: Chardonnay, Cabernet Franc, Merlot and Dunkelfelder.Dunkelfelder is a teinturier grape variety known in Changins, whereas the other three grape varieties are well-known and widely cultivated elsewhere.Samples were provided by t h e University of Changins winery (Switzerland) during the 2012 and 2013 harvests.To limit the influence of external factors, and to obtain a better comparison among results, all samples were from the same geographical area (46°23'56.4»N6°13'58.9»E).The grapes used were harvested at the optimum technological ripeness.While the fourth variety considered, Chardonnay grape pomace was collected the day of grape harvest after destemming and pressing of the grapes under identical conditions.
Once pressed, all the grape pomaces were combined and homogenised to ensure representative sampling of the whole grape pomace.For the three red varieties, pomaces were collected immediately after pressing at 2 105Pa (RPS 50, Bucher Vaslin SA, France) and were treated with 50 mg of SO 2 per kg of raw material (RM).Samples were stored at -20 °C under vacuum until further processing.
In order to avoid complications with repeatability due to the heterogeneity of the raw material, the skins and the seeds were separated with a vibrating sifter (Retsch GmbH, Germany).Small fractions (diameter Ø<2.8 mm) and large fractions (Ø>5.5 mm) were t h e n removed and the two standardised fractions were manually and homogeneously mixed (49 % of seeds and 51 % of skins fresh weight).

Process of extraction and parameters
The schematic diagram of apparatus used for the extraction of polyphenolic compounds using subcritical water is shown in Figure 1.
In the extraction system, a HPLC pump (I.C.S.National 1100) was used for deionised water delivery, pressurisation and s y s t e m p r e s s u r e control.A pressure transducer (Swagelock NG160) and thermocouple (Eurotherm Automation 90) were installed in the custom-made high-pressure vessel to monitor both pressure and temperature of t h e system.T h e extract was collected in a n inerted vessel (65 mL volume) once it had passed through an ice bath.
In each run, the pomace (13.00 g) was loaded into the high-pressure vessel, with a capacity of 26 cm3 of material (Figure 1).The liquid-tosolid ratio was maintained at the value of 5, to obtain 65 mL of extract.The vessel was placed in an oven at several temperatures (100 °C, 150 °C, 200 °C).The outlet valve of the extraction vessel was then closed and the system was pressurised to the desired pressure of 25 10 5 Pa at a constant flow rate of 6 mL/min.The p o m a c e a n d t h e solution collected in the inerted sampling vessel were then stored at 4 °C for further analysis without any preliminary preparation steps.

Conventional extraction experiments
The extraction of polyphenols from grape pomace (100.0 ± 0.1 g) was carried out in a mixture of ethanol and water (50/50, v/v) and maintained at ambiant temperature in a cylindrical extraction cell.The liquid-to-solid Sami Yammine et al.

FIGUrE 1. Schematic diagram of the p ressurised liquid extraction process
ratio was maintained at a value of 5. A gentle agitation at 160 rpm (16.8 rad•s −1 ) was maintained using a round incubator of 12.5 mm shaking throw (Infors HT Aerotron, Bottmingen, Switzerland).For untreated samples, the same protocol of extraction was used.Regular sampling was carried out during t h e 420 min of extraction.At the end of t h e extraction process, the juice was separated from grape pomace by centrifugation (Model 3-16P, Sigma Laborzentrifugen GmbH, Germany) at 3076 g for 10 min, and stored at −18 °C for further analysis.

Total polyphenol content
The total phenolic content was spectrophotometrically measured according to a modified Folin-Ciocalteu method and applied on 96-well microplates.Stock solutions (10 mg/mL) of the grape pomace extracts were prepared in EtOH/H2O (25:75, v/v), and a microplate spectrophotometer (MultiSkan Spectrum, Thermo Scientific) was used for incubation and measurement.Each well was filled with 184 µL of distilled water and 24 µL of the sample solution, followed by 12 µL of the Folin-Ciocalteu reagent and 30 µL of 20 % (w/v) Na2CO3 solution.Prior to the measurement of the absorbance at 765 nm, the mixture was incubated for 1 h under dark conditions at 25 °C.Gallic acid (0−24 mg/L) was used as a standard for calibration.The results, expressed as milligrams of gallic acid per 100 g of grape pomace sample (on a dry matter basis, DM), were given as the mean of six determinations.
The total proanthocyanidin content of grape pomace by-products were measured using the Bate-Smith reaction (Bate-Smith, 1954).

Antioxidant activity
Polyphenols extracted from grape are wellknown for their antioxidant capacity.This antioxidant activity is not a single reaction, but comprises a wide range of multiple mechanisms.It i s usually recommended using several techniques, since no single technique is able to take into consideration all antioxidant mechanisms.Therefore, four different antioxidant capacity assays were used: one fluorometric assay based on hydrogen transfer, (ORAC) and three spectrophotometric assays based on electron transfer ( ABTS, CUPRAC and FRAP).FLUOstar Optima (BMG LabTech) was used for the first assay, and an automated microplate reader (MultiSkan Spectrum (Thermo Scientific) for the other three.For the spectrophotometric methods, stem extract solutions (4 mg/10 mL) were prepared in EtOH/H 2 O (25:75, v/v).Additional diluted stock solutions of the sample extracts (20 mg/1L) were prepared in 75 mM phosphate buffer (pH 7.4) for the ORAC measurement.The difference in absorbance between the final reading and the reagent blank reading was correlated with Trolox standard curves in all assays.Because the moisture level of each pomace extracted sample was quite different, the antioxidant capacity was reported on a dry weight basis to enhance comparison with the literature.Thus, the results were expressed as milligrams of Trolox per gram of grape sample (DM).Each resulting value was given as a mean of six determinations.

ABTS Assay
The ABTS radical cation (ABTS•+) was prepared i n 9 6 -w e l l m i c r o p l a t e s by reacting equivalent volumes (1:1) of both aqueous solutions of 7 mM ABTS and 2.45 mM potassium persulfate.This stock solution was left to react for 12-16 h at room temperature in the dark and w a s t h e n s t o r e d i n t h e s a m e thermal and light conditions to be used within two days.At the moment of the analysis 8 mL of the ABTS solution was diluted with EtOH/H 2 O (25:75, v/v) in a 100 mL volumetric flask to obtain an absorbance of 1.00 ± 0.02 units at 734 nm.The extract solutions and ABTS reagent (190 µL in each well) were prewarmed in a 96-well microplate at 25 °C for 20 min.Next, a b l a n k reagent reading was taken at a wavelength of 734 nm.The reaction was carried out by adding 10 µL of the pomace extract solution to each well.After 3 min of shaking, the mixture was incubated at the same temperature for a 30 min period, and then the absorbance decrease was measured at the same wavelength.

CUPRAC Assay
The cupric reducing antioxidant capacity a n a l y s i s of the sample extracts was carried out in 96-well microplates.The CUPRAC reagent was prepared just before the analysis by reacting equal volumes (1:1:1) of 10 mM Cu(II) aqueous solution, 7.5 mM neocuproine ( in 9 6 % f r e s h l y p r e p a r e d EtOH), and ammonium acetate buffer (1 M, pH 7).Pomace extract solution and 190 µL of CUPRAC reagent for each determination were incubated in a 96well microplate under the same conditions as the ABTS assay.Once the initial absorbance had been read at 450 nm, 10 µL of the pomace extract solution was added to each well.After 3 min of shaking, the mixture was incubated at 25 °C for 30 min, and then the absorbance increase was measured at the same wavelength.The Trolox standard curve was linear between 0 and 1.3 mM (R2=0.996).

FRAP Assay
The ferric reducing antioxidant power assay was carried out in 96-well microplates.The fresh working FRAP reagent was prepared by mixing a 0.01 M TPTZ solution in 0.04 M HCl, a 0.02 M FeCl3•6H2O aqueous solution, and an acetate buffer (pH 3.6, 3.1 g of sodium acetate and 16 mL of acetic acid glacial per liter of buffer solution) at a ratio of 1:1:10.All of these solutions were prepared on the day of analysis, except for the buffer and hydrochloric solutions.For the measurement of the antioxidant activity by the FRAP method, the protocol and experimental conditions were exactly the same as those reported for the ABTS and CUPRAC assays.However, the increase in absorbance was measured at 593 nm and the Trolox calibration curve was obtained using concentrations of 0 to 1.6 mM (R2=0.996).

ORAC Assay
The oxygen radical absorbance capacity analysis was carried out using 96-well fluorescence microplates.The reaction was carried out in a phosphate buffer (75 mM, pH 7.4).In this order, 30 µL of the pomace extract solution, 180 µL of fluorescein (117 nM final concentration), and 90 µL of AAPH (40 mM) were added to each well.The mixture was shaken and left to stand for 1.5 h at 37 °C.Fluorescence was recorded every minute during this period at excitation and emission wavelengths of 485 and 530 nm respectively.A blank sample (phosphate buffer replaced the sample) and Trolox calibration solutions (1-40 µM) were also performed (R2=0.983)simultaneously on the same microplate.The area under the curve (AUC) was calculated for each extract sample by integrating its relative fluorescence curve.By subtracting the AUC of the blank, the net AUC of the pomace extracts was calculated and correlated with Trolox concentrations.

Anthocyanin analysis
Quantitative and qualitative analyses of anthocyanins were performed on the extracts with high performance liquid chromatography (HPLC) after 420 min of extraction.Samples of extracts were diluted (ratio 1/10) in acidified water (0.1 % formic acid) and then filtered through Polyamide filters (pores diameter Ø = 0.45 µm).The system used for the anthocyanin analysis was an Agilent 1200 HPLC Series (Agilent Technologies) equipped with a diode array detector.The separation was carried out with a Prontosil C18AQ column (4.6 × 250 mm, 5 µm, Bischoff Chroma-tography, Germany) operated at 25 °C in reverse phase.

Statistics
The Variance analysis was used for data analysis.
The statistical significance of the differences in the data was obtained using the Tukey's test (α = 5 %).Data processing was carried out using XLSTAT (Addinsoft SARL, France) software.
rESULTS AnD DISCUSSIOn

Total polyphenol content
The total phenolic content of subcritical water extracts of grape pomace of four different grape varieties according to e x t r a c t o n temperature is shown in Temperature had a high influence on the extracted total polyphenols.For example, for Merlot pomace an increase of 1.08 ± 0.2 g to 2.29 ± 0.18 g of GA/100g DM i n p o l y p h e n o l content was observed with a 100 °C to 200 °C temperature increase, and the polyphenol content w a s above that of conventional solvent extraction at temperatures higher than 150 °C.
Another important factor was the influence of harvest year on the Dunkelfleder variety, with significant differences for total polyphenol content being linked to all three extraction temperatures; for example, the extraction of the 2012 harvest and 2013 harvest at 200 °C resulted in 3,66 ± 0,19 and 2,72 ± 0,09 g of GA/100 g DM respectively.
In other studies using subcritical extraction and conventional solvent extraction, polyphenol content h as been f o u n d to largely vary from 1.06 to 6.23 g of GA/100 g depending on the grape cultivar, geographical origin a n d vintage winemaking process (Ju and Howard, 2003;García-Marino et al., 2006;Luque-Rodríguez et al., 2007;Casazza et al., 2012;Vergara-Salinas et al., 2013;Duba et al., 2015a;Vergara-Salinas et al., 2015).The results obtained in the present study are in agreement with the aforementioned range.Specifically, total phenolic content of weight (g) of GA/100 g was given by Aliakbarian et al. (2012) for grape red Cortina pomace.These results a r e also comparable w i t h t h o s e i n s t u d i e s which utilised organic solvent for the extraction (González-centeno et al., 2012;Ky et al., 2014).

Total proanthocyanidins content
The total proanthocyanidin content of grape pomace by-products, obtained with t h e Bate-Smith reaction, are shown in Table 2. Similar to phenolic content quantification using the Folin Ciocalteu method, Dunkelfelder subcritical water extraction at 200 °C gave the highest value of 72.52 ± 2.43 mg/g DM, whereas Cabernet Franc pomace had the lowest value of 11.67 ±1.67 mg/g DM.Significant differences (p<0.05) were observed a m o n g the grape varieties, years of harvest and extraction temperatures.
Extraction temperature had a high influence on total extracted proanthocyanidins.In our case, for example, polyphenols extracted from Chardonnay pomace showed an increase of 54.20±1.33 to 68.37±4.17mg/g DM in content of total proanthocyanidins by increasing the temperature from 100 °C to 200 °C, and the polyphenol content was above conventional solvent extraction at temperatures higher than 100 °C.
The year of harvest a l s o had an important influence on proanthocyanidin content.F o r t h e Dunkelfleder variety, t h e r e w e r e significant differences in proanthocyanidin content depending on the harvest year for each of the three extraction temperatures; for example, at 150 °C t h e extractions of t h e 2012 harvest a n d 2 0 1 3 h a r v e s t resulted in 68.76 ± 2.55 and 52.31 ± 0.59 g of tannins/100g DM respectively.
The total proanthocyanidin Bate-Smith test is a coloration method used to detect the presence of condensed tannins, which comprise an important fraction of the extract that is usually overlooked when quantifying subcritical water pomace extracts.For this reason, it was difficult to compare results to other subcritical water extracts.Nonetheless, t h e results obtained in the present study were similar to those previously reported in the literature for pomace by-products from white and red grape varieties extracted using an organic solvent (Rockenbach et al., 2011;Mandic et al., 2008;Obreque-Slier et al., 2010;Travaglia et al., González-centeno et al., 2012).Nonetheless, t h e observed total tannins values of 68 mg/g DM for Chardonnay pomace was 2.2-fold higher than those obtained by González-centeno et al. (2012) using a solvent of MeOH/water (60:40,v/v) for extraction.These differences can be attributed to the different vintage and viticulture conditions of the samples.
As previously observed in several studies (Mandic et al., 2008;Lorrain et al., 2013;Ky et al., 2014) a highly significant correlation was found between the total phenolic and total proanthocyanidin contents of the grape pomace extracts (r=0.94,p <0.05).

HPLC analysis of monomeric and oligomeric flavan-3-ols
The monomeric and oligomeric flavan-3-ol composition of grape pomace by-products from four grape varieties extracted via subcritical water at different temperatures was investigated and is given in Table 2.All the extracts were analysed using HPLC to identify and quantify the flavan-3-ols procyanidin B1, (+)-catechin, (− )-epicatechin, and the trimer C1, in that order of elution.
The combined amount of the above flavan-3-ols in grape pomace by-products ranged from 27.90 to 198.86 mg/100 g DM, for t h e Merlot (SWE 100 °C) and Dunkelfelder (2012, SWE 200 °C) varieties.These results are in accordance with those published b y L u q u e -R o d r í g u e z e t a l .( 2 0 0 7 ) w i t h a total flavan-3-ol range of 29 to 199 mg/100 g DM for a red grape pomace by-product (8 MPa, 120 °C, 1:1 (v/v) ethanol, 0.8 % (v/v) HCl).Significant differences were found among the four varieties, year of harvest, and extraction temperature (p<0.05),w i t h both Dunkelfelder and Chardonnay exhibiting the highest total flavan-3-ol content of 198 mg/100 g DM and 97 mg/100 g DM respectively at 200 °C.
The amount of extracted Flavan-3-ols can be linked to the temperature of t h e subcritical water extraction ; for example, increasing the temperature from 100 °C to 200 °C led to a 1.37-to 1.91-fold increase in F l a v a n -3 -o l extracts.Temperature had a differential influence on individual compounds; (+)-catechin and (-)epicatechin were optimally extracted at 200 °C for all grape pomaces, while Proanthocyanidins B1 and C1 were optimally extracted at 150 °C for Cabernet Franc, Merlot and Chardonnay.
The content ratio of both monomers, (+)catechin and (-)-epicatechin, accounted for 65 to 81 % of the total flavan-3-ol quantified content of grape pomaces, depending on the grape variety, extraction temperature and year of harvest.Apart from the Chardonnay variety, the monomeric fraction was g e n e r a l l y greater than the dimeric and trimeric ones at high temperatures.This observation agrees with that reported by Monrad et al. (2014) for red grape pomace (V.labrusca L.) , as well as with the results described by different authors for red grape pomace (V.vinifera L.) (Vergara-Salinas et al., 2015;Duba et al., 2015b).The optimal extraction conditions are shown in Table 1.
These ratios of higher quantity of monomers with respect to dimer and trimer have been previously observed in the literature for skins and/or seeds of different grape pomaces extracted by subcritical water (Srinivas et al., 2011;García-Marino et al., 2006;Vergara-Salinas et al., 2013;Bucić-Kojić et al., 2011;Monrad et al., 2014).

HPLC Analysis of anthocyanins for red grape by-products
The anthocyanin content of skin extracts was analysed u s i n g HPLC and the concentrations of the obtained compounds are shown in Table 3.
For total anthocyanins, Dunkelfelder 2012 and 2013 extracted at 100 °C (47.94 mg/g DM and 40.04 mg/g DM respectively), and Cabernet Franc (12.1 mg/g DM) superior amounts to lower temperatures.Grape variety had a stronginfluence on quantity of total anthocyanins and the ratios of individual anthocyanins.
Undeniably, "teinturier" cultivars ( i .e ., Dunkelfelder) had higher anthocyanin content than "non-teinturier" grapes (i.e., Merlot and Cabernet Franc).As can be seen in Table 3, the Dunkelfelder pomace m a i n l y contained malvidin-3-O-glucoside (72-78 %) o u t of the total anthocyanins found in the extract in contrast to t h e other two cultivars used for the extraction.It also contained unusually higher amounts of peonidin-3-O-glucoside (6.7.1 mg/g DM at SWE 100 °C) than Merlot and Cabernet Franc under the same e x t r a c t i o n conditions.These results are in accordance with several studies that have been carried out on teinturier grape varieties (Hermosín-Gutiérrez and García-Romero, 2004;Ky et al., 2014), in which higher ratios of peonidin-3-O-glucoside w e r e f o u n d in teinturier grape varieties than in other varieties.
In the 2012 and 2013 vintages, pomace from Dunkelfelder contained variable levels of anthocyanins the ratio of the anthocyanins stayed the same.For both D u n k e l f e l d e r vintages, the major anthocyanin was malvidin-3-O-monoglucoside, accounting for 72 % total anthocyanins, while f o r t h e other varieties i t ranged from 21 to 65 %.In t h e 2013 byproducts, less anthocyanins were observed.
Values ranged from 1.4 mg/g DM to 10.6 mg/g DM for glycosylated anthocyanins, The temperature of subcritical water extraction was the most important factor to contribute to a varied amount of extracted anthocyanins; for example, increasing the temperature from 100 °C to 200 °C led to a 1.37-to 1.91-fold decrease.Temperature had a differential influence on individual compounds; (+)-catechin and (-)-epicatechin were optimally extracted at 200 °C for all grape pomaces.
The optimum e x t r a c t i o n depended on the molecules; with average optimum temperatures (0.47 mg/100 mg) of around 100 °C.These molecules were optimally extracted at a lower temperature than flavonoid and phenolic acids.The presence of a sugar molecule in glycoside anthocyanins tends to make them more soluble in polar substances, and they can therefore be extracted better at a lower temperature than less-polar flavonoid and phenolic acids (Monrad et al., 2010b).Furthermore, anthocyanins are highly thermolabile compounds, due to the presence of the glucoside function that leads degradation at high temperatures during extraction (Ko et al., 2014).Proanthocyanidins B1 and C1 were optimally extracted at 150 °C for Cabernet Franc, Merlot and Chardonnay.

Antioxidant capacity
The antioxidant potential of each sample was determined in order to select the most active grape pomace among studied varieties.T h e antioxidant capacity of each extract cannot be assessed by applying a single method.Indeed, antioxidant measurements can be related to the capacity of extracts to either directly transfer hydrogen to a radical (ABTS, FRAP) or to act as competitors for the peroxy radicals (ORAC, CUPRAC) (Roginsky and Lissi, 2005).Hence, more than one type of antioxidant measurement needs to be performed to take into account the various mode of action of antioxidants (Huang et al., 2005).In that context, the free radical scavenging capacities of seed and skin extracts were evaluated using the f o l l o w i n g four tests: the FRAP, ABTS decolorisation tests and the CUPRAC and ORAC assays.
The results of these tests are shown in Table 4. Similar behavior patterns were observed i n a l l f o u r t e s t s , regardless of their action mechanism.The Dunkelfelder variety produced the highest antioxidant capacities extracted at 200 °C.
Meanwhile, the Merlot variety had the lowest values at the same temperature, with an antioxidant potential 1.6 times lower than that observed for the Dunkelfelder variety.Furthermore, i n the CUPRAC assay, significant differences (p<0.05) were found among the antioxidant capacity values of all four grape pomaces.In the ORAC assay, h o w e v e r, n o s i g n i f i c a n t d i ff e r e n c e w a s found between the Merlot and Cabernet Franc varieties, which had the lowest antioxidant potential (p>0.05).A c o n s i d e r a b l e r i s e i n a n t i o x i d a n t a c t i v i t y w a s o b s e r v e d f o r all the examined grape pomace extracts when the temperature was increased from 100 °C to 200 °C.Out of the the four varieties studied, total phenolic content was particulary high in the Dunkelfelder variety.
The comparison of literature on the antioxidant capacity of winemaking by-products is quite challenging, due to the fact that we used different analytical methods (such as CUPRAC, ABTS and FRAP, etc.), a variety of standards and reference units, and importantly, differing grape materials of reference.The antioxidant capacity is also affected by other factors, such as the winemaking procedure, geographical origin of the samples and the extraction methodology.
Nonetheless, in our study, the same orders of magnitude as those previously described in literature for the antioxidant capacity of grape pomace extracts were observed, irrespective of the analytical method applied or grape variety used.It was difficult to find an antioxidant assay on extracts from grape pomace to compare with.Ju and Howard (2005) reported antioxidant capacity ranges measured by the ORAC assay to be higher t h a n those extracted at 160 °C in the present research (1105 mg of Trolox/g DM).A larger scope of comparison to grape extracted solvent gave lower antioxidant values: Sánchez-Alonso et al. (2007) reported the antioxidant capacity of Airén white grape pomace, measured by ABTS (71.1 mg of Trolox/g DM) and FRAP assays (116.6 mg of Trolox/g DM) as being similar to the extracts obtained at 100 °C.
Pearson's correlation coefficients were calculated to evaluate the agreement of the expression of the grape pomace antioxidant capacity among the four assays applied.Regardless of the pair of methods considered, a high, significant and positive correlation was observed (r≥0.84,p<0.05), suggesting that ABTS, CUPRAC, FRAP, and ORAC assays give comparable and interchangeable antioxidant capacity values for grape pomaces.Correlation coefficients among antioxidant capacities based on ORAC and FRAP assays were the highest (r≥0.96),whereas ABTS data exhibited little to low correlation values ranging from 0.62 to 0.89.The different degrees of correlation among these four assays may be due to the different chemical information provided, depending on the electron or hydrogen transfer mechanism on which they are based.
A correlation with total phenolic content was exhibited by the ORAC, ABTS, FRAP and CUPRAC assays (r=0.350.53, 0.22 and 0.15 respectively at p<0.05); this low correlation has not previously been observed in the literature for skins, seeds, and grape pomaces (Ju and Howard, 2005;Lafka et al., 2007;Aliakbarian et al., 2012).When comparing the total tannin content and the antioxidant capacity of the grape pomace extracts , a lower correlation was observed (0.12≥r≥0.51, p <0.05).Similar results were obtained when correlating total anthocyanin content and the antioxidant capacity of the grape pomace extracts : a lower correlation was observed (0.05≥r≥0.16, p<0.05).This i s unexpected, because the anthocyanins were degraded at temperatures above 100 °C, while t h e antioxidants increased with temperature and peaked at 150 to 200 °C.

COnCLUSIOn
The present research carried out a detailed evaluation of the phenolic composition (total phenolic and total proanthocyanidin content, monomeric and oligomeric flavan-3-ol composition, and proanthocyanidin profile, anthocyanins) and antioxidant potential of white grape pomace by-products derived from the vinification process.To the best of our knowledge, no studies addressing this variability of the by-product in such a detailed way for multiple grape varieties have been previously published.
In conclusion, by using the subcritical water extraction method, high amounts of anthocya- nins and Flavan-3-ols were recovered from fermented grape pomace a t different temperatures with a high variability between by-products.Contrary to anthocyanins, high extraction temperatures (about 200 °C) yielded higher amounts of tannins.Overall, we found that grape pomace antioxidant activity and total polyphenols, quantified by a p p l y i n g t h e Folin Ciocalteau method, were not directly related to the main polyphenol content in SWE extracts; this critical point needs to be investigated further.The data obtained here from laboratory-scale equipment could be useful for developing industrial scale SWE processes.
Finally, it has been shown that grape pomace by-products can be considered as an important source of polyphenols.They could therefore potentially provide a basis for the sustainable and integrated exploitation of winemaking byproducts, and be used as inexpensive and r e a d i l y available sources of bioactive compounds for the pharmaceutical, cosmetic and food industries.