Properties of Wine Pomace Flours from Koshu and Cabernet Sauvign on Grapes Grown in Japan

Wine pomace is a solid byproduct of winemaking and is often destined for waste. Wine pomace can be an upcycle resource because it contains compounds with properties functional for human health. Characteristics of pomace from Japanese Koshu grapes (for white wine) and macerated Cabernet Sauvignon red grapes were compared by evaluating the levels of various phenols, antioxidant activity, inhibitory effects on angiotensin-converting enzyme, heat tolerance, and resistanceto in vitro gastrointestinal digestion. In addition, the stability of pomace resveratrol in cookies stored for a year was investigated. Cabernet Sauvignon pomace was superior in the levels of total polyphenols, resveratrol, and proanthocyanidins compared with Koshu pomace. There was no significant difference in γ-aminobutyric acid levels between Koshu and Cabernet Sauvignon pomace extracts. Antioxidant activity and the inhibitory effect on angiotensin-converting enzyme were higher in Cabernet Sauvignon pomace extracts than in Koshu extracts. Heating reduced the levels of components, but the total polyphenols content of Koshu pomace was the least affected by heating. Correspondingly, there was no decrease in the antioxidant activity of Koshu pomace extract after heating. In vitro small intestine digestion decreased the levels of total polyphenols and proanthocyanidins, but not resveratrol levels. Resveratrol levels in cookies containing pomace flour and stored dry at ambient temperature were stable for at least a year. Overall, Cabernet Sauvignon pomace would be more useful as an upcycled resource based on the levels of useful ingredients, in comparison with Koshu pomace.

Introduction

Grape (Vitis spp.) is one of the most valuable fruits in the world, and the total harvest of 75% is used for producing wine [1]. The most common to wine grape is Vitis vinifera. In Japan, though most grape cultivars are used for table grapes, 66.5% of the total grape production, regardless of cultivar, is used for wine production [2]. Wine shipments in Japan account for about 4% of all alcoholic beverages, but domestically produced wines account for 34% of the wine shipments [3]. One of the typical red wine grapes of Vitis vinifera cultivated in Japan is cv. Cabernet Sauvignon which production is second to cv. Merlot as a vinifera variety. Leading cultivar for white wine is cv. Koshu that is an indigenous Japanese variety of Vitis vinifera [4].

The generation of a substantial quantity of waste is unavoidable in the wine-making process. Improved waste management is an important issue for the wine industry to enhance its environmental performance. One such waste product is the solid residue, called “grape pomace,” amounting to 10%−30% of the total weight of grapes [5]. Grape pomace comprises skin, pulp, seeds, and stalks, generated after pressing grapes in white winemaking or after maceration and fermentation in red winemaking. Until now, a limited amount of grape pomace has been traditionally reused to produce spirits such as grappa, extract seed oils, recover pigments, composting and provide livestock feed [5], [6].

Grape pomace includes a broad range of phenolic compounds, such as phenolic acids, flavonoids, proanthocyanidins (PAs), anthocyanins, and stilbenes [7]. There is reliable evidence that phenolic compounds and polyphenol-rich foods contribute positively to human health, especially in the prevention of chronic diseases [8]. Excessive oxidative stress induced by reactive oxygen species and free radicals is likely to be a primary or secondary cause of many chronic diseases, such as cardiovascular disease [9]. The preventive and therapeutic effect of polyphenols on chronic diseases is associated with their ability to chemically reduce reactive oxygen species. They are, in short, potent antioxidants. Notably, resveratrol (RSVL), one of the characteristic polyphenol stilbenes of grapes contained in the skin and seed, has been widely investigated for its functionality in human health [10].

In vascular cells, oxidative stress leads to impaired endothelial function, an etiology associated with cardiovascular diseases such as hypertension and diabetes. One of the factors involved in the development of oxidative stress and endothelial dysfunction is the overactivation of the renin–angiotensin–aldosterone system [11]. Angiotensin I-converting enzyme (ACE) is a crucial enzyme in this system; ACE cleaves angiotensin I to angiotensin II, a potent vasoconstrictor. Previous studies generally confirm that polyphenols directly inhibit ACE as well as scavenge reactive oxygen species [12].

γ-Aminobutyric acid (GABA), a four-carbon and nonprotein amino acid, is another component beneficial to health in pomace. Biologically, GABA plays a fundamental role in inhibitory neurotransmission in the central nervous system and peripheral tissues (autonomic nervous system) in animals, including humans, and may exert various positive effects, such as stress relief and hypotensive effects [13].

Reuse of pomace as a beneficial food additive may be one useful way to avoid waste. Dried and powdered grape pomace has been previously studied as an ingredient in wheat flour products such as cookies [14]. This study concluded that the addition of grape pomace flour increases product value by adding polyphenols to increase antioxidant capacity.

Production processes, such as heating food containing grape pomace flour, may adversely affect the quality and quantity of the beneficial ingredients and thus impair their value. However, there is only limited information on the influence of heating on polyphenols and their antioxidant properties. Volf et al. [15] investigated the heat tolerance of standard polyphenol solutions and vegetal extracts at 60°C, 80°C, and 100°C, and found that degradation rates of polyphenols depend on the type of polyphenol. Heating red wine pomace at 90°C affected different phenolic types unequally, negatively impacting anthocyanins and flavonol-3-O-glycosides but positively impacting phenolic acids and flavonol aglycons [16]. Overall, these studies reported that decreases in the antioxidant capacity of heated wine pomace were not severe.

Beneficial food ingredients obviously need to be bioavailable to exert their effects. Not all dietary components are effectively utilized by organisms in their intact form. Gastrointestinal digestion affects the structure and activity of dietary components, resulting in modification or deterioration of beneficial activity. In vitro gastrointestinal digestion models have been used to determine the metabolic fate of phytochemicals from several foods [17]. The effects of in vitro gastrointestinal digestion on the quality and quantity of phenolic compounds—flavonols, flavanols, and anthocyanins, as well as their antioxidant capacity—were investigated in crude extracts of white and red wine pomaces [18], [19]. They reported that pomace-derived phenolic compounds are qualitatively changed and quantitatively reduced by gastrointestinal digestion, with a related decrease in antioxidant capacity.

In this study, we investigated potentially useful compounds in white wine pomace from Koshu and red wine pomace from Cabernet Sauvignon grapes grown in Japan. The compound classes and single compounds analyzed were total polyphenols (TPs), RSVL, PAs, and GABA. Our study assessed antioxidant properties, ACE inhibitory activity of grape pomace flours, resistance to synthetic digestive solutions, and tolerance to oven heating. In addition, cookies containing these pomace flours were produced, and the storage kinetics of RSVL were examined. We have discussed the differences in the characteristics of ingredients and their potential usefulness as food additives from both pomace types.

Materials and Methods

Wine Pomace Flour and Heating Treatment

Freeze-dried, powdered white and red wine pomaces were provided by a commercial winery (Katashimo, Osaka). The white Koshu grapes were grown in Osaka Prefecture, whereas the red Cabernet Sauvignon grapes were from Yamagata Prefecture. The solids after squeezing the juice from Koshu grapes for fermentation were used to produce white wine pomace (KWP). Red wine pomace (CSRP) was made from the solids remaining after 10 days of maceration of Cabernet Sauvignon grapes.

To evaluate the tolerance of wine pomace to heating, simulating food preparation, the pomace flours were spread evenly in a glass Petri dish (90 mm diameter) and baked in a household electric oven at 180°C for 15 min and 60 min. The controls were non-heated pomace flours.

Phytochemical Properties of Non-Heated and Heated Wine Pomace Flour

Total Polyphenols Analysis

A 0.5 g pomace flour sample was stirred with 50 mL of 70% aqueous MeOH in the dark at 25°C for 18 h. After homogenization with a Bio-mixer (Nihon Seiki, Osaka), the sample was filtered by suction through No. 1 filter paper (Advantec, Tokyo). The filtration residue was washed twice with 25 mL of 70% MeOH, and all filtrates were combined. The aqueous MeOH extract was reduced to the aqueous phase by vacuum and made up to 30 mL with deionized water.

Aqueous samples, suitably diluted with deionized water, were subjected to TPs analysis. A 2 mL diluted sample was mixed with an equal volume of 1 M Folin–Ciocalteu reagent and held for 1 min. Then, 1 mL of 7.5% Na₂CO₃ solution was added and stirred, followed by holding in the dark at 22°C for 1 h. The absorbance of the mixture was measured at 530 nm using a spectrophotometer (UV mini-1240, Shimadzu, Kyoto). TPs were quantified with a standard curve prepared using chlorogenic acid.

Resveratrol Analysis

Pomace flour (1 g) was extracted with 50 mL of 70% aqueous MeOH and concentrated to the aqueous phase as described for TPs. The aqueous phase was partitioned once with an equal volume of hexane to remove fats. Subsequently, the pH of the aqueous phase was adjusted to 8.25 ± 0.25 with 1 M NaOH and partitioned three times with an equal volume of ethyl acetate. The combined ethyl acetate extracts containing RSVL were reduced to dryness under vacuum. The RSVL extract was then purified using a Waters Sep-Pak C₁₈-ENV cartridge, previously activated with 6 mL of 100% MeOH, as described by Shiozaki et al. [20]. After purification, RSVL was eluted from the cartridge with 5 mL of ethyl acetate and concentrated to dryness with a centrifugal concentrator (VC-96N; Taitec, Koshigaya).

The dried samples containing RSVL were redissolved in 500 μL of ethyl acetate and centrifuged at 1800 g for 3 min. A 20 μL aliquot of the supernatant was analyzed using liquid chromatography (HPLC 576, GL Science, Tokyo) equipped with a UV–Vis detector (502U, GL Science) under the conditions described by Shiozaki et al. [20]. (E)-RSVL and (Z)-RSVL were detected at 306 nm. (E)-RSVL was identified based on retention time compared with an authentic standard. A (Z)-RSVL standard was prepared from a standard (E)-RSVL solution in ethyl acetate by irradiation with UV-C light. The concentrations of (E)- and (Z)-RSVL were quantified based on the external standard of (E)-RSVL.

γ-Aminobutyric Acid Analysis

Pomace flour (0.5 g) was suspended in 50 mL of hot water and boiled for 120 min; a period found to be optimal in a preliminary experiment. After cooling, the extract was filtered by suction through filter paper and made up to 50 mL with water. GABA was analyzed after precolumn derivatization. A 100 μL aliquot was diluted with 300 μL of 25 mM boric acid buffer (pH 10.4) and mixed with 200 μL of a derivatization reagent (32 mg of o-phthalaldehyde and 40 mg of N-acetyl-L-cysteine in 2 mL of MeOH). The mixture was heated at 30°C for 7 min. A 20 μL aliquot of the derivatized sample was immediately analyzed by HPLC, equipped with a fluorometric detector (Ex. 365 nm, Em. 490 nm; RF550, Shimadzu). The samples were eluted at 1.0 mL min⁻¹ from a reverse-phase column (Inertsil ODS-SP; 4.6 mm × 250 mm, GL Sciences). The elution solvents were 100 mM phosphate buffer (pH 6.2) (A) and 40% (v/v) acetonitrile in A (B). Separation was performed by a linear gradient as follows: 6% B + 94% A at 0 min–25 min; 40% B at 25 min–42 min; 100% B at 42 min. GABA in the samples was identified and quantified using a derivatized standard of GABA.

Proanthocyanidins Analysis

Pomace (0.5 g) was suspended in 50 mL of 80% methanol and stirred in the dark at 25°C for 18 h. The sample was homogenized with the Bio-mixer and filtered by suction through filter paper. The filtration residue was washed with a total of 50 mL of 75% acetone in several batches. Deionized water (20 mL) was added to the filtrate, which was then concentrated to water only with a vacuum evaporator at 40°C (N-1, Rikakikai, Tokyo). The pH of the aqueous phase was adjusted to 7.0 with 0.1 N NaOH and made up to 50 mL with 0.05 M phosphate buffer (pH 7.0).

PAs were purified according to the method of Sun et al. [21], where a Sep-Pak tC-18 cartridge (Waters, Milford) was mounted on a Sep-Pak C-18 cartridge, followed by mounting with a 10 mL glass syringe. The assembly was activated with 10 mL of methanol and washed with 20 mL of distilled water before loading with grape extracts. The loaded volumes of red and white pomace extract were 2 and 5 mL, respectively. Phenolic acids were eluted with 10 mL of distilled water and discarded. The assembly was dried with nitrogen for 1 min; then, monomeric PA and oligomeric PA were eluted with 25 mL of ethyl acetate, and polymeric PAs were eluted with 10 mL of methanol. These eluates were vacuum dried.

A second Sep-Pak assembly, likewise activated and washed with water, was used for the separation and purification of monomeric and oligomeric PAs. The dried PAs redissolved in 5 mL of water were similarly loaded, and the assembly was dried with nitrogen. Monomeric PAs were eluted with 25 mL of diethyl ether, and oligomeric PAs with 10 mL of methanol. Each eluate was vacuum dried.

Quantification of PAs was conducted using the vanillin assay. The purified extracts were dissolved in 1.5 mL of methanol. One mL dissolved samples were vortexed with 1 mL of methanol, 2.5 mL of 1% vanillin in methanol, and 2.5 mL of 25% concentrated sulfuric acid in methanol. After incubation for 15 min at 30°C (monomeric PA) and 25°C (oligomeric PA), absorbances were measured at 500 nm with the spectrophotometer, using methanol as a blank. PAs were determined from a standard curve prepared using (+)-catechin as the standard.

Analysis of Antioxidant Activity

A 0.5 g sample of pomace flour was suspended in 50 mL of 70% methanol and stirred in the dark at 25°C for 18 h. The sample was homogenized with the Bio-mixer and filtered by suction. The residue was washed with three portions of 70% methanol, totaling 50 mL. Distilled water (20 mL) was added and the mixture was concentrated to an aqueous phase by vacuum and finally adjusted to 50 mL with distilled water.

Antioxidant activity of the extract was determined by the 1,1-diphenyl-2-picrylhydrazyl (DPPH) assay, the ferric-reducing antioxidant power (FRAP) assay, and the 2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) assay. In each assay, the antioxidant activity of the sample was calculated as the amount of Trolox equivalents using a calibration curve.

In the DPPH assay, 0.7 mL of extract diluted with distilled water was mixed by vortexing in a test tube with 2.8 mL of 0.1 M Tris–HCl buffer (pH 7.4) and 3.5 mL of 500 μM DPPH in methanol. The mixture was incubated in the dark at 25°C for 60 min, and the absorbance at 517 nm was measured with a spectrophotometer, using a blank where 0.7 mL of distilled water was used instead of the extract. The suitable dilution rate was defined as the absorbance at 517 nm between 1.5 and 2.0 after mixing the diluted extract and each reagent for 1 min.

In the FRAP assay, 5 mL of acetate buffer (0.3 M, pH 3.6), 0.5 mL of 10 mM 2,4,6-tris(2-pyridyl)-s-triazine/40 mM HCl solution, 0.5 mL of 20 mM FeCl₃·6H₂O, and 0.6 mL of distilled water were mixed by vortexing, and the tube was held at 37°C in an incubator for 10 min. Suitably diluted extract (0.2 mL) was added to the heated mixture, which was incubated at 37°C for 60 min. Absorbance was measured at 593 nm. The blank used 0.2 mL of distilled water instead of the extract.

For the ABTS assay, ABTS radicals were generated by incubating a mixture of 72 mg of ABTS in 20 mL of HPLC-grade water with 13.2 mg of potassium persulfate in the dark at 25°C for 12–16 h. The ABTS radical solution was diluted with methanol so that the absorbance at 734 nm was 0.7 (± 0.02). The diluted ABTS solution (6.93 mL) was vortexed with 70 μL of the suitably diluted extract and incubated at 37°C for 60 min, and the absorbance was measured at 734 nm. The blank contained distilled water instead of the extract.

Analysis of Angiotensin-Converting Enzyme Inhibitory Activity

The sample for ACE activity was extracted by boiling wine pomace (0.5 g) in the same manner as for GABA analysis, except the boiling time was set at 60 min.

The 5 mM ACE substrate was prepared by dissolving 25 mg of hippuryl-L-histidyl-L-leucine in 11.64 mL of 0.2 M Tris–HCl (pH 8.3). A quarter unit of ACE (from rabbit lung) was dissolved in 10 mL of 50% glycerol. Substrate solution (100 μL) and 100 μL of diluted extract were mixed in a centrifuge tube and heated at 37°C for 5 min. After adding 100 μL of ACE solution, the mixture was incubated at 37°C for 30 min and terminated by adding 250 μL of 1 M HCl. Ethyl acetate (1.5 mL) was added with vortexing, followed by centrifugation at 3000 rpm for 10 min. One milliliter aliquot of ethyl acetate containing hippuric acid (the reaction product) was dispensed into a centrifuge tube and completely dried by vacuum.

The dry residue was dissolved in 1 mL of distilled water, and absorbance at 228 nm was measured in a quartz cuvette, defined as As. A no-enzyme control also had ACE solution included, but 100 μL of distilled water was added after termination with HCl, defined as Asb. A parallel blank test group used distilled water instead of the extract, yielding Aw and Awb absorbances. The ACE inhibition rate (%) was calculated by: (1−(As−Asb)/(Aw−Awb)) × 100.

In Vitro Gastrointestinal Digestion

A 2.0 g sample of wine pomace flour was suspended in 200 mL of 70% methanol and stirred in the dark at 25°C for 18 h. The sample was then homogenized with the Bio-mixer and filtered by vacuum through filter paper. The residue was washed with three portions of 70% methanol, totaling 100 mL. Distilled water (30 mL) was added and the mixture was concentrated to an aqueous phase by vacuum, then made up to 200 mL with distilled water. Aliquots were designated for the untreated control, pH adjustment (2.0 and 7.5), and their subsequent respective digestion under gastric and intestinal conditions.

For in vitro gastric digestion, 50 mL of extract was pH-adjusted to 2.0 and deoxygenated with nitrogen for 10 min in a conical flask. After deoxygenation, 15,800 units of porcine pepsin were added. The flask was sealed with parafilm and incubated at 37°C for 2 h with shaking at 100 rpm. After incubation, the sample was filtered, and the pH was readjusted to 2.0, followed by making the volume up to 60 mL with distilled water. A 30 mL aliquot was used for the analysis of various phenol classes and the determination of antioxidant activity.

A 25 mL sample of pepsin-digested matter was adjusted to pH 7.5 with NaHCO₃ and made to 30 mL with distilled water. After degassing in a conical flask as described above, 7.5 mL of a pancreatic and bile salt mixture was added. (This mixture comprised 0.2 g of porcine pancreatin and 1.25 g of bile salt dispersed in 50 mL of distilled water at 30°C, then filtered.) The flask was sealed with parafilm and incubated at 37°C for 2 h with shaking at 100 rpm. After incubation, the sample was filtered, the pH readjusted to 7.5, and the volume made up to 50 mL with distilled water. The entire 50 mL was analyzed as the post-gastrointestinal digestion sample.

In order to investigate the effect of pH only, the sample was subjected to the same conditions as in vitro digestion treatment except adding digestion enzymes. The TP, (E)-RSVL, PAs, and antioxidant activity (DPPH) were analyzed as described in other sections.

Wine Pomace Added to Cookies

The ingredients used were margarine (200 g), granulated sugar (200 g), a raw egg (58 g), and low-gluten wheat flour (300 g), all mechanically mixed to form a dough. Wine pomace flour (30 g) was then added and blended in. The dough was shaped into a bar about 3 cm in diameter, wrapped in plastic film, and held in the freezer for 2 h. The cookie dough was then sliced with a sharp knife into discs about 3 mm thick, placed on cooking paper, and baked in a domestic oven at 180°C for 15 min. After baking, the weight of the dough containing wine pomace flour decreased by 87.6% and 86.7% of their raw dough weight for KWP and CSRP, respectively.

The (E)-RSVL content of the cookie dough was determined before and immediately after baking, and after storing under dry conditions for 3, 6, and 12 months at ambient temperature. Based on weights before and after baking, samples of raw dough and cookies equivalent to 0.5 g of wine pomace flour were used for RSVL analysis as described earlier.

Statistical Analysis

All analysis was conducted in triplicate (except resveratrol analysis: five sample repetitions) and the data expressed as mean ± standard deviation. StatView 5.0 (SAS Institute Inc.) was used for student’s t-test, analysis of variance and post-hoc Tukey test.

Results

Phytochemical Properties of Wine Pomace Flour

The levels of TPs were 2.4 times greater in the red wine pomace flour (CSRP) than in the white wine pomace flour (KWP) (Table I). CSRP contained 3.9 and 9.8 times higher amounts of (E)- and (Z)-RSVL, respectively, than KWP. No significant difference in the levels of GABA was found between CSRP and KWP. PAs were abundant in each pomace flour, in the increasing order of monomer, oligomer, and polymer. CSRP contained roughly 1.8, 2.7, and 2.6 times higher levels of monomers, oligomers, and polymers, respectively, compared with KWP.

Table I. Contents of Total Polyphenols, Resveratrol, γ-Amino Butylic Acid and Proanthocyanidins in the Wine Pomace Flour of Kosyu and Cabernet Sauvignon

		Pomace		p
		KWP	CSRP
TPs (mg CHL equiv./g)		16.3	39.4	<0.001
RSVL (µg/g)	(E)	0.8	3.1	<0.001
	(Z)	0.5	4.9	<0.001
GABA (µg/g)		389.5	349.6	0.869
	Monomer	1.9	3.5	0.005
PAs (mg/g)	Oligomer	4.5	12.2	<0.001
	Polymer	19.1	49.5	<0.001

Between the three assay methods, DPPH, FRAP, and ABTS, there was no change in the order of radical scavenging activity over 720 min (data not shown). Therefore, the absorbances at 60 min of incubation were used to compare the radical scavenging activity of wine pomace extracts (Table II). With the extract, the highest radical scavenging activity shown by Trolox equivalents was with the DPPH assay. However, the ratios of antioxidant activities between CSRP and KWP were almost the same for the three assays. The antioxidant activity of CSRP was roughly 2.4 times higher than that of KWP.

Table II. Antioxidant Activity of Methanol Extracts of Wine Pomace Flour

	DPPH	FRAP	ABTS
Pomace	µmol Trolox equiv./g
KWP	103.4	77.1	56.5
CSRP	256.3	186.6	141.2
p	<0.0001	<0.0001	<0.0001

Fig. 1 shows changes in the inhibition of ACE according to the amount of the extract included. ACE was more strongly inhibited by CSRP extract than by KWP extract. The minimum addition of extract needed to show a difference in the inhibitory activity of ACE was 0.4 μL. With 25 μL of extract, KWP and CSRP inhibited ACE by 87.1% and 98.5%, respectively.

Fig. 1. Inhibitory effect of wine pomace extract on the activity of angiotensin-converting enzyme in vitro. Note: * indicates a significant difference between KWP and CSRP at p < 0.05 obtained by Student’s t-test.

Components Antioxidant Activity After Heating

The TPs in KWP retained the level of the non-heated control after 15 min of heating (Fig. 2). By contrast, the TPs decreased to 78% after 15 min of heating in CSRP. Heating for 60 min decreased the TPs by 12% and 46%, respectively, for KWP and CSRP compared with their non-heated flours. There were no significant differences in the levels of (E)-RSVL between the non-heated samples and those heated for 15 min in each wine pomace flour. Heating for 60 min significantly decreased (E)-RSVL levels to 30% and 41% of the non-heated control, respectively, for KWP and CSRP.

Fig. 2. TPs (a), (E)-RSVL (b), GABA (c), and PAs (d) contents of oven-heated wine pomace. Proanthocyanidins were from CSRP only. Different letters within each pomace indicate a significant difference between the heating times at p < 0.05 obtained by Tukey’s test.

In both pomace flours, the levels of GABA markedly decreased after 15 min of heating. After 60 min of heating, the levels of GABA decreased by roughly 85% in each pomace flour. Comparing the three subfractions of PAs in CSRP, the larger the molecule, the greater the decrease due to heating. Even after 60 min of heating, 65% of the monomeric PA was retained. Oligomeric and polymeric PAs decreased to 6.5% and 2.9%, respectively, of the non-heated controls after 60 min.

The antioxidant activity of the aqueous methanol extract of the heated wine pomace flours was estimated by the scavenging capacity of DPPH. Heating had no influence on the antioxidant activity of the extract of KWP (Fig. 3). By contrast, heating significantly decreased the antioxidant activity of the CSRP extract; the Trolox equivalence was 85% and 54% of the non-heated control at 15 and 60 min of heating, respectively.

Fig. 3. Antioxidant activity of the extracts of the oven-heated wine pomaces. Note: Different letters within the same fraction indicate a significant difference between heating times at p < 0.05 obtained by Tukey’s test.

Components and Antioxidant Activity after In Vitro Gastrointestinal Digestion

As to the effect of pH, adjusting to pH 2.0 and successive adjustment from pH 2.0 to 7.5 had no effect on the levels of TPs, (E)-RSVL, and PAs in the pomace extracts, but there was a significant decrease in TPs in the KWP extract at pH 7.5 (Table III). In gastric digestion tests (pH 2.0 + pepsin), the levels of all analytes were unaffected by pepsin. In a post-gastrointestinal digestion sample (pH 7.5 + pancreatin and bile), the levels of TPs and PAs decreased significantly. TPs in the extracts of KWP and CSRP decreased by 42.3% and 38%, respectively. The decrease in PAs of the CSRP extract was greatest in the polymeric class, followed in order by oligomers and monomers. For example, monomeric PA was 23.7% of the control, but polymeric PA was only 8.8%. By contrast, the levels of (E)-RSVL were not affected by any treatment.

Table III. Total Polyphenols, (E)-Resveratrol, and Proanthocyanidins of Wine Pomace Extracts Treated with Digestion Enzyme

		Pomace	Cont	pH 2.0		pH 7.5
				−Pepsin	+Pepsin	−Pancreatin, bile	+Pancreatin, bile
TPs		KWP	34.3 a	34.7 a	33.6 a	26.7 b	19.8 c
(mg CHL equiv./g)		CSRP	59.2 a	59.4 a	58.7 a	48.7 a	36.7 b
(E) RSVL (µg/g)		KWP	1.9 a	2.0 a	0.9 a	1.8 a	1.8 a
		CSRP	3.3 a	3.3 a	3.2 a	3.3 a	3.2 a
	Monomer	CSRP	6.5 a	7.6 a	6.4 a	6.5 a	1.6 b
PAs (mg/g)	Oligomer		15.7 a	16.7 a	17.3 a	16.3 a	3.0 b
	Polymer		35.7 a	31.2 a	34.7 a	30.9 a	3.1 b

The antioxidant activity of KWP and CSRP extracts subjected to in vitro gastrointestinal digestion was estimated by the DPPH method (Table IV). Without any treatment, pH adjustment to 2.0 had no effect on the antioxidant activity of either extract. However, pH adjustment to 7.5 after treatment at 2.0 decreased the antioxidant activity in both pomace extracts. Pancreatin plus bile treatment at pH 7.5 caused a further decrease in the antioxidant activity in each extract; the antioxidant activities of KWP and CSRP extracts were reduced to 87.1% and 73.3%, respectively, of their adjustment by pH alone.

Table IV. Antioxidant Activity (DPPH) of Pomace Extracts Treated with Digestion Enzymes

Pomace	Antioxydant activity (µmol Trolox equiv./g)
	Cont	pH 2.0		pH 7.5
		−Pepsin	+Pepsin	−Pancreatin, bile	+Pancreatin, bile
KWP	235.6 ab	237.4 ab	242.4 a	218.2 b	190.1 c
CSRP	362.1 a	369.9 a	368.3 a	318.0 b	233.1 c

Changes in (E)-RSVL in Cookies During Storage

The levels of (E)-RSVL in the raw cookie dough containing the wine pomace flour were 0.69 ± 0.06 (for KWP) and 2.91 ± 0.03 (for CSRP) μg/g of dough. There were no significant differences in the levels of (E)-RSVL in uncooked dough compared with freshly baked cookies (p values by Student’s t-test were 0.89 and 0.86 for KWP and CSRP, respectively). As can be seen in Fig. 4, over 12 months of storage, there was no significant decrease in the levels of (E)-RSVL for either pomace.

Fig. 4. Changes in (E)-RSVL content over 12 months in cookies containing wine pomace flour. Month 0: freshly baked. p values are obtained by one-way ANOVA for each wine pomace.

Discussion

The TP levels in the aqueous organic solvent extracts of white and red wine pomace derived from Vitis vinifera cultivars have been determined in several previous reports [22]–[24]. In our study, the TP level of methanolic extracts of CSRP was 39.4 mg/g (Table I) and is comparable to the results from an acetonic extract of freeze-dried Merlot pomace [22]. In KWP, TP levels were close to those in the white wine pomaces of Sauvignon Blanc and Chardonnay [23]. The levels were 41% of the CSRP in this experiment. Although the phenolic components in Koshu grapes were previously characterized by Kobayashi et al. [25], ours is the first analysis of the components in Koshu pomace. Because white wine pomace results directly from grape pressing, but red pomace is recovered after maceration and fermentation, the residual components in the white pomace may be higher than in the red assuming equal initial levels. A comparison of the TPs content in white and red wine pomace from grapes grown in the same region revealed that white wine pomace contains a same level of TPs [24] or higher levels of TPs than red wine pomace [23]. Kobayashi et al. [25] reported that although the total phenols in the grape tissues of Koshu were higher than those in the white cultivars Sauvignon Blanc and Chardonnay, the total phenols in the skin and the seed of Koshu were 62% and 77%, respectively, of those in Merlot. Thus, the lower levels of TPs in KWP should result from the initially lower phenolic content in Koshu grapes.

Comparison of the levels of (E)-RSVL, a typical stilbene in grapes, also showed lower levels in KWP than in CSRP (Table I). In CSRP, (E)- and (Z)-RSVL levels were 3.1 μg/g and 4.9 μg/g, respectively. Previous reports have also confirmed that the RSVL levels of red wine pomace are in the range of 1.72 μg/g–36 μg/g [24], [26]. Although we have no information about RSVL levels in Koshu grapes, the lower levels of RSVL in the pomace may reflect the levels in the grapes, similar to the levels of TPs.

PAs were abundant in each pomace flour, in the increasing order of monomer, oligomer, and polymer (Table I). The total PA levels in Koshu and Merlot grapes were reported to be 2.4 and 2.7 mg/g, respectively, in their skins, and 48.0 and 80.0 mg/g in their seeds [25]. In our work, the differences in the levels of fractionated PAs between KWP and CSRP were more pronounced. The levels of monomeric, oligomeric, and polymeric PAs were all higher in CSRP than those in KWP. The differences in the levels between KWP and CSRP were more marked for the oligomers and polymers. Compared with high molecular weight PAs, low molecular weight PAs are easily extracted into wines during maceration/fermentation, so necessarily, the remaining red wine pomace has a higher percentage of higher molecular weight PAs [27]. Our results may reflect differences in the fermentation process, in addition to original differences in the levels of PAs in Koshu and Cabernet Sauvignon grapes.

Unlike phenolic compounds, there was no significant difference in the level of GABA between KWP and CSRP (Table I). Grapes contain relatively high levels of GABA in comparison to other fruits, and GABA is also present in wine [28]. Carullo et al. [29] measured the levels of GABA in the skin and seed residues of Vitis vinifera grapes after red wine fermentation and found that although the remaining seeds had no GABA, the skin contained a relatively high level of GABA, 392 μg/g. Our findings on GABA in KWP and CSRP suggest that both wine pomaces can be used as sources of GABA irrespective of fermentation.

We analyzed the antioxidant activity of aqueous methanol extracts of pomace by three Trolox methods (Table II). Between the three assay methods, DPPH, FRAP, and ABTS, there was no change in the order of radical scavenging activity over 720 min (data not shown). Therefore, the absorbances at 60 min of incubation were used to compare the radical scavenging activity of wine pomace extracts. The antioxidant activity of CSRP was 2.4-fold greater than that of KWP in all methods: DPPH, FRAP, and ABTS. The antioxidant activity of the CSRP extract assessed by DPPH was 256 μmol of Trolox/g dry weight. This activity is almost twice that of skin pomace from Cabernet Sauvignon grape from Morocco [27]. Unlike in our study, where the differences between white and red grapes were marked, extracts of white Sauvignon Blanc and Chardonnay pomaces showed higher antioxidant capacity than red wine pomaces from Cabernet Sauvignon and Carmenère grapes [23]. In that study, there were higher total phenolic and total PAs in white pomaces compared with red pomaces. In general, the antioxidant activity of plant extracts is strongly correlated with the content of phenolic compounds [30]. Our results also support this relationship.

A potent ACE inhibitor class in foods is phenolic compounds. Inhibition of ACE activity is associated with the levels of flavanol and PAs in foods [31]. A study on the relationship between the polymerization of PAs extracted from the skins and seeds of grapes and the ACE inhibitory activity confirmed that a higher average polymerization of PAs corresponds to stronger inhibitory activity [32]. We found that although a greater volume of pomace extract resulted in relatively less ACE inhibitory activity due to CSRP, the inhibitory activity of CSRP was always higher than that of KWP; it ranged from 3.3 to 1.13 times higher (Fig. 1). On average, the ACE inhibitory activity of CSRP was about double that of KWP. TP, oligomer, and polymer procyanidin levels were 2.4–2.7 times higher in CSRP than in KWP. This may indicate that the differences in the levels of TPs and larger molecules of PAs correlate with the differences in ACE inhibition between KWP and CSRP.

Although RSVL is a polyphenol in wine pomace, its involvement in ACE inhibition might be negligible, at least its inhibitory effects on the enzyme’s activity. Olszanecki et al. [33] reported no effect of RSVL on ACE inhibition in a study with fragments of the rat aorta. It has been demonstrated that the blood pressure-lowering effect of RSVL depends on the suppression of ACE enzyme gene expression, not on the inhibition of ACE enzyme activity [34].

To concentrate nutrients, make them less susceptible to microbial attack, reduce storage volume, and increase shelf life, high-moisture raw materials such as wine pomace are generally heat dried. Wine pomace flour can be subjected to high temperatures during processing. Thus, the effect of heating wine pomace flour on the useful components cannot be ignored. Some previous reports investigated the influence of heating at temperatures ranging from 40°C to 125°C for 90 min to several hours on standard phenolic compounds, extracts of grape seeds, and phenolics in wine pomace [15], [16], [35]. In our work, the effect of heating cookie dough to 180°C for 15 and 60 min was investigated to simulate typical cookie baking temperatures.

Oven heating decreased all the component levels analyzed in pomace (Fig. 2), but to varying extents. There were remarkable differences in the degree of decrease in TPs caused by heating time between CSRP and KWP. Although TPs in KWP were static after 15 min of heating, those in CSRP decreased to 78% of the non-heated pomace. TPs in KWP retained 88% of the non-heated pomace even after 60 min of heating, whereas TPs in CSRP resulted in only 53.7% of the non-heated pomace level remaining. The decrease in TPs in CSRP might be mainly caused by the degradation of the red anthocyanins. In red wine pomace from Tempranillo, the effect of heating at 90°C for 90 min was very pronounced for anthocyanins, decreasing to 39.8% of the non-heated pomace [16].

Decreases in the levels of RSVL and GABA by oven heating were similar in KWP and CSRP; RSVL maintained its level during 15 min of heating, while GABA levels were significantly different between heating times. Heating for 60 min significantly decreased (E)-RSVL levels to 30% and 41% of the non-heated control, respectively, for KWP and CSRP. In both pomace flours, the levels of GABA markedly decreased after 15 min of heating. After 60 min of heating, the levels of GABA decreased by roughly 85% in each pomace flour. Tempranillo pomace heated at 90°C for 90 min maintained 87.3% of the RSVL of the non-heated pomace [16]. Thus, RSVL appears to be a thermally stable compound. PAs of CSRP also decreased with oven heating; the decrease in levels was greatest for the polymers, followed by oligomers and monomers (Fig. 2). Degradation of a catechin standard at 100°C for 4 h was less than 25%, and its thermal stability was similar to that of the phenolic acid standard [15]. Likewise, hexamers and heptamers of PAs were less stable than monomers in peach [35]. Our results of the thermal stability of PAs in wine pomace flour are in general agreement with that report.

The antioxidant activity of the aqueous methanol extract of the heated wine pomace flours was estimated by the scavenging capacity of DPPH. Heating had no influence on the antioxidant activity of the extract of KWP (Fig. 3). By contrast, heating significantly decreased the antioxidant activity of the CSRP extract; the Trolox equivalence was 85% and 54% of the non-heated control at 15 min and 60 min of heating, respectively. Changes in the antioxidant activity of the extract of wine pomace after oven heating paralleled the changes in the TPs.

We investigated the metabolic fate of compounds in KWP and CSRP extracts by an in vitro gastrointestinal digestion model. The results show that low pH and pepsin—the major digestive enzyme in the stomach—had no influence on either TPs, RSVL, or PAs (Table III). However, small intestine digestion with pancreatin plus bile salts, following gastric digestion, significantly decreased the levels of TPs and PAs. This was particularly true for PAs. Interestingly, pH 7.5 conditions before the addition of pancreatin and bile significantly decreased TPs in the KWP extract, with a similar tendency observed in the CSRP extract. Decreases in antioxidant activity of the KWP and CSRP extracts almost paralleled the decreases in PAs through gastrointestinal digestion (Table IV). Correa et al. [18] reported that phenolic compounds in the extract of Merlot pomace greatly decreased after in vitro digestion by a contiguous artificial saliva, gastric, and intestinal fluid treatment. In blackberries, a decrease in TPs and anthocyanins was found not only after intestinal digestion but also after gastric digestion [36]. In chokeberry juice, polyphenols such as anthocyanins, flavonols, and caffeic acid derivatives were not affected by in vitro gastric digestion but decreased with in vitro pancreatin digestion [37]. A similar digestive tolerance property of polyphenols was also reported in anthocyanins from red wine [38]. Our findings and these previous reports suggest that mildly alkaline conditions and pancreatin + bile in intestinal digestion could have a great influence on the chemical properties of polyphenols and thus their bioaccessibility and bioavailability.

By contrast, RSVL showed stability against in vitro gastrointestinal digestion (Table III). To our knowledge, this is the first report on the stability of RSVL in extracts of wine pomace under gastrointestinal conditions. In general, there is limited information on the stability of stilbenes against in vitro gastrointestinal digestion. In aqueous acetone extracts of grape cane, different outcomes of RSVL in digestion simulations have been reported: RSVL levels decreased by human saliva digestion (oral phase) and later reached non-detectable levels with gastric and intestinal digestion, although ε-viniferine, a dimer of RSVL, increased by gastrointestinal digestion [39]; pure RSVL degraded by intestinal digestion but not by gastric digestion [40]. The different results between the former studies and ours might be caused by the differences in some research methodologies and the constituents of the extracts.

In previous studies resulting in a decrease in TPs and antioxidant activity after gastrointestinal digestion, some individual phenolic compounds showed high stability against digestion, while some phenolic compounds, such as phenolic acids and flavanols, have also been reported in some cases to increase after digestion [19], [37]. RSVL, at least from wine pomace, has high digestive stability, so it might have beneficial functions for human health as a polyphenol with high bioaccessibility.

One of the possible uses of wine pomace flour in food products is in baked confectionery. In addition to the quantitative and qualitative changes caused by heating during production, the changes in the composition of pomace-based foods during storage are of interest. The levels of (E)-RSVL in the raw cookie dough containing the wine pomace flour were 0.69 ± 0.06 (for KWP) and 2.91 ± 0.03 (for CSRP) μg/g of dough. There were no significant differences in the levels of (E)-RSVL in uncooked dough compared with freshly baked cookies (p values by Student’s t-test were 0.89 and 0.86 for KWP and CSRP, respectively). As can be seen in Fig. 4, RSVL levels in cookies containing wine pomace remained static for at least 12 months after production. Theagarajan et al. [14] investigated the storability of grape pomace cookies for 60 days and found that the addition of pomace flour increased the antioxidant activity, thereby inhibiting lipid peroxidation, which is the main cause of quality deterioration in cookies. The addition of wine pomace to food products, such as baked confectionery, might contribute to their nutritional function and storability.

Conclusions

This study provides the first characterization of wine pomace flours from Koshu grapes, an indigenous Japanese Vitis vinifera used for white wine, and Cabernet Sauvignon, a variety of Vitis vinifera used for red wine, both grown in Japan. Both pomace flours contained ingredients useful for human health, such as polyphenols and GABA. In comparison with the pomace flour of Koshu, that of Cabernet Sauvignon was superior in TP levels, resulting in higher antioxidant and anti-ACE activities. Although the in vitro gastrointestinal digestion tolerance was similar for both pomace flours, the heat tolerance of TP and antioxidant activity of their extracts was higher in Koshu pomace. RSVL in cookies containing pomace flours did not deteriorate over a year. The valorization of pomace for human health is higher in Cabernet Sauvignon pomace than in Koshu based on the levels of useful ingredients. However, the higher heat tolerance of TPs may be a beneficial feature of Koshu pomace flour.