Myricetin

Assessment of in-vitro bioaccessibility and antioxidant capacity of phenolic compounds extracts recovered from grapevine bunch stem and cane by-products

Susana Ferreyra a, Carolina Torres-Palazzolo b, Rub´en Bottini a,c, Alejandra Camargo b, Ariel Fontana a,d,*

Abstract

Grapevine woody by-products contain bioactive substances, mainly phenolic compounds (PCs), whose beneficial health effects initially depends on their levels of intake and bioavailability. Therefore, in-vitro simulated gastrointestinal digestion (GID; oral, gastric and intestinal phases) was performed to evaluate the bioaccessibility and antioxidant capacity (AC) of PCs extracts recovered from grapevine bunch stem and cane from Malbec grape cultivar. The total PCs in cane and bunch stem extracts were 74 and 20% bioaccessible, respectively. Syringic acid, cinnamic acid, ε-viniferin, naringenin and myricetin were highly bioaccessible, noticeably ε-viniferin in cane extract with 137%. The high bioaccessibility observed, particularly for compounds at high concentration such as ε-viniferin, will help to better understand the bioactive potential of these by-products. In this sense, bunch stems and canes can be considered as new and sustainable sources of bioactive substances for applications as functional ingredients or nutraceuticals in food and pharmaceutical industries.

Keywords:
Winemaking by-products
Grape-derived products
Polyphenols
In-vitro digestion Bioaccessibility
Antioxidant activity
Grape cane
Bunch stem

1. Introduction

Every year, viticulture produces more than 70 million tons of berries coming from a worldwide cultivated area of 7.5 million ha (International Organization of World Vitiviniculture (OIV), (2017)). This agro- economic activity produces a significant volume of grapevine woody derivatives, such as bunch stems and canes, which constitutes potential sources of a wide range of phenolic compounds (PCs) with purported biotechnological applications. Canes are the result of a management practice on grapevine plants performed every year and aimed to prune the plants for improving grapes yield stability and quality of berries. This activity generates an average amount of roughly 2.5 tons per ha per year (Ferreyra, Antoniolli, Bottini, & Fontana, 2020). Bunch stems are the other lignocellulosic by-product that accumulate along the winemaking process and constitutes 5% of the processed grapes at harvest time (Ferreyra, Bottini, & Fontana, 2019). Normally, these grapevine derivatives are composted or burned thus limiting their potential application as a font of bioactive compounds for food, pharmaceutical and cosmetic industries.
Recently, the study of grapevine residues has been increased due to their potential as source of health-promoting bioactive compounds. Antioxidant, cardio-protective, anti-inflammatory, anti-microbial, anti- fungal, anti-aging and anti-cancer properties have been reported (Barros, Girones-Vilaplana, Texeira, Baenas, ´ & Domínguez-Perles, 2015; Teixeira et al., 2014). Most researches are focused on the study of PCs (flavonoids, phenolic acids and stilbenes) and dietary fiber together with other minor compounds which allows assessing nutritional quality of these wastes (Çetin, Altinoz, Tarçan, ¨ & Goktürk Baydar, 2011; Gonz¨ alez- ´ Centeno, Rossello, Simal, Garau, L´ opez, ´ & Femenia, 2010). Although there is an overall trend in the phenolic profiles of these by-products in grapevine, levels of individual PCs depend on plant variety and phenology, environmental growth conditions, vineyards management, and winemaking process (Ferreyra et al., 2020, 2019).
Previous reports showed that grape cane extracts are characterized by high levels of stilbenes, particularly ε-viniferin (Ferreyra et al., 2020; Guerrero et al., 2016; Lambert et al., 2013), although other oligomeric derivatives have been also reported (Peˇsi´c et al., 2019). ε-viniferin has remarkable valuable health properties, conspicuously for its cardio- protective and antioxidant functions in comparison to trans-resveratrol (Saez et al., 2018; Zghonda et al., 2012´ ). Although previous reports informed that ε-viniferin has difficulties in passing the intestinal barrier and to be metabolized as compared with trans-resveratrol, it may act locally on the gastrointestinal tract (Willenberg, Michael, Wonik, Bartel, Empl, & Schebb, 2015). During an in-vitro assay, ε-viniferin exhibited stronger inhibitory potential of intestinal glucose uptake than trans- resveratrol (Guschlbauer, Klinger, Burmester, Horn, Kulling, & Breves, 2013). Besides of the potential bioactivity of ε-viniferin, there are no studies reporting its digestive stability and bioavailability. In the case of bunch stem extracts, flavanols, flavonols and phenolic acids have been reported as the main constituents (Barros et al., 2015; Goufo, Singh, & Cortez, 2020). In this matrix, the strong antioxidant activity is due to their high levels of (+)-catechin, which protect macromolecules from the action of free radicals (Barros et al., 2015). In turn, the hydro- methanolic extracts of grape stems have also been shown inhibition of the growth of foodborne pathogens including E. coli and Salmonella sp. (Barros et al., 2015).
To perform a positive biological function in the organism, bioactive compounds or moieties from them must be bioavailable, that is, digested and efficiently assimilated. Therefore, to understand the potential health benefits of novel grapevine by-products it is essential the investigation of the digestion effect on the PCs bioaccessibility (Lorenzo et al., 2019). It is known that bioaccessibility differs from one compound to another, even from the same family, due to the influence of chemical structure, like type and native form in which they are found in the ingested matrix
(Manach, Williamson, Morand, Scalbert, & Rem´ ´esy, 2005). The in-vitro simulated gastrointestinal digestion (GID) allows to study the bioavailability of compounds concerning their potential for intestinal absorption (Haas, Toaldo, Gomes, Luna, De Gois, & Bordignon-Luiz, 2019; Tagliazucchi, Verzelloni, Bertolini, & Conte, 2010). During this process, the bioactive components can be retained inside their matrix or released into intestinal fluids remaining stable or changing into other compounds. Thereby, digestion process can change the profile of bioactive compounds in certain product and thus its properties, such as the overall AC. Currently, in-vitro digestion is the most employed methodology in studies of bioavailability. This in-vitro method is safer, faster and cheaper than other in-vivo approaches, and also has no ethical restrictions (Hur, Lim, Decker, & McClements, 2011). As an additional characteristic, the in-vitro methods are specially designed for initial screening of the compounds performance in different matrices, in complementation with in-vivo studies. However, only few researches have studied the effect of GID on the chemical composition and the bioactivity of some winemaking by-products in the past years (Correa ˆ et al., 2017; Garbetta et al., 2018; Laurent, Besançon, & Caporiccio, 2007; Sanz-Buenhombre, Villanueva, Moro, Tomas-Cobos, Viadel, ´ & Guadarrama, 2016; Tagliazucchi et al., 2010). Most of them have been focused on grape pomace extracts. A low number of studies about woody by-products, specifically for bunch stems have been reported (Jara- Palacios, Gonçalves, Hernanz, Heredia, & Romano, 2018; Li, Loo, Cheng, Howell, & Zhang, 2019). So far, there are not reports on bioaccessibility of PCs recovered from grape cane extracts for any variety, nor for bunch stem extracts from cv. Malbec have been published.
The present work is aimed to evaluate the effect of digestion process upon bunch stem and grape cane extracts of the Malbec variety, the most cultivated Vitis vinifera of Argentina. As well, to establish the bioaccessibility of recovered PCs and AC variations throughout the digestive phases. With the objective to mimic the digestive process in the mouth, stomach (gastric digestion) and small intestine (duodenal digestion), a three-step procedure was performed. Before digestion and after each step, the stability of individual PCs was evaluated by LC-DAD. Moreover, the total phenolic content (TPC) and the in-vitro AC were determined by ABTS and ORAC methods.

2. Materials and methods

2.1. Reagents and standards

Standards of gallic acid (99%), syringic acid (≥95%), cinnamic acid (99%), caftaric acid (≥97%), p-coumaric acid (98%), trans-resveratrol (≥99%), (+)-ε-viniferin (≥95%), procyanidin B1 (≥90%), procyanidin B2 (≥90%), (+)-catechin (≥99%), (− )-epicatechin (≥95%), (− )-gallocatechin (≥98%), (− )-gallocatechin gallate (≥99%), (− )-epigallocatechin gallate (≥95%), naringin (≥95%), naringenin (≥95%), astilbin (≥98), quercetin 3-β-D-glucoside (≥90%), quercetin 3-β-D- galactoside (≥97%), myricetin (≥96%), kaempferol (≥90%) and 3- hydroxytyrosol (≥99.5%) were acquired from Sigma-Aldrich (Steinheim, Germany). Individual standard solutions of PCs were prepared in methanol at a concentration level of 1000 mg mL− 1. Calibration standards were dissolved in the initial mobile phase condition of the utilized chromatographic method.
Porcine pepsin (P7000), porcine pancreatin (P7545), porcine bile salts (B8756), Trolox reagent (6-hydroxy-2,5,7,8-tetramethylchroman- 2-carboxylic acid), NaH2PO4⋅2H2O, Na2HPO4⋅12H2O, fluorescein, ABTS, K2S2O8 and AAPH were purchased from Sigma-Aldrich (Steinheim, Germany). The acetonitrile (MeCN), acetone, methanol (MeOH) and formic acid (FA) were of HPLC-grade and purchased from Mallinckrodt Baker (Inc. Pillispsburg, NJ, USA). Ethanol and Folin-Ciocalteu reagent were acquired from Merck (Sao Paulo, Brazil). Additional re˜ agents were of analytical grade. Ultrapure water (H20) was obtained from a Milli-Q system (Millipore, Billerica, MA, USA).

2.2. Samples and extraction procedure

Canes and bunch stems of Vitis vinifera L. cv. Malbec were collected in 2017/18 season from vineyards of Gualtallary region, in Mendoza province, Argentina. This variety was selected for being the most cultivated and emblematic of Argentina’s winemaking industry. As well, according to previous works (Ferreyra et al., 2020, 2019), Malbec extracts have greater PCs content and AC among other studied varieties. Grape cane samples were collected in the vineyard after pruning and cut into small pieces (2–4 cm long). Afterward, sample were dried at 60 ◦C until constant weight was achieved (5 days). Bunch stem sample was collected after de-stemming berries at the beginning of the winemaking process. The samples were taken in the winery and immediately freeze- dried (5 days). Finally, samples were milled and stored in hermetic plastic tubes at room temperature till extraction of PCs.
A solid–liquid extraction, according to previous protocols reported by our group (Ferreyra et al., 2020, 2019) was performed to extract the PCs from samples. Briefly, 50 mL (50:50, v/v) acetone/water for cane samples and ethanol/water for stem bunch samples, in ultrasonic washer at 50 Hz and 60 ◦C during 60 min were used to extract 1 g of pulverized sample. After centrifuging the mixture for 10 min at 806 g, it was filtered through filter paper and finally the solvent was removed by concentrating the extract in a rotary evaporator at 40 ◦C. Then, the obtained extract was kept in closed dark-glass bottles at − 20 ◦C until analysis.

2.3. In-vitro simulated gastrointestinal digestion

In-vitro GID was carried out according to Torres-Palazzolo, Ramirez, Locatelli, Manucha, Castro and Camargo (2018) and Lingua, Wunderlin and Baroni (2018) with slight modifications. Three separate experiments were performed for both samples. PCs profile, AC and TPC were evaluated before and after individual experiment by taking aliquots after finalizing each digestive step. To discard any influence of the digestion reagents on the results blank samples were processed and analyzed.
In-vitro GID model comprised three sequential stages: oral phase (OP), gastric phase (GP) and intestinal phase (IP). In the OP, 1 mL of extract was homogenized in a closed flask with fresh human saliva (1 mL, 100 D unit’s mL− 1) at 37 ◦C by orbital agitation (250 rpm, 30 s). After that, an aliquot of 2 M HCl was immediately added to adjust the pH to 2. Addition of HCl inactivates amylase and conditions the medium to continue with the GP. In the second phase, 5 mL of simulated gastric fluid (porcine pepsin 1,600 U mL− 1; HCl 0.1 M; CaCl2⋅2H2O 3.6 mM; MgCl2⋅6H2O 1.5 mM; NaCl 49 mM; KCl 12 mM; KH2PO4 6.4 mM) were added to the mixture obtained from the first step, continuing the incubation on the same conditions for 60 min. When GP was finished, 2 M NaOH was added to adjust the pH to 6.8. After that, 3 mL of simulated intestinal fluid (pancreatin 16,000 USP L-1; NaHCO3 0.1 M; bile salts 25 g L-1) were added. Samples were hatched for 120 min. Finally, aliquots arising from OP, GP and IP were centrifuged at 9677 g for 10 min and the supernatants kept in closed tubes at − 80 ◦C until analysis.

2.4. Total phenolic content

TPC was determined in an UV–vis spectrophotometer Cary-50 (Varian Inc., Mulgrave, VIC, Australia) analyzing an aliquot of the extracts obtained before and after each step of digestion. To measure the TPC, a Folin-Ciocalteu assay at 750 nm of the sample properly diluted was used according to the report by Ferreyra et al. (2020). Results were informed as mg of gallic acid equivalents (GAE) per gram of dry weight by-product (mg GAE g− 1 DW) by means of a calibration curve prepared with the standard solutions of gallic acid (three replicates) in the 0–200 mg L− 1 range (r2 = 0.999). 2.5. Chromatographic method
Analysis of PCs were done with a Dionex Ultimate 3000 LC-DAD equipment (Dionex Softron GmbH, Thermo Fisher Scientific Inc., Germering, Germany) using a Kinetex C18 reversed phase column (3.0 mm × 100 mm, 2.6 µm) (Phenomenex, Torrance, CA, USA). Ultrapure water with 0.1% FA (A) and acetonitrile (B) were used as mobile phases of the chromatographic method. Separation of analytes was performed with a previously described method (Fontana, Antoniolli, & Bottini, 2016) using the subsequent gradient: 0–1.7 min, 5% B; 1.7–11 min, 30% B; 11–14 min, 95% B; 14–15.5 min, 95% B; 15.5–17 min, 5% B; 17–20, 5% B. The flow rate of mobile phase was set at 0.8 mL min− 1, the column temperature at 35 ◦C and the injection volume was 5 µL. Analytes of different phenolic sub-classes were quantified by using the following conditions: 254 nm for quercetin 3-β-D-glucoside, quercetin 3-β-D- galactoside, and (− )-gallocatechin; 280 nm for (+)-catechin, procyanidin B2, procyanidin B1, (− )-epicatechin, (− )-gallocatechin gallate, (− )-epigallocatechin gallate, astilbin, naringin, naringenin, syringic acid, gallic acid and hydroxytyrosol; 320 nm for cinnamic acid, caftaric acid, p-coumaric acid, trans-resveratrol and ε-viniferin; and 370 nm for kaempferol and myricetin.
Analytes identity was assigned by retention times (Rt) and absorbance values evaluation between the peaks detected in samples and the obtained after injecting pure diluted standards. Analytes were quantified with an external calibration prepared with individual pure standards of compounds. The obtained linear ranges were between 0.5 and 40 μg mL− 1. For naringin, naringenin and kaempferol, the linear ranges were between 0.5 and 20 μg mL− 1. Coefficients of determination (r2) higher than 0.991 for all the studied analytes were achieved. Chromeleon, version 7.1 (Dionex Softron GmbH, Thermo Fisher Scientific Inc.) was utilized to control the LC-DAD system and to processing the data.

2.6. Calculation of recovery index and bioaccessibility of PCs

The effect of each digestion step (OP, GP and IP) on the composition of extracts was determined by calculating the recovery index (RI) according to the reported by Jara-Palacios et al. (2018), as follows: where A was the phenolic content (µg g− 1) of sample extract next each digestion step and B was the initial phenolic content (µg g− 1) in the sample extract quantified before the GID. The percentage of bioaccessibility was the value obtained when A was the phenolic content (µg g− 1) of the extract after the duodenal step, and thus the recovered PCs will be available to be absorbed into the systemic circulation (Torres-Palazzolo et al., 2018).

2.7. Antioxidant capacity

The AC of extracts and digests at each stage was evaluated by ORAC and ABTS assays. Trolox was employed as standard and results were put as µmol of Trolox equivalents per gram of by-product dry weight (μmol TE g− 1 DW). Each sample was studied in triplicate and results were expressed as the mean ± standard deviation (SD).
The ORAC assay was performed according to Antoniolli, Fontana, Piccoli and Bottini (2015), with some modifications. Crude and digested extracts from Malbec bunch stems and canes were diluted with 75 mmol L-1 NaH2PO4-Na2HPO4 to achieve satisfactory concentration relative to the Trolox standard. Later, diluted aliquots of samples (50 μL) and Trolox standards (0–50 μmol L-1) were spiked to a 96-well plate. At that point, fluorescein solution (100 μL) were added. The mix was incubated during 7 min at 37 ◦C and then, 50 μL of 140 mmol L-1 peroxyl radical generator of AAPH were added. The fluorescence of samples was monitored by using a microplate fluorometer (Fluoroskan Ascent FL, Thermo Fisher Scientific Inc, Wilmington, DE), at 485 nm excitation and 538 nm emission wavelengths, with 1 min intervals during 90 min. The area under the curve of the fluorescence diminution during 90 min was calculated for individual samples by the area integration under the curve of relative fluorescence.
The method employed for the ABTS assay was the described by Ferreyra et al. (2020) . The radical cation ABTS•+ was formed by mixing 2.5 mL of 7 mmol L–1 ABTS standard solution and 44 μL of 140 mmol L–1 K2S2O8. The prepared combination was kept 12–16 h in obscurity and then diluted with 80% methanol solution up to achieve an absorbance value of 0.70 ± 0.02 at 734 nm. Finally, 10 μL of Trolox (0–2,000 μmol L− 1) or the properly diluted sample were added to 2.5 mL of ABTS•+ solution and the absorbance was read with a Cary-50 UV–vis spectrophotometer after 7 min of the initial mixing of reagents and sample extracts.

2.8. Statistical analysis

The software package INFOSTAT v. 2018 was used for statistical analysis. One-way analysis of variance (ANOVA) using Tukey’s test (p ≤ 0.05) was performed to analyze if PCs and AC differed significantly between extracts and their digests for each type of by-product. For the treatment of data, a correlation analysis by multiple linear regression (p ≤ 0.05) was also performed. Moreover, principal component analysis (PCA) was designed to study the effect of phenolic composition (TPC and individual PCs) and AC (ORAC and ABTS) for each type of by-product. The determinations were performed in triplicate and the results expressed as mean ± SD.

3. Results and discussion

3.1. Effect of in-vitro digestion of Malbec cane and bunch stem extracts on the total bioaccessibility of PCs

The RI of total PCs quantified by LC-DAD (expressed as the sum of individual compounds) after each GID step for bunch stem and cane extracts is shown in Fig. 1. The RI was calculated using Eq. (1) presented in section 2.6 and the IP value obtained corresponds to the percentage of bioaccessibility. For bunch stem extracts, the PCs were relatively stable in the OP, although a decreasing trend was found as the digestive process progressed. Otherwise, the PCs of cane extracts showed an increasing trend from the OP to the IP. As a consequence, after the GID the total bioaccessibility of PCs was higher in cane (74%) than in bunch stem (20%) extracts. This finding suggests that the matrix plays an important role on the release of PCs from conjugated components of the matrix, or interconversion of other complex PCs. On the light of these results, PCs of cane extracts showed more stable in in-vitro GID conditions, giving some clues on their potential bioaccessibility. However, additional studies using mass spectrometry (MS) will be necessary in the future with the aim to evaluate the presence of oligomeric derivatives of proathocyanidins and stilbenoids phenolic sub-classes. This information could help to the deep characterization of extracts before in-vitro digestion and better justify the increases or diminutions of bioaccessibility of compounds.
Our results showed different response for each matrix, which could be explained by the PCs profile, their concentration and the overall composition of extracts. In addition, these variations in responses may be determined by the effect of factors such as pH, composition of gastric medium and interactions between enzymes in the digestion system. Perhaps, since the studied extracts had different chemical composition, some physicochemical transformations of PCs such as oxidation may happen to a variable extent depending on the interaction with the other matrix components (Ortega, Macia, Romero, Reguant, ` & Motilva, 2011). As a result, the bioaccessibility of some compounds may be modified in the presence of other nutrients, such as proteins, fiber, lipids, carbohydrates, minerals and others (Mellado-Ortega, Zabalgogeazcoa, Vazquez ´ de Aldana, & Arellano, 2017; Ortega et al., 2011). Previous reports observed that soluble carbohydrates could have a protecting character for polyphenols against pH changes and enzymatic activities through the digestion process, augmenting their stability and possible bioaccessibility (Ortega et al., 2011). Furthermore, soluble carbohydrate content might too take a significant role in the interruption of polyphenol-protein complexation (Jakobek, 2015). Grape canes constitute woody perennial organs that function as reservoirs for carbohydrates in grapevines, and their total content could vary from 350 to 450 g kg− 1 (Çetin et al., 2011).
As showed in Fig. 1, important differences regarding the level of PCs among the different digestion phases were observed, indicating that PCs may undergo structural changes or degradation during the digestion process (Haas et al. 2019). In this sense, some authors reported that bioaccessibility of PCs is also affected by pH conditions (Tagliazucchi et al., 2010). In fact, many PCs (mainly phenolic acids and anthocyanins) have been described to be stable in the acidic GP medium, while in the mild alkaline medium of IP they suffered some degradation (Tagliazucchi et al., 2010). The latter agrees with what we found in bunch stem extracts. Some authors reported stable values of PCs after in- vitro GID for whole grape and grape skin extracts (about 62–66%) (Garbetta et al., 2018; Tagliazucchi et al., 2010), fairly similar to those obtained in our work for cane extracts. Nevertheless, there are no previous reports on bioaccessibility of grape cane extracts for direct comparison and the results presented here give new knowledge about potential bioaccessibility of compounds in this matrix. 3.2. Modifications in the levels of different PCs through in-vitro digestion
Profile and concentrations of PCs in bunch stem and cane extracts before and after in-vitro GID are summarized in Table 1. A total of 22 PCs coming from different phenolic sub-classes were quantified in the crude extracts, as well as in their digested fractions. Specifically, 18 PCs were quantified in each sample, although differing in some compounds according to the matrix. In fact, cinnamic acid, trans-resveratrol, kaempferol and hydroxytyrosol were only found in cane extracts, while p- coumaric acid, (− )-gallocatechin, (− )-epigallocatechin gallate and quercetin-3-galactoside were solely found in bunch stem extracts. In grape canes, the stilbene ε-viniferin followed by the flavanols (+)-catechin and (− )-epicatechin were the most abundant PCs. For bunch stems, the flavanols (+)-catechin and procyanidin B1 together with caftaric acid were the predominating PCs constituents. As can be observed from Table 1, the PCs levels of the different phenolic sub- classes varied importantly among the matrices and their digested fractions. For better understanding, the performance of the phenolic sub- classes and some selected individual PCs at each GID step is further discussed in the text (Fig. 2).

3.2.1. Phenolic acids

In cane extracts concentrations of hydroxybenzoic acids did not change remarkably amongst the first GID phases, although a decrease of 45% was observed after the IP in comparison to the undigested extract (Fig. 2B). For bunch stem extracts, concentrations of hydroxybenzoic acids decreased importantly during digestion (Fig. 2A).
For hydroxycinnamic acids, the concentrations in digested cane extracts remained stable and only a moderate increase was observed in the GP (33%) respect to the initial concentration of the undigested extract. Some studies have reported that the GP increase the bioaccessibility of some PCs, while during IP their levels could be decreased. However, this behavior is closely related to the stability and structural changes that each type of PCs suffers (Tagliazucchi et al., 2010). Whereas, in digested bunch stem extracts, acid levels decreased 74% in the IP (Fig. 2A and B). Jara-Palacios et al. (2018) reported a slight decrease of phenolic acid contents at the finish of the process of digestion for stem extract (RI of 92%).
After GID of bunch stem extracts, the bioaccessibility of gallic acid was 39% (Fig. 2C). Contrarily, this compound was not detected in the last phase of GID in cane extracts. Tagliazucchi et al. (2010) also described the degradation (43%) of pure gallic acid after GID. A previous study also showed the total degradation for gallic acid from grape, skin and stem extracts after digestion (Jara-Palacios et al., 2018). Concerning uptake issues, Manach et al. (2005) showed that gallic acid is better absorbed as compared to other PCs due to its low molecular weight, which makes it highly bioaccessible.
For syringic acid, significant changes were only observed in IP (Fig. 2C and 2D). For cane extracts, the passage from the GP to IP produced a concentration increase of 73%. On the other hand, for bunch stem extracts its concentration was reduced up to 34%. Thereby, the bioaccessibility of this compound was substantially different for cane and bunch stem extracts (237% and 57%, respectively, Fig. 2C and D). Ortega et al. (2011) also reported increases of syringic and cinnamic acids concentrations for carob flour after the IP. This fact may occur because phenolic acids could be liberated from complexes of matrix components such as proteins and carbohydrates, and then become bioaccessible. Similar results were reported for rice hull extracts (Nenadis, Kyriakoudi, & Tsimidou, 2013). In this case, the authors proposed that syringic acid was expected to be formed from degradation of hydroxycinnamic acid, process that may be reversibly associated with matrix- extracted minerals. Regarding the above, grape canes are rich in some minerals such as K, Ca, Fe, Mg, P and Zn (Çetin et al., 2011). Another explanation may be that grape cane extracts could contain other phytochemicals such as syringyl lignin, the major component of plant cell walls, which constitutes an immediate source of syringic acid.
The concentrations of caftaric acid remained stable in cane extracts, but decreased in bunch stem extracts after the digestion process (Fig. 2C and 2D). The bioaccessibility of caftaric acid was higher in canes (94%) than in bunch stems (27%), where the level of this compound decreased 73% in the last step of in-vitro GID. Garbetta et al. (2018) reported a similar bioaccessibility for caftaric acid (about 80%) in grape skin extracts, which is in agreement with our finding for cane extracts. On the other hand, cinnamic acid increased in the last step in cane extracts, but p-coumaric acid was completely missed in bunch stem extracts after GP.

3.2.2. Stilbenes

The compound trans-resveratrol was only detected in grape cane extracts, with low digestive stability since it was not detected after GP and IP. Tagliazucchi et al. (2010) indicated a similar behavior for pure trans-resveratrol, which suffered significant degradation after GID. Other monomeric stilbenes such as trans-piceatannol and piceid were not detected in the studied extracts. Previous works reported the presence of some oligomeric stilbenoids in bunch stems and wines from different grape varieties (Moss, Mao, Taylor, & Saucier, 2013; Püssa, Floren, Kuldkepp, & Raal, 2006).
On the other hand, ε-viniferin showed high stability after GID, with a completely different performance for both matrices (Table 1). For bunch stems, the levels of stilbenes (Fig. 2A), particularly ε-viniferin, decreased, being its bioaccessibility of 51%. For cane extracts however, an opposed trend was observed, since the bioaccesibility of this compound showed a substantial increase (Fig. 2D). In fact, ε-viniferin reported an overall bioaccesibility of 137%. As far as ε-viniferin is referred, no earlier reports on its stability under digestive conditions have been published. A possible explanation about the high bioaccessibility of ε-viniferin could be related to the depolymerisation of oligomeric stilbendoids during in-vitro digestion. There are reports that have identified several polymeric stilbenes in grapevine canes of different Vitis species at lower concentration levels than ε-viniferin (Aliano-Gonz˜ alez, Richard, ´ & Cantos-Villar, 2020). In fact, these complex compounds could suffer depolymerisation during in-vitro digestion increasing the levels and, consequently, the bioaccessibility of ε-viniferin. In this sense, additional studies with MS detectors should be performed in the future to identify the presence of polymeric compounds and evaluate their stability after digestion steps. Our data from cane extracts digestion indicate that ε-viniferin is released to fluids under simulated intestinal conditions. Therefore, the ε-viniferin high bioaccessibility encourages further studies on the by-product bioactivity, especially taking into account its high concentration in crude cane extracts (3,596 µg g− 1 DW). As an additional point to highlight, some studies reported that ε-viniferin is more efficient than trans-resveratrol in getting better some functions of the cardiovascular system (Zghonda et al., 2012). Besides, it has been proven that ε-viniferin exhibits remarkably higher antioxidant activity than trans-resveratrol (Saez ´ et al., 2018).

3.2.3. Flavanols

Most flavanols were not detected or detected in relatively small amounts after the GID in both matrices. As an example, the (+)-catechin levels decreased a 99% and 70% in bunch stem and cane extracts, being 1% and 30% bioaccessible, respectively (Fig. 2C and D). The initial content of this compound was higher in bunch stem extracts, but suffered the largest decrease during digestive process in comparison to cane extracts. This fact may be linked to the different matrix composition respect to canes as previously commented (Mellado-Ortega et al., 2017; Ortega et al., 2011). Additionally, these differences may be also due to the interaction between (+)-catechin and digestive enzymes used in GID, making undetectable the compound by chromatographic analysis (Tagliazucchi et al., 2010). Previous studies in grape seed and skin extracts, obtained higher bioaccessibility values for (+)-catechin (about 54–59%), so this difference could be due to the changes in the chemical composition of the studied matrices (Garbetta et al., 2018; Laurent et al., 2007). Peˇsi´c et al. (2019) presented a similar result to our work studying grape seed extracts. These authors observed that flavanol sub-class, including monomeric, dimeric and trimeric procyanidins isomeric forms were almost missed after digestion. Opposed, Jara-Palacios et al. (2018) showed a different result in stem extracts, informing a 201% increase of flavanols levels after GID, but reported a similar behavior that our work for skin extracts. Reasons for the difference may be due to dissimilar matrix composition of extracts, since theirs came from a different grapevine cultivar (white) and were obtained by macerating the samples at room temperature overnight. Meanwhile, our extracts came from a red cultivar and were made by using an ultrasonic bath at 60 ◦C during 60 min. Additionally, in the Jara-Palacios et al. (2018) work, the digestion protocol had slight differences that could enlarge divergence between results, such as gastric pH that was less acidic and intestinal pH that was less alkaline than ours. The human gastrointestinal tract can present a wide pH range from 1.5 to 7.4, thereon it was reported that by increasing pH above 5.2 progressively rose (+)-catechin degradation (Ho, Thoo, Young, & Siow, 2017). These facts could modify the nature of digestion medium, altering the PCs stability and their interaction with other matrix components (Tagliazucchi et al., 2010).
For procyanidins, only 32% of initial procyanidin B2 from bunch stems was bioaccesible. On the other hand, the procyanidin B1 was completely missed after the GID of both matrixes. It was interesting that in bunch stem extracts this compound showed high level in the crude extract, but it diminished nearly 26% in GP, to finally being not detected at IP. It is convenient to mention that grape derived matrixes could have more complex oligomeric forms of proanthocynidins which have not been studied in this work. In fact, Jara-Palacios et al. (2018) reported the presence of procyanidin trimer and tetramer which were not detectable after the intestinal step of bunch stem extract. In previous studies where grape seed extracts were subjected to GID, it was also observed a diminution of procyanidins B1 and B2, which is in accordance with our results (Janisch, Olschl¨ ager, Treutter, ¨ & Elstner, 2006; Laurent et al., 2007). Jara-Palacios et al. (2018) presented an increase of monomeric and dimeric proanthocyanidins levels in stem extracts after in-vitro assay, which might be due to degradation of their trimer and tetramer as was commented below. Other authors also described the formation of monomeric units after the simulated digestion due to the possible hydrolysis of proanthocyanidin (Lingua et al., 2018). However, several authors also described that a poor bioaccessibility of flavanols is expected because proanthocyanidins strongly interact with digestive enzymes and food ingredients (proteins, carbohydrates and lipids), and the interactions are enhanced with the increase of polymerization degree (Jakobek, 2015; Peˇsi´c et al., 2019). Proanthocyanidins are considered one of the most abundant family of PCs in our nutrition, but they are not absorbed as such, given that the polymerization greatly impairs their intestinal absorption. Many studies reported that the absorption of dimers B1 and B2 is generally lower than of the monomers (Manach et al., 2005). Besides of that, proanthocyanidins may exert their beneficial effect directly on the intestinal mucosa by avoiding oxidative stress (Manach et al., 2005). In-vitro and in-vivo assays in mice and humans have also demonstrated the proanthocyanidins ability to inhibit lipoprotein lipase activity, which resulted in a decrease of triglyceride absorption (Oteiza, Fraga, Mills, & Taft, 2018).

3.2.4. Flavanones

For this family, almost no significant changes were observed during the OP. However, after GP the concentration of naringin decreased by 50% in both samples, and finally bioaccessibility differed between bunch stem (70%) and cane extracts (10%). In the case of naringenin, the levels decreased to 29% after GP, but after IP the initial levels (those of the crude extract) were recovered in both matrices (see Fig. 2C and 2D). The naringenin recovery in the digestion mixture may be due to the protecting result of the soluble fraction (sugars + soluble fiber) contained in samples (Ortega et al., 2011).

3.2.5. Flavonols

Different flavonol profiles between samples were found, quercetin-3- galactoside was only in bunch stem extracts while kaempferol in cane extracts. Nevertheless, the initial traces of kaempferol and quercetin-3- galactoside were not detected after the GID.
The most abundant flavonol in bunch stems was quercetin-3- glucoside, followed by astilbin (di-hydroquercetin 3-rhamnoside), with maximum concentrations of 2,651 and 1,666 μg g− 1 DW, respectively. Quercetin-3-glucoside showed good stability after GID, presenting 71% of bioaccessibility (Fig. 2C). Jara-Palacios et al. (2018) reported that the level of quercetin-3-O-glucoside was also reduced (47%) after IP in stem extracts. Other authors reported similar results in grapes (Corrˆea et al., 2017; Garbetta et al., 2018). Some studies have reported that quercetin glucosides are more efficiently absorbed than quercetin itself. Therefore, the bioavailability depends on food sources, specifically the type of glycosides that they contain (Manach et al., 2005).
For myricetin, similar levels were found in both initial samples, but bioaccessibility percentages varied between bunch stem (454%) and cane (0%) extracts, as can be seen in Fig. 2C. The augmented value observed here for myricetin were also detected in the study of Ortega et al. (2011), who hypothesized about possible hydrolysis during GID of the glycoside forms present in the sample matrix, yielding the aglycone forms of this compound.

3.2.6. TPC vs AC

TPC and AC were assayed in bunch stem and cane extracts before and during the in-vitro GID (Fig. 3). TPC levels showed a dissimilar performance depending on the by-product. Furthermore, significant differences in TPC values were also observed amid the digestion phases in each sample. Initially, bunch stems showed higher levels of TPC (142 mg GAE g− 1 DW) than canes (27 mg GAE g− 1 DW). In bunch stems, a decreasing trend in TPC levels was observed during OP and GP (132 to 51 mg GAE g− 1 DW). Contrarily, an increasing tendency in TPC quantities was assessed in cane extracts (16 to 24 mg GAE g− 1 DW). Besides, IP enhanced TPC levels compared to GP, even at values higher than the initial ones in the case of cane extracts (39 mg GAE g− 1 DW). TPC results are in agreement with the data for the overall bioaccessibility. In fact, for cane extracts the increase of TPC after IP could be associated to the high amount of ε-viniferin released at this phase. On the other hand, for bunch stem the reduction of TPC may be due to the substantial diminution of specific abundant compounds in crude extract such as (+)-catechin and procyanidin B1 (Table 1). Previous data for other winemaking by-products (seeds and pomace extracts), reported a decrease in TPC after in-vitro GID (Jara-Palacios et al., 2018). Nevertheless, Jara-Palacios et al. (2018) reported an increase of TPC levels after GID for bunch stems. Besides, as it was previously discussed, some methodological variations could cause the different results. Other authors observed a similar behavior to the presented here for cane extracts but with whole grape extracts (Tagliazucchi et al., 2010). As Pavan, Sancho and Pastore (2014) described, the reduction in TPC during GID could be related to PCs sensitivity to high pH, since at that pH, monomers obtained by hydrolysis from larger molecules may be less stable. Whereas, the increase in the TPC was associated to the release of complexed PCs as a result of the digestive process (Haas et al., 2019; Pavan et al., 2014; Tagliazucchi et al., 2010).
Taking into account the direct relationship between PCs content and their AC in a given sample, the latter was assessed by ORAC and ABTS methods and associated with the TPC achieved for both by-product extracts during the different digestion phases (Fig. 3). This direct relationship between TPC and AC (independently of the method used) was clearly observed in our results, where AC presented similar tendency as TPC for each step of digestion. In agreement with the high TPC levels found in bunch stems, this matrix also showed considerably high levels of AC. A previous study reported that ORAC values increased notably during in-vitro GID of stem extracts (Jara-Palacios et al., 2018), which was not in accordance with our results. Some authors also noted an important increase in the AC of the bioaccessible fractions by ORAC for grape pomace extracts (Sanz-Buenhombre et al., 2016). This difference might be related to in-vitro conditions of GID used and/or changes of PCs availability due to release of matrix associated compounds (Hur et al., 2011; Ortega et al., 2011; Pavan et al., 2014). In fact, free PCs have shown higher AC than iron-phenol chelates (Bouayed, Hoffmann, & Bohn, 2011; Ferreyra et al., 2020). In addition to the enzymatic action, the pH effect within the GID and the presence of compounds not analyzed (e.g. peptides or complex PCs) may be involved in this increase of activity (Haas et al., 2019; Tagliazucchi et al., 2010). The step between acidic GP and alkaline IP environments notably increases the antioxidant power of PCs by deprotonating the hydroxyl groups of their aromatic rings (Bouayed et al., 2011).
The correlations between TPC and AC were also analyzed (data not shown). In both matrices, significant positive correlations between TPC with ORAC assay (p ≤ 0.05, r ≥ 0.81) were observed. On the other way, correlation between TPC and ABTS levels was significant for bunch stems (r = 0.97) but not for canes (r = 0.70). This lower correlation may be attributed to different reaction mechanism, the action of other phytochemical compounds, and/or possible interaction (either synergism or antagonism) between compounds (Ferreyra et al., 2019). Considering the literature, the release of certain compounds after GID depends on complexity of the extracts (Oliveira & Bastos, 2011).
Remarkably, TPC and ORAC levels were improved after in-vitro GID of cane extracts, which encourage to consider this by-product as a promissory ingredient of novel functional foods. On the other hand, for bunch stem extracts the ORAC levels were maintained and TPC amounts were slightly reduced after GID, therefore this by-product may also be suitable for development of functional foods.

3.2.7. Principal component analysis

As it was presented in previous paragraphs, differences in the performance of compounds in both matrixes were observed after each step of GID. A principal component analysis (PCA) was performed to assess the variations and to examine the relationship between the individual PCs, TPC and AC data of each step of GID. Firstly, the analysis showed a clear separation of different digestion phases (OP, GP and IP) for both extracts (Fig. 4). The main components (PC1 × PC2) explicated 100% of data variability after GID of bunch stem and cane extracts.
For bunch stem, CP1 explained 74.1% and CP2 25.9% of the total dispersion. The majority of PCs were powerfully linked with the OP, displaying the minor changes in the potential bioaccessibility of extracts associated to this phase. The IP had a strong association with myricetin followed by naringenin. As was commented below, these compounds showed high bioaccessibility, being available to be absorbed in the intestine. Regarding ABTS assay, the most influential phenolic compounds were naringin and (− )-epicatechin, mostly associated with the OP.
For cane extracts, PC1 explained 73.3% and CP2 26.7% of the total data dispersion. In the OP, gallic acid, naringin, astilbin and myricetin were strongly associated. (+)-catechin, (− )-epicatechin and caftaric acid were greatly associated with the GP. Finally, naringenin, TPC, syringic acid, ε-viniferin and quercetin-3-glucoside were associated to the IP. This results completely agree with the high bioaccessibility that these compounds showed, particularly ε-viniferin. As regards ABTS assay, gallic acid, naringin and astilbin singly influenced AC associated with the OP. On the other hand, regarding ORAC assay, syringic acid significantly influenced AC associated with the IP (Fig. 4). The exploratory study showed that the matrix composition influence PCs performance during digestion phases. Cane was the matrix showing high association with PCs in the IP, suggesting a greater availability of compounds in this matrix for their absorption after GID.

4. Conclusions

The current study stated the effect of in-vitro GID on the phenolic profile and AC of bunch stem and cane extracts of the Malbec variety. Both matrices showed different results in terms of bioaccessibility. Furthermore, phenolic profile and RI of PCs were differently affected by each digestion phase. The TPC and AC of extracts and their digested fractions showed a similar behavior than phenolic profile and a direct association of these variables was observed. Additionally, the levels of PCs were influenced for the matrices during the digestion process. Notably, the digested extracts showed high levels of bioaccessible PCs, mainly syringic acid, cinnamic acid, ε-viniferin, naringenin and myricetin. It is important to highlight that this is the first time that the bioaccessibility of cane PCs was assessed. Particularly interesting was the result presented for ε-viniferin, the most abundant compound in the studied cane extract, which was not significantly degraded under simulated digestive conditions. The high bioaccessibility observed, specifically in cane extracts, will help to better understand the bioactive potential of this by-product. Additional studies using MS techniques will be necessary to achieve a deeper characterization of by-products, particularly to evaluate the potential depolymerisation of oligomeric stilbenoids during in-vitro digestion. As well, the possibility of absorption of high bioaccessible compounds such as ε-viniferin needs to be understood for promoting new applications of extracts and individual compounds. In this sense, bunch stems and canes can be considered a novel and sustainable potential font of bioactive compounds for exploring new food and pharmaceutical industry applications by their use as functional ingredients or nutraceuticals.

References

Aliano-Gonz˜ alez, M. J., Richard, T., & Cantos-Villar, E. (2020). Grapevine Cane Extracts: ´ Raw Plant Material, Extraction Methods, Quantification, and Applications. Biomolecules, 10(8), 1195. https://doi.org/10.3390/biom10081195.
Antoniolli, A., Fontana, A. R., Piccoli, P., & Bottini, R. (2015). Characterization of polyphenols and evaluation of antioxidant capacity in grape pomace of the cv. Malbec. Food Chemistry, 178, 172–178. https://doi.org/10.1016/j. foodchem.2015.01.082.
Barros, A., Giron´es-Vilaplana, A., Texeira, A., Baenas, N., & Domínguez-Perles, R. (2015). Grape stems as a source of bioactive compounds: Application towards added-value commodities and significance for human health. Phytochemistry Reviews. https://doi. org/10.1007/s11101-015-9421-5.
Bouayed, J., Hoffmann, L., & Bohn, T. (2011). Total phenolics, flavonoids, anthocyanins and antioxidant activity following simulated gastro-intestinal digestion and dialysis of apple varieties: Bioaccessibility and potential uptake. Food Chemistry, 128(1), 14–21. https://doi.org/10.1016/j.foodchem.2011.02.052.
Çetin, E. S., Altinoz, D., Tarçan, E., & G¨ oktürk Baydar, N. (2011). Chemical composition ¨ of grape canes. Industrial Crops and Products, 34(1), 994–998. https://doi.org/ 10.1016/j.indcrop.2011.03.004.
Correa, R. C. G., Haminiuk, C. W. I., Barros, L., Dias, M. I., Calhelha, R. C., Kato, C. G., ˆ Ferreira, I. C. F. R. (2017). Stability and biological activity of Merlot (Vitis vinifera) grape pomace phytochemicals after simulated in vitro gastrointestinal digestion and colonic fermentation. Journal of Functional Foods, 36, 410–417. https://doi.org/ 10.1016/j.jff.2017.07.030.
Ferreyra, S. G., Antoniolli, A., Bottini, R., & Fontana, A. (2020). Bioactive compounds and total antioxidant capacity of cane residues from different grape varieties. Journal of the Science of Food and Agriculture, 100, 376–383. https://doi.org/10.1002/ jsfa.10065.
Ferreyra, S., Bottini, R., & Fontana, A. (2019). Assessment of grapevine stems as source of phenolics with antioxidant properties. Retrieved from Rev. FCA UNCUYO, 51(2), 417–426 http://revistas.uncu.edu.ar/ojs3/index.php/RFCA/article/view/2728/ 1973.
Fontana, A. R., Antoniolli, A., & Bottini, R. (2016). Development of a high-performance liquid chromatography method based on a core–shell column approach for the rapid determination of multiclass polyphenols in grape pomaces. Food Chemistry, 192, 1–8. https://doi.org/10.1016/j.foodchem.2015.06.101.
Garbetta, A., Nicassio, L., D’Antuono, I., Cardinali, A., Linsalata, V., Attolico, G., & Minervini, F. (2018). Influence of in vitro digestion process on polyphenolic profile of skin grape (cv. Italia) and on antioxidant activity in basal or stressed conditions of human intestinal cell line (HT-29). Food Research International, 106, 878–884. https://doi.org/10.1016/j.foodres.2018.01.072.
Gonzalez-Centeno, M. R., Rossell´ o, C., Simal, S., Garau, M. C., L´ opez, F., & Femenia, A. ´ (2010). Physico-chemical properties of cell wall materials obtained from ten grape varieties and their Myricetin byproducts: Grape pomaces and stems. LWT – Food Science and Technology, 43(10), 1580–1586. https://doi.org/10.1016/j.lwt.2010.06.024.
Goufo, P., Singh, R. K., & Cortez, I. (2020). A Reference List of Phenolic Compounds (Including Stilbenes) in Grapevine (Vitis vinifera L.) Roots, Woods, Canes, Stems, and Leaves. Antioxidants, 9(5), 398. https://doi.org/10.3390/antiox9050398.
Guerrero, R. F., Biais, B., Richard, T., Puertas, B., Waffo-Teguo, P., Merillon, J.-M., & Cantos-Villar, E. (2016). Grapevine cane’s waste is a source of bioactive stilbenes. Industrial Crops and Products, 94, 884–892. https://doi.org/10.1016/j. indcrop.2016.09.055.
Guschlbauer, M., Klinger, S., Burmester, M., Horn, J., Kulling, S. E., & Breves, G. (2013). trans-Resveratrol and ε-viniferin decrease glucose absorption in porcine jejunum and ileum in vitro. Comparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology, 165(3), 313–318. https://doi.org/10.1016/j.cbpa.2013.03.040.
Haas, I. C., Toaldo, I. M., Gomes, T. M., Luna, A. S., de Gois, J. S., & Bordignon-Luiz, M. T. (2019). Polyphenolic profile, macro- and microelements in bioaccessible fractions of grape juice sediment using in vitro gastrointestinal simulation. Food Bioscience, 27, 66–74. https://doi.org/10.1016/j.fbio.2018.11.002.
Ho, S., Thoo, Y. Y., Young, D. J., & Siow, L. F. (2017). Cyclodextrin encapsulated catechin: Effect of pH, relative humidity and various food models on antioxidant stability. LWT – Food Science and Technology, 85, 232–239. https://doi.org/10.1016/ j.lwt.2017.07.028.
Hur, S. J., Lim, B. O., Decker, E. A., & McClements, D. J. (2011). In vitro human digestion models for food applications. Food Chemistry, 125(1), 1–12. https://doi.org/ 10.1016/j.foodchem.2010.08.036.
International Organization of World Vitiviniculture (OIV). (2017). OIV 2017 report on the world vitivinicultural situation. Retrieved May 26, 2020, from http://www.oiv. int/en/oiv-life/oiv-2017-report-on-the-world-vitivinicultural-situation.
Jakobek, L. (2015). Interactions of polyphenols with carbohydrates, lipids and proteins. Food Chemistry, 175, 556–567. https://doi.org/10.1016/j.foodchem.2014.12.013.
Janisch, K. M., Olschl¨ ager, C., Treutter, D., & Elstner, E. F. (2006). Simulated digestion of ¨ Vitis vinifera seed powder: Polyphenolic content and antioxidant properties. Journal of Agricultural and Food Chemistry, 54(13), 4839–4848. https://doi.org/10.1021/ jf052989k.
Jara-Palacios, J. M., Gonçalves, S., Hernanz, D., Heredia, F. J., & Romano, A. (2018). Effects of in vitro gastrointestinal digestion on phenolic compounds and antioxidant activity of different white winemaking byproducts extracts. Food Research International, 109, 433–439. https://doi.org/10.1016/j.foodres.2018.04.060.
Lambert, C., Richard, T., Renouf, E., Bisson, J., Waffo-T´eguo, P., Bordenave, L., Ollat, N., M´erillon, J.-M., Cluzet, S. (2013). Comparative Analyses of Stilbenoids in Canes of Major Vitis vinifera L. Cultivars. Journal of Agriculture and Food Chemistry, 61(47), 11392–11399. https://doi.org/10.1021/jf403716y.
Laurent, C., Besançon, P., & Caporiccio, B. (2007). Flavonoids from a grape seed extract interact with digestive secretions and intestinal cells as assessed in an in vitro digestion/Caco-2 cell culture model. Food Chemistry, 100(4), 1704–1712. https:// doi.org/10.1016/j.foodchem.2005.10.016.
Li, M. J., Loo, Y. T., Cheng, L., Howell, K., & Zhang, P. Z. (2019). Impacts of supplementation of probiotics on the prevalence of grape marc derived polyphenols in colonic digesta using in vitro digestion model. IOP Conference Series: Earth and Environmental Science, 346(1), Article 012075. https://doi.org/10.1088/1755-1315/ 346/1/012075.
Lingua, M. S., Wunderlin, D. A., & Baroni, M. V. (2018). Effect of simulated digestion on the phenolic components of red grapes and their corresponding wines. Journal of Functional Foods, 44, 86–94. https://doi.org/10.1016/j.jff.2018.02.034.
Lorenzo, J. M., Estevez, M., Barba, F. J., Thirumdas, R., Franco, D., & Munekata, P. E. S. ´ (2019). Polyphenols: Bioaccessibility and bioavailability of bioactive components. In
Innovative Thermal and Non-Thermal Processing, Bioaccessibility and Bioavailability of Nutrients and Bioactive Compounds (pp. 309–332). Elsevier. https://doi.org/10.1016/ B978-0-12-814174-8.00011-1.
Manach, C., Williamson, G., Morand, C., Scalbert, A., & R´em´esy, C. (2005). Bioavailability and bioefficacy of polyphenols in humans. I. Review of 97 bioavailability studies. The American Journal of Clinical Nutrition. https://doi.org/ 10.1093/ajcn/81.1.230s.
Mellado-Ortega, E., Zabalgogeazcoa, I., Vazquez de Aldana, B. R., & Arellano, J. B. ´ (2017). Solutions to decrease a systematic error related to AAPH addition in the fluorescence-based ORAC assay. Analytical Biochemistry, 519, 27–29. https://doi. org/10.1016/j.ab.2016.12.009.
Moss, R., Mao, Q., Taylor, D., & Saucier, C. (2013). Investigation of monomeric and oligomeric wine stilbenoids in red wines by ultra-high-performance liquid chromatography/electrospray ionization quadrupole time-of-flight mass spectrometry: Study of wine stilbenoids in red wines by UPLC/ESI-QTOFMS. Rapid Communications in Mass Spectrometry, 27(16), 1815–1827. https://doi.org/10.1002/ rcm.6636.
Nenadis, N., Kyriakoudi, A., & Tsimidou, M. Z. (2013). Impact of alkaline or acid digestion to antioxidant activity, phenolic content and composition of rice hull extracts. LWT – Food Science and Technology, 54(1), 207–215. https://doi.org/ 10.1016/j.lwt.2013.05.005. de Oliveira, D. M., & Bastos, D. H. M. (2011). Biodisponibilidade de acidos fen´ olicos. ´ Química Nova, 34(6), 1051–1056. https://doi.org/10.1590/S0100- 40422011000600023.
Ortega, N., Macia, A., Romero, M.-P., Reguant, J., & Motilva, M.-J. (2011). Matrix ` composition effect on the digestibility of carob flour phenols by an in-vitro digestion model. Food Chemistry, 124(1), 65–71. https://doi.org/10.1016/j. foodchem.2010.05.105.
Oteiza, P. I., Fraga, C. G., Mills, D. A., & Taft, D. H. (2018). Flavonoids and the gastrointestinal tract: Local and systemic effects. Molecular Aspects of Medicine, 61, 41–49. https://doi.org/10.1016/j.mam.2018.01.001.
Pavan, V., Sancho, R. A. S., & Pastore, G. M. (2014). The effect of in vitro digestion on the antioxidant activity of fruit extracts (Carica papaya, Artocarpus heterophillus and Annona marcgravii). LWT – Food Science and Technology, 59(2), 1247–1251. https:// doi.org/10.1016/j.lwt.2014.05.040.
Peˇsi´c, M. B., Milinˇci´c, D. D., Kosti´c, A.Z., Stanisavljeviˇ ´c, N. S., Vukoti´c, G. N., Koji´c, M. O., Gaˇsi´c, U. M., Bara´c, M. B., Stanojevi´c, S. P., Popovi´c, D. A., Banjac, N. R., Teˇsi´c, Z. L. (2019). In vitro digestion of meat- and cereal-based food matrix enriched ˇ with grape extracts: How are polyphenol composition, bioaccessibility and antioxidant activity affected? Food Chemistry, 284, 28–44. https://doi.org/10.1016/ j.foodchem.2019.01.107.
Püssa, T., Floren, J., Kuldkepp, P., & Raal, A. (2006). Survey of Grapevine Vitis vinifera
Stem Polyphenols by Liquid Chromatography− Diode Array Detection− Tandem Mass Spectrometry. Journal of Agriculture and Food Chemistry, 54(20), 7488–7494. https://doi.org/10.1021/jf061155e.
Saez, V., Pastene, E., Vergara, C., Mardones, C., Hermosín-Guti´ errez, I., G´ omez- ´ Alonso, S., Gomez, M. V., Theoduloz, C., Riquelme, S., ´ von Baer, D. (2018). Oligostilbenoids in Vitis vinifera L. Pinot Noir grape cane extract: Isolation, characterization, in vitro antioxidant capacity and anti-proliferative effect on cancer cells. Food Chemistry, 265, 101–110. https://doi.org/10.1016/j. foodchem.2018.05.050.
Sanz-Buenhombre, M., Villanueva, S., Moro, C., Tomas-Cobos, L., Viadel, B., & ´ Guadarrama, A. (2016). Bioavailability and the mechanism of action of a grape extract rich in polyphenols in cholesterol homeostasis. Journal of Functional Foods, 21, 178–185. https://doi.org/10.1016/j.jff.2015.11.044.
Tagliazucchi, D., Verzelloni, E., Bertolini, D., & Conte, A. (2010). In vitro bio- accessibility and antioxidant activity of grape polyphenols. Food Chemistry, 120(2), 599–606. https://doi.org/10.1016/j.foodchem.2009.10.030.
Teixeira, A., Baenas, N., Dominguez-Perles, R., Barros, A., Rosa, E., Moreno, D., & Garcia- Viguera, C. (2014). Natural Bioactive Compounds from Winery By-Products as Health Promoters: A Review. International Journal of Molecular Sciences, 15(12), 15638–15678. https://doi.org/10.3390/ijms150915638.
Torres-Palazzolo, C., Ramirez, D., Locatelli, D., Manucha, W., Castro, C., & Camargo, A. (2018). Bioaccessibility and permeability of bioactive compounds in raw and cooked garlic. Journal of Food Composition and Analysis, 70, 49–53. https://doi.org/10.1016/ j.jfca.2018.03.008.
Willenberg, I., Michael, M., Wonik, J., Bartel, L. C., Empl, M. T., & Schebb, N. H. (2015). Investigation of the absorption of resveratrol oligomers in the Caco-2 cellular model of intestinal absorption. Food Chemistry, 167, 245–250. https://doi.org/10.1016/j. foodchem.2014.06.103.
Zghonda, N., Yoshida, S., Ezaki, S., Otake, Y., Murakami, C., Mliki, A., Ghorbel, A., Miyazaki, H. (2012). ε-Viniferin Is More Effective Than Its Monomer Resveratrol in Improving the Functions of Vascular Endothelial Cells and the Heart. Bioscience, Biotechnology, and Biochemistry, 76(5), 954–960. https://doi.org/10.1271/ bbb.110975.