Predicting the effects of in-vitro digestion in the bioactivity and bioaccessibility of antioxidant compounds extracted from chestnut shells by supercritical fluid extraction – A metabolomic approach

Chestnut ( Castanea sativa ) shells (CS) are an undervalued antioxidant-rich by-product. This study explores the impact of in-vitro digestion on the bioaccessibility, bioactivity

Currently, the search for efficient and clean extraction techniques is a hot topic in the food research field.For instance, supercritical fluid extraction (SFE) has gained ground as an alternative technology committed to the sustainable recovery of bioactive compounds from agro-industrial wastes (Pinto, Cádiz-Gurrea, Vallverdú-Queralt, et al., 2021;Pinto et al., 2020).Several factors influence the performance of SFE, including temperature, time, pressure, and solid-liquid ratio (Pinto, Cádiz-Gurrea, Vallverdú-Queralt, et al., 2021;Pinto et al., 2020).Pinto et al. (2020) optimized the extraction of phenolic compounds from CS by SFE using a response surface methodology (RSM).In addition to CO 2 used as a solvent to enable the extraction of low polar molecules without additional clean-up procedures, ethanol was employed as a cosolvent to improve the solubility of polar molecules and enhance the phenolics recovery yield.The optimal extract was obtained at 60 • C,  350 bar, and 15% of ethanol (co-solvent) and revealed extraordinary antioxidant/antiradical and antimicrobial properties ascribed to the phenolic acids (ellagic acid and caffeic acid derivative), flavonoids (catechins) and procyanidin polymers identified (Pinto, Cádiz-Gurrea, Garcia, et al., 2021;Pinto et al., 2020).The safety of the extract was proven up to 100 μg/mL in two intestinal cell lines (Pinto et al., 2020).
This study proposed the valorization of CS extract prepared by SFE as a new active ingredient for nutraceuticals, proving its bioactivity and safety.Nevertheless, the in-vitro digestibility of the antioxidants-rich extract from CS and its impact on the phenolic composition and bioactivity have not been evaluated.In fact, during human digestion, pH, temperature, digestive enzymes, and intestinal microbiota directly affect the bioaccessibility of micronutrients (such as vitamins and phenolic compounds) and, consequently, their bioactivity, emphasizing the complexity of the in-vivo environment (Hu et al., 2023).Therefore, the implementation of in-vitro models that mimic the physiological conditions of the human digestive tract is increasing among the scientific community as robust and useful tools for predicting the impact of digestion on phenolic compounds (Hu et al., 2023;Pinto, Silva, et al., 2023).Although previous research has proven the efficacy of SFE in the recovery of phenolics-rich extracts from CS (Pinto et al., 2020), no studies have demonstrated its bioactivity after gastrointestinal digestion.The present study aims to evaluate the effects of in-vitro gastrointestinal digestion on total phenolic and flavonoid contents (TPC and TFC, respectively), antioxidant/antiradical properties, scavenging activity against reactive oxygen and nitrogen species (ROS and RNS, respectively), antioxidant enzymes activities, LPO, and inhibition of acetylcholinesterase (AChE) and α-amylase activities of the CS extract prepared by SFE, considering valorizing it as an active ingredient for nutraceutical products.The metabolic profile was assessed by liquid chromatography coupled to Orbitrap-mass spectrometry (LC-ESI-LTQ-Orbitrap-MS) targeted on phenolic compounds.The recovery and bioaccessibility of phenolic compounds were estimated after each digestion phase.Notably, this paper provides, for the first time, new insights into the phenolic profile and bioactivity of SFE extract from CS digested in an in-vitro simulated model, proposing their potential industrial valorization as raw material for nutraceutical industry.

Castanea sativa shells
CS were generously offered by Sortegel located in Sortes, Bragança, Portugal (latitude 41 • 42′18.6″Nand longitude 6 • 48′36.6″W) in October 2018.After dehydration at 40 • C for 24 h (Excalibur Food Dehydrator, Sacramento, CA, USA), the shells were ground to 1 mm particle size using an ultra-centrifugal grinder (Retsch model ZM200, Düsseldorf, Germany) and stored in sealed flasks at room temperature in the dark.

Preparation of chestnut shells extract by supercritical fluid extraction
SFE was performed according to Pinto et al. (2020) using a supercritical fluid extractor (Waters Prep SFE System SFE-100, Milford, MA, USA) containing automated back pressure regulator, CO 2 and co-solvent pumps, heating exchangers for low and high pressures, and pressurized extraction and collection vessels.An Accel 500 LC chiller (Thermo Scientific TM , Leicestershire, UK) was attached to maintain CO 2 in the liquid state.The extraction vessel was filled with a three-layer sandwich containing: Ottawa sand (5 g) in the first layer, a blend of powdered shells (15 g) and Ottawa sand (30 g) in the second layer, and Ottawa sand (5 g) in the third layer.To avoid sample projections, wool glass (Sigma-Aldrich, Steinheim, Germany) was added at the top and bottom of the extraction cell.CO 2 and ethanol supplied by Air Product and Chemicals (Allentown, PA, USA) and VWR chemicals (Radnor, PA, USA) were used, respectively, as supercritical fluid and co-solvent.The extraction conditions were 60 • C temperature, 350 bar pressure, and 15% co-solvent, based on the optimization study performed by Pinto et al. (2020).The extraction was conducted for 90 min at a constant flow rate (30 g/min).Afterward, the extract was evaporated (Rotavapor Buchi model R210, Flawil, Switzerland) and kept at − 20 • C in the dark until further analysis.Three batches of extract were prepared.

In-vitro simulated gastrointestinal digestion
In-vitro simulated digestion was performed following the method implemented by Minekus et al. (2014), with slight alterations (Pinto, Silva, et al., 2023).For the gastric phase, the CS extract dissolved in distilled water (50 mg/mL) was mixed with simulated gastric fluid in a 1:1 (v/v) ratio and pepsin (2000 U/mL).The pH was adjusted to 3 and the mixture was incubated in a water bath (Ovan, BSC127-E model, Barcelona, Spain) at 37 • C for 2 min under stirring.Finally, the intestinal D. Pinto et al. phase was performed by mixing the gastric digest with simulated intestinal fluid in a 1:1 (v/v) ratio, pancreatin (100 U/mL) and bile salts (10 mM).The pH was adjusted to 7 and the mixture was incubated in a water bath at 37 • C for 2 h under stirring.After each digestion phase, aliquots were collected, centrifuged at 10,000 g for 10 min (Sigma 3-30KS, Osterode am HarzSigma, Germany), and kept at − 80 • C until further analysis.The simulated fluids were prepared using KCl, KH 2 PO 4 , NaHCO 3 , NaCl, MgCl 2 (H 2 O) 6 and (NH 4 ) 2 CO 3 (Minekus et al., 2014).Three independent experiments were carried out for each digestion stage.The phenolic recovery rate was determined by applying the equation (1): where the PC digested fraction is the phenolic content determined after each digestion phase, while the PC undigested extract is the phenolic content determined in the undigested extract.The bioaccessibility (%) refers to the recovery rate after all digestion phases.

Total phenolic and flavonoid contents
TPC and TFC were estimated by the Folin-Ciocalteu and aluminum chloride assays, respectively, using the methods previously reported by (Singleton et al., 1965) and Zhishen et al. (1999), with minor modifications (Pinto, Moreira, Švarc-Gajić, et al., 2023).The TPC results were presented in µg of gallic acid equivalents (GAE) per mg of dry weight (DW) (µg GAE/mg DW).The TFC results were expressed as µg of catechin equivalents (CE) per mg of DW (µg CE/mg DW).

Scavenging activity against reactive oxygen and nitrogen species
The scavenging capacity of digested and undigested samples was evaluated following the procedures described by Gomes et al. (2007), with slight modifications (Pinto, Moreira, Švarc-Gajić, et al., 2023), against ROS and RNS, including superoxide anion radical (O 2 •− ), hydrogen peroxide (H 2 O 2 ), hypochlorous acid (HOCl), peroxyl radical (ROO • ), and peroxynitrite (ONOO − ) in the presence and absence of 25 mM NaHCO 3 to simulate physiological CO 2 conditions.A Synergy HT Microplate Reader (BioTek Instruments, Winooski, VT, USA) was used for absorbance and fluorescence measurements.Gallic acid and catechin were employed as positive controls.The undigested extract was tested up to 500 µg/mL.The results were expressed as inhibition, in % or halfmaximal inhibitory concentration (IC 50 , µg/mL), apart from the ROO • scavenging assay whose results were presented as µmol of TE per mg of DW (µmol TE/mg DW).

In-vitro biological activities 2.9.1. Acetylcholinesterase activity inhibition
AChE activity was evaluated after exposure to the undigested extract and its digests.A commercial kit (Sigma-Aldrich, St. Louis, MO, USA) estimated the amount of the colorimetric product formed by the reaction between 5,5′-dithiobis(2-nitrobenzoic acid) and thiocholine.The absorbance was measured at 412 nm.The results were presented in inhibition percentage (%).

Amylase activity inhibition
Amylase activity was evaluated after exposure to the undigested extract and its digests using a commercial kit (Sigma-Aldrich, St. Louis, MO, USA) that estimates the amount of the colorimetric product formed by the cleavage of ethylidene-pNP-G7 (substrate) by amylase.Nitrophenol was the standard used.The absorbance was measured at 405 nm.The results were presented in inhibition percentage (%).

Targeted metabolomic profile by LC-ESI-LTQ-Orbitrap-MS
The metabolic profile was analyzed using a LC-ESI-LTQ-Orbitrap-MS consisting of Accela chromatograph (Thermo Scientific, Hemel Hempstead, UK) with a quaternary pump, a photodiode array detector and a thermostated autosampler coupled to a LTQ Orbitrap Velos mass spectrometer (Thermo Scientific, Hemel Hempstead, UK) attached to an electrospray ionization (ESI) source operated in negative mode.A full scan mode was employed with a resolving power of 30,000 at m/z 600 and data-dependent MS/MS events were acquired with a resolving power of 15,000.Fourier transform mass spectrometry (FTMS) modetriggered data-dependent scanning was applied to detect the most intense ions.MSn mode was applied to the ions not intense enough for a data-dependent scan.Precursors were fragmented by collision-induced dissociation using a C-trap with normalized collision energy (35 V) and an activation time of 10 ms.In FTMS mode, the mass range was m/z 100 to 600.An Acquity TM UPLC® BEH C18 Column (2.1 × 100 mm, i.d., 1.7 μm particle size) (Waters Corporation, Wexford, Ireland) was used for the chromatographic elution.Water (A) and acetonitrile (B) both containing 0.1% formic acid were the mobile phases used.The solvent gradient (v/v) of B (t (min), %B) was defined as follows: (0, 0); (2, 0); (3, 30); (4, 100); (5, 100); (6, 0); (9, 0).The column temperature was 30 • C.
The injection volume was 5 μL and the flow rate was 0.450 mL/min.The instrumental conditions were defined as described in previous studies (Pinto, Almeida, et al., 2023;Pinto, López-Yerena, et al., 2023).Phenolic compounds were identified using commercial standards, while metabolites were identified considering chemical composition, elution time, MS/MS fragmentation, and comparison with similar compounds.The remaining molecules (whose standards were not available) were identified in comparison with data published by our research group (Pinto, Almeida, et al., 2023;Pinto, López-Yerena, et al., 2023) considering retention times, chemical formula, and MS fragments, and confirmed with databases, namely food database (https://foodb.ca)and phenol database (https://phenol-explorer.eu).Identical procedures were followed in previous studies for targeted metabolic profiling in foods and extracts (Pinto, Almeida, et al., 2023;Pinto, López-Yerena, et al., 2023).MS n measurements were performed to collect fragment ions produced in the linear ion trap.For the elemental composition of the metabolites, accurate masses, and isotopic patterns were considered.The quantification was performed using calibration curves (concentration range = 0.1-3 μg/mL, R 2 > 0.994) with ellagic acid, gallic acid, methyl gallate, protocatechuic acid, and pyrogallol that were previously identified as main phenolic compounds in CS (Ferreira et al., 2022;Pinto, Silva, et al., 2021;Pinto, Vieira, et al., 2021;Lameirão et al., 2020;Pinto et al., 2020).Semi-quantification was performed for the D. Pinto et al. remaining phenolic compounds and metabolites.XCalibur 3.0 software (ThermoFisher Scientific, Hemel Hempstead, UK) was used for system control and data processing.The peak area of the parent molecules was used for calculations and the results were expressed as µg of each phenolic compound equivalents per mg of DW.

Statistical analysis
Three independent experiments were performed for each assay.Oneway ANOVA and Tukey's HSD test were applied for statistical analysis using IBM SPSS Statistics 24.0 software (Chicago, IL, USA).A p < 0.05 was established for significant differences.Multivariate data statistics were also performed by principal component analysis (PCA) and heatmap correlations using GraphPad Prism v9 software (La Jolla, CA, USA).

Effects of gastrointestinal digestion in total phenolic and flavonoid contents
The richness of CS in phenolic compounds and flavonoids has been demonstrated by previous studies, exploring the efficacy of different extraction techniques and conditions (Pinto, Silva, et al., 2021;Pinto, Vieira, et al., 2021;Lameirão et al., 2020;Pinto et al., 2020).Nonetheless, the effects of in-vitro gastrointestinal digestion on the phenolic composition of CS extract have not been explored.The TPC and TFC of CS extract prepared by SFE were determined before and after in-vitro simulated digestion (Table 1).
The results showed significant differences in the TPC and TFC results after in-vitro digestion of CS extract, emphasizing the substantial impact of gastric and intestinal phases in the phenolic and flavonoid concentrations.Similar TPC was estimated for CS extract after gastric and intestinal digestion (37.96 and 38.57µg GAE/mg DW, respectively), revealing non-significant differences (p > 0.05).Oppositely, the undigested extract (127.72 µg GAE/mg DW) had a significantly (p < 0.05) higher TPC when compared to its digests.The present results indicate that gastric conditions (i.e., pepsin and acidic pH) allow the release of most the phenolic compounds from the extract to the digestion medium, remaining after intestinal digestion (highlighting their stability under intestinal conditions) or being metabolized into other phenolic compounds that become potentially available for intestinal absorption.
The TFC of the CS extract also significantly decreased (p < 0.05) after in-vitro digestion, following this order: undigested extract (54.15 µg CE/ mg DW) > intestinal digest (15.03 µg CE/mg DW) > gastric digest (10.65 µg CE/mg DW).From gastric to intestinal digestion, the TFC increased by 41%.In contrast to TPC, flavonoids were recovered more effectively under intestinal environment, highlighting that the action of pancreatin, bile salts and neutral pH enhanced the flavonoids release from the extract to the digestive medium, which agrees with recent studies on cupuassu peels, blueberry fruits, roasted coffee beans, and functional cookies enriched with phenolic extracts (Andrade et al., 2022;Muñoz-Fariña et al., 2023;Pinto, Moreira, Švarc-Gajić et al., 2023;Wu et al., 2022).This explanation is corroborated by the significantly different results (p < 0.05) between the gastric and intestinal phases.
The decrease in TPC and TFC after the in-vitro gastrointestinal digestion was already expected, as proven in previous studies for different food by-products (Andrade et al., 2022;Kashyap et al., 2022;Ruíz-García et al., 2022;Tu et al., 2021).This may be explained by the phenolic compounds' instability to digestive enzymes and pH changes that lead to their degradation or biotransformation into metabolites.Another hypothesis could be associated to possible interactions with other constituents of the digestive media or extract (e.g., amino acids, fatty acids, sugar moieties) that may form complexes with some phenolics and interfere with the TPC quantification (Hu et al., 2023).Recent studies have reported lower TPCs after in-vitro digestion, particularly for cherry pomace, cupuassu seeds, and grape skins (Andrade et al., 2022;Kashyap et al., 2022;Ruíz-García et al., 2022), which may be explained by the low solubility of phenolic compounds in the digestive media or their metabolization into compounds resulting from interactions with other constituents from the digestive media (e.g., amino acids) (Hu et al., 2023).Recently, Tu et al. (2021) investigated the bioaccessibility of phenolic compounds from Chinese chestnut (Castanea molissima) shells and examined their metabolic transformations upon in-vitro simulated digestion.The results were in line with the current study, confirming a decrease in TPC during gastrointestinal digestion (gastric phase: 83.58 and 47.14 mg GAE/100 mL, respectively, for chestnut outer and inner shells; intestinal phase: 56.69 and 32.47 mg GAE/100 mL, respectively) (Tu et al., 2021).

Bioaccessibility of phenolic compounds
Bioaccessibility corresponds to the concentration of phenolic compounds released from the extract after gastrointestinal digestion, becoming potentially available for intestinal absorption (Pinto, Silva et al., 2023).The phenolic recovery rates were 29.73% and 30.25%, respectively, after gastric and intestinal digestion of CS extract (Table 1).These results were identical (p > 0.05), emphasizing the hypothesis that most of the phenolic compounds present in the extract were recovered in the gastric environment (i.e., acidic pH and pepsin), while a similar fraction of these compounds was also retained in the intestinal medium after the action of pancreatin, bile salts and neutral pH (Pinto, Moreira, Švarc-Gajić et al., 2023).The flavonoid recovery rates showed a distinct behavior, increasing significantly (p < 0.05) from gastric (19.80%) to intestinal (27.74%) digestion.Although a substantial portion was retained under gastric conditions, flavonoids were released mostly by gut enzymes and under neutral pH.
The bioaccessibility of phenolic and flavonoid compounds indicated a satisfactory release profile of the CS extract, with a maximum bioaccessibility of 30% and following the same trends of TPC and TFC.These findings corroborated the TPC and TFC results, indicating that phenolic compounds and flavonoids become more bioaccessible upon invitro digestion due to the slight increase in their recovery rates.This trend was also attested for other fruit by-products after in-vitro digestion, such as grape skin and cherry pomace, reporting bioaccessibility rates of up to 50% (Kashyap et al., 2022;Ruíz-García et al., 2022).Recent studies have implemented encapsulation techniques to improve the bioaccessibility of phenolic compounds recovered from fruit by-

Table 1
Total phenolic and flavonoid contents, bioaccessibility, and antioxidant activity of chestnut shells (CS) extract prepared by supercritical fluid extraction (SFE) before and after in-vitro digestion.products, such as acerola residue and pomegranate peel (Andishmand et al., 2023;Silva et al., 2022).

In-vitro simulated digestion
The recovery rates may be justified by the interaction of phenolic compounds and flavonoids with other CS extract components, such as minerals, pigments, fatty acids, and sugars, that can be sensitively activated by the acidic and alkaline environment (Hu et al., 2023).Thus, phenolic compounds could precipitate with other substances present in the digests under acidic and alkaline pH (Hu et al., 2023;Pinto, Silva et al., 2023).Furthermore, it is important to highlight that phenolic compounds not released in gastric and intestinal digestions may be retained in the indigestible fraction and be further biotransformed in colon, releasing aglycones and microbial metabolites with interesting bioactive properties (Pinto, Almeida, et al., 2023;Pinto, López-Yerena, et al., 2023).

Effects of gastrointestinal digestion on antioxidant/antiradical properties
Natural antioxidants from CS have been documented for food and nutraceutical applications due to their disease-preventive properties on chronic diseases mediated by oxidative stress (e.g., cancer, diabetes, premature aging, neurological, cardiovascular, and metabolic pathologies) as well as food preservation properties (replacing synthetic preservatives to extend shelf-life, and preventing lipid oxidation and microorganisms' proliferation) (Pinto, Cádiz-Gurrea, Vallverdú-Queralt et al., 2021).Different studies have provided in-vitro and in-vivo evidence of the disease-preventing properties of CS, including against oxidative stress-triggered diseases, type 2 diabetes, microbial infections, and gastritis (Pinto, Almeida, et al., 2023;Pinto, Ferreira, et al., 2023;Pinto, López-Yerena, et al., 2023;Sangiovanni et al., 2018;Tsujita et al., 2008;Pinto, Silva, et al., 2021;Pinto et al., 2023).The use of CS extract as a functional ingredient requires the assessment of its bioactivity after digestion.To the best of our knowledge, this is the first study that evaluates the impact of digestion on the antioxidant/antiradical properties of CS extract prepared by SFE (Table 1).
As shown in Table 1, the antioxidant/antiradical properties of the CS extract increased in the following order: gastric digest < intestinal digest < undigested extract, highlighting a better antioxidant response for the undigested extract, along with an improvement in the antioxidant effects during in-vitro digestion.The three antioxidant assays revealed similar patterns.
Regarding antiradical activity, the ABTS response was twice as high after intestinal digestion (57.85 µg AAE/mg DW) when compared to the gastric phase (27.15 µg AAE/mg DW), with significant differences (p < 0.05) between the two phases.Compared with the undigested extract (119.80 µg AAE/mg DW), the antiradical activity after gastric and intestinal digestion was, respectively, 4.4-fold and 2.1-fold lower.An identical response was observed in the DPPH radicals scavenging assay, with a 49% increase from gastric to intestinal digestion, achieving significant differences (p < 0.05).The antiradical potential was 2.2 and 1.5 times lower, respectively, after the gastric and intestinal phases (33.20 and 49.44 µg TE/mg DW, respectively), when compared to the undigested extract (73.33 µg TE/mg DW).Additionally, significantly different antiradical responses (p < 0.05) were observed between the two digestive phases and the undigested extract in the ABTS and DPPH assays.
The antioxidant activity assessed by the FRAP assay also increased 37% from gastric to intestinal digestion, with significant differences (p < 0.05) between digests.Compared with the undigested extract (318.05 µg FSE/mg DW), the antioxidant potential was 68% and 22% lower, respectively, after gastric and intestinal digestion (189.25 and 260.12 µg FSE/mg DW, respectively).The FRAP responses were significantly different (p < 0.05) between gastric and intestinal digests and the undigested extract.
Overall, the undigested CS extract revealed better antioxidant/antiradical properties, which corroborates recent studies on the digestibility of antioxidants from food by-products (Andrade et al., 2022;Muñoz-Fariña et al., 2023;Wu et al., 2022).Considering only the digests, better antioxidant/antiradical effects were obtained after intestinal digestion, proposing that higher concentrations of antioxidants were recovered at this phase or that the bioactive molecules retrieved had better antioxidant/antiradical properties (Pinto, Moreira, Švarc-Gajić et al., 2023).
In general, antioxidant/antiradical results are in close agreement with TPC and TFC, reinforcing the promising role of phenolic and flavonoid compounds in the bioactivity of undigested and digested CS extract.The hydroxyl groups released from phenolic monomers or aglycones in gastric digestion that provide effective antioxidant properties may be a plausible explanation for the mild antioxidant properties after gastric digestion (Wu et al., 2022).Furthermore, phenolic compounds are biotransformed into small molecules recognized as metabolites and endowed with potent antioxidant properties under intestinal conditions (by the action of gut enzymes and neutral pH), explaining the more pronounced increase in antioxidant/antiradical responses after intestinal digestion (Muñoz-Fariña et al., 2023).
According to Tu et al. (2021), lower antioxidant results were reported for Chinese CS before (50.27 and 50.85 mg AAE/100 mL, respectively, for chestnut outer and inner shells in DPPH assay) and after gastric (99.71 and 74.79 mg AAE/100 mL, respectively) and intestinal (90 and 70 mg AAE/100 mL, respectively) digestion.Previous reports have shown similar patterns with an increase in antioxidant/antiradical properties of phenolic extracts during in-vitro digestion (Muñoz-Fariña et al., 2023;Pinto, Silva, et al., 2023).
The antioxidant properties of the CS extract digested fractions may be due to the phenolic compounds released from the extract to the digestive media, including free phenolics and soluble-bound phenolics (e.g., esterified) that are linked to sugars through a hydroxyl group or carbon-carbon linkages (Hu et al., 2023).Additionally, other soluble molecules with antioxidant properties (e.g., minerals, vitamins, fatty acids, fibers) may be present in the CS extract and could be released to the digestive media, affecting the ability to scavenge free radicals (in ABTS and DPPH assays) and reduce ferric ions (in FRAP assay) (Andrade et al., 2022;Hu et al., 2023;Wu et al., 2022).The differences observed in the antioxidant/antiradical results may be due to the different methodologies principles as well as the bioaccessibility of the phenolic compounds since the chemical structures are directly related to the antioxidant efficiency.

Scavenging efficiency against reactive oxygen and nitrogen species
Natural antioxidants have been described as promising anti-aging molecules with potential uses as nutraceuticals or functional ingredients for foods.CS are excellent sources of phenolic compounds and vitamin E that deliver protective effects against oxidative stress and inflammation-induced injuries in biomolecules (Pinto, Almeida et al., 2023;into, Cádiz-Gurrea, Vallverdú-Queralt, et al., 2021).The results of scavenging assays against ROS and RNS produced in the human body before and after in-vitro digestion of CS extract are presented in Supplementary Table 1.
The highest scavenging efficiency was achieved against HOCl and ONOO − in the presence and absence of sodium bicarbonate.The scavenging capacity of the samples improved as follows: gastric digest < intestinal digest < undigested extract, except for O 2 •− and ROO • in which the digests showed better results when compared to the undigested extract.
Considering O 2 •− scavenging potential, the CS extract showed promising results after gastric and intestinal digestion (55.71 and 57.41% inhibition, respectively), while the undigested extract inhibited 49.42% of the O 2 •− generated.Among the positive controls, catechin reached a 4.4-fold higher IC 50 value (48.21 µg/mL) when compared to gallic acid (10.95 µg/mL).Significant differences (p < 0.05) were achieved between undigested extract and its digests, while non-significant differences (p > 0.05) were observed between gastric and intestinal D. Pinto et al.

digests.
Regarding the H 2 O 2 quenching assay, the undigested extract (69.97% inhibition) and its intestinal digest (59.78% inhibition) scavenged this species more efficiently, followed by gastric digest (50.38% inhibition).However, the positive controls tested showed significantly higher H 2 O 2 scavenging ability.Catechin displayed a 5-fold lower IC 50 value (20.78 µg/mL) when compared to gallic acid (106.03µg/mL).Additionally, the results of the undigested extract and its digests were significantly different (p < 0.05).
The ROO • scavenging potential was estimated through the oxygen radical absorbance capacity (ORAC) assay to assess the ability of undigested and digested CS extract to protect biological tissues against LPO (Pinto et al., 2021).The undigested CS extract and its gastric and intestinal digests achieved identical ROO • scavenging responses (p > 0.05), with results ranging between 0.05 and 0.07 µmol TE/mg DW.Catechin (1.84 µmol TE/mg DW) and gallic acid (1.09 µmol TE/mg DW) showed the highest ROO • quenching ability.
The ONOO − scavenging activity assay was performed in the absence and presence of sodium bicarbonate to simulate physiological bicarbonate concentrations (≈25 mM) (Pinto et al., 2021).Among samples, ONOO − counteracting responses improved in the following order: gastric digest < intestinal digest < undigested extract, with better results in the presence of sodium bicarbonate (64.85% inhibition, 71.09% inhibition, and IC 50 of 3.20 µg/mL, respectively) than in its absence (62.28% inhibition, 70.15% inhibition, and IC 50 of 3.80 µg/mL).Nonetheless, catechin and gallic acid were better ONOO − quenchers in the presence of sodium bicarbonate (IC 50 of 0.23 and 0.29 µg/mL, respectively) as well as in its absence (IC 50 of 0.16 and 0.15 µg/mL).The undigested extract and its digests showed significantly different results (p < 0.05), while catechin and gallic acid revealed similar (p > 0.05) scavenging efficiency.
The promising results observed were probably attributed to the phenolic composition of the CS extract and respective digests, namely phenolic acids, flavonoids, hydrolyzable tannins, and lignans (identified in this study by metabolomic analysis), whose scavenging potential has been previously reported (Ketsawatsakul et al., 2000;Pinto, Ferreira, et al., 2023).These outcomes are in line with the ones reported for CS extracts prepared by other eco-friendly technologies (Pinto, Silva, et al., 2021;Pinto, Vieira, et al., 2021;Lameirão et al., 2020;Pinto et al., 2020).

Antioxidant enzymes activities and lipid peroxidation
Antioxidant enzymes protect the human body against the deleterious effects of pro-oxidant species, minimizing the impact of oxidative stress on biomolecules (Pinto, Reis et al., 2021).However, their activities decrease with age and poor diet (Pinto, Reis et al., 2021).In this sense, antioxidants-rich nutraceuticals may offer a valuable ally against oxidative stress, enhancing the activity of these antioxidant enzymes  (Pinto, Almeida, et al., 2023;Pinto, Cádiz-Gurrea, Vallverdú-Queralt, et al., 2021).The effects of undigested CS extract and its digests on antioxidant enzymes' activities and LPO are shown in Fig. 1.
The present results indicate an improvement in the activities of antioxidant enzymes, namely CAT, SOD, and GSH-Px, and a protective effect against LPO.These findings highlight the antioxidant properties of CS extract before and after in-vitro digestion, corroborating previous results.
Considering the CAT activity (Fig. 1A), the undigested CS extract (426.61nmol/min/g DW) revealed a 2.2-fold higher result when compared to its digests, while the gastric (189.06 nmol/min/g DW) and intestinal (192.47 nmol/min/g DW) digests achieved similar capacities (p > 0.05) to improve the CAT activity.Additionally, significant differences (p < 0.05) were observed between the undigested extract and its digests.
An identical response was observed for GSH-Px activity (Fig. 1B).Gastric and intestinal digests of the CS extract revealed similar capacities to improve GSH-Px activity (416.40 and 422.83 µmol/min/g DW, respectively), without significant differences (p > 0.05).Furthermore, CS extract stimulated 4 times more the GSH-Px activity before in-vitro digestion (1688.10µmol/min/g DW).The result of the undigested extract was significantly different (p < 0.05) from its gastric and intestinal digests.
The SOD activity increased in the presence of gastric and intestinal digests (174.26 and 197.10 µmol/min/g DW, respectively), without significant differences (p > 0.05) (Fig. 1C).In contrast to the CAT and GSH-Px assays, the undigested extract showed no effect on SOD activity.
The LPO is an effective marker of oxidative damage, playing a central role in the pathogenesis of chronic diseases.The CS extract after gastric and intestinal digestion prevented efficiently the LPO-induced injuries, disclosing the lowest results (1.60 and 1.53 nmol MDA/mg DW, respectively) 1D).No significant differences (p > 0.05) were observed between gastric and intestinal digests, highlighting similar effects on LPO.A lower protective response was obtained for the undigested extract (3.53 nmol MDA/mg DW), with significant differences (p < 0.05) when compared to both digests.
Overall, the CS extract improved the activities of antioxidant enzymes and prevented LPO before and after in-vitro digestion.These results suggest a slow release of antioxidant molecules from the extract during digestion, probably retaining a significant concentration of antioxidants after the gastric and intestinal phases, which explains the antioxidant properties of these digests.Moreover, these results are in agreement with TPC, TFC and antioxidant/antiradical results previously discussed, supporting that phenolic compounds may be effective molecules in the activation of antioxidant enzymes as well as in the prevention of LPO, as suggested by other authors (Gawlik-Dziki, 2014;Pinto, Almeida, et al., 2023).In a recent study, Pinto, Almeida, et al. (2023) demonstrated an improvement in the activities of antioxidant enzymes along with a protective response against LPO in the blood serum, liver, and kidney from rats treated with 50 and 100 mg/kg body weight of CS extract prepared by subcritical water extraction.Likewise, Gawlik-Dziki (2014) proved the activation of antioxidant enzymes induced by different spice extracts, including basil, ginger, marjoram, pepper, and thyme.Ginger, marjoram, and thyme extracts stimulated the SOD activity after in-vitro digestion, with better results after the intestinal phase (<1 mg/g DW of extract required to induce 50% activation of SOD activity), while a similar trend was observed in the CAT response after exposure to the marjoram extract, with a 50% decrease in the inhibitory activity after intestinal digestion when compared to the undigested extract (Gawlik-Dziki, 2014).

Inhibition of acetylcholinesterase and α-amylase activities
Besides antioxidant properties, phenolics-rich extracts have been explored as promising nutraceutical ingredients useful in the prevention and co-therapy of metabolic and neurological pathologies, due to their hypoglycemic and neuroprotective effects (Pinto, Silva, et al., 2021).To the best of our knowledge, this is the first study evaluating the hypoglycemic and neuroprotective properties of CS extract prepared by SFE after in-vitro digestion.The inhibition results of AChE and α-amylase activities induced by CS extract before and after in-vitro digestion are shown in Fig. 2.
Considering the AChE activity, the undigested CS extract reached more than 70% inhibition at a concentration of 125 µg/mL, which was significantly (p < 0.05) higher when compared to its digests.A 32% and 30% decrease in the AChE inhibition was obtained, respectively, after the gastric and intestinal phases.Nonetheless, the CS extract after gastric (38.47%) and intestinal (40.34%) digestion still effectively inhibited the AChE activity, revealing similar results (p > 0.05).These outcomes agree with the TPC, TFC, and bioaccessibility results, reinforcing the contribution of phenolic compounds and flavonoids in the extract as well as their slow release during digestion.In addition, the identical results of gastric and intestinal digests suggest that the phenolic compounds endowed with neuroprotective properties are mainly released in the gastric phase and preserved under intestinal conditions.In another study, Pinto, Silva, et al. (2021) proved the anticholinergic activity of CS extract prepared by microwave-assisted extraction (MAE), reaching up to 29.94% inhibition at 1000 µg/mL.Murugan et al. (2016) demonstrated lower inhibitory enzymatic properties in fruit and seed extracts of Algerian dates (Phoenix loureirii) on AChE activity before (<20% and < 40%, respectively) and after intestinal digestion (<30% and < 35%, respectively).
The undigested CS extract also induced a moderate inhibitory response on α-amylase activity, with 15.09% inhibition at 125 µg/mL.
The α-amylase inhibitory potential was maintained even after in-vitro digestion, revealing mild hypoglycemic properties after the gastric and intestinal phases (14.12% and 12.98% inhibition, respectively).Notably, similar inhibitory effects (p > 0.05) were observed before and after gastric and intestinal digestion of CS extract.The hypoglycemic effects observed are possibly attributed to the phenolic and flavonoid compounds present in the CS extract.Furthermore, the identical responses on α-amylase activity before and after digestion highlight that most of the bioactive molecules in CS extract were efficiently released during digestion, exerting their pro-healthy benefits (particularly hypoglycemic properties).Pinto, Silva, et al. (2021)   significantly after gastric digestion (9.03-19.56%)and, to a higher extent, after intestinal digestion (2.12-14.13%).The authors also attributed this bioactivity of honey extracts to their phytochemical composition rich in phenolic compounds and flavonoids (Peláez-Acero et al., 2022).The promising inhibitory potential observed is probably ascribed to the phenolic composition of the CS extract and its digests (mainly phenolic acids, flavonoids, and hydrolyzable tannins identified in the metabolomic analysis) that demonstrated mild to strong anti-amylase and anti-AChE activities, exerting hypoglycemic and neuroprotective effects (Peláez-Acero et al., 2022;Pinto, Moreira, Švarc-Gajić, et al., 2023;Pinto, Cádiz-Gurrea, Vallverdú-Queralt, et al., 2021).The anti-AChE activity showed positive correlations with TPC (r 2 ≤ 0.42) and TFC (r 2 ≤ 0.57), while anti-amylase activity only revealed mild positive correlations with TPC (r 2 ≤ 0.65), reinforcing the prevailing contribution of phenolic and flavonoid compounds for the neuroprotective and hypoglycemic properties observed in undigested and digested CS extract.Noteworthy, these results are in close agreement with the TPC and TFC values of digested and undigested CS extract.

Targeted metabolomic profile by LC-ESI-LTQ-Orbitrap-MS
During metabolism, phenolic compounds are biotransformed into metabolites, influencing their bioaccessibility and, subsequently, their bioactivity in human health (Pinto, Almeida, et al., 2023).Various invitro models have been designed to simulate human gastrointestinal digestion, aiming to evaluate the effects of metabolism on the phenolic composition and overcome the limitations of in-vivo assays (such as restraining the interindividual variability and interactions with other food nutrients) (Pinto, Almeida, et al., 2023;Pinto, López-Yerena, et al., 2023).Table 2 presents the phenolic compounds identified in the CS extract prepared by SFE before and after in-vitro digestion.Table 3 presents the concentrations of phenolic compounds in the undigested CS extract and its gastric and intestinal digests.
A total of 45 compounds were identified in the undigested extract, representing 45.78 µg/mg DW.The total phenolic concentrations decreased significantly (p < 0.05) after in-vitro digestion.Similar phenolic concentrations (p > 0.05) were determined after gastric (36.75 µg/mg DW) and intestinal (35.54 µg/mg DW) digestion, with recovery rates of 80.29% and 77.64%, respectively.Only 42 and 24 compounds were identified, respectively, in gastric and intestinal digests.These results highlight substantial variations in the phenolic profile of the CS extract upon in-vitro digestion, reinforcing the significant impact of pH and digestive enzymes on phenolic compounds retained after each digestive phase that probably suffered molecular modifications (Pinto, Moreira, Švarc-Gajić, et al., 2023).
The main polyphenolic class was phenolic acids, representing 92.8%, 97.6%, and 99.2% of the total content, respectively, in the undigested extract, and its gastric and intestinal digests.Among phenolic acids, phenylpropanoic acids (44.9%, 57.2%, and 83.7%, respectively, in undigested extract, gastric and intestinal digests), hydroxybenzoic acids (33.1%, 25.3%, and 8.0%, respectively), and phenylacetic acids (13.9%, 14.5% and 7.0%, respectively) were the most abundant subclasses.The differences in the phenolic composition of digested and undigested CS extract may be due to the presence of gastric and intestinal enzymes (such as pepsin and pancreatin) that induce changes in the original phenolic compounds, with subsequent influence on the bioaccessible fraction (Andrade et al., 2022).Furthermore, the phenolic compounds released can suffer complexation with other molecules from the CS extract or the digestive media (e.g., fatty acids, amino acids, sugars) during digestion, resulting in additive, neutralization, or synergistic interactions with bioactive compounds (Hu et al., 2023;Wu et al., 2022).
Considering hydroxybenzoic acids, gallic acid was the main compound in the undigested extract (9.17 µg/mg DW).However, the gallic acid concentration decreased by half after gastric digestion and only reached 4.08% recovery after the intestinal phase, suggesting its metabolization under gastric conditions, which originates 2,5-dihydroxybenzoic acid, 3,5-dihydroxybenzoic acid, and 3-hydroxybenzoic acid (via dehydroxylation) together with catechol (via decarboxylation and dehydroxylation), explaining the recovery rates above 100% for these compounds in the gastric phase (Bento-Silva et al., 2020).Significant differences (p < 0.05) were observed between both digests and undigested extract.This explanation may also justify the presence of 3hydroxybenzoic acid as the main compound in the intestinal digest.Methyl gallate may be also biotransformed into simpler hydroxybenzoic acids, such as 3-hydroxybenzoic acid (Pinto, López-Yerena, et al., 2023).However, these compounds showed low stability in intestinal conditions, as evidenced by the low recovery rates, probably being degraded or metabolized into smaller molecules by gut enzymes acting at neutral pH.Syringic acid was also quantified at higher concentrations in the gastric digest, with a recovery of 165% that may have resulted from sinapic acid metabolism or be released by hydrolysis of syringic acid-Oglucoside (Bento-Silva et al., 2020).According to Table 3, sinapic acid was quantified in the undigested extract at 0.1 µg/mg DW.Additionally, two sinapic acid derivatives were also identified, namely sinapic acid-Oglucoside and sinapoylquinic acid, which can lose glucose and quinic acid moieties, respectively, originating sinapic acid that may be further metabolized into syringic acid (which was detected in the undigested extract and respective digests) (Bento-Silva et al., 2020).Alternatively, sinapic acid may also be produced from ferulic acid (also detected in the undigested extract) by hydroxylation and subsequent O-methylation, and directly metabolized to syringic acid (Bento-Silva et al., 2020).Protocatechuic acid, also known as 3,4-dihydroxybenzoic acid, was not identified after in-vitro digestion, probably due to its bioconversion into β-ketoadipate via ring cleavage that further originates succinylcoenzyme A (CoA) and acetyl-CoA (Alvarez-Rodríguez et al., 2003).Glycosidic and quinic acid derivatives of hydroxybenzoic acids were not identified or were only quantified at trace levels after digestion, which may be explained by the hydrolysis of these phenolic acids derivatives, releasing their aglycones, glucose, and quinic acids (Bento-Silva et al., 2020;Pinto, López-Yerena, et al., 2023).The concentrations of hydroxybenzoic acids decreased significantly (p < 0.05) in the following order: undigested extract (15.15 µg/mg DW) > gastric digest (9.28 µg/ mg DW) > intestinal digest (2.85 µg/mg DW), with recovery rates of 61.25% and 18.83%, respectively, for gastric and intestinal digests.
Regarding hydroxycinnamic acids, low concentrations were determined in the undigested extract and its digested fractions (up to 0.43 µg/ mg DW).Caffeic acid, chlorogenic acid, ferulic acid, and p-coumaric acid were the only hydroxycinnamic acids identified and quantified in the intestinal digest.Furthermore, caffeic acid-O-glucoside and sinapoylquinic acid were identified, but not quantified in the gastric digest.Sinapic acid was quantified in higher concentrations in the gastric digest than in the undigested extract, revealing a recovery rate above 100% that can be justified by the hydrolysis of sinapoylquinic acid and sinapic acid-O-glucoside (present in the undigested extract) into sinapic acid, quinic acid and glucose by glycosidases under acidic pH of the stomach (Sova & Saso, 2020).The same applies to p-coumaric acid with recovery rates above 100% in both digests that may derived from coumaroylquinic acid and coumaric acid-O-glucoside (Sova & Saso, 2020).Glycosidic derivatives of caffeic, coumaric, ferulic, and sinapic acids, as well as quinic acid derivatives of coumaric, ferulic, and sinapic acids, were not identified after intestinal digestion, possibly due to their hydrolysis into aglycones, glucose and quinic acid (Pinto, López-Yerena, et al., 2023;Sova & Saso, 2020).This may explain the almost 100% recovery rates of caffeic, p-coumaric, ferulic, and sinapic acids after gastric and intestinal digestion.Although quinic acid was not identified due to the lack of available standard, different studies performed in similar matrices identified quinic acid as a metabolite resulting from quinic acids derivatives of hydroxybenzoic and hydroxycinnamic acids (Bento-Silva et al., 2020;Pinto, Ferreira, et al., 2023;Pinto, López-Yerena, et al., 2023;Sova & Saso, 2020).Nevertheless, the identification  of quinic acid should be investigated in further studies to support this explanation.The total concentration of hydroxycinnamic acids decreased in the same order as hydroxybenzoic acids.A higher bioaccessibility was determined for hydroxycinnamic acids after intestinal digestion (44.14%) when compared to hydroxybenzoic acids.
Phenylpropanoic acids were probably the major metabolites of hydroxycinnamic acids (i.e., caffeoylquinic, feruloylquinic, caffeic, and ferulic acids) and flavanols (i.e., (epi)catechin, (epi)catechin-O-gallate, (epi)gallocatechin, and (epi)gallocatechin-O-gallate), which may justify the recovery rates of hydroxyphenylpropionic acid above 100% after gastric and intestinal digestion (Pint o , Ló pez-Yerena, et al., 2023).The highest concentrations of hydroxyphenylpropionic acid in the intestinal digest agree with the lowest concentrations of hydroxycinnamic acids and flavanols, suggesting that this metabolite originated during the intestinal phase by the action of gut enzymes and neutral pH.In addition, dihydroxyphenylpropionic acid was not quantified in the undigested extract, reinforcing its role as a phenolic metabolite produced during invitro digestion.
Flavanols were the only subclass of flavonoids identified at significantly (p < 0.05) lower concentrations after digestion.Only trace levels of (epi)catechin, (epi)gallocatechin, and methyl-(epi)catechin were determined.
Ellagic acid was the only hydrolyzable tannin identified, revealing a significantly (p < 0.05) higher concentration in the undigested extract (2.52 µg/mg DW), followed by gastric (0.26 µg/mg DW) and intestinal (0.05 µg/mg DW) digests, with low recovery rates (10.34% and 2.16%, respectively).These results suggest the extensive metabolization of ellagic acid by the intestinal microbiota probably into urolithins due to its instability under gastric and intestinal conditions (Pinto, Almeida, et al., 2023).
Concerning the lignans, lariciresinol was identified in the digested and undigested extract, with decreasing concentrations from the undigested extract (0.05 µg/mg DW) to its intestinal digest (0.01 µg/mg DW).Secoisolariciresinol was identified only in the undigested extract and its gastric digest.Lignans are also metabolized by the intestinal microbiota, originating enterolignans (i.e., enterodiol and enterolactone) with proven pro-healthy benefits, including antioxidant, anticancer, and modulating effects on hormone metabolism (Pinto, López-Yerena, et al., 2023).
Moreover, other phenolic compounds were quantified, including catechol, pyrogallol, and hydroxybenzaldehyde, revealing higher concentrations in the undigested extract and lower after intestinal digestion.Noteworthy, hydroxybenzaldehyde was quantified only in the digests, proposing that this compound is a metabolite derived from the reduction of hydroxybenzoic acids during digestion (Bento-Silva et al., 2020).
Hippuric acid and hydroxyhippuric acid were also identified as phenolic metabolites in the digested fractions of CS extract.Hippuric acid is an acyl glycine formed by the conjugation of benzoic acid and glycine (Pinto, López-Yerena, et al., 2023).The benzoic acid may be derived from quinic acid released by hydrolysis of caffeoylquinic and galloylquinic acids (Pinto, Ferreira, et al., 2023).However, it was not possible to quantify these phenolic metabolites.
Besides the identical in-vitro bioactivity proven by previous assays, the present results reinforced the distinct phenolic profile of CS extract and its gastric and intestinal digests, allowing to understand the molecular modifications and degradation of phenolic compounds that occur during metabolism by digestive enzymes under acidic and alkaline conditions.Furthermore, certain phenolics may bind to proteins under acidic and alkaline conditions via covalent, hydrogen, and hydrophobic bonds, reducing their concentrations (Hu et al., 2023).Overall, the decrease of total phenolic concentrations after intestinal digestion can be associated with the sensitivity of these molecules to higher pH values, becoming less stable and leading to their hydrolysis into smaller molecules during digestion.

Multivariate data analysis
Multivariate data analysis was performed by PCA and heatmap correlation to outline differences between undigested CS extract and respective gastric and intestinal digests (Fig. 3).
Considering Fig. 3A, the scores plot indicates three individual clusters with 92.19% of cumulative variance.The principal component (PC) 1 justifies the major fraction of the variance of the results (70.88%), while the remaining 21.31% of the variance is explained by PC2.The differences between undigested extract and both gastric and intestinal digests are mostly elucidated by PC1, while PC2 explains the variance between gastric and intestinal digests.In summary, the PCA model points out distinct responses for the undigested CS extract and its gastric and intestinal digests, highlighting clear differences in the metabolic profile, bioaccessibility, and bioactivity of phenolic compounds extracted from CS before and after in-vitro digestion.
The heatmap correlations depicted in Fig. 3B-3D highlight positive correlations between most of the variables studied for the undigested CS extract and its gastric and intestinal digests.Even though different correlation patterns were observed between samples, these results emphasize the exceptional role of phenolic compounds and flavonoids (namely hydroxycinnamic acids, hydroxybenzoic acids, phenylpropanoic acids, phenylacetic acids, flavanols, and hydrolyzable tannins) in the antioxidant/antiradical effects, radicals scavenging efficiency, activation of antioxidant enzymes activities, protection against LPO, and hypoglycemic and neuroprotective properties of CS extract before and after in-vitro digestion, exhibiting mild to strong correlations.In general, significant correlations (p < 0.05) were mainly obtained between ROS and RNS scavenging potential, SOD, CAT and GSH-Px activities, LPO, and AChE and α-amylase inhibitory activity with the polyphenolic classes identified in the undigested and digested CS extract, highlighting a remarkable contribution of these phenolic compounds to the bioactivity of CS extract and its digests.In summary, the The multivariate statistical analysis indicates a pronounced heterogeneity between the undigested CS extract and its gastric and intestinal digests considering the in-vitro bioactivity and bioaccessibility of phenolic compounds.Furthermore, the phenolic concentrations and their bioactivity can fluctuate in response to different metabolic pathways that affect their bioaccessibility.

Conclusion
This study attempted, for the first time, to evaluate the impact of invitro gastrointestinal digestion in the bioaccessibility of phenolic compounds extracted from CS by SFE and their biological activities, n.i., non-identified.n.q., non-quantified.CS, chestnut shells.DW, dry weight.Results are expressed as mean ± standard deviation (n = 3).Different letters (a, b, and c) in the same line indicate significant differences between groups (p < 0.05) regarding phenolics concentrations.Different numbers at superscript (1, and 2) in the same line indicate significant differences between groups (p < 0.05) regarding recovery rates.
exploring the metabolic profile by LC-ESI-LTQ-Orbitrap-MS and considering a final application as a nutraceutical ingredient.The main findings of this study were: • The phenolic concentrations retained after gastric and intestinal digestion decreased significantly when compared to the undigested extract, reaching a maximum bioaccessibility of 30%.• The metabolic profile revealed the biotransformation of complex phenolic acids and flavonoids into smaller metabolites, mainly phenylpropanoic acids.• Gallic acid, ellagic acid, 3-hydroxyphenylacetic acid, and hydroxyphenylpropionic acid were the main phenolic compounds identified in the digested and undigested fractions.
• Despite the higher bioactivity of the undigested extract, its gastric and intestinal digests still exhibited promising antioxidant/antiradical, hypoglycemic, and neuroprotective properties, as well as upmodulating effects of antioxidant enzymes activities, and protection against LPO.• PCA and heatmap pointed out the exceptional contribution of the phenolic composition to the bioactivity of CS extract upon simulated digestion, along with distinct responses for the undigested extract and its gastric and intestinal digests.
These findings prove the antioxidant efficacy of the phenolics-rich CS extract prepared by SFE even after gastrointestinal digestion, highlighting the pro-healthy benefits of its incorporation into functional foods and nutraceuticals.Further research should focus on

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

D
.Pinto et al.

Table 2
Identification of phenolic compounds and their metabolites in chestnut shells (CS) extract prepared by supercritical fluid extraction (SFE) before and after in-vitro digestion explored by LC-ESI-LTQ-Orbitrap-MS.

Table 2
(continued ) D.Pinto et al.metabolic profile of phenolic compounds greatly contributes to the antioxidant/antiradical properties, ROS/RNS scavenging capacity, activation of CAT and GSH-Px activities, and anti-cholinergic activity, offering protective effects against oxidative stress injuries.

Table 3
Quantification of phenolic compounds and their metabolites in undigested chestnut shells (CS) extract and respective gastric and intestinal digests by LC-ESI-LTQ-Orbitrap-MS.