Phytochemical stability and bioavailability of black mulberry, chokeberry, and elderberry during gastrointestinal digestion and their implications for human nutrition

2.1 Plant material

This investigation employed plant materials derived from elderberry (Sambucus nigra L. cv. Haschberg), chokeberry (Aronia melanocarpa (Michx.) Elliot, cv. Nero), and Black mulberry (Morus nigra L. cv. Gümüşhacıköy Horum). Both fruit and leaf samples were collected from each species. Materials were obtained from experimental parcels at the Apricot Research Institute located in Malatya. Leaves were collected during the plants’ flowering stage. Fruits were harvested upon reaching full maturity (Yaman, 2022).

2.2 Sample preparation and extraction

Fruit and leaf samples were dried using a lyophilizer (Christ Alpha 1-2 LD plus, Germany). The dried samples underwent fine grinding into powder form. These powdered samples were sealed in polyethylene bottles and stored at −20°C until analysis. For the extraction process, approximately 1 g of lyophilized samples was weighed into tubes. The samples were mixed with 20 mL of solvent consisting of methanol:water:hydrochloric acid mixture in 70:29.9:0.1 ratio. This solvent composition was selected based on its proven effectiveness for extracting phenolic compounds from plant materials, as demonstrated in previous studies (Ghafoor et al., 2009). The methanol component serves as the primary extractant for phenolic compounds, while the acidic conditions (HCl) help stabilize anthocyanins and improve extraction efficiency. The water component enhances the polarity of the solvent system, facilitating the extraction of hydrophilic compounds. The preparations were mixed thoroughly and incubated in the dark for 2 h. After incubation, the mixtures underwent centrifugation. Supernatants were transferred to separate tubes. Fresh solvent was introduced to the remaining solid residues. The extraction procedure was repeated twice more to complete a three-step process. Combined supernatants from all steps were filtered through a 0.45 µm polyvinylidene difluoride (PVDF) filter (Zengin et al., 2025b; Zengin et al., 2025c).

2.3 In Vitro bioaccessibility

The simulation of gastrointestinal digestion involved creating synthetic oral, gastric, and intestinal environments for bioaccessibility assessment, following established protocols (Minekus et al., 2014; Brodkorb et al., 2019). For the oral phase, 0.5 g of lyophilized sample was combined with 20 mL of distilled water, 14 mL of synthetic saliva solution, 2 mL of α-amylase solution (1500 U/mL in phosphate buffer; Sigma-Aldrich, St. Louis, MO, USA), 100 µL of 0.3 M CaCl₂, and 3.9 mL of distilled water. The synthetic saliva solution was prepared according to Minekus et al. (2014) and incorporated several compounds at specific concentrations: 0.50 mol/L KCl (15.10 mL), 0.50 mol/L KH₂PO₄ (3.70 mL), 1.00 mol/L NaHCO₃ (6.80 mL), 0.15 mol/L MgCl₂ (H₂O)₆ (0.50 mL), 0.50 mol/L (NH₄)₂CO₃ (0.06 mL), and 6.00 mol/L HCl (0.09 mL). The solution was adjusted to a final volume of 400 mL with distilled water. The mixture was adjusted to pH 7 and underwent incubation at 37°C and 100 rpm for 2 min. After incubation, the mixture was filtered through a 0.22 µm PVDF filter and stored at −20°C pending analysis. The gastric solution was prepared according to Minekus et al. (2014) by dissolving 3.2 g pepsin (Sigma-Aldrich, St. Louis, MO, USA), 2 g NaCl, and 7 mL 37% HCl in distilled water. The pH was adjusted to 1.2 with 1 M HCl and diluted to 1 L. The gastric phase involved mixing 20 mL of the digested sample from the oral phase with 20 mL of the pre-prepared gastric solution. The mixture was adjusted to pH 3 using 1 M NaOH. The preparation was incubated at 37°C and 100 rpm for 2 h, following the standardized gastric digestion time established by Minekus et al. (2014). For the intestinal phase, 30 mL of the digested sample from the gastric phase underwent 10-min incubation at 37°C. The pH was adjusted to 7 using 1 M NaOH. Subsequently, 2.5 mL pancreatin lipase solution (4.8 mg/mL in phosphate buffer; Sigma-Aldrich, St. Louis, MO, USA), 4 mL bile salt solution (5 mg/mL), and 1 mL CaCl₂ solution (750 mM) were added. The pH was readjusted to 7. The pH was adjusted to 7 using 1 M NaOH. The mixture was incubated at 37°C and 100 rpm for 2 hours, following the standardized intestinal digestion time established by Minekus et al. (2014). Bioaccessibility evaluation involved analyzing samples collected after each digestion phase (oral, gastric, and intestinal). The analyses included total phenolic content (TPC), antioxidant capacity (via ABTS and CUPRAC assays), and total flavonoid content (TFC). Individual phenolic and anthocyanin compounds were also identified and quantified.

2.4 Analytical methods

TPC determination followed Bae and Suh’s (2007) procedure. The method involved combining 30 µL extract, 2370 µL distilled water, and 150 µL diluted Folin-Ciocalteu reagent (1:10 with distilled water). The mixture was allowed to stand for 3 minutes. Then, 450 µL of 7% Na₂CO₃ solution was added. The mixture was incubated for 2 hours. Absorbance was measured at 765 nm. Results were expressed as gallic acid equivalents (mg GAE/g dry weight). Antioxidant capacity assessment employed both ABTS and CUPRAC assays. The ABTS assay (Re et al., 1999) combined 50 µL extract with 150 µL extraction solvent and 3800 µL ABTS⁺ radical solution. The mixture was incubated in the dark for 15 min. Absorbance was measured at 734 nm. The CUPRAC assay (Apak et al., 2004) mixed 1 mL each of Cu+2 chloride solution (10⁻² M), neocuproine solution (7.5 × 10⁻³ M), ammonium acetate buffer (pH 7). The mixture was incubated in darkness for 1 h. Absorbance was measured at 450 nm. Results were expressed as Trolox equivalents (mg TE/g dry weight). TFC determination involved mixing 1 mL extract with 4 mL distilled water and 0.3 mL 5% NaNO₂ solution. The mixture was allowed to stand for 5 min. Then, 0.3 mL 10% AlCl₃·6H₂O solution was added and incubated for 6 min. Next, 2 mL 1 M NaOH solution was added. The volume was adjusted to 10 mL with distilled water. The mixture was incubated in darkness for 15 min. Absorbance was measured at 510 nm. Results were expressed as catechin equivalents (mg CE/g dry weight) following Kim et al., (2003). Individual phenolic compounds underwent quantification via ultra-fast liquid chromatography (UFLC) (Shimadzu Technologies, Kyoto, Japan). The system was equipped with DGU-20 A5 vacuum degasser, 20 ADXR solvent pump, and SPD M20 A photodiode array detector. A Clipeus C18 5 μm reversed-phase column (250 mm × 4.6 mm) was used. Gradient elution was performed with 1 mL/min flow rate, 30°C column temperature, and 20 μL injection volume. The mobile phase consisted of 4.5% acetic acid (solvent A) and acetonitrile (solvent B). Results were expressed as mg per 100 g dry weight (Uğur et al., 2023). Anthocyanins and K3R quantification employed liquid chromatography-tandem mass spectrometry (LC-MS/MS) 8040 system. The system was equipped with LC-20AD XR gradient pumps, CTO-10AS vp column oven, DGU-20A5R degasser, and SIL-20AC HT autosampler. A reverse-phase C18 Inertsil ODS-4 column (2.1 × 50 mm, 3 µm particle size) was used. Gradient elution was performed with 2% acetic acid in water as solvent A and methanol as solvent B. The flow rate was 0.3 mL/min, and the column temperature was 40°C. The method was adapted from Domínguez et al. (2020).

2.5 Statistical analyses

Statistical analyses involved triplicate measurements presented as mean ± standard deviation, using one-way analysis of variance (ANOVA) with IBM SPSS Statistics 22 and Duncan’s multiple range test at 0.001 significance level for evaluating genotype differences. Results are presented as mean values with their standard errors (SE). Detailed data for all treatments, including means and standard deviations (SD), are available in the supplementary material. Furthermore, a hierarchical clustering heat map was generated to visualize the relationships and intensities among the factors and studied properties. This visualization was created using the SRPLOT online platform (https://www.bioinformatics.com.cn, accessed on February 10, 2025).

3.1 Effects of digestion methods and gastrointestinal environments on phytochemical content and antioxidant capacity of fruit and leaves samples

The results showed that in the pre-digestion fruit samples, the total phenolic content (TPC) in the BMF remained constant across the oral, gastric, and intestinal phases (16.8 ± 0.2 mg GAE/g). In contrast, in the digested samples, there was a noticeable increase in the BMF fruit from the gastric phase (15.8 ± 0.1 mg GAE/g) to the intestinal phase (22.3 ± 2.3 mg GAE/g). For the Chokeberry fruit (CF), the TPC was initially high in the pre-digestion samples (41.7 ± 0.6 mg GAE/g) but decreased after digestion, with the intestinal phase reaching 20.3 ± 2.1 mg GAE/g. In the Elderberry fruit (EF), the highest TPC value was observed in the digested intestinal phase (44.0 ± 2.0 mg GAE/g). In the pre-digestion samples, total flavonoid content (TFC) values remained constant across all phases: 8.0 ± 0.1 mg CE/g in BMF, 40.7 ± 2.8 mg CE/g in CF, and 18.1 ± 0.5 mg CE/g in EF. After digestion, TFC decreased in all fractions, with the lowest levels observed in the gastric phase. The intestinal EF fraction showed a slight recovery (12.9 ± 0.2 mg CE/g), but overall, digestion led to a marked reduction in total flavonoid content. In the pre-digestion samples, ABTS radical scavenging activity stayed the same across all gastrointestinal phases: 29.6 ± 0.7 mg TE/g in BMF, 104.8 ± 0.9 mg TE/g in CF, and 63.0 ± 1.2 mg TE/g in EF. After digestion, BMF values dropped slightly, while the CF and EF fractions showed notable changes, especially higher antioxidant activity in the intestinal EF fraction (68.9 ± 1.6 mg TE/g). In the pre-digestion samples, cupric reducing antioxidant capacity (CUPRAC) antioxidant activity stayed constant across all phases: 33.3 ± 0.8 mg TE/g in BMF, 134.3 ± 4.3 mg TE/g in CF, and 87.2 ± 0.2 mg TE/g in EF. After digestion, values decreased notably in CF and EF fractions, though the intestinal EF fraction reached a higher level (96.1 ± 3.4 mg TE/g). Overall, digestion reduced CUPRAC activity in most fractions but slightly boosted it in the intestinal EF fraction. In the pre-digestion leaf samples, TPC values remained stable across all phases, with the highest in the CL fraction (69.2 ± 1.0 mg GAE/g), followed by EL (55.3 ± 0.7) and BML (45.8 ± 0.2). After digestion, TPC decreased in all fractions; the oral phase showed the lowest values, while the gastric and intestinal phases showed moderate recovery, especially in EL (45.9 ± 1.8 and 43.5 ± 0.8). Overall, digestion significantly reduced total phenolic content, particularly in the BML fraction. TFC values also remained unchanged in pre-digestion samples, highest in CL (66.0 ± 1.5 mg CE/g), followed by EL (47.1 ± 0.6) and BML (37.7 ± 0.9). After digestion, TFC dropped markedly in BML, while CL and EL fractions showed moderate decreases, with the highest value observed in the gastric CL fraction (33.9 ± 0.6). Digestion significantly reduced total flavonoid content, especially in BML. ABTS antioxidant activity was highest in CL (105.2 ± 0.4 mg TE/g) in pre-digestion samples, followed by EL (59.0 ± 0.4) and BML (47.1 ± 0.7). After digestion, values generally decreased, though the oral EL fraction showed higher activity (80.2 ± 0.3). Digestion reduced antioxidant activity overall, but the intestinal EL fraction retained relatively strong activity. CUPRAC measurements showed BML with the lowest, CL with the highest, and EL at intermediate levels. CUPRAC values generally decreased across pre-digest and digestion phases (oral, gastric, intestinal), with a notable drop during intestinal digestion in all groups. In summary, the CL group had the highest antioxidant capacity, followed by EL and BML (Table 1, Supplementary Tables 1 and 2).

3.2 Anthocyanin stability and distribution in fruit samples during simulated gastrointestinal digestion

With respect to digestion methods and gastrointestinal compartments, fruit samples displayed distinct anthocyanin profiles throughout the study. From the perspective of pre-digestion, Cyanidin-3-O-glucoside (Cy3G) exhibited remarkably higher concentrations pre-digest 2106.12±1.8 mg/100g) environments when compared to post-digestion values in oral (652.5±1.2 mg/100g), gastric (623.9±1.5 mg/100g), and intestinal (364.4±1.7 mg/100g) conditions, which represented 68.9%, 70.4%, and 82.7% reductions respectively. In the pre-digest phases (pre-digest, oral, gastric, intestinal), cyanidin-3-O-arabinoside (Cy3A) levels in Black Mulberry Fruit (BMF) remained constant at 2.9 ± 0.0 mg/100g, while Chokeberry fruit (CF) maintained a very high level at 475.6 ± 0.9 mg/100g. Cy3A was not detected in Elderberry fruit (EF). In the digest phases, Cy3A was not detected in BMF during oral, gastric, and intestinal digestion. In the CF digest, Cy3A sharply decreased to 16.3 ± 0.2 mg/100g in the oral phase, then rose to 160.8 ± 0.8 mg/100g in the gastric and was not detected in the intestinal phase. In the EF digest, Cy3A was only detected in the gastric phase at 86.2 ± 1.0 mg/100g and not detected in other phases. In the pre-digest phases, cyanidin-3-O-galactoside (Cy3Gal) was not detected in BMF and EF, but was highly present in CF with 2234.2 ± 6.3 mg/100g, constant across oral, gastric, and intestinal phases. In the digest phases, Cy3Gal was not detected in BMF and EF in the oral phase. In the CF digest, Cy3Gal decreased sharply to 41.1 ± 0.1 mg/100g in the oral phase, then increased to 870.5 ± 1.5 mg/100g in the gastric phase and decreased again to 448.5 ± 4.3 mg/100g in the intestinal phase. Cy3Gal was not detected in BMF and EF during the digestion phases. Also, pelargonidin-3-O-glucoside (P3G), cyanidin-3-O-rutinoside (Cy3R), and kaempferol-3-O-rutinoside (K3R) contents were measured across different gastrointestinal digestion phases in BMF, CF, and EF. In the pre-digestion phases, P3G and Cy3R were found at consistently high levels in BMF, with P3G at 65.3 ± 1.3 mg/100g and Cy3R at 771.1 ± 7.7 mg/100g, while K3R was relatively low and stable at 0.5 ± 0.0 mg/100g. In contrast, CF and EF exhibited low but stable amounts of P3G and Cy3R throughout pre-digestion, with P3G around 0.7 ± 0.0 and Cy3R at 0.5 ± 0.0 mg/100g in CF, and slightly higher K3R in EF at 3.4 ± 0.0 mg/100g. Upon digestion, the concentrations of these compounds generally decreased in all fractions. Specifically, P3G in BMF dropped significantly from 30.1 ± 0.7 mg/100g in the oral phase to 14.2 ± 0.9 mg/100g in the intestinal phase, while CF and EF showed only minor fluctuations at much lower levels. Similarly, Cy3R in BMF decreased markedly from 192.2 ± 7.7 mg/100g during oral digestion to 126.5 ± 1.0 mg/100g in the intestinal phase, with CF and EF maintaining low levels with slight variations. Interestingly, K3R in BMF initially increased after digestion, reaching 2.7 ± 0.1 mg/100g in the oral phase before declining through the gastric and intestinal phases, whereas EF retained relatively higher K3R levels during digestion, albeit with some decrease, and CF’s K3R content remained consistently low throughout all phases. Overall, the data suggested that BMF contained the highest levels of these anthocyanins before digestion, which declined significantly during the digestive process, whereas EF exhibited relatively high K3R content that was somewhat stable, and CF maintained low concentrations of all analyzed compounds across all phases (Table 2, Supplementary Tables 1 and 2).

3.3 Bioactive compound stability and distribution in fruit samples during simulated gastrointestinal digestion

In this study, the phenolic compound content of fruit samples was determined according to pre-digest and digest stages, with oral, gastric, and Intestinal applications. When considering both fruit and leaves samples in terms of digestion method, gastrointestinal system, and bioactive compounds, significant differences (p < 0.001) were observed in the concentrations of quercetin, catechin, chlorogenic acid, epicatechin, p-coumaric acid, and rutin. Quercetin content (mg/100g) was measured in three fruit fractions: BMF, CF, and EF across various gastrointestinal digestion phases. In the pre-digest phases (oral, gastric, and intestinal), quercetin levels remained stable and low in all fractions, with values around 0.8 ± 0.0 mg/100g in BMF, and 0.9 ± 0.0 mg/100g in both CF and EF. After digestion, quercetin levels significantly increased in BMF and EF during the oral and gastric phases, reaching 8.5 ± 0.4 mg/100g and 9.2 ± 0.4 mg/100g, respectively, during oral digestion, and peaking at 13.5 ± 0.4 mg/100g in BMF, 13.4 ± 0.0 mg/100g in CF, and 13.1 ± 0.0 mg/100g in EF during the gastric phase. The highest quercetin concentration was observed in the intestinal phase of BMF at 16.8 ± 0.0 mg/100g, while quercetin was not detected in CF and EF fractions during intestinal digestion. Regarding our results, in the pre-digest stages (oral, gastric, intestinal), catechin levels remained stable with BMF at 10.2 ± 0.6 mg/100g, CF at 46.4 ± 0.2 mg/100g, and EF showing the highest content at 1151.4 ± 1.0 mg/100g. During digestion, catechin was not detected in the oral phase of BMF and EF but was present at 8.3 ± 1.5 mg/100g in CF. In the gastric phase, catechin increased to 13.0 ± 0.6 mg/100g in BMF, 302.6 ± 24.1 mg/100g in CF, and decreased to 10.8 ± 1.1 mg/100g in EF. In the intestinal phase, catechin was detected at low levels in BMF (1.7 ± 0.1 mg/100g), was absent in CF, and showed a moderate level of 166.5 ± 13.6 mg/100g in EF. Overall, EF had the highest catechin content pre-digestion and retained significant levels through digestion, whereas CF showed a marked increase during gastric digestion, and BMF maintained low and relatively stable catechin levels throughout. In the pre-digest phases (oral, gastric, intestinal), chlorogenic acid levels remained stable, with the highest content found in CF at 292.1 ± 1.1 mg/100g, followed by EF at 97.1 ± 1.0 mg/100g, and BMF showing the lowest level at 61.4 ± 2.7 mg/100g. After digestion, chlorogenic acid content decreased significantly in all fractions during the oral phase, dropping to 11.0 ± 0.6 mg/100g in BMF, 162.1 ± 7.3 mg/100g in CF, and 4.0 ± 0.2 mg/100g in EF. During the gastric phase, content increased somewhat, especially in CF and EF, reaching 223.5 ± 2.9 mg/100g and 34.4 ± 2.1 mg/100g, respectively, while BMF rose moderately to 18.9 ± 2.6 mg/100g. In the intestinal phase, chlorogenic acid levels declined again but remained detectable: 15.9 ± 0.5 mg/100g in BMF, 178.3 ± 9.9 mg/100g in CF, and 27.3 ± 0.3 mg/100g in EF. Overall, CF consistently showed the highest chlorogenic acid content throughout digestion, while BMF and EF demonstrated lower levels with fluctuations during digestive phases (Table 3, Supplementary Tables 1 and 2).

In the pre-digest stages (oral, gastric, intestinal), epicatechin levels remained stable, with the highest concentration in CF at 702.3 ± 1.0 mg/100g, followed by BMF at 143.7 ± 2.2 mg/100g, and the lowest in EF at 10.6 ± 0.1 mg/100g. Epicatechin content decreased markedly in all fractions during digestion, dropping to 22.4 ± 2.1 mg/100g in BMF, 20.9 ± 3.9 mg/100g in CF, and 6.7 ± 0.3 mg/100g in EF during the oral phase. Levels rose moderately in the gastric phase to 36.1 ± 5.9 mg/100g in BMF, 44.6 ± 1.0 mg/100g in CF, and 13.7 ± 0.7 mg/100g in EF, then decreased again by the intestinal phase to 15.3 ± 0.4 mg/100g in BMF, 27.3 ± 2.0 mg/100g in CF, and 18.4 ± 0.2 mg/100g in EF. CF maintained the highest epicatechin levels throughout, although significantly reduced compared to pre-digestion, while BMF and EF showed lower and fluctuating levels during digestion. On the other hand, in the pre-digest stages (oral, gastric, intestinal), p-coumaric acid levels remained stable and low in all fractions, with values of approximately 0.7 ± 0.0 mg/100g in BMF and CF, and 0.5 ± 0.0 mg/100g in EF. After digestion, p-coumaric acid content increased significantly in all fractions during the oral phase, reaching 4.7 ± 0.3 mg/100g in BMF, 6.5 ± 0.9 mg/100g in CF, and 1.6 ± 0.1 mg/100g in EF. Levels further increased during the gastric phase to 3.0 ± 0.1 mg/100g in BMF, 6.3 ± 0.3 mg/100g in CF, and 3.3 ± 0.1 mg/100g in EF. The highest concentrations were observed in the intestinal phase, where BMF reached 4.2 ± 0.0 mg/100g, CF peaked at 7.2 ± 0.1 mg/100g, and EF had 4.6 ± 0.0 mg/100g. In addition, in the pre-digest stages (oral, gastric, intestinal), rutin levels remained stable, with EF showing the highest concentration at 421.0 ± 2.9 mg/100g, followed by CF at 29.2 ± 0.1 mg/100g, and BMF at 27.6 ± 2.3 mg/100g. During digestion, rutin content decreased in all fractions during the oral phase to 8.6 ± 0.4 mg/100g in BMF, 17.6 ± 3.2 mg/100g in CF, and 316.1 ± 6.5 mg/100g in EF. In the gastric phase, rutin content partially recovered, reaching 24.1 ± 0.2 mg/100g in BMF, 23.1 ± 0.3 mg/100g in CF, and 345.0 ± 1.3 mg/100g in EF. The intestinal phase showed a slight decline or stabilization, with rutin levels at 23.9 ± 1.8 mg/100g in BMF, 22.7 ± 0.1 mg/100g in CF, and 340.8 ± 10.4 mg/100g in EF. EF consistently contained the highest rutin content throughout digestion, with BMF and CF maintaining much lower but relatively stable levels (Table 4, Supplementary Tables 1 and 2).

3.4 General evaluation

The correlation analysis of fruit samples showed variations in the strength of relationships between bioactive compounds. A strong positive correlation was observed between quercetin and p-coumaric acid (r > 0.80), particularly after gastric digestion. In contrast, catechin and rutin exhibited a weak negative correlation (r < -0.30), indicating a simultaneous decrease in their concentrations post-digestion. Chlorogenic acid showed a moderate correlation with epicatechin (r ≈ 0.50), suggesting partial retention of these compounds across digestion phases. Overall, digestion influenced the correlation patterns, with stronger relationships observed among phenolic acids compared to flavonoids (Fig. 1). In the correlation analysis of leaf samples, a weak positive correlation indicated a slight tendency for one variable to have increased as the other increased, but the relationship remained weak (e.g., r ≈ 0.1–0.3). In contrast, a strong positive correlation signified a robust association, where an increase in one variable was strongly linked to an increase in the other (e.g., r > 0.7). Similarly, a weak negative correlation suggested a minor inverse relationship, meaning that as one variable increased, the other tended to decrease slightly (e.g., r ≈ -0.1 to -0.3). On the other hand, a strong negative correlation represented a pronounced inverse relationship, where a rise in one variable was strongly associated with a decline in the other (e.g., r < -0.7) (Fig. 2). The correlation analysis of fruit samples revealed distinct clustering patterns among different extraction groups, indicating variations in metabolite composition. The hierarchical clustering grouped samples into BMF, EF, and CF categories, each further divided into PD, D, and I subgroups. Strong positive correlations were observed between TPC, TFC, and CUPRAC, particularly in BMF_PD_O and EF_PD_O groups, suggesting a high antioxidant potential. Catechin, quercetin, and rutin clustered together, indicating their co-occurrence and possible contribution to antioxidant activity. In contrast, chlorogenic acid and p-coumaric acid formed a separate cluster, showing distinct distribution patterns across extraction groups. Negative correlations were observed between certain phenolic acids and antioxidant assays, particularly in CF_PD and CF_D subgroups, indicating variations in antioxidant response depending on extraction conditions (Fig. 3). The heatmap analysis for leaves samples revealed several important findings about phenolic compound distributions across different sample groups. CL_PD samples showed significantly higher concentrations of TPC, TFC, CUPRAC, and Catechin compared to other groups. EL_PD samples clustered together and exhibited elevated Rutin levels. Clear distinctions were observed between PD and D treatment groups, indicating treatment-specific effects on phenolic profiles. BML_PD samples displayed moderate to high levels of specific phenolics like ABTS and Chlorogenic acid. CL_D samples showed notable concentrations of p-coumaric acid and Epicatechin. EL_D samples generally contained lower concentrations of most compounds but had moderate Quercetin levels. The hierarchical clustering demonstrated that samples grouped primarily by their first designation (CL, BML, EL) and secondarily by treatment type (PD, D) (Fig. 4).

Close

Fig. 1.

Pearson correlation matrix of total phenolic content (TPC), total flavonoid content (TFC), antioxidant capacity (ABTS, CUPRAC), and individual phenolic compounds in black Mulberry (BMF), Chokeberry (CF), and Elderberry (EF) fruit samples. Cy3G; Cyanidin-3-O-glucoside, Cy3A; Cyanidin-3-O-arabinoside, Cy3Gal; Cyanidin-3-O-galactoside, P3G; Pelargonidin-3- O-glucoside, Cy3R; Cyanidin-3-O-rutinoside, K3R; Kaemferol-3-rutinoside.

Export to PPT

Close

Fig. 2.

Pearson correlation matrix of total phenolic content (TPC), total flavonoid content (TFC), antioxidant capacity (ABTS, CUPRAC), and individual phenolic compounds in black Mulberry, Chokeberry, and Elderberry leaf samples.

Export to PPT

Close

Fig. 3.

Heatmap representation of total phenolic content (TPC), total flavonoid content (TFC), antioxidant capacity (ABTS, CUPRAC), and individual phenolic compounds in black Mulberry (BMF), Chokeberry (CF), and Elderberry (EF) fruit samples. Cy3G; Cyanidin-3-O-glucoside, Cy3A; Cyanidin-3-O-arabinoside, Cy3Gal; Cyanidin-3-O-galactoside, P3G; Pelargonidin-3- O-glucoside, Cy3R; Cyanidin-3-O-rutinoside, K3R; Kaemferol-3-rutinoside BMF_PD_O; Black Mulberry Pre-digest Oral, BMF_PD_G; Black Mulberry Pre-digest Gastric, BMF_PD_I; Black Mulberry Pre-digest Intestinal, BMF_D_G; Black Mulberry Digest Gastric, BMF_D_I; Black Mulberry Digest Intestinal, BMF_D_O; Black Mulberry Digest Oral, EF_PD_O; Elderberry Pre-digest Oral, EF_PD_G; Elderberry Pre-digest Gastric, EF_PD_I; Elderberry Pre-digest Intestinal, EF_D_O; Elderberry Digest Oral, EF_D_G; Elderberry Digest Gastric, EF_D_I; Elderberry Digest Intestinal, CF_PD_O; Chokeberry Pre-digest Oral, CF_PD_G; Chokeberry Pre-digest Gastric, CF_PD_I; Chokeberry Pre-digest Intestinal, CF_D_O; Chokeberry Digest Oral, CF_D_I; Chokeberry Digest Intestinal, CF_D_G; Chokeberry Digest Gastric.

Export to PPT

Close

Fig. 4.

Heatmap representation of total phenolic content (TPC), total flavonoid content (TFC), antioxidant capacity (ABTS, CUPRAC), and individual phenolic compounds in black Mulberry (BMF), Chokeberry (CF), and Elderberry (EF) leaf samples. BML_PD_O; Black Mulberry Pre-digest Oral, BML_PD_G; Black Mulberry Pre-digest Gastric, BML_PD_I; Black Mulberry Pre-digest Intestinal, BML_D_G; Black Mulberry Digest Gastric, BMF_D_I; Black Mulberry Digest Intestinal, BML_D_O; Black Mulberry Digest Oral, EL_PD_O; Elderberry Pre-digest Oral, EL_PD_G; Elderberry Pre-digest Gastric, EL_PD_I; Elderberry Pre-digest Intestinal, EL_D_O; Elderberry Digest Oral, EL_D_G; Elderberry Digest Gastric, EL_D_I; Elderberry Digest Intestinal, CL_PD_O; Chokeberry Pre-digest Oral, CL_PD_G; Chokeberry Pre-digest Gastric, CL_PD_I; Chokeberry Pre-digest Intestinal, CL_D_O; Chokeberry Digest Oral, CL_D_I; Chokeberry Digest Intestinal, CL_D_G; Chokeberry Digest Gastric.

Export to PPT

4.1 Effects of digestion methods and gastrointestinal environments on phytochemical content and antioxidant capacity of fruit and leaves samples

The effectiveness of the in vitro digestion model employed in this study relies fundamentally on the enzymatic actions of α-amylase, pepsin, and lipase, which serve distinct yet interconnected roles in polyphenol bioavailability. α-amylase, responsible for starch hydrolysis, can influence polyphenol release through the breakdown of carbohydrate-polyphenol complexes, while pepsin’s proteolytic activity in the gastric environment facilitates the liberation of phenolic compounds bound to protein matrices. Lipase action in the intestinal phase not only hydrolyzes dietary fats but also affects the solubility and accessibility of lipophilic polyphenols. However, it is important to acknowledge that polyphenols can reciprocally inhibit these digestive enzymes, as demonstrated by studies showing that tea polyphenols can inhibit α-amylase, pepsin, and lipase activities by 61%, 32%, and 54%, respectively (Zhang et al., 2006). While the three-enzyme system provides a valuable framework for understanding polyphenol bioavailability, the complexity of human digestion extends far beyond these primary enzymes. The in vivo digestive process involves numerous additional factors including gastric and intestinal motility, varying residence times, microbial fermentation, absorption kinetics, and the presence of bile salts and other digestive secretions. Furthermore, individual variations in enzyme activity, pH fluctuations, and the influence of dietary matrices create a multifactorial environment that cannot be fully replicated in vitro. The limitations of static in vitro models have been extensively documented, with researchers emphasizing that while these models provide valuable preliminary data, they cannot fully capture the dynamic and adaptive nature of human digestion (Minekus et al., 2014; Brodkorb et al., 2019).

Our analysis of different digestion methods and gastrointestinal environments revealed notable differences in the extraction efficiency and stability of phenolic compounds and their antioxidant capacities in both black mulberry, chokeberry, and elderberry samples. Our findings for elderberry fruits demonstrated consistently higher TPC values compared to BMF, aligning with previous studies on Sambucus species (Mandrone et al., 2014; Mikulic-Petkovsek et al., 2016). The observed TPC values are comparable to those reported by Duymuş et al. (2014) and Mandrone et al. (2014), who documented TPC ranges in S. nigra extracts. The reduction in phenolic compounds after simulated digestion corresponds with findings by Olejnik et al. (2016), who reported similar degradation rates upon in vitro digestion of S. nigra berries. This degradation can be attributed to pH variations and interactions with dietary constituents during gastrointestinal digestion, as suggested by Gullón et al. (2015). The mechanistic basis for these observations lies in the structural vulnerability of phenolic compounds to enzymatic hydrolysis and pH-dependent transformations. The gastric environment, characterized by low pH and pepsin activity, initiates the breakdown of protein-polyphenol complexes, while the subsequent alkaline intestinal environment, rich in pancreatic enzymes including lipase, facilitates further matrix disruption. However, the alkaline conditions simultaneously promote the degradation of pH-sensitive phenolic compounds, creating a complex balance between release and degradation. Interestingly, the intestinal environment showed the highest TPC values, which contrasts with some previous reports indicating significant phenolic losses in intestinal conditions. This discrepancy might be explained by the findings of Liu et al. (2008) and Viuda-Martos et al. (2018), who suggested that digestion rather than processing induces the release of polyphenols from fiber matrices in berries. The higher CUPRAC values observed for BMF compared to EF suggest that the type of phenolics may be more important for antioxidant activities than the total phenolic amounts, consistent with observations by Mandrone et al. (2014) and Corrado et al. (2023) for Sambucus species.

In addition, the antioxidant capacity measurements revealed interesting trends, with pre-digest samples generally showing higher values than digest samples. This pattern aligns with findings by Huang et al. (2014) and Olejnik et al. (2016), who documented losses in antioxidant capacity after in vitro digestion of various berries. The decrease in antioxidant activity can be attributed to the structural modifications of polyphenols under alkaline conditions in the small intestine, particularly affecting anthocyanins as noted by Gullón et al. (2015). However, like observations by Pinto et al. (2017) for elderberries, our digested samples still maintained considerable antioxidant activity, suggesting potential health benefits even after gastrointestinal transformation. The varying antioxidant capacity across different assays (ABTS, CUPRAC) reflects the diverse mechanisms of action of different phenolic compounds present in chokeberry, elderberry, and black mulberry, consistent with comparative studies by Namiesnik et al. (2014) on various berry species. Leaf samples demonstrated distinct phytochemical profiles and antioxidant capacities compared to their corresponding fruits. The TPC values for chokeberry leaves (CL) ranged across different gastrointestinal sections, consistently higher than black mulberry leaves (BML) values. This trend of leaves having higher TPC than fruits aligns with findings reported in the Sambucus lanceolata study, where TPC and TFC, measured by colorimetric assays, showed an opposite trend compared to TPC (leaves > berries) (Mandrone et al., 2014; Corrado et al., 2023). The significant differences observed between pre-digest and digest methods for BML indicate that leaf phenolics may be more susceptible to degradation during gastrointestinal digestion than fruit phenolics. This phenomenon corresponds with observations from the S. lanceolata study, which noted that berry components were more affected than leaves, with a reduction of TPC upon simulated digestion (Gullón et al., 2015;2020). The varying susceptibility to digestion conditions may be attributed to differences in the structural composition of phenolics or their complexation with other leaf components such as proteins or fiber, as suggested by Gullón et al. (2015), who found that the increased concentration might be due to this compound being bound to proteins or fiber in the original matrix and, as a result of enzymatic digestion, it was released from these structures. Regarding antioxidant capacity, elderberry leaves (EL) demonstrated substantial activity across different assays. This robust antioxidant profile aligns with previous observations where leaves extract was the most active, regardless of the assayed method (Mandrone et al., 2014). The higher CUPRAC values for leaf samples compared to fruits suggest superior copper-reducing abilities, which may be related to specific phenolic compositions in leaf tissues. This finding is supported by the referenced literature stating that this trend is opposite to previous HPLC quantification, suggesting that in this case the type of phenolics seems more important for antioxidant activities than total phenolic amounts (Mandrone et al., 2014). The effects of gastrointestinal conditions on leaf antioxidant capacity were noticeable but less pronounced than for fruits, with CUPRAC values for EL remaining relatively stable. This relative stability might be attributed to the presence of more complex phenolic structures in leaves, as suggested by López-Alarcón and Denicola (2013). Importantly, despite the decrease in antioxidant activity after digestion, which was more notorious for leaves than for berries according to the literature, digested samples were still active towards targeted free-radicals (Gullón et al., 2015; Huang et al., 2014; Dou et al., 2022), indicating that leaf extracts maintain biological relevance even after gastrointestinal transformation. As noted in the referenced studies, although in vitro antioxidant assays do not provide an accurate estimation of the in vivo situation, they are highly useful for preliminary screening of the antioxidant potential of natural products (López-Alarcón and Denicola, 2013; Dou et al., 2022).

4.2 Anthocyanin stability and distribution in fruit samples during simulated gastrointestinal digestion

The overall mean results presented in Table 2 reveal critical insights into the general behavior of anthocyanins across different fruit matrices during digestion. This comprehensive analysis reveals that anthocyanin degradation follows consistent patterns regardless of fruit type, with the most pronounced losses occurring during the transition from acidic gastric to alkaline intestinal environments. The overall mean data particularly emphasizes the protective effect of acidic conditions compared to alkaline environments, supporting the established understanding that anthocyanin stability is fundamentally governed by pH-dependent equilibria. These findings suggest that despite matrix-specific protective effects observed in individual fruits, the fundamental chemical instability of anthocyanins in alkaline conditions represents a universal challenge for their bioavailability. The present our study of anthocyanin stability and bioaccessibility across black mulberry (BMF), chokeberry (CF), and elderberry (EF) samples reveals significant variations in the retention patterns throughout the simulated gastrointestinal digestion process. The substantial reduction in cyanidin-3-glucoside (Cy3G) content observed after digestion aligns with multiple studies documenting the pH-dependent instability of anthocyanins during gastrointestinal transit (Castañeda-Ovando et al., 2009; Huang et al., 2014; Olejnik et al. 2016). This degradation pattern follows the established understanding that anthocyanins undergo structural transformations in response to pH variations throughout the digestive tract. The observed degradation profile where Cy3G concentrations diminished progressively from oral to gastric to intestinal environments corresponds with findings by Olejnik et al. (2016), who reported similar degradation rates in elderberry (S. nigra) with approximately 88.4% anthocyanin loss upon in vitro digestion. As noted in the reference material, anthocyanins are very reactive compounds and susceptible to multiple factors such as temperature, light, pH, and enzymes/oxygen action (Huang et al., 2014). This progressive degradation through the gastrointestinal tract corroborates Castañeda-Ovando et al. (2009) explanation that anthocyanins exist in equilibrium between different molecular species depending on pH, the flavylium cation (predominantly at pH 1), quinoidal species (pH 2-4), and colorless carbinol pseudobase and chalcone forms (pH 5-6), with complete degradation at alkaline pH. The intestinal environment (pH 7.5-8.0) therefore represents the most challenging condition for anthocyanin stability, explaining the lowest recovery observed in the intestinal compartment.

The pronounced differences in anthocyanin concentrations between BMF and CF samples suggest significant matrix-dependent effects on anthocyanin stability. This observation is consistent with Liu et al. (2008), who noted that digestion, rather than thermomechanical treatment, primarily induces the release of polyphenols from plant matrices such as apple pomace. The higher anthocyanin concentrations in BMF compared to CF and EF fractions suggest a potentially protective effect of the black mulberry matrix. This finding aligns with the observation that chokeberry and blueberry pomace extracts exhibit distinct cytotoxic and antioxidative properties, influencing colorectal carcinoma cells through different signaling pathways (Stanca et al., 2024). Similarly, the reference document notes that when only mulberry puree was introduced into the gastrointestinal system, recovery of TPC was lower compared with those from the acidified and fermented samples, indicating that matrix components can significantly influence polyphenol stability. On the other hand, the varying degradation patterns observed among different anthocyanins across fruit samples highlight the importance of anthocyanin structure in determining digestive stability. This differential stability corroborates findings from the reference material that glycation seems to increase the stability of anthocyanins in neutral pH conditions as sugar molecules avoid the degradation of instable intermediaries into phenolic acid and aldehyde compounds (Castañeda-Ovando et al., 2009). The remarkably higher retention of certain anthocyanins in BMF compared to CF could be attributed to differences in fruit matrix composition, including potential interactions with dietary fiber, proteins, or other phenolic compounds. This parallels observations by Gullón et al. (2015), who noted that some phenolic compounds might be bound to proteins or fiber in the original matrix and released through enzymatic digestion. The present findings suggest extensive transformation of anthocyanins rather than complete destruction. This is consistent with studies showing that anthocyanins generate smaller phenolic compounds during digestion. As noted in the reference material, researchers have identified protocatechuic acid as a degradation product of anthocyanins formed by B/C ring cleavage of cyanidin-3-O-glucoside (Seeram et al., 2001a, b). Perez-Vicente et al. (2002) similarly proposed that the apparent low bioavailability of anthocyanins could be attributed to their transformation into smaller molecular phenolic compounds that may themselves be bioavailable and bioactive. The digestion process can lead to the formation of various phenolic acids, including protocatechuic acid, caffeic acid, ferulic acid, and tartaric acid derivatives, as observed in the reference study.

4.3 Bioactive compound stability and distribution in fruit samples during simulated gastrointestinal digestion

In our results, the observed stability and bioaccessibility patterns of bioactive compounds in fruit samples during simulated gastrointestinal digestion revealed significant differences among black mulberry (BMF), chokeberry (CF), and elderberry (EF) samples. Quercetin content displayed a notable increase during digestion, particularly in the intestinal phase for BMF, aligning with findings from Viuda-Martos et al. (2018) who reported enhanced bioaccessibility of flavonols in maqui berries after consecutive gastric and intestinal digestion. This suggests that the digestion process may facilitate the release of quercetin from the fruit matrix. However, the contrasting behavior of catechin, which decreased dramatically in EF while increasing in CF, indicates fruit-specific matrix effects on polyphenol stability, supporting Liu et al. (2008), who determined that digestion rather than processing induced the release of polyphenols in apple pomace. The observed decrease in epicatechin during digestion, particularly in CF, corresponds with findings from Olejnik et al. (2016) who reported significant degradation of flavonoids in elderberry during in vitro digestion, attributable to structural modifications under varying pH conditions. The stability of anthocyanins and other phenolic compounds during gastrointestinal digestion varied considerably among the fruit samples, with rutin showing a substantial increase in CF but a decrease in EF. This differential behavior aligns with Castañeda-Ovando et al. (2009), who explained that anthocyanins exist in various pH-dependent forms, with the flavylium cation predominating at acidic pH but transforming to colorless forms at higher pH. The remarkable stability of chlorogenic acid in CF during gastric digestion compared to BMF and EF supports the findings of Pinto et al. (2017), who reported an important increase in 5-O-caffeoylquinic acid in elderberries after in vitro digestion. Furthermore, the observed increase in p-coumaric acid during digestion in CF is consistent with Gullón et al. (2015), who attributed similar increases in phenolic acids to the release of bound compounds from fiber matrices during enzymatic digestion. These results collectively suggest that while the digestion process generally reduces total phenolic content, specific compounds may become more bioaccessible due to matrix disruption and structural transformation.

The bioactive compound stability in leaf samples during simulated gastrointestinal digestion demonstrated distinct patterns compared to their fruit counterparts. The observed differences in quercetin, catechin, chlorogenic acid, epicatechin, p-coumaric acid, and rutin content between pre-digest and digest stages indicate that leaf matrices significantly influence the behavior of phenolic compounds during digestion. These findings align with Topuzović et al. (2016), who reported higher total phenolic content in Sambucus leaves compared to berries, suggesting that leaf tissues may contain more digestible phenolic compounds. The varying stability of these compounds during oral, gastric, and intestinal phases supports Alminger et al. (2014), who emphasized that static in vitro digestion models, while useful for preliminary studies, may not fully replicate the complex dynamic processes occurring during human digestion. The bioaccessibility of phenolic compounds in leaf samples was generally higher than in fruit samples, particularly for rutin and catechin, which corresponds with findings from Namiesnik et al. (2014) who reported higher total flavonoid content in leaf extracts compared to various berry fruits. The observed reduction in antioxidant activity after digestion aligns with López-Alarcón & Denicola (2013), who noted that structural modifications of polyphenols under alkaline conditions in the small intestine significantly affect their bioactivities. Additionally, the differential stability of phenolic acids in leaf samples during digestion suggests that, as observed by Denev et al. (2010), the overall antioxidant capacity might provide more relevant biological information compared to individual compound analysis. These results underscore the importance of considering the complete phenolic profile of leaf extracts when assessing their potential health benefits, as the type of phenolics appears more influential for antioxidant activities than their total amounts.

4.4 General evaluation

The correlation analysis and hierarchical clustering of bioactive compounds in different berry samples revealed significant insights into their relationships and behaviors during the digestive process. In our results, the strong positive correlation observed between quercetin and p-coumaric acid (r > 0.80) after gastric digestion aligns with previous observations on phenolic compound stability in acidic environments. As noted by Pérez-Vicente et al. (2002), gastric conditions can promote the hydrolysis of polymerized polyphenols to monomers or aglycones due to acidic gastric juice and pepsin action. This could explain why certain phenolic compounds show enhanced correlation patterns specifically during the gastric phase (Fig. 1). The weak negative correlation between catechin and rutin (r < -0.30) suggests differential stability during digestion, which corresponds with the findings of Gullón et al. (2020) who reported that different classes of polyphenols exhibit varying susceptibility to digestive conditions. The observed moderate correlation between chlorogenic acid and epicatechin (r ≈ 0.50) further supports the matrix-dependent retention patterns discussed by Correa-Betanzo et al. (2014), who noted that digestion processes, rather than mechanical treatments, primarily drive the release and retention of polyphenols (Fig. 1). In the correlation analysis of leaf samples, a weak positive correlation indicated a slight tendency for one variable to increase as the other increased, but the relationship remained weak (e.g., r ≈ 0.1-0.3). In contrast, a strong positive correlation signified a robust association, where an increase in one variable was strongly linked to an increase in the other (e.g., r > 0.7). Similarly, a weak negative correlation suggested a minor inverse relationship, meaning that as one variable increased, the other tended to decrease slightly (e.g., r ≈ -0.1 to -0.3). On the other hand, a strong negative correlation represented a pronounced inverse relationship, where a rise in one variable was strongly associated with a decline in the other (e.g., r < -0.7) (Fig. 2).

The hierarchical clustering patterns revealing distinct groupings of samples into BMF, EF, and CF categories reflect significant matrix effects on bioactive compound stability. This observation is consistent with findings from Helal and Tagliazucchi (2018), who demonstrated that different food matrices can provide varying degrees of protection for phenolic compounds during gastrointestinal transit. The further subdivision into PD, D, and I subgroups highlights the impact of digestive phases on compound stability and interactions (Fig. 3). Particularly noteworthy is the strong positive correlation between TPC, TFC, and CUPRAC in BMF_PD_O and EF_PD_O groups, which suggests that these samples maintain high antioxidant potential even during digestion. This finding parallels observations from Noguer et al. (2008), who verified that many simple phenolic acids appear after gastric and intestinal digestion, extending the interpretation of in vitro antioxidant values for nutritional purposes. The clustering of catechin, quercetin, and rutin indicates their co-occurrence and potential synergistic contributions to antioxidant activity. Conversely, the separate clustering of chlorogenic acid and p-coumaric acid demonstrates distinct distribution patterns, which may reflect their differential binding to matrix components as suggested by Scalbert et al. (2002), who noted that bioaccessible phenolics can be present both in soluble free form and partially bound to other constituents like proteins or polysaccharides (Fig. 3). The clear distinctions observed between PD and D treatment groups in leaf samples highlight treatment-specific effects on phenolic profiles, supporting Alminger et al.’s (2014) assertion that digestion represents a complex process with significant impacts on phenolic compound structures and interactions. The higher concentrations of TPC, TFC, CUPRAC, and catechin in CL_PD samples compared to other groups suggest matrix-specific protective effects that may preserve these compounds during the initial stages of digestion (Fig. 4).