Inhibitory Activity of Red and Yellow Araçá Genotypes Towards Carbohydrate-Hydrolyzing Enzymes: Putative Role of Ellagitannins

: Psidium cattleianum Sabine (araçá) is a species native to Southeast Brazil that grows under abiotic stress conditions conferring high content of bioactive compounds to its fruits. The presence of these compounds is thought to be responsible for the many health-promoting effects including antioxidant, anti-inflammatory, anti-aging and antidiabetic activities. In this study, we evaluated the inhibitory potential of 10 (red and yellow) araçá genotypes towards carbohydrate-hydrolyzing enzymes (CHEs) using cell-free ( α -glucosidase, α -amylase) and cell-based assays (sucrase). Araçá extracts displayed stronger inhibition towards α -glucosidase than α -amylase, and only 3 inhibited sucrase activity. The high variability towards the in vitro inhibitory CHEs activity was reflected in the total phenolics content with values ranging between 38.9 and 117 mg/100 g. Of the thirty compounds identified by High-Performance Liquid Chromatography-Diode Array Detection-Electrospray Ionization-Tandem Mass Spectrometry (HPLC-DAD-ESI-MS/MS), including caffeic acids (9), organic acids (3) ellagitannins (15) and flavonoids (3), ellagitannins were the most abundant class. Statistical analysis showed ellagitannins were the main discriminators to the CHEs inhibitory activity. In summary, by expanding the panel of red and yellow araçá varieties studied, our results show that not all araçá genotypes inhibit CHE as only YA-23, RA-29, and RA-87 inhibited all 3 CHE which were related to the presence of ellagitannins. Information on the araçá genotypes with greater CHE inhibitory activity allied with the health-promoting effects of ellagitannin-rich foods, can be used to scale-up commercially exploitable genotypes with the aim to develop araçá-containing food formulations targeted to the pre-diabetic population.


Introduction
Climate changes and the increase of atmospheric carbon dioxide are changing temperature and rainfall patterns worldwide leading to an increase in the intensity, duration and frequency of sunshine and drought periods.In adaptation to these changes, plants adjust their development, including yield and quality of its fruits [1].
Araçá (Psidium cattleianum Sabine) from the Myrtaceae family is a species native to Brazil that grows under temperature and water stress conditions in a delimited area in the Southeast of the country.Due to the abiotic conditions in which araçá plants grow, their fruits have high content of bioactive compounds (e.g.carotenoids, anthocyanins and phenolic compounds) thought to be responsible for the many health-promoting effects including antioxidant, antiaging, anti-inflammatory and anti-diabetic properties [2].Due to its lower content of carbohydrates, when compared to widely consumed apples [2], araçá fruits are less caloric making it highly appealing to be included in the eating plans of diabetic patients.This is particularly relevant, considering the increasing prevalence of type-2 diabetes (T2DM) worldwide, but even more concerning the rising incidence of diabetes among children and young adults [3] likely to increase the economic burden of T2DM in the near future.
In addition to its lower carbohydrate content, studies have shown that the intake of Psidium guineensis Sw clarified juice prepared from commercial frozen pulp, had a positive effect on postprandial glycaemia in healthy individuals after the consumption of carbohydrates [4] thus supporting the in vivo anti-diabetic properties of araçá fruits.Other in vitro studies conducted on a small panel of red and yellow araçá extracts revealed that compounds present in extracts were able to inhibit carbohydrate hydrolyzing enzymes (CHE) such as α-glucosidase [5][6] and α-amylase [6].The results observed were partly attributed to the presence of anthocyanins and other non-flavonoid compounds [5][6] that prevented the degradation of starch by salivary and pancreatic α-amylase during the digestive process and subsequent liberation of glucose units by intestinal CHE and absorption by the organism [7].In spite of the strong association between the presence of bioactive phenolics and flavonoids and the inhibition of CHE observed in single red and yellow araçá fruits [5][6], work on a wider panel of 6 (red and yellow) araçá fruit genotypes over consecutive harvest seasons have shown that the phytochemical composition (e.g.ascorbic acid content, total anthocyanins, phenolics, tannins and total carotenoids) contributing to the antioxidant capacity was mostly influenced by the (red and yellow) genotypes, and less by harvest seasons [8].The variability in the phytochemical composition reported for the wider panel [8] may explain the subtle differences in total antioxidant activity and the cell-free α-glucosidase and α-amylase inhibitory activities (IC 50 ) observed when single red and yellow accessions were studied [6].Following the work by Vinholes and colleagues (2018), the authors found that simulated digested extracts of red araçá still retained antioxidant capacity and α-glucosidase inhibitory activity [9].Considering that during metabolization phenolic compounds lose the molecular traits responsible for their antioxidant capacity [10][11], the α-glucosidase inhibitory activity reported in red araçá digested extracts, and not found in digested extracts of other tropical butiá and pitanga (purple, red and orange) native fruits [9], was attributed to the release of ellagitannin compounds entangled in the carbohydrate matrix surviving the digestive process [9].While the link between araçá phenolic composition and the inhibition of CHEs was evidenced in studies using single red and yellow araçá fruits [6] comparative studies on a wider panel of araçá genotypes are still missing.This is relevant as targeting CHE by dietary inhibitors may be an effective strategy to delay the intestinal absorption of glucose, reduce the postprandial blood glucose levels and help manage the postprandial hyperglycemia whilst reducing the side effects caused by treatment with drugs such as acarbose.
In view of this, the main goal herein is to study the in vitro inhibition towards hydrolytic enzymes involved in sugar metabolism in an expanded panel of red (6) and yellow (4) araçá genotypes using cell-free and cell-based assays grown in the same geographical region thus minimizing the contribution of environmental conditions (soil, humidity, sun) on the araçá's phytochemical composition.Characterization of the phytochemical profile by liquid chromatography coupled to mass spectrometry detection (LC-MS) and liquid chromatography coupled to diode array detection (LC-DAD) was used to evaluate the impact of phenolic panel and its content on the inhibition of CHE, and statistical analysis was performed to identify the main discriminators on the inhibitory activity.
Extracts were prepared using the whole fruit.At least ten fruits from each sample were thawed at room temperature, sliced and homogenized (5 g) with 20 mL of ethanol (95%) for 5 min using an Ultra-Turrax homogenizer.The homogenates were filtered (Whatman no. 4, Darmstadt, Germany), evaporated under reduced pressure, re-dissolved in water (20 mL), freeze-dried (L101, Liobrás, São Carlos, SP, Brazil) and stored in a desiccator, protected from sunlight for a maximum of 4 weeks [9].

Inhibition of carbohydrate-hydrolyzing enzymes
The inhibitory potential of extracts for α-glucosidase activity was done as described earlier [9].An aliquot of 20 µL of fruit extract or 20 µL of water (control) was added to a vial containing 100 µL of PNP-G (3.25 mM) in PBS (pH 7.0).The reaction was initiated by the addition of 100 µL of enzyme (72 mU/mL in PBS, pH 7.0) and vials were incubated at 37 °C for 10 min.The reaction was stopped by adding 600 µL Na 2 CO 3 (1 M) and the absorbance at λ = 405 nm was measured by spectrophotometry (Genesys, Thermo, Brooklyn, NY, USA).IC 50 values were calculated using at least five concentrations (serial dilution) for each extract.The inhibition percentage (I%) for α-glucosidase assay was calculated using equation 1: Where A control is the absorbance of the control (water) reaction, and A sample is the absorbance of the extract tested in

Volume 5 Issue 2|2024| 303
Food Science and Engineering the reaction mixture.α-Amylase inhibition was determined using the method described elsewhere [12][13].Briefly, 60 µL of araçá extract or acarbose was pre-incubated for 5 min at 37 °C with 50 μL of α-amylase (6 U/mL) dissolved in PBS, pH 7.0 and 200 μL of the same buffer.Then, 250 µL starch solution (4 mg/mL in PBS, pH 7.0) was added as the substrate and the mixture incubated for 15 min at 37 °C.The reaction was stopped by adding 50 μL HCl (1 M).Following this, 100 μL of a mixed solution of I 2 and KI (both at 0.005 M) was added to the reaction mixture, and absorbance was measured at 690 nm by spectrophotometry (Genesys, Thermo, Brooklyn, NY, USA).IC 50 values were calculated using at least five concentrations (serial dilution) for each extract.The inhibition percentage (I%) for the α-amylase assay was calculated using equation 2: Where A control is the absorbance of 100% of enzyme activity (+ enzymeinhibitor); A sample is the sample (+ enzyme + inhibitor); A test1 is the absorbance of starch due to reducing sugar (-enzymeinhibitor), and A test2 is absorbance of the inhibitor and the starch (-enzyme + inhibitor).Acarbose was used as positive control in both assays.
The sucrase (α-glucosidase) inhibitory activity was determined by cell-based assay using Caco-2 cell intestinal model, which expresses sucrase and isomaltase [14].The method is based on the addition of sucrose to the apical side, which in contact with sucrase is hydrolyzed producing free glucose.The released glucose in the apical and basal sides of Caco-2 cells is used to calculate the efficiency of inhibition of carbohydrate digestion compared to controls (no added extract).
Caco-2 cell culture.Caco-2 cell line was provided by the Rio de Janeiro Cell Bank (Banco de Células do Rio de Janeiro-BCRJ) and were used between passage 40-46.Cells were routinely maintained in a humidified atmosphere of 5% CO 2 plus 95% air and grown in 20% RPMI medium containing 5.55 mM glucose supplemented with 20% fetal bovine serum (FBS, Gibco), 25 mM HEPES (N-2-hydroxyethylpiperazine-N-2-ethane sulfonic acid), 0.1% fungizone and 100 U/L penicillin/streptomycin. Culture medium was changed every 2-3 days and was split every 7 days.For sub-culturing, the cells were removed mechanically and sub-cultured in plastic culture dishes (21 cm 2 ; Ø 60 mm; Corning Costar, NY, USA) and were used between passage numbers 40 and 46.
Inhibition of sucrase.Cells were seeded (5 × 10 4 cells/cm 2 ), grown and differentiated in 6-well plates (Corning® Transwell® polyester membranes, pore size 0.4 μm, Ø:24 mm).After cells reached 100% confluency (5 days) cells were grown for 14-20 days more.The culture medium was removed, and apical and basal chambers were washed 3 times with PBS (2 mL).As araçá extracts of YA-23, RA-29 and RA-87 displayed no cytotoxicity (MTT assay) up to 24 h (Appendix Figure S1a-d) the medium in the apical chamber was replaced with a mixture of 200 µL of extracts of YA-23; RA-29; or RA-87 at 50, 100 and 150 µg/mL, as well as 28 mM sucrose solution in PBS (800 µL) as a substrate for sucrase inhibitory activity assay.Acarbose (50 and 150 µg/mL) was used as a positive control.In the basal chamber, 1 mL of PBS was added instead of the culture medium.The assay plate was incubated at 37 °C in 5% CO 2 atmosphere for 2 h.After incubation, the solution of the apical chamber was collected and dipped in boiling water (100 °C) for 1 min (enzyme inactivation).The liberated extracellular glucose concentration in the cell free solution was measured by the glucose oxidase method (test kit Megazime®) using a SpectraMax Plus plate-reader (Molecular Devices Corp, San Jose, CA, USA).

Statistical analysis
All experiments were done in triplicate (n = 3) and results were expressed as mean ± standard error of the mean (SEM).Dataset of 30 phenolic compounds estimated in 10 araçá samples was processed using Principal Component Analysis (PCA).Prior to analysis, variables were normalized to the total area for each sample.All calculations were made using Matlab R2023a (The MathWorks, Inc., USA).Statistical significance of the differences between two groups of araçá accessions was evaluated using Analysis of Variance-Simultaneous Component Analysis (ASCA) [16].The significance of the difference between araçá fruits was evaluated using a permutation test with 1,000 permutation.Percentage of the variance explained by the model was used as a quality-of-fit criterion [17].ASCA and permutation tests were implemented in MATLAB R2023a using the algorithms described earlier [16].Calibration model for prediction of the α-glucosidase inhibition using phenolic compounds concentrations was calculated using Partial Least Square regression (PLS) and validated using leave-one-out validation.Prior to calculation, decimal logarithm of the concentration estimates was calculated and dataset standardized.Ranking of the variables that contributed most to the model was done using Variable Importance in Prediction (VIP) [18].

α-Glucosidase and α-amylase inhibitory activity
The potential of red and yellow araçá fruit extracts for CHE inhibition was assessed using cell-free α-glucosidase and α-amylase assays.Figure 1 shows the dose-dependent inhibitory response of α-glucosidase (Figure 1a and b) and α-amylase enzymes (Figure 1c and d) to araçá extracts.The IC 50 values for α-glucosidase and α-amylase enzymes, calculated from curves in Figure 1, are shown in Table 1, with all samples showing IC 50 values lower than the positive control acarbose.Among all araçá genotypes, YA-23 outperformed the other yellow and all red araçá extracts showing 80% inhibition at 45 µg/mL (Figure 1a).The IC 50 values of α-glucosidase varied from 60.9 ± 3.20 to 153 ± 13.0 µg/ mL.These values were in general three to eight-fold higher than values previously reported [5][6].Studying ethanolic extracts, Vinholes and colleagues previously reported lower IC 50 values for red and yellow araçá fruits, with the yellow genotype displaying lower IC 50 values than the red genotype [5].In another study, methanolic extracts of red araçá fruits exhibited lower IC 50 values for the α-glucosidase than the yellow Bicudo accession [6].The IC 50 values estimated in this study for the edible fraction of red accessions RA-44 and RA-87 (Table 1) were higher than those reported earlier for red accessions AC-44 and AC-87 fruit parts [6].The difference between IC 50 values observed in this study and others previously reported may be multifold, namely to different agricultural araçá fruit harvests [5], different fruit harvesting/ collection strategies [5], different fruit sampling approaches as seeds were separated from pulp/peel [6] and different sample extraction protocols [5][6].
Regarding the α-amylase inhibition assays, the IC 50 values obtained for YA-23, RA-29, RA-9 and RA-87 extracts varied from 111 ± 10 to 149 ± 7.0 µg/mL (Table 1).These values were all higher than that of the positive control (acarbose) and in general 12 to 23-fold lower than that found for crude extracts of araçá pulp-peel [6].Of note, extracts of yellow YA-86, YA-102 and YACI and red fruits RA-19, RA-44 and RA-93 showed very low inhibition (Figure 1c  and d) and IC 50 values were not estimated.Overall, YA-23, RA-29, RA-9 and RA-87 araçá extracts showed higher inhibitory activity towards α-glucosidase than α-amylase (Table 1).This is in agreement with the work in crude pulp-peel araçá extracts [6] and in other fruits from Macaronesia region [19] suggesting that compounds in these extracts may act as CHE inhibitors suitable for diabetes management and treatment.Ideally, such compounds should possess high inhibitory activity towards α-glucosidase avoiding the release of glucose into circulation, and mild inhibitory activity towards α-amylase avoiding the accumulation of undigested starch, a source of gas in the intestine, causing gastrointestinal discomfort (diarrhea, abdominal pain and flatulence) [20].Even though the reasons behind the differences between IC 50 values observed in our study and others may be related to sample preparation approaches, i.e. analysis of the whole fruit or separation of seeds from pulp/peel [6], and extraction protocols [5][6], results obtained in 10 araçá genotypes highlight the high variability in the inhibition of CHE.

Sucrase inhibitory activity
In view of the strong inhibitory activity of YA-23, RA-29, RA-9 and RA-87 extracts in cell-free assays (Table 1) these were further studied with respect to sucrase inhibition by cell-based assays using Caco-2 cells.Caco-2 cells monolayer is a suitable and widely used model to investigate the absorption, transport, and metabolism of compounds due to its similarity to enterocytes as it expresses most of the morphological and functional characteristics of the small intestine such as α-glucosidase (sucrase and isomaltase) and lactase activities on the apical membrane [21].However, as previous studies had already shown red and yellow araçá extracts with anti-proliferative and cytotoxic effects [22][23] preliminary viability assays were carried out in YA-23, RA-29, RA-9 and RA-87 genotypes up to 72 h (Appendix).With the exception of RA-9 that displayed cytotoxicity at low concentrations (Appendix Figure S1) YA-23, RA-29, and RA-87 were screened for their sucrase inhibitory activity.
The inhibitory effect of YA-23, RA-87 and RA-29 extracts on sucrase activity measured in apical and basal sides of Caco-2 monolayers is shown in Figure 2. As shown, YA-23, RA-29 and RA-87 araçá extracts tested under non-cytotoxic concentrations inhibited sucrase activity and were even more effective in the inhibition of sucrase than acarbose, a widely prescribed drug with hypoglycemic effect.The yellow araçá genotype (YA-23) showed the highest inhibitory activity against the enzyme in the apical side of monolayer, reaching 67.6% inhibition at the lowest concentration tested (50 µg/mL) and 76.7% at the highest concentration (Figure 2a).Curiously, this extract was more effective than the purified extract of Eucommia ulmoides Oliv.leaves (IC 50 of 70 µg/mL) [24].RA-29 displayed values higher than 50% inhibition for all tested concentrations, while RA-87 displayed 50% inhibition only at higher concentrations (100 and 150 µg/mL, Figures 2b and 2c).All araçá extracts showed higher sucrase inhibition than the isolated feruloylated arabinoxylan mono-and oligosaccharides from corn bran and wheat aleurone (IC 50 of 1,030 and 1,280 µg/mL, respectively) [25].The inhibition of sucrase in the apical side of Caco-2 monolayer may result in decreased glucose transport across the monolayer [14,26] and can be responsible for reduced glucose levels in the basal side.

Characterization of composition in araçá extracts and statistical analysis
To further understand the distinct inhibitory activity displayed by both red and yellow araçá fruits, extracts were characterized by LC-based approaches.Identification of compounds was carried out by LC-ESI-MS based on the retention time (RT), UV absorption maxima and mass spectrometric data against authentic standards and data available in the literature for Psidium and other species.A representative LC-MS chromatogram of red and yellow araçá extracts is shown in Appendix Figure S2.Compounds identified (listed in Appendix Table S1) were numbered by their order of elution in reverse-phase column and further quantified by LC-DAD against calibration curves (as detailed in the Experimental section).Thirty compounds belonging to four classes were identified (as shown in Table 2) including caffeic acid derivatives, organic acids, ellagitannins and flavonoids.Of these, compounds 5, 7, 8-11, 13, 18-20, 25, 28-30 were common to all araçá samples, while caffeic acid derivatives (1-4, 6, 8-11) and vanillic acid di-hexoside ( 12) are reported for the first time in araçá.The total content of phenolic compounds estimated in araçá fruit extracts ranged from 38.9 to 117 mg/100 g of fresh weight (Table 2).Our findings are in agreement with the results reported by Teixeira and colleagues who investigated 6 genotypes over 6 harvest seasons reported that genotype had a stronger influence than the harvest season on the variability on the phytochemical composition of araçá fruits, and hence on the antioxidant capacity [8].Ellagitannins were the predominant class in both yellow and red araçá extracts ranging from 62% to 89% of total compounds followed by organic acids, caffeic acid derivatives and flavonoids (Table 2).Our results show that, in general, red genotypes contained higher amounts of ellagitannins than yellow genotypes, except for RA-19 that contained the lowest amount among all genotypes studied (24.1 mg/100 g).Even though the content of phenolic compounds found in this study is consistent with previous reports [27] and only slightly smaller than those reported for crude methanolic extracts of red and yellow araçá genotypes [6], expansion to a broader panel of yellow ( 4) and red (6) genotypes carried out for the first time showcases the variability of phenolic content that can occur among red and yellow araçá genotypes and overlooked in previous studies.Principal Component Analysis (PCA) allowed the discrimination of araçá genotypes (Figure 3) based on the phytochemical composition (Table 2).The first two principal components account for 43% of the data variance.Araçá extracts formed two separate clusters along the PC1 according to the color of the fruits except for red accession RA-19 (Figure 3a).Separation along PC1 axis of all yellow genotypes and red genotype RA-19 was attributed to the higher concentration of caffeic acid derivatives, organic acids, vanillic acid, some ellagitannins (18,19,21,24,26) and two flavonoids (28,30) in these extracts (Figure 3b).The remaining red genotypes were distinguished by their higher concentration of ellagitannins (13-17, 20, 22, 23 and 25) (Figure 3b).Sample distribution along PC2 axis was related to the differences in composition of individual accessions and cultivar independently of fruit color.As identified by PCA two groups of araçá fruits were considered: one comprising all yellow accessions and one red accession RA-19, and the other comprising red accessions.Statistical significance of the differences between the two groups of araçá accessions was evaluated using Analysis of Variance-Simultaneous Component Analysis (ASCA).According to the permutation test, the difference between the two groups of araçá accessions was significant (p-values of 0.009).Variables that contributed most to the difference between the two groups were citric acid (7), 4,5-O-dicafeoylquinic acid isomer (9), caffeic acid galloyl-hexoside (10), caffeic acid hexoside derivative (11) and vanillic acid di-hexoside (12) present at higher concentration in yellow accessions and RA-19 group, and hexahydroxydiphenoyl (HHDP)-tri-galloyl-Glc isomer (20) and Galloyl-di-HHDP-Glc isomer (23) present at higher concentration in red accessions group.Further calibration
Based on the results here presented, the inhibitory activity observed for YA-23, RA-29, RA-9 and RA-87 (Table 1) can be attributed to ellagitannins as these predominate in araçá extracts YA-23, RA-29, RA-9 and RA-87 (Table 2).Statistical analysis carried out on the extract's composition (Section 3.3.)identified several compounds contributing to α-glucosidase inhibitory activity in araçá fruits (and shown in Table S2) that include galloyl-and HHDP-containing compounds together with caffeic acid derivatives.The yellow genotype YA-23 contains high amounts of caffeic acid hexoside and Castalagin/Vescalagin, residual in the remaining yellow genotypes, which could account for the highest alpha-glucosidase inhibition among the yellow genotype (YA-23).However, the contribution of other compounds (not identified in this study) cannot be excluded nor the synergistic and additive effect of these to the overall alphaglucosidase inhibition.In fact, previous studies conducted with in vitro simulated digested extracts of pomegranate, another ellagitannin-rich food [28], had already assigned ellagitannins, namely castalagin/vescalagin isomers, also present in araçá fruit genotypes (Table 2), with α-glucosidase inhibition properties [29].The authors attributed the reminiscent high α-glucosidase inhibitory activity in in vitro digested extracts to the ability of ellagitannins to survive the digestive process [29].In another study, Camu-camu (Myrciaria dubia) seed extract with high content of castalagin (17) and vescalagin (18) displayed strong inhibition of α-glucosidase and α-amylase (above 82%) at low concentrations (1 µg/mL) [30].The strong inhibition of castalagin and vescalagin and other ellagitannins seems to be related to the presence of a hexahydroxydiphenoyl (HHDP) unit at the 4-O-and 6-O-glucose positions crucial for enzyme interaction and protein binding changing the enzyme conformation [29], as vescalin and castalin compounds lacking this unit in their structure were inactive towards CHE [31].Ellagitannins enzymatic inhibition mode of action is likely related to their protein binding properties (association and precipitation) that change the enzyme conformation and reduce the catalysis velocity and the accessibility to the active site of the substrate [29].Nevertheless, the contribution of other compounds, such as caffeic acid hexoside (1), citric acid (7), and caffeoyl quinic acids derivatives also present in araçá genotypes (Table 2), to the inhibition of CHE in araçá fruits cannot be excluded as these compounds have previously been associated with α-glucosidase and α-amylase inhibitory activity [26,32].Likewise, the contribution of other phenolic compounds such as caffeic acid hexoside (compound 1), identified in all genotypes (except for RA-19, Table 2) and reported as a main constituent of Geoffroea decorticans extract had previously shown high α-glucosidase inhibition [33].
For instance, caffeoyl quinic acids derivatives, including 4-O-caffeoylquinic and 4,5-O-dicaffeoylquinic acids (compounds 2 and 8, respectively) present in all genotypes (Table 2), were previously found to reduce the catalytic activity of the enzyme by establishing hydrophobic interactions thus altering its molecular conformation and, in this way, able to inhibit the α-glucosidase [32].In a similar manner, citric acid (compound 7), identified in all genotypes (Table 2), was also reported with the ability to inhibit the pancreatic porcine α-amylase enzyme activity with an IC 50 of 0.91 mM [26].
In this study, investigation on the inhibitory activity towards CHE on a wide panel of red and yellow araçá fruits genotypes shows that not all araçá fruits possess anti-diabetic properties.To this date, studies advocating to the potential anti-diabetic properties of araçá fruits [5][6]11] had mainly focused on individual red and yellow genotypes preventing from obtaining a comprehensive view on the variability of araçá fruits inhibitory activity towards CHE and the real antidiabetic properties.Our study shows that of the 6 red and 4 yellow araçá fruit genotypes screened, all red araçá extracts and one yellow exhibited strong inhibition towards α-glucosidase, where only 4 exhibited α-amylase inhibitory activity (Table 1) and just 3 (YA-23, RA-29, and RA-87) araçá genotypes exhibited inhibition towards sucrase activity (Figure 2).The high variability in the CHE inhibitory activity was reflected in the phenolic content.Chemical characterization by advanced LC-MS screening approaches showed that araçá extracts with higher inhibitory activity towards CHE were particularly rich in ellagitannins (Table 2).Statistical analysis identified galloyl and HHDP-containing compounds together with caffeic acid derivatives as key contributors to α-glucosidase inhibitory activity in araçá fruits (Appendix Table S2).The yellow genotype YA-23 contains high amounts of caffeic acid hexoside and Castalagin/Vescalagin and residual in the remaining yellow genotypes, which could account for the highest alpha-glucosidase inhibition among the yellow genotypes.This is in accordance with recent reported findings conducted on purified galloyl-based polyphenols with free and unfree galloyl moieties where the authors found that both free galloyl and intramolecularly-linked galloyl (HHDP) groups contributed to α-glucosidase enzyme inhibition [34].
Based on the findings from this study, our results suggest that, allied with its low sugar content, the inclusion and ingestion of certain araçá fruits rich in ellagitannins in the eating habits of (pre)diabetic patients may turn out to be a valuable nutritional strategy to help manage hyperglycemia status and keeping the postprandial glucose levels low in (pre)diabetic patients.While the link between araçá phenolic composition (and antioxidant capacity) with the inhibition of CHE was evidenced in previous studies, though most of the research has been devoted to low molecular weight compounds such as Vit C, Vit E, carotenoids, phenolic acids and flavonoids.To date, very little interest has been devoted to unveil the role of high molecular weight compounds on the inhibition of CHE.However, the innovative findings described by Vinholes and colleagues were pivotal showing that after in vitro digestion, araçá fruits but not pitanga and butiá, still retained antioxidant capacity and the ability to inhibit alpha-glucosidase activity [9].This was attributed to the high content of ellagitannins (identified by advanced LC-MS) which entangled in the carbohydrate matrix of araçá survived the digestive process, unlike low molecular weight flavonoids are readily metabolized with a marked decrease in their antioxidant capacity [10][11].Upon reaching the intestine, ellagitannins containing the HHDP moiety are metabolized by gut microbiota with formation of urolithins as already shown by ex vivo studies on individual ellagitannins [35].The ability of ellagitannin-rich araçá fruits to inhibit CHE as shown in this study is in agreement with others on pomegranate another ellagitannin-rich fruit [29].Even though the nutraceutical effect of each of the ellagitannins is not yet fully understood, they occur as mixtures in fruits suggesting that ellagitannins may exert their health benefits through a combined and synergistic effect able to ameliorate metabolic disease [36][37].
The health-promoting effects of ellagitannin-rich foods such as araçá fruits goes beyond the obvious glucose lowering (anti-hyperglycemic) effect attributed to the inhibition of CHE as urolithins, the intestinal metabolites of ellagitannins, not only have high antioxidant capacity [28] and hypertensive properties [37] but have been shown to bind to estrogen receptors and thus key players in the gut-brain-endocrine interactome and valuable nutritional alternatives in most hormone/endocrine-dependent diseases (cardiovascular disorders, osteoporosis, muscle health, neurological disorders, and cancers of breast, endometrium, and prostate) [38].In addition to this, the ingestion of ellagitannin-rich foods may be associated with a more diverse microbiota ecology and improved gut functionality as (ellagi)tannins have been found to modulate the growth of Gram(+) and Gram(-) bacteria [39][40] thus ultimately shaping the profile of microbiota-derived end products in circulation.In fact, micromolar levels of other (poly)phenol microbial metabolites have been shown to reduce the release of pro-inflammatory cytokines in endothelial cell cultures grown under normo [41] and even under hyperglycemic conditions [42].
The CHE inhibiting potential displayed by YA-23, RA-29 and RA-87 araçá genotypes, contributes to the valorization of native tropical fruits which are currently farmed locally in family orchards [43].This allied with its resistance to diseases and pests, and its ability to adapt and grow under abiotic conditions, makes the YA-23, RA-29, and RA-87 araçá fruits highly desirable raw material in the development of functional foods [44].Even though ellagitannins were shown to survive the digestive process, the microencapsulation of araçá pulp to maintain the fruits sensorial characteristics and to increase the bioavailability of bioactive compounds are surely worth exploring not just by Food (Bio)technology industry but also by pharmaceutical companies.

Conclusions
In the present study, a large panel of red and yellow araçá fruit extracts was investigated towards the inhibition of carbohydrate-hydrolyzing (CHE) enzymes.Of 10 araçá genotypes studied, YA-23, RA-9, RA-87 and RA-29 showed the lowest IC 50 values, and only YA-23, RA-29 and RA-87 displayed strong inhibition towards sucrase activity.By expanding the panel of red and yellow araçá genotypes to include a wider revealed that not all araçá genotypes are able to inhibit CHE.Inhibition values were partially attributed to high content of ellagitannins.Based on the LC-DAD-ESI-MS/MS data, 10 compounds were described for the first time in araçá extracts with ellagitannins as the most

Volume 5 Issue 2|2024| 313
Food Science and Engineering abundant class.Improved knowledge on which araçá genotypes display stronger inhibition of CHE and thus higher antihyperglycemic effect provide novel information on the potential genotypes to be commercially exploited as raw material in the development of functional foods (e.g.juices, nectars, yoghurts and ice creams) for the management of T2DM onset in pre-diabetic population.In parallel, given the great adaptability to the growing climate changes of native araçá fruit and being typically cultivated on a small-scale, valorisation of specific araçá genotypes with the best performance to the inhibition of CHE, will surely contribute to the sustainability of local and regional economy.
where A control was measure of the formazan formed in negative control cells and A sample was the measure of the formazan formed after extracts exposure.

Figure S2 .
Figure S2.Representative LC-MS chromatogram of araçá accession YA-23.Compounds were numbered as identified Table 1.(a) LC-MS chromatogram showing retention time from 2 to 6.9 minutes and (b) retention from 7 to 34 minutes

Table 1 .
IC 50 values (µg/mL) determined for α-glucosidase and α-amylase inhibition of araçá (yellow and red genotypes) ethanolic extracts and positive controls.Results are expressed as mean ± standard error mean (SEM) of triplicate experiments (n = 3)

Table 2 .
Content of phenolic compounds (mg/100 g fresh weight) in ethanolic extracts of yellow and red araçá genotypes.Results are expressed as mean values of triplicates (n = 3).

Table S1 .
Phenolic compounds identified in red and yellow araçá extracts.Compounds were assigned based on their retention time (RT), UV and mass spectrometric data from authentic standards and data available in literature for Psidium and other species.Compounds were numbered according to their order of elution