An Arylbenzofuran, Stilbene Dimers, and Prenylated Diels–Alder Adducts as Potent Diabetic Inhibitors from Morus bombycis Leaves

Morus bombycis has a long history of usage as a treatment for metabolic diseases, especially, diabetes mellitus (DM). Thus, we aimed to isolate and evaluate bioactive constituents derived from M. bombycis leaves for the treatment of DM. According to bioassay-guided isolation by column chromatography, eight compounds were obtained from M. bombycis leaves: two phenolic compounds, p-coumaric acid (1) and chlorogenic acid methyl ester (2), one stilbene, oxyresveratrol (3), two stilbene dimers, macrourin B (4) and austrafuran C (6), one 2-arylbenzofuran, moracin M (5), and two Diels–Alder type adducts, mulberrofuran F (7) and chalcomoracin (8). Among the eight isolated compounds, the anti-DM activity of 3–8 (which possess chemotaxonomic significance in Morus species) was evaluated by inhibition of α-glucosidase, protein tyrosine phosphatase 1B (PTP1B), human recombinant aldose reductase (HRAR), and advanced glycation end-product (AGE) formation as well as by scavenging peroxynitrite (ONOO−), which are crucial therapeutic targets of DM and its complications. Compounds 4 and 6–8 significantly inhibited α-glucosidase, PTP1B, and HRAR enzymes with mixed-type and non-competitive-type inhibition modes. Furthermore, the four compounds had low negative binding energies in both enzymes according to molecular docking simulation, and compounds 3–8 exhibited strong antioxidant capacity by inhibiting AGE formation and ONOO− scavenging. Overall results suggested that the most active stilbene-dimer-type compounds (4 and 6) along with Diels–Alder type adducts (7 and 8) could be promising therapeutic and preventive resources against DM and have the potential to be used as antioxidants, anti-diabetic agents, and anti-diabetic complication agents.


Introduction
Diabetes mellitus (DM) is one of the most serious health problems worldwide. According to the Diabetes Federation's Global Diabetes Overview, there were 463 million people aged 20 to 79 years with diabetes in 2019, and this number is expected to continue to rise, reaching 700.2 million in 2045 [1]. Therefore, DM could cause hundreds of millions of individuals to experience serious health problems around the world in the future. DM, which is broadly divided into type 1 and type 2 DM (T2DM), is a metabolic disease caused by defects in insulin secretion and action [2]. In particular, dysfunction of α-glucosidase and protein tyrosine phosphatase 1B (PTP1B) was the main mechanism associated with T2DM [3]. The final stage of carbohydrate metabolism involves enzymatic breakdown into monosaccharides by α-glucosidase at the brush boundary of small intestine cells, and glucose uptake causes an increase in blood glucose [4]. Inhibiting carbohydrate digestion to have bioactive compounds inhibiting β-secretase, tyrosinase, and cholinesterase. In addition, this plant is reported to harbor prenylated flavonoids (morusin, morusinol, and flavones), flavones (norartocarpetin and kuwanon C), and phenolic compounds (mulberrosides A and C), exhibiting these bioactivities [29,30]. Morus bombycis, called San-sang in Korean, is a wild-type plant from the mountains. Although its appearance is similar to that of M. alba, its fruit is smaller, and its pistil is divided into two parts, the stigma and ovary. Morus bombycis is reported to exhibit anti-inflammatory [31], anti-diabetic [32,33], antiobesity [34], skin-whitening [35], and anti-Alzheimer's disease effects [36]. Phytochemical studies on M. bombycis have demonstrated the presence of moracinoside M, mulberrofuran K, kuwanon V, oxyresveratrol, such as Diels-Alder type adducts, moracin glycoside derivatives, flavone, flavonoid glycoside derivatives, and chalcone derivatives [31,37]. In particular, 1-deoxynojirimycin (1-DNJ) and N-methyl-1-DNJ have been found in the leaves of M. bombycis, and they have been shown to have a strong anti-diabetic effect [38].
Although Morus species were traditionally used as medicines due to their pharmacological properties, there are limited studies on the bioactivity of M. bombycis leaves and their pharmacological compounds. Thus, a more detailed physiological action and phytochemical analysis of M. bombycis were performed in this study. In particular, two stilbene dimers and two Diels-Alder type adducts were isolated from M. bombycis leaves at first along with four compounds. We investigated the anti-diabetic and antioxidant properties of M. bombycis and its major constituents as part of our ongoing efforts to identify potent inhibitors against PTP1B, α-glucosidase, AGEs formation, and antioxidant agents from natural sources. The mode of inhibition or molecular interactions of active compounds with corresponding enzymes such as PTP1B and α-glucosidase were investigated. Furthermore, various in vitro anti-diabetic complication assays were used to evaluate the inhibitory effects of active compounds on AGE formation and HRAR. Overall, we sought to substantiate the anti-diabetic, anti-diabetic complications, and antioxidant effects of the compounds isolated from M. bombycis leaves.

Plant Material
The leaves of M. bombycis were collected at Jeju in September 2021 and purchased from JEJU SAN YA CHO (Jeju, Korea). A voucher specimen as leaves is registered (MB202109002) and deposited at the Department of Food Science and Human Nutrition, Jeonbuk National University, Jeonju, South Korea (Professor H. A. Jung).

UPLC-QToF/ESI-MS Analysis
UPLC-QToF/ESI-MS analysis was performed to identify and quantify components from the methanol extract and its organic solvent fractions from M. bombycis leaves. According to the operating protocol [47], LC chromatogram and mass spectra were simultaneously measured. Briefly, 1 µL of the sample (2 µg/µL) was injected into column (30 °C) and run for 40 min at a flow rate of 0.35 mL/min. Solvent system consists of mobile phase A (0.5% formic acid in water) and B (0.5% formic acid in acetonitrile), eluting gradient condition. Mass spectra were operated within the range of m/z 50-800 in positive ionized mode using a positive ESI probe, and their parameters were capillary voltage 3.5 kV, sampling cone voltage 40 V, source temperature 120 °C, desolvation temperature 400 °C, and desolvation N2 gas flow 1000 L/h.

UPLC-QToF/ESI-MS Analysis
UPLC-QToF/ESI-MS analysis was performed to identify and quantify components from the methanol extract and its organic solvent fractions from M. bombycis leaves. According to the operating protocol [47], LC chromatogram and mass spectra were simultaneously measured. Briefly, 1 µL of the sample (2 µg/µL) was injected into column (30 • C) and run for 40 min at a flow rate of 0.35 mL/min. Solvent system consists of mobile phase A (0.5% formic acid in water) and B (0.5% formic acid in acetonitrile), eluting gradient condition. Mass spectra were operated within the range of m/z 50-800 in positive ionized mode using a positive ESI probe, and their parameters were capillary voltage 3.5 kV, sampling cone voltage 40 V, source temperature 120 • C, desolvation temperature 400 • C, and desolvation N 2 gas flow 1000 L/h.

Determination of Total Phenolic Content (TPC) and Total Flavonoids Content (TFC)
TPC and TFC measurements of the extract and each fraction obtained from M. bombycis leaves were conducted according to previous literature with some modifications [48].

Assay for Scavenging Activity against ABTS Radical and DPPH Radical
The ABTS and DPPH radical scavenging activity of the extract and each fraction obtained from M. bombycis leaves were measured according to previous literature with modifications [48].

In Vitro Assay for ONOO − Scavenging Activity
ONOO − scavenging activity was measured using the method in previous literature involving measuring highly fluorescent rhodamine 123 that is converted from non-fluorescent DHR123 in the presence of ONOO − [49].
2.9. In Vitro Assay for Inhibitory Activity of α-Glucosidase and PTP1B Enzyme The enzyme inhibition study was executed spectrophotometrically following the previous literature [49]. Acarbose and ursolic acid were used as the positive controls for α-glucosidase and PTP1B, respectively.

Kinetic Parameters of Isolated Compounds for Inhibition of α-Glucosidase and PTP1B Using Lineweaver-Burk and Dixon Plots
The two kinetic methods, Lineweaver-Burk plots and Dixon plots, were used to determine the inhibition mechanism [49][50][51][52]. The α-glucosidase inhibition type was measured at various concentrations of substrate (pNPG, 0.625, 1.25, and 2.5 mM) and several con-

In Silico Molecular Docking Analysis for α-Glucosidase and PTP1B Inhibition
Before the docking analysis to investigate the binding poses of compounds inside the active receptor pockets, the crystal protein structures for PTP1B (PDB ID: 1NNY for the catalytic site; 1T49 for the allosteric site) and α-glucosidase (PDB ID: 3A4A) were downloaded from the Protein Data Bank (PDB) [53]. These protein structures were confirmed using X-ray diffraction. The reported heteroatom compounds and water molecules were removed, and the protein was regarded as ligand-free for the docking simulation using Accelrys Discovery Studio 19.1 (http://www.accelrys.com, accessed on 1 January 2023; Accelrys Inc., San Diego, CA, USA). Polar hydrogen atoms were added to the protein using an automated docking tool, AutoDock 4.2.6. [54]. The docking studies for macrourin B (4), austrafuran C (6), chalcomoracin (7), mulberrofuran F (8), acarbose, and co-crystalline ligands were performed without modifying the default parameters. The 2D structures of all the compounds were drawn with MarvinSketch (www.chemaxon.com, accessed date 1 January 2023; Chemaxon, Life Science, Informatics, Cheminformatics, Budapest, Hungary); Chemaxon, Budapest, Hungary). Energy minimization of each ligand was carried out using the molecular mechanics 2 (MM2) force field, and the docking analysis was conducted using AutoDock Vina [55]. A grid box size of 60 × 60 × 60 points with a spacing of 1.0 Å between the grid points was executed to cover almost all the favorable protein-binding sites. The X, Y, Z centers were PTP1B (56.019, 31.36, and 22.48), and α-glucosidase (21.28, −0.75, and 18.63). In the docking studies, the selected ligands (all compounds) were examined to find qualified binding poses for each compound. The binding aspects of the PTP1B and α-glucosidase residues and their corresponding binding affinity scores are regarded as the best molecular interactions.

In Vitro Assay for Inhibitory Activity of HRAR and AGEs Formation
The inhibitory activity of HRAR was examined according to previous literature with modifications [56]. First, 150 µL of 100 mM sodium phosphate buffer (pH 6.2), 20 µL of 0.3 mM NADPH as the co-enzyme, 5 µL of the test samples (50, 10, and 2 mg/mL or 100% DMSO), and 20 µL of 10 mM DL-glyceraldehyde as the substrate were added to each of the 96 wells (final volume 200 µL). Quercetin was used as a positive control. The inhibitory activity of AGEs formation was examined according to the modified method [49]. Aminoguanidine hydrochloride was used as a positive control for the AGEs formation inhibition assay.

Statistics
All results are expressed as the mean ± SD of triplicate experiments. Statistically significant differences were determined by analysis of variance (ANOVA) and Duncan's test (Systat Inc., Evanston, IL, USA). A p-value < 0.05 was considered statistically significant.

Preliminary Experiment of Three Dominant Morus Species
In the preliminary experiments of three dominant and widely cultivated species, including M. alba, M. lhou, and M. bombycis, the MeOH extract of the last species exhibited α-glucosidase inhibitory activity and a higher TPC value, while the MeOH extracts of the first two species showed good antioxidant capacity and higher content in total flavonoids (Table 1). Therefore, the leaves of M. bombycis were selected as promising candidates for anti-diabetic therapy, and further research on the evaluation of anti-diabetic activity and phytochemical analysis was performed.

Antioxidant and Anti-Diabetic Activities of the Leaves of Morus bombycis
To evaluate antioxidant activity, the MeOH extract of three species and its organic solvent fractions from M. bombycis were tested via DPPH and ABTS radicals ( Table 2). Among its organic solvent fractions, the EtOAc fraction showed the strongest scavenging activities against ABTS and DPPH. In vitro inhibitory activity assays by α-glucosidase and PTP1B were performed to evaluate the anti-diabetic effect of MeOH extract and four organic solvent fractions of M. bombycis leaves and isolated compounds. As given in Table 2, the MeOH extract and its four organic solvent fractions showed significant α-glucosidase inhibitory activities, compared to acarbose as a positive control. The EtOAc fraction, which showed significant α-glucosidase, exhibited good PTP1B inhibitory activity, compared to ursolic acid, although the CH 2 Cl 2 fraction showed stronger inhibitory activity. According to the results of antioxidant and anti-diabetic activities, the EtOAc fraction was selected as a potent candidate, and further phytochemical isolation experiments were performed.

Evaluation of Bioactivities of Compounds Derived from the Leaves of Morus bombycis
3.2.1. Antioxidant, Anti-Diabetic, and Anti-Diabetic Complication Activities of Compounds As given in Table 3, tested compounds exhibited significant ONOO − scavenging activity, with IC 50 values ranging from 0.92 to 8.64 µM. In particular, compound 5 showed strong ONOO − scavenging activity, compared to L-penicillamine as a positive control. Interestingly, tested compounds exhibited a significant α-glucosidase inhibitory effect, compared to acarbose as a positive control: Compound 4 showed the highest α-glucosidase inhibitory effect, followed by compounds 8, 6, 7, 3, and 5. As for anti-diabetic activity by evaluation of the tested compounds on PTP1B inhibitory activities, compound 6 showed the highest inhibitory activity, followed by compounds 8, 4, and 7. Compounds 4, 6, and 8 showed stronger inhibitory activity compared to ursolic acid, a positive control. In order to evaluate anti-diabetic complication activity, inhibitory activities of the tested compounds against BSA-AGEs formation and HRAR were determined. As given in Table 4, the test compounds except for compounds 7 and 8 demonstrated strong inhibitory activity, when compared to the positive control. In the case of HRAR inhibitory activity, compound 4 showed strong inhibitory activity, followed by compound 6. These can be compared to quercetin as a positive control with an IC 50 value of 16.67 µM. With regard to the above results, compounds 4, 6, 7, and 8 might be promising candidates for anti-diabetic and anti-diabetic complication remedies, and further investigation was accomplished.

Enzyme Kinetic Study of Isolated Compounds Derived from Morus bombycis Leaves
Enzyme kinetic analysis was performed with different concentrations of substrate (pNPG and pNPP) and various concentrations of compounds to determine the type of inhibition on the compounds. Lineweaver-Burk and Dixon plots were used to determine the type of inhibition in enzyme kinetics. Each line of inhibitors intersected at the xy-side, indicating mixed-type inhibitors. On the other hand, the lines penetrated the same point on the x-intercept, representing non-competitive inhibitors in Lineweaver-Burk plots, and the Dixon plot was also used to calculate the K i value for the enzyme inhibitor complex with the value shown on the x-axis indicating the -K i value [50][51][52]. Figure 3 depicts the enzyme kinetic analysis for α-glucosidase inhibition of each compound (4 and 6-8), with A representing the Lineweaver-Burk plot and B representing the Dixon plot. As displayed in Table 3 and Figure 3, compounds 4 and 6-8 exhibited mixed-type inhibition against α-glucosidase with respective K i values of 0.19, 0.75, 1.71, and 1.84. In the enzyme kinetic analysis for PTP1B inhibition (Table 3 and  enzyme kinetic analysis for α-glucosidase inhibition of each compound (4 and 6-8), with A representing the Lineweaver-Burk plot and B representing the Dixon plot. As displayed in Table 3 and Figure 3, compounds 4 and 6-8 exhibited mixed-type inhibition against αglucosidase with respective Ki values of 0.19, 0.75, 1.71, and 1.84. In the enzyme kinetic analysis for PTP1B inhibition (Table 3 and Figure 4), compounds 4, 6, and 7 represented mixed-type inhibition with Ki values of 1.54, 1.45, and 8.90, respectively, while compound 8 exhibited non-competitive-type inhibition with Ki values of 4.41.

Docking Interaction between Compounds and Key Binding Ligands of α-Glucosidase
Since compounds 4 and 6-8 exhibited significant inhibitory activities against α-glucosidase and PTP1B (which play important enzymes in therapeutic strategy against DM), four candidates were subjected to a molecular docking analysis. All the docked active compounds overlapped within the α-glucosidase (PDB: 3A4A) pocket sites, and α-D-glucose was used as a co-crystalline ligand for α-glucosidase. Using AutoDock Vina, the ligand-enzyme complexes of the four test compounds, acarbose and α-D-glucose, were stably posed in the catalytic pocket of α-glucosidase ( Figure 5A−D). Hydrogen bonds, hydrophobic interactions, and electrostatic interactions were used to calculate the binding energies of test compounds. The predicted binding energies and binding residues are pro-

Discussion
About 95% of people with DM have T2DM, which is caused by the inefficient use of insulin in the body. Insufficient insulin production and insulin resistance are the causes of T2DM, and these affect the control of the metabolism of proteins, lipids, and carbohydrates. Deterioration of insulin-producing pancreatic ß-cells and insulin resistance present in diverse organs contribute to microvascular and macrovascular problems [1,2]. Consequently, chronic and accelerated hyperglycemia cause cardiovascular disease, coronaropathy, and other problems, especially diabetic retinopathy, and diabetic foot [9]. Moreover, free radicals and ROS are produced by living things as part of regular physiological and biochemical processes, and their overproduction can lead to oxidative damage to biomolecules (such as lipids, proteins, and DNA) and many chronic diseases in people, including DM, Alzheimer's disease, cardiovascular disease, and chronic inflammation [16]. Since the enzymes PTP1B, α-glucosidase, and AR (as well as non-enzymatic glycation products known as AGEs) play critical roles in T2DM, much research has been conducted to develop therapeutic inhibitors. Unfortunately, clinical trials using enzyme inhibitors, which are important targets of T2DM mechanisms, have recently failed to produce effective therapy agents [58]. For example, aminoguanidine has been used to inhibit DM complications, it has adverse effects on the heart and lungs, and may cause histamine buildup in the system [59]. Therefore, therapeutic agents isolated from natural products that are utilized in conventional medicine or functional foods may be effective treatments for DM.
Several active compounds and the EtOAc fraction derived from M. bombycis leaves against DM were found, and comparisons were made on the antioxidant and anti-diabetic effects among three Morus species (e.g., M. alba and M. lhou). The antioxidant and antidiabetic effects of M. bombycis extract showed significantly higher inhibitory activity than those of the two Morus species, such as M. alba and M. lhou (Table 1). In the phytochemical content evaluation of the leaves of M. bombycis, the EtOAc fraction indicated the highest value in TPC and TFC; potent inhibitory activities against both α-glucosidase and PTP1B (Table 2). Repeated column chromatography of potent bioactive EtOAc fractions led to the isolation of compounds 1-8, and we further evaluated their bioactivities. Overall, the goal of this study was to quantitatively analyze the EtOAc fraction from M. bombycis leaves and evaluate its antioxidant, anti-diabetic, and anti-diabetic complication effects.
Although there have been many studies on compounds isolated from M. bombycis leaves, compound 6 has not been investigated by bioactivity screening, and little research has been conducted on a 2-arylbenzofuran-type compound (4), a stilbene-dimer-type compound (6), and the Diels-Alder type adducts (7,8), which are known to have antioxidant and anti-diabetic properties. Compounds 4-6 and 8 exhibited potent scavenging activities against ONOO − . All test compounds demonstrated significant α-glucosidase inhibitory activity when compared to a positive control, acarbose. Compounds 4 and 6-8 were strong PTP1B inhibitors by compared to a positive control, ursolic acid. The inhibitory activities of AGE formation were tested to confirm the anti-diabetic complications effect. All the compounds except for compound 7 demonstrated significant inhibitory effects, and compounds 3 and 4 exhibited extremely potent inhibitory activities against the formation of AGEs compared to the positive control. Moreover, compounds 4 and 6 exhibited significant inhibitory activities against HRAR. Compounds 3 and 5 were recently reported to have anti-diabetic and antioxidant activities by inhibiting α-glucosidase, PTP1B, AGEs, and ONOO − [49], which is consistent with our current studies.
Since compounds 4 and 6-8 have the potential to be effective α-glucosidase and PTP1B inhibitors, we focused on the anti-DM effects of four key compounds by performing enzyme kinetic studies and molecular docking simulations. Studying the impact of the inhibitory concentration on enzyme kinetics is crucial to comprehending the mechanism of inhibitor-mediated enzyme inhibition. The inhibitors have an affinity for the enzyme binding site in which greater affinities are indicated by lower values of K i . All the test compounds used in the enzyme kinetic study against α-glucosidase were mixed-type inhibitors. Mixed-type compounds may be posed at active and/or allosteric sites, while non-competitive compound 8 could be posed at the allosteric site on PTP1B.
Molecular docking is a way to determine how a ligand will fit into a protein's binding site. A scoring function is used to determine the binding energy values for each structure to predict the activity of the bound ligand [65]. In silico docking simulation studies on test compounds were performed to demonstrate the α-glucosidase and PTP1B inhibition mechanisms of the potent bioactive compounds (Figures 5-8). The docking scores of the binding energies on each enzyme were estimated and are listed in Tables 5 and 6. The results of the docking simulation against α-glucosidase confirmed that all test compounds had high affinity and lower binding energies within the enzyme catalytic site compared to acarbose ( Figure 5 and Table 5). Previous research has supported the existence of catalytic residues on α-glucosidase, such as Asp215, Glu277, Asp352, His112, Asp242, Gln279, and His280, which play an important role in inhibiting enzyme activation [26]. Among them, the Diels-Alder type adducts (7 and 8) and two stilbene dimers (4 and 6) interacted via hydrogen bonds to α-glucosidase catalytic residues. Furthermore, the 12-OH and 6-OH of compound 4 interacted with the Asp442 residue and the hydrogen bonding with Asp352 and Glu411 residues on the α-glucosidase catalytic site. Compound 8 interacted with Tyr158 residue via the hydrophobic interaction as well as Pro312 residue via the hydrogen bonds. None of the test compounds were bound to the residue to which the substrate, α-Dglucose, binds via hydrogen bonding. Compounds 4 and 8 interacted with binding residues similar to acarbose in terms of hydrogen bond interactions. In other words, the findings of this study showed that compounds 4 and 8 interact with major residues without causing adverse effects compared to acarbose (which is known to have side effects). Compounds 4 and 8 had the lowest energy and inhibited the α-glucosidase when the docking score of residues that interacted with the target was calculated in this manner.
To demonstrate the interaction and binding modes of active compounds with PTP1B, these test compounds were compared with the reported compounds 2 (allosteric inhibitor) and 23 (catalytic inhibitor). The calculated docking score of binding energies indicated high affinity and lower binding energies within the enzyme catalytic site and allosteric site. All test compounds posed within the allosteric site on PTP1B interacted with multiple hydrogen bonds for compounds 4 and 6-8 (Figures 6-8). Similar to previous α-glucosidase studies, compound 4 was combined with the same residue as allosteric and catalytic inhibitors to inhibit PTP1B (Table 6). In particular, stilbene dimer 4 and Diels-Alder type adduct 8 interacted with hydrogen bonds. As shown in Figures 7 and 8, compounds 6 and 8 showed interactions by binding to Asn 193 and Glu 276-the same residues as allosteric inhibitors. Previous research by Jung et al. [66] found that Cys215, His214, Arg221, Thr177, Pro189, Glu186, Glu200, Ser201, Gly209, Ala264, and Ile281 residues play a role in the catalytic loop of PTP1B, which is consistent with our findings.
We have demonstrated that compounds 4 and 6-8, which were isolated at first from M. bombycis leaves, have anti-diabetic and anti-diabetic complication effects. Kinetic analyses and a molecular docking study were used to determine the interaction mechanisms within the enzyme sites. The implication of these findings is noteworthy in that the inhibition mechanism of stilbene dimers (4 and 6) and Diels-Alder adducts (7 and 8) against DM and its complications corroborate their potential as therapeutic or preventive agents and functional foods.

Conclusions
Morus bombycis has been used as traditional medicine, and many studies have been conducted on other parts of the plant such as the root, bark, and cortex. However, few studies have explored M. bombycis leaves. In the present study, we found that M. bombycis and its isolated compounds were very effective scavengers/inhibitors against ONOO − , α-glucosidase, PTP1B, AGEs, and HRAR. Among the test compounds, stilbene-dimer-type compounds 4 and 6 exhibited strong antioxidant, anti-diabetic, and anti-diabetic complication effects, whereas Diels-Alder type adducts compounds 7 and 8 effectively inhibited α-glucosidase and PTP1B. All the tested compounds showed mixed-type inhibition against α-glucosidase in the enzyme kinetic study. Compounds 4 and 6-7 were confirmed as mixed-type inhibitors, while compound 8 was determined to be a non-competitive inhibitor against PTP1B. Compounds 4 and 6-8 docked within the catalytic site of α-glucosidase, whereas compounds 4 and 8 were bound within both the catalytic and allosteric sites of PTP1B, and compounds 6 and 7 were bound only within the allosteric site. In conclusion, these findings imply that stilbene dimers and Diels-Alder type adducts could be novel and/or important natural inhibitors or preventive resources as antioxidants, anti-diabetic agents, and anti-diabetic complication agents.