Novel Potent Hypoglycemic Compounds from Euonymus laxiflorus Champ. and Their Effect on Reducing Plasma Glucose in an ICR Mouse Model

α-Glucosidase inhibitors (aGIs) have been used as an effective therapy for type-2 diabetes, which remains a global health issue. The aim of this study was to achieve bioactivity-guided isolation, identification and evaluation of hypoglycemic compounds from Euonymus laxiflorus Champ. trunk bark (ELCTB). Eleven active compounds were isolated and identified as walterolactone A/B β-d-pyranoglucoside (1), 1-β-d-glucopyranosyloxy-3,5-dimethoxy-4-hydroxybenzene (9), (−)-gallocatechin (10), schweinfurthinol 9-O-β-d-pyranoglucoside (11), 1-O-(3-methyl)-butenoyl-myo-inositol (12), leonuriside (14), (+)-catechin (19), methyl galloate (20), (−)-catechin (23), and condensed tannins (5 and 18). Of these 11, novel 4 compounds (1, 11, 12, and 14) were found as new α-glucosidase inhibitors. Notably, in vitro results indicated that compounds 1, 5, 10–12, 18, and 19 showed potent activity (IC50 = 0.076−31 µg/mL), and their activities were at a higher level than that of acarbose, a commercial inhibitor (IC50 = 1345 µg/mL). In animal tests, the major inhibitor, condensed tannin (18), demonstrated significant reduction of plasma glucose in mice with no symptoms of diarrhea at the dose of 100 mg/kg bw. The results suggest that Euonymus laxiflorus Champ. is a rich source of bioactive compounds for development as health food or drugs with potent hypoglycemic effect. The results of this study also enriched the current novel biological activities of constituents from Euonymus laxiflorus species.


Introduction
Natural bioactive products are of great interest due to their beneficial use as health foods or drugs to manage significant numbers of diseases including type-2 diabetes (T2D), a serious current global
In the comparison, this herbal extract demonstrated comparable or much higher α-glucosidase inhibitory activity compared to those of extracts from other reported herbals also collected in the central highlands of Vietnam due to its smallest IC 50 values against α-glucosidases from rat (Table 1).  [7] α-Glucosidase from rat was used for testing; results are means ± SD of multi tests (n = 3); coefficient of variation = 12.35314; the means of IC 50 values with the different letter are significantly different in comparison based on Duncan s multiple range test (alpha = 0.01) using SAS version 9.4, Statistical Analysis Software analysis.
Molecules 2018, 23,1928 3 of 14 The methanol extract of this herbal also possesses significant effect on reducing plasma glucose in diabetic rats [17]. Thus, it was used for bioactivity-guided isolation of active hypoglycemic compounds via several columns, such as Diaion, Octadecylsilane, and preparative HPLC columns in the current report.
Compounds 5 & 18 were confirmed as condensed tannins by using total phenolic determination, biological assays [25], and comparison of their 13 C-NMR data with those of reported condensed tannins [18,26,27]. These 2 compounds were pre-confirmed as tannins since they contained ≥94% total phenolic acid and also exhibited tested biological activities, including antioxidant activity (≥96%), αamylase inhibition (100%), and protease inhibition (≥95%) at their tested concentration of 1 mg/mL. In addition, the 13 C-NMR data on these compounds (materials and methods section) were similar to those of condensed tannin ELC3.1-d isolated and identified from the same herbal species [18], and also closely similar to those of other reported condensed tannins, including persimmon tannin [26], and condensed tannins from Delonix regia [27]. Thus, compounds 5 & 18 were confirmed as condensed tannins.
Compounds 5 & 18 were confirmed as condensed tannins by using total phenolic determination, biological assays [25], and comparison of their 13 C-NMR data with those of reported condensed tannins [18,26,27]. These 2 compounds were pre-confirmed as tannins since they contained ≥94% total phenolic acid and also exhibited tested biological activities, including antioxidant activity (≥96%), α-amylase inhibition (100%), and protease inhibition (≥95%) at their tested concentration of 1 mg/mL. In addition, the 13 C-NMR data on these compounds (materials and methods section) were similar to those of condensed tannin ELC3.1-d isolated and identified from the same herbal species [18], and also closely similar to those of other reported condensed tannins, including persimmon tannin [26], and condensed tannins from Delonix regia [27]. Thus, compounds 5 & 18 were confirmed as condensed tannins.
Compounds 1, 11, and 12 were newly isolated and identified as new compounds from the same medicinal plant in the previous report by Nguyen et al. [18]. However, the α-glucosidase inhibitory activity of these compounds was investigated for the first time in this study. They were determined as new aGIs based on the current literature review, and all their NMR spectrums, including 1 H-MNR, 13 C-NMR, DEPT 135 , COSY, HSQC, and HMBC were presented in the supplementary materials section ( Figures S1-S21). Compound 14, leonuriside, was reported to possess several bioactive properties, including α-amylase inhibition [18], COX-1 inhibition [24], and potent anti-NO [22]. However, the α-glucosidase inhibition of leonuriside had not been reported before; this compound was also determined as a new aGI.

Comparision of α-Glucosidase Inhibitory Activity of Identified Compounds
To screen the most active α-glucosidase inhibitors, all the isolated inhibitors were tested for their activity at various concentrations; the activity was then expressed as IC 50 (µg/mL) value. As shown in Table 3, compounds 1, 5, 14, 18, and 19 demonstrated the most effective inhibition due to their smallest IC 50 values (0.076-0.926 µg/mL). These inhibitors also showed great maximum inhibition (99-100% at 250 µg/mL). Compounds 10, 11, and 12 were also found as potent inhibitors, which possessed low IC 50 values (11.9-31.6 µg/mL) and high maximum inhibition (87-96%). Compounds 9 and 23 showed weak activity (≤38%). Overall, these novel inhibitors had their activity ranked sequentially based on Duncan's Multiple Range Test (alpha = 0.01):

The Effect of Condensed Tannin (CT) on Reducing Plasma Glucose in a Mouse Model
Condensed tannin-ELCTB-3.1 (CT) was isolated at a large amount (~2000 mg) and showed efficient inhibition against α-glucosidase. Thus, this major inhibitor was conducted to test its effect on reducing plasma glucose in mice. To evaluate the effect of the samples on reducing plasma glucose in animals, sucrose and starch tolerance tests were used [11,36]. In this study, starch (3 g/kg bw) was chosen for the assay, since it is abundant in cereals, the daily food of people in Asian countries. Therefore, in the case that the compound (CT) showed significant effect on the reducing plasma glucose in mice, it may show potent hypoglycemic effect in humans.
Two doses of isolated CT (50 and 100 mg/kg bw) were orally administered to mice to evaluate the effect on their plasma glucose level. CT at the dose of 50 mg/kg bw showed significant reduction of plasma glucose in mice at 0.5 h after CT administration; thereafter, the hypoglycemic effect showed less significant difference than that of the control group (mice administered water only) ( Figure 3A). On the other hand, the effect of CT on the reduction of plasma of mice at the dose of 100 mg/kg bw was clearly observed from 0.5 to 2 h after CT administration ( Figure 3B). Notably, the effect of CT at 100 mg/kg bw was comparable to that of acarbose at 50 mg/kg bw ( Figure 3B), and much higher than that of acarbose at 25 mg/kg bw ( Figure 3A). The mice showing symptoms of diarrhea during the tests were recorded. CT at all treated doses on mice led to no symptoms of diarrhea in mice, while mice with this illness symptom were recorded at 20% and 50% of the mice treated with acarbose doses of 25 and 50 mg/kg bw, respectively. chosen for the assay, since it is abundant in cereals, the daily food of people in Asian countries. Therefore, in the case that the compound (CT) showed significant effect on the reducing plasma glucose in mice, it may show potent hypoglycemic effect in humans. Two doses of isolated CT (50 and 100 mg/kg bw) were orally administered to mice to evaluate the effect on their plasma glucose level. CT at the dose of 50 mg/kg bw showed significant reduction of plasma glucose in mice at 0.5 h after CT administration; thereafter, the hypoglycemic effect showed less significant difference than that of the control group (mice administered water only) ( Figure 3A). On the other hand, the effect of CT on the reduction of plasma of mice at the dose of 100 mg/kg bw was clearly observed from 0.5 to 2 h after CT administration ( Figure 3B). Notably, the effect of CT at 100 mg/kg bw was comparable to that of acarbose at 50 mg/kg bw ( Figure 3B), and much higher than that of acarbose at 25 mg/kg bw ( Figure 3A). The mice showing symptoms of diarrhea during the tests were recorded. CT at all treated doses on mice led to no symptoms of diarrhea in mice, while mice with this illness symptom were recorded at 20% and 50% of the mice treated with acarbose doses of 25 and 50 mg/kg bw, respectively. Natural products such as tea and coffee with high polyphenolic compounds content [37], were reported to show good effect on the reduction of plasma glucose in mice [38], as well as on postprandial plasma glucose in healthy humans [37]. Some herbal extracts rich in condensed tannins [6,17,39] were also tested for their anti-hyperglycemic effects in diabetic rats. Psidium littorale Raddi leaf extract at the dose of 150 mg/kg bw reduced fasting plasma glucose levels in streptozotocininduced diabetic rats [6]. The methanolic extract of Euonymus laxiflorus Champ. trunk bark (the crude sample of CT) showed significant reduction of blood glucose in diabetic rats at the dose of 200 mg/kg bw, comparable to that of acarbose at 120 mg/kg bw [17]. The pinhão coat extract, and the A. mearnsii tannin also demonstrated significant effectiveness in diminishing the post-prandial glycemic levels in rats at their doses of 250 mg/kg bw after starch administration [39]. In this study, Euonymus laxiflorus Champ. condensed tannins demonstrated significant effect on reducing plasma glucose at their low doses of 50, and 100 mg/kg bw.
Condensed Tannins possess vast beneficial bioactivities, including cardio-protective, antioxidative, antitumor, antiviral, antibacterial, immune-modulatory, anti-inflammatory activities, antiobesity, antidiabetic [40,41], and hypoglycemic effects [39]. Euonymus laxiflorus Champ. Natural products such as tea and coffee with high polyphenolic compounds content [37], were reported to show good effect on the reduction of plasma glucose in mice [38], as well as on postprandial plasma glucose in healthy humans [37]. Some herbal extracts rich in condensed tannins [6,17,39] were also tested for their anti-hyperglycemic effects in diabetic rats. Psidium littorale Raddi leaf extract at the dose of 150 mg/kg bw reduced fasting plasma glucose levels in streptozotocin-induced diabetic rats [6]. The methanolic extract of Euonymus laxiflorus Champ. trunk bark (the crude sample of CT) showed significant reduction of blood glucose in diabetic rats at the dose of 200 mg/kg bw, comparable to that of acarbose at 120 mg/kg bw [17]. The pinhão coat extract, and the A. mearnsii tannin also demonstrated significant effectiveness in diminishing the post-prandial glycemic levels in rats at their doses of 250 mg/kg bw after starch administration [39]. In this study, Euonymus laxiflorus Champ. condensed tannins demonstrated significant effect on reducing plasma glucose at their low doses of 50, and 100 mg/kg bw.
Condensed Tannins possess vast beneficial bioactivities, including cardio-protective, antioxidative, antitumor, antiviral, antibacterial, immune-modulatory, anti-inflammatory activities, antiobesity, antidiabetic [40,41], and hypoglycemic effects [39]. Euonymus laxiflorus Champ. condensed tannins (condensed tannin-ELCTB-3.1) were found to possess potent hypoglycemic effect in this study. It is well known that starch, a polysaccharide, is degraded by amylases to dextrin oligomers; these oligomers are then further degraded to α-D-glucoses by α-glucosidase. These monomeric may enter the blood circulation via intestinal epithelial absorption. Therefore, the use of α-amylase and α-glucosidase inhibitors may block or slow down this process, leading to the hypoglycemic effect after meals. Euonymus laxiflorus Champ. condensed tannins (ELC-CT) showed effective inhibition both on α-amylase [16] and α-glucosidase, since this resulted in their significant hypoglycemic effect.
To determine the inhibition mode of ELC-CT, three concentrations of ELC-CT: 0.5, 0.244, 0 µg/mL, and 0.05, 0.03, 0 µg/mL against α-amylase, and α-glucosidase, respectively, were tested. According to the Lineweaver-Burk plots of enzymatic inhibition kinetics of ELC-CT against α-amylase ( Figure 4A), and α-glucosidase ( Figure 4B), ELC-CT was determined as mix (non-competitive-uncompetitive) inhibition against α-amylase, and non-competitive inhibition against α-glucosidase via comparison to the typical Lineweaver-Burk plots [42]. condensed tannins (condensed tannin-ELCTB-3.1) were found to possess potent hypoglycemic effect in this study. It is well known that starch, a polysaccharide, is degraded by amylases to dextrin oligomers; these oligomers are then further degraded to α-D-glucoses by α-glucosidase. These monomeric may enter the blood circulation via intestinal epithelial absorption. Therefore, the use of α-amylase and α-glucosidase inhibitors may block or slow down this process, leading to the hypoglycemic effect after meals. Euonymus laxiflorus Champ. condensed tannins (ELC-CT) showed effective inhibition both on α-amylase [16] and α-glucosidase, since this resulted in their significant hypoglycemic effect. To determine the inhibition mode of ELC-CT, three concentrations of ELC-CT: 0.5, 0.244, 0 µg/mL, and 0.05, 0.03, 0 µg/mL against α-amylase, and α-glucosidase, respectively, were tested. According to the Lineweaver-Burk plots of enzymatic inhibition kinetics of ELC-CT against α-amylase ( Figure 4A), and α-glucosidase ( Figure 4B), ELC-CT was determined as mix (noncompetitive-uncompetitive) inhibition against α-amylase, and non-competitive inhibition against αglucosidase via comparison to the typical Lineweaver-Burk plots [42]. These results indicated that ELC-CT did not combine with enzymes in the active sites of both enzymes (α-amylase, and α-glucosidase); ELC-CT combined with α-glucosidase to produce deadend enzyme-inhibitor complex regardless of the substrate (pNPG) was bound; while some ELC-CT molecules could combine with α-amylase-starch complex, but did not combine with free amylase, and some ELC-CT molecules could combine with both α-amylase-starch complex, and free amylase to block the amylase activity [42]. According to the results of this study and the previous report [17], suggest that Euonymus laxiflorus Champ. is a rich source of condensed tannins, which could be developed as health food with potent hypoglycemic effect.

Materials
The methanolic extract of Euonymus laxiflorus Champ. trunk bark was obtained from the previous study [4]. Saccharomyces cerevisiae (yeast) α-glucosidase and acarbose were purchased from Sigma Chemical Co., St. Louis City, MO, USA; p-nitrophenyl glucopyranoside (pNPG) was obtained from Sigma Aldrich, 3050 Spruce Street, St. Louis, MO, USA. All the solvents and common chemicals were obtained at their highest grade. These results indicated that ELC-CT did not combine with enzymes in the active sites of both enzymes (α-amylase, and α-glucosidase); ELC-CT combined with α-glucosidase to produce dead-end enzyme-inhibitor complex regardless of the substrate (pNPG) was bound; while some ELC-CT molecules could combine with α-amylase-starch complex, but did not combine with free amylase, and some ELC-CT molecules could combine with both α-amylase-starch complex, and free amylase to block the amylase activity [42].
According to the results of this study and the previous report [17], suggest that Euonymus laxiflorus Champ. is a rich source of condensed tannins, which could be developed as health food with potent hypoglycemic effect.

Materials
The methanolic extract of Euonymus laxiflorus Champ. trunk bark was obtained from the previous study [4]. Saccharomyces cerevisiae (yeast) α-glucosidase and acarbose were purchased from Sigma Chemical Co., St. Louis City, MO, USA; p-nitrophenyl glucopyranoside (pNPG) was obtained from Sigma Aldrich, 3050 Spruce Street, St. Louis, MO, USA. All the solvents and common chemicals were obtained at their highest grade.

α-Glucosidase Inhibitory Activity Determination
The α-glucosidase inhibitory activity was closely detected following the method described in detail by   [12]. The mixture of 50 µL α-glucosidase, 50 µL sample solutions, 100 µL buffer was pre-incubated at 37 • C for 20 min; the reaction then started when 50 µL of p-nitrophenyl glucopyranoside (10 mmol/L) was added to the mixture. After incubation at the same temperature for 30 min, the reaction was stopped by adding 100 µL Na 2 CO 3 solution (1 mol/L) to the reaction mixture; the absorbance of this final mixture was then measured at 410 nm (A). The control group also underwent the same described method with the use of 50 µL buffer instead of 50 µL sample solutions; the absorbance was recorded at 410 nm (B). The aGI activity (%) was calculated using the following equation: The inhibition was also expressed as IC 50 value determined as per the previous study [2]. The enzyme and the samples were prepared in 0.1 mol/L potassium phosphate buffer (pH 7). The purified compounds were tested at concentrations in the range of 0.015-250 µg/mL and the IC 50 plots for all tested compounds (1, 5, 10, 11, 12, 14, 18, 19, and 20), and acarbose were presented in the supplementary materials section (Figures S22-S31). The corresponding % inhibition at each concentration of all tested compounds were also recorded in Figure S32 . Forty ICR mice were randomly divided into 5 groups (8 mice/group), including a control group orally administered with distilled water, and 4 experimental groups orally administered with 50 mg CT/kg bw, 100 mg CT/kg bw, 25 mg acarbose/kg bw, and 50 mg acarbose/kg bw, respectively. Distilled water was used to prepare the condensed tannins and acarbose solutions.
Assay: The effect of condensed tannin-ELCTB-3.1.1 on the reduction of plasma glucose in mice was performed according to the experimental animal protocol described by   [11] with the use of starch solution (3 g/kg bw) instead of sucrose solution administered to ICR mice. All the mice groups were fasted overnight (16 h), and then orally administered water (control group), CT or acarbose (4 experimental groups); thereafter (20 min) starch solution (3 g/kg bw) was orally administered to mice; blood was then sampled and measured after 0.5, 1.0, 1.5, and 2.0 h.

Determination of Enzymatic Inhibition Modes of Isolated Condensed Tannin
The inhibition mode of condensed tannin was determined by performing as the reported assay with minor modification [10]. Enzyme kinetics of the isolated condensed tannin was determined using the α-glucosidase and α-amylase inhibitory activity assay mentioned above. The concentration range used was 2-18 mmol/L pNPG, and 0.0625-2% starch for the α-glucosidase, and α-amylase inhibitory activity assays, respectively. The inhibition modes of the sample were determined by analyzing the Vmax and comparing the Lineweaver-Bur plots of condensed tannin to the standard typical Lineweaver-Bur plots [42].
The chemical structures of the isolated compounds were identified via analysis of their NMR data, coupled with the comparison of those of reported compounds. The 1 H and 13 C-NMR spectra, and 2D-NMR spectra (COSY, HMQC, HMBC, and NOESY), were recorded in MeOH-d 4 on a Bruker AVX NMR spectrometer (Bruker, Karlsruhe, Germany) operating at 600 MHz for 1 to 12 h and 150 MHz for 13 C using the MeOH-d 4 solvent peak as internal standard (δ H 3.317, δ C 49.1 ppm).

Statistical Analysis
Statistical Analysis Software (SAS) version 9.4, provided by SAS Institute Taiwan Ltd., Minsheng East Road, Section 2, Taipei, Taiwan 149-8, was used to analyze the differences between the means of inhibition, and blood glucose level via Duncan's Multiple Range Test (alpha = 0.01 or 0.05). All tests were repeated in triplicate.

Conclusions
Eleven hypoglycemic compounds were isolated and identified from the methanolic extract of Euonymus laxiflorus Champ. trunk bark. Of these, four novel compounds of walterolactone A/B β-D-pyranoglucoside (1), schweinfurthinol 9-O-β-D-pyranoglucoside (11), 1-O-(3-methyl)-butenoyl-myo-inositol (12), and leonuriside (14) were determined as new α-glucosidase inhibitors. The results of in vitro tests indicated that most of the isolated compounds (1, 5, 10-12, 17, and 18) showed much higher activity than that of acarbose. Codensed tannin (18) demonstrated significant reduction of blood glucose in mice at the dose of 100 mg/kg bw. The results could enrich the current biological activities of constituents isolated from Euonymus laxiflorus Champ. species, and also suggest that this medicinal plant is a valuable source of bioactive compounds for development as health food or drugs with potent hypoglycemic effect.