Xanthones as α-glucosidase inhibitors from the antihyperglycemic extract of Securidaca inappendiculata

Abstract

Context: Securidaca inappendiculata Hassk. (SI) is used to cure fractures and rheumatoid arthritis in China. Also, it is a potential antidiabetes drug; however, there are no reports on this.

Objective: The study was designed to evaluate the antihyperglycemic activities of fractions and compounds from SI, and attempt to explore the mechanism.

Materials and methods: Antihyperglycemic activities were evaluated by the suppression on serum glucose levels in vivo and α-glucosidase inhibition assays in vitro. Fractions were given to mice by gastric intubation for 8 d. The high, medium, and low doses of fractions were equal to 10, 5, and 2.5 g/kg of the herb [SID (dichloromethane fraction) and SIE (ethyl acetate fraction) were doubled]. The serum glucose was monitored at 1 and 12 h after feeding. The silica gel and LH-20 chromatography were used to isolate active compounds. Structure–activity relationship analysis was based on IC₅₀s and structures.

Results: The IC₅₀s of SID, SIE, SIA (acetone fraction), SIM (methanol fraction), and acarbose were 712, 446, 1123, 1418, and 735 μg/mL. The postprandial and fasting serum glucose levels of SID, SIE, SIA, and SIM (high dose) were 5.5, 5.9, 6.2, 6.3 and 3.7, 3.5, 4.0, 5.0 mmol/L, while those of vehicle control were 7.5 and 5.6 mmol/L. Eleven xanthones isolated all exhibited inhibitory activities, mainly in a non-competitive reversible manner. The IC₅₀s varied from 3.2 to 77.3 μg/mL. Structure–activity relationship analysis exhibited free hydroxyls contributed the most importance to the activity.

Conclusion: The results indicated that xanthones from SI were powerful agents for antidiabetes.

• Activity screening
• antidiabetes
• material base
• structure–activity relationship analysis

Introduction

Diabetes is one of the most prevalent chronic diseases in the world. Between 2010 and 2030, there will be an increase of 69% in number of adults with diabetes in developing countries and an increase of 20% in developed countries (Shaw et al., 2010). Diabetes is ultimately due to chronically high levels of blood glucose. The persistent hyperglycemic condition in diabetes plays a predisposing role in dysfunction and failure of various organs (American Diabetes Association (ADA), 2011). Thus, the sustained control of hyperglycemia is considered to be important to the effective treatment of diabetes. For a long time, diabetics have been treated by insulin and synthetic drugs, which are available at present, but they produced serious side effects. Hypoglycemic episodes cause blackouts and when severe, they are life-threatening (Lopez-Candales, 2001). Herbal medicines play an important role in this part to prevent side effects. Located in the brush-border surface membrane of intestinal cells, α-glucosidase plays a crucial role in dietary carbohydrates digestion and post-translational processing of glycoprotein (Carrascosa et al., 2001). By inhibiting the function of α-glucosidase, many herbal components can delay glucose absorption as inhibitors (Andrade-Cetto et al., 2008).

Securidaca inappendiculata Hassk. (SI), also known as Chan Yi Teng, is mainly yielded in the south of China. The stem, leaf, and radix are used as traditional medicines to alleviate pains, heal wounds, cure fractures, and rheumatoid arthritis (Yunnan Institute of Botany, 1983). It also is a potential herbal remedy for many other diseases, such as hepatitis, gastritis, and HIV (Yong-wen, 2005). SI is known to generate an abundance of bioactive compounds, especially polyphenolic compounds (Li-jie, 2005; Xue-dong, 2001; Yong-wen, 2005). Among them, xanthones are the most important active components, which exert antioxidant, anticancer, anti-inflammatory, immunoregulation, and α-glucosidase inhibitory activities (Ghias et al., 2012; Jing et al., 2013).

There are no reports on the antidiabetes activity of SI so far. In ongoing research, we found SI can ameliorate diabetes, so we screened the active fractions hampering glucose absorption in vivo and in vitro. Furthermore, we isolated the compounds from the active fractions. In this paper, we report the active fractions screening results, and the isolation of 11 xanthones. The compounds were all assayed on inhibitory activities and inhibitory manners against α-glucosidase. A brief structure–activity relationship analysis was also done.

Materials and methods

Reagents and materials

Stem of Securidaca inappendiculata was collected in November 2012 at Nanning, Guangxi Province, China. The plant was authenticated by Professor Jian-Wei Chen (College of Pharmacy, Nanjing University of Chinese Medicine, China). A voucher specimen of the plant (ID: CYT2012-005) was deposited in the Herbarium Center, Nanjing University of Chinese Medicine, China. α-Glucosidases from Saccharomyces cerevisiae, p-nitrophenyl-α-d-glucopyranoside (PNPG) and acarbose were purchased from Sigma-Aldrich (St. Louis, MO). Reagent-grade petroleum level, dichloromethane, methanol, ethanol, and ethyl acetate (for extraction and column chromatography) were purchased from Guanghua Sci-tech Co. Ltd. (Guangdong, China). Reagent-grade sodium dihydrogen phosphate, disodium hydrogen phosphate, and sodium hydroxide were purchased from Xilong Chemical Factory (Guangdong, China). Silica gel for column chromatography was purchased from Qingdao Ocean Chemical Factory (Shandong, China). Ultra-pure water was prepared using a Millipore Milli-Q purification system by Eped Science Co. Ltd. (Jiangsu, China).

Experimental animals

Male Kun Ming mice (4 weeks of age) were purchased from Medical Comparison Centre of Yangzhou University, China. All animals were housed individually in a light- (12 h on/12 h off) and temperature-controlled room with pelleted food and water available ad libitum. All animal procedures were approved by the Ethical Committee of Nanjing University of Chinese Medicine, and strictly in accordance with the guide for the care and use of laboratory animals (US National Research Council, 2011).

Fraction preparation

The dried stem of SI (10 kg) was extracted three times with 95% alcohol, and filtered. The residue was re-extracted with 50% alcohol by one time. The filtrates were evaporated together by a rotation evaporator and obtained the total extract (2026 g). The extract was mixed with silica gel (1000 g) and heated to dryness. Then it was partitioned with dichloromethane, ethyl acetate, acetone, and methanol in turn. Afterwards, each part was concentrated to dryness in vacuo to afford four fractions: dichloromethane fraction (SID) 56 g, ethyl acetate fraction (SIE) 79 g, acetone fraction (SIA) 311 g, and methanol fraction (SIM) 206 g for biological assay and isolation.

Inhibition assays of fractions against α-glucosidase

The α-glucosidase enzyme reaction was performed using PNPG as a substrate. The inhibitory activity of enzyme was measured spectrophotometrically through monitoring the nitrophenyl which produced by the hydrolysis of the substrate (Shelly et al., 2010). α-Glucosidase, acarbose, sodium carbonate, and the four fractions (samples) were all dissolved with 0.1% PBS individually. α-Glucosidase (0.2 unit/mL) (20 μL) was treated with 10 μL of sample solutions for 15 min at 37 °C, with the same volume of DMSO as a negative control and 20 mg/mL acarbose as a positive control. The final volume of the reaction solution was 0.16 mL added by 0.1% PBS. About 20 μL of PNPG (2.5 mmol/L) was added to initiate the enzyme reaction. After incubating the mixture at 37 °C for 15 min, 80 μL of sodium carbonate solution was used to cease the reaction. OD values were detected at 405 nm by plate reader (BIO-RAD Science Co. Ltd., Wilmington, DE).

Enzymatic inhibitory activity = (1 − A_x/A₀) × 100% (A₀ represents the OD value of DMSO control and A_x represents those of samples being tested).

According to the preliminary results, similar experiments were undertaken at different concentrations using the same method.

Inhibitory mechanistic analyses of fractions

α-Glucosidase was treated without or with samples for 15 min at 37 °C and then the substrate (1.25, 1.67, 2.5, 3.75, or 8.33 mmol/L PNPG) was added to initiate the enzyme reaction. The main procedures were the same as described above. Velocity of the reaction was observed for double reciprocal plots analysis (Zhi-Wei et al., 2013).

Suppressions of fractions on serum glucose levels in normal mice

Normal mice were randomly divided into 14 groups. The mice (with eight in each group) received various doses of fractions by gastric intubation for 8 d. The mice of vehicle control were given 0.5% CMC-Na. After the last administration, the serum glucose levels were monitored at 1 and 12 h after feeding by tail bleeding using a glucometer (Roche Diagnostics Gmbh, Mannheim, Germany) to compare postprandial serum glucose and fasting serum glucose levels among groups (the mice were fasted in 12 h after last feeding). Fasted animals were deprived of food but allowed free access to water (Kazuhiro et al., 2010).

The doses of fractions were in accord with the percentages in the crude drug. The exactly doses were as follows: SID 112, 56, 28 mg/kg, SIE 158, 79, 39.5 mg/kg, SIA 311, 155.5, 77.75 mg/kg, and SIM 206, 103, 51.5 mg/kg. The high, medium, and low doses were equal to 10, 5, and 2.5 g/kg of the crude drug (the doses of SID and SIE were doubled because of the relative low percentages).

Isolation of active compounds

The mixture combined by SID and SIE (20 g and 30 g) was chromatographed on a silica gel (200 g) column (4 i.d. × 40 cm) by medium pressure liquid chromatography (MPLC) (Lisure Science Co. Ltd., Zhejiang, China). The column was gradient eluted with petroleum level and methanol, and yielded 26 sub-fractions. Each sub-fraction was then chromatographed on a Diaion LH-20 (Mitsubishi Chemical Co. Ltd., Tokyo, Japan) column (2 i.d. × 40 cm) with dichloromethane and methanol. Sub-fractions 5–13 yielded 1,3,7-trihydroxyl-xanthone (A, 17 mg, 97%), 1,3,7-trihydroxyl-2-methoxyl-xanthone (B, 41 mg, 96%), 1,5-dihydroxyl-2,6,8-trimethoxyl-xanthone (C, 13 mg, 98%), 1,3,7-trihydroxyl-4-methoxyl-xanthone (D, 5 mg, 94%), 1,7-dihydroxyl-3,4-dimethoxyl-xanthone (E, 98 mg, 99%), 1,7-dihydroxyl-4-methoxyl-xanthone (F, 8 mg, 98%), 1,7-dihydroxyl-xanthone (G, 132 mg, 95%), 1,3,7-trihydroxyl-2,8-dimethoxyl-xanthone (H, 19 mg, 96%), 2-hydroxyl-1,7-dimethoxyl-xanthone (I, 117 mg, 97%), 1,3,6-trihydroxyl-2,7-dimethoxyl-xanthone (J, 6 mg, 93%), and 7-hydroxyl-1,2-dimethoxyl-xanthone (K, 35 mg, 98%) (Li-jie, 2005; Xue-dong, 2001). The purity of the compounds was checked by HPLC. The structures are shown in Figure 1.

Figure 1. The structures of xanthones isolated from Securidaca inappendiculata.

Inhibition assays of active compounds against α-glucosidase

The methods and procedures were the same as described above.

Inhibitory mechanistic analyses of active compounds

The methods and procedures were the same as described above.

Results

Fractions inhibited α-glucosidase activity

Inhibitory activities of different fractions and acarbose agaist α-glucosidase were detected in the same concentration (0.833 mg/mL) with results shown in Figure 2. All the fractions inhibited α-glucosidase activity. SIE inhibited the activity by 70.6%, whereas SIA and SIM showed relatively low inhibitory activities (≤40.0%). All the fractions inhibited the enzyme activity in a dose-dependent manner. The IC₅₀s of SID, SIE, SIA, SIM, and acarbose were 712, 446, 1123, 1418, and 735 μg/mL, respectively.

Figure 2. The inhibitory effects of fractions from SI on yeast α-glucosidase activity. Acarbose served as the positive control. All the fractions and drugs tested were the same concentration (0.833 mg/mL). The α-glucosidase activity was determined by measuring p-nitrophenol released from PNPG at 405 nm. The reaction was conducted at 37 °C for 15 min. Results are expressed as mean ± SD, n = 3. SID, SIE, SIA, and SIM were the abbreviations of dichloromethane fraction, ethyl acetate fraction, acetone fraction, and methanol fraction.

Inhibitory mechanistic analyses of fractions

The results of inhibitory analyses are shown in Figure 3. There were three kinds of inhibitory effects according to the results. Intersection of the double reciprocal plots of SID seated in the deuto-quadrant, indicating that it acted as a mixed-type inhibitor (Kazuhiro et al., 2010). Intersections of SIA and SIM both seated on the longitudinal axis, indicating that they acted as competitive reversible type inhibitors, while SIE acted as a non-competitive reversible type inhibitor.

Figure 3. Double reciprocal plots for inhibitory analysis against α-glucosidase by fractions. α-Glucosidase (0.2 unit/mL) was treated with fractions for 15 min at 37  °C and then 20 μL of PNPG (1.25, 1.67, 2.5, 3.75, or 8.33 mmol/L) was added to initiate the enzyme reaction. The enzyme reaction was performed by incubating the mixture at 37 °C for 15 min. SID, SIE, SIA, and SIM were the abbreviations of dichloromethane fraction, ethyl acetate fraction, acetone fraction, and methanol fraction.

Suppression of serum glucose levels in normal mice by different fractions

All the four fractions suppressed serum glucose levels, both at fasting (Figure 4) and postprandial conditions (Figure 5). SID and SIE suppressed the serum glucose levels efficiently, while the rest exhibited similar but weaker antihyperglycemic effects. Because of the much lower doses administered, SID and SIE were believed to be the main active components for antidiabetes.

Figure 4. Inhibition of fractions on the increase of fasting serum glucose levels in normal mice. The mice fasted for 12 h were received various doses of fractions by gastric intubation. The mice of vehicle control group were given 0.5% CMC-Na. Each value represents mean ± SD, n = 8. **p < 0.01 versus vehicle control. SID, SIE, SIA, and SIM were the abbreviations of dichloromethane fraction, ethyl acetate fraction, acetone fraction, and methanol fraction.

Figure 5. Inhibition of fractions on the increase of postprandial serum glucose levels in normal mice. The mice received various doses of fractions by gastric intubation. Mice of the vehicle control group were given 0.5% CMC-Na. The postprandial serum glucose was measured 1 h after feeding. Each value represents mean ± SD, n = 8. *p < 0.05, **p < 0.01 versus vehicle control. SID, SIE, SIA, and SIM were the abbreviations of dichloromethane fraction, ethyl acetate fraction, acetone fraction, and methanol fraction.

Efficiencies of inhibitory activities against α-glucosidase varied among compounds

The inhibition efficiencies varied greatly among compounds. Compound A inhibited α-glucosidase most efficiently, while compound J could not be tested for IC₅₀ because of extremely weak activity and dissolubility. The exact activity sequence of the compounds (based on the inhibition efficiency from the strongest to weakest) was A, B, D, F, G, K, C, H, I, and J (Figure 6).

Figure 6. IC₅₀s of α-glucosidase inhibition of compounds. IC₅₀ was measured by monitoring p-nitrophenol released from PNPG at 405 nm. The enzyme (0.2 unit/mL) was treated by compounds at five different concentrations each for 15 min at 37 °C. The reaction was conducted at 37 °C for 15 min. Compounds A–K represented.

Inhibitory mechanistic analyses of active compounds

The results of inhibitory analyses demonstrated that nearly all the compounds acted in the same manner. Of 11 compounds, nine inhibited α-glucosidase in a non-competitive reversible manner judged by the intersection of the double reciprocal plots seated on the horizontal axis. Just two compounds, J and C, acted as competitive reversible type inhibitors (Figure 7).

Figure 7 Double reciprocal plots for inhibitory analysis of α-glucosidase by xanthones. (A) Inhibitory characteristics of compounds C and J. (B) Inhibitory characteristics of compounds H, K, D, F, I, E, A, B, and G. α-Glucosidase (0.2 unit/mL) was treated with compounds for 15 min at 37 °C and then 20 μL of PNPG (1.25, 1.67, 2.5 3.75, or 8.33 mmol/L) was added to initiate the enzyme reaction. The enzyme reaction was performed by incubating the mixture at 37 °C for 15 min.

Discussion

The inhibition assays against α-glucosidase demonstrated that SID and SIE had inhibitory activities similar to acarbose. SID and SIE also showed remarkable serum glucose level suppressing activities compared with the vehicle control group, both at postprandial and fasting conditions in vivo. Based on these results, we came to the conclusion that SI can ameliorate diabetes by slowing down glucose absorption as a α-glucosidase inhibitor. The isolation of active fractions afforded a series of xanthones. Most of them inhibited α-glucosidase efficiently. Hence, we concluded that xanthones are the material base of antidiabetes and antihyperglycemic in SI.

Structure–activity relationship analyses of xanthones as α-glucosidase inhibitors mainly focus on the pi-system of the aromatic rings, the branching of the molecule and the number of the hydroxyl groups. It is believed that the increase in the number of aromatic rings, double bonds, or other conjugate structures enhances the interaction between compounds and the hydrophobicity surface of the enzyme protein, hence enforcing the inhibitory activity (Vesna et al., 2012; Vijay et al., 2012). An increase in the number of the hydroxyl groups is favorable for the inhibitory activity because molecular charge transfers among atoms cause stabilization of the aromatic rings (Vesna et al., 2012). The hydroxyl at position number 7 is supposed to be the most important functional group (Vijay et al., 2012). In our research, we found some clues which are beneficial to clarifying the relationship. Compound A exhibited stronger activity than G, hinting that more hydroxyls are beneficial to the activity. The conclusion is coincident with former reports. Compound J was the weakest inhibitor among the xanthones tested but with three hydroxyls. The similar phenomenon also existed in compound H, which has three hydroxyls. Besides, there is a hydroxyl at position number 7 in the structure of compound H, but it was a weak inhibitor. More hydroxyls will enhance activity only if all the other factors are fixed, and no reciprocal effects exist. If the hydroxyls are surrounded by other oxygen-containing groups, the increase of hydroxyls will not be essential. The activities of compound E, F, and G decreased in turn in accord with the number of methoxyls possessed. We can come to the conclusion that given the same number of hydroxyls, the increase of methoxyls will enhance activity too. The results can be elucidated by H-bond hypothesis. The hydroxyl on xanthones binds hydrophilic groups of catalytic domain A and B of α-glucosidase to inhibit activity (Ghias et al., 2012). While it binds the nearby oxygen-containing groups on its own structure, the binding affinity between the compound and enzyme will be weak. Methoxyl interact with the H-bond donor in the same way but with weaker activity. In sum, more free hydroxyls are essential for an efficient α-glucosidase inhibitor.

Conclusion

We investigate SI as a potential antidiabetes medicine for the first time, and found all the fractions from it suppressed the serum glucose levels in normal mice. In addition, SID and SIE inhibited α-glucosidase activity efficiently. The xanthones from SI exhibited much stronger inhibitory activities than fractions. All these results together hinted that xanthones were the powerful agents in SI for antidiabetes, by inhibiting α-glucosidase activity.

Declaration of interest

The authors report no declarations of interest.