α-Glucosidase Inhibitory Activity and Quantitative Contribution of Phenolic Compounds From Vietnamese Aquilaria crassna Leaves

Abstract

Aquilaria crassna Pierre ex Lecomte, Thymelaeaceae, is cultivated for producing resinous heartwood, also called agarwood. Its leaves are a source of herbal tea in Vietnam due to its rich content of polyphenols. However, the α-glucosidase inhibition activity and the contents of phenolic compounds in leaves of different ages have not yet been determined. In the current study, 7 polyphenols [iriflophenone 3,5-C-β-D-diglucoside (1), iriflophenone 3-C-β-D-glucoside (2), mangiferin (3), iriflophenone 2-O-α-rhamnoside (4), genkwanin 5-O-β-primeveroside (5), genkwanin 4′-methyl ether 5-O-β-primeveroside (6), and genkwanin (7)] were isolated from the leaves of A crassna. Among them, genkwanin (7), an O-methylated flavone, was the most active compound that inhibited α-glucosidase activity, with an IC₅₀ value of 24.0 μM. Molecular docking studies were performed to understand the binding interactions of the active compounds. In addition, a reliable and straightforward reversed-phase HPLC method was developed to determine the content of compounds in different leaves of A crassna. Mangiferin (3) showed the highest content. The contents of 1-4 contributed to the total polyphenolic contents and significantly decreased from the youngest to the oldest leaf. The contents of 5-7 fluctuated through various ages of leaves. Compounds 5 and 6 showed a low accumulation in the first and second leaves, then obtained high contents among middle leaves and declined in the oldest. These results suggested that A crassna and its polyphenols may prevent the development and progression of diabetes through α-glucosidase inhibition. Also, the analysis of the polyphenol content in A crassna may be helpful for tea product manufacture.

Introduction

Aquilaria is one of the most widespread genera of the Thymelaeaceae family, which is native to Southeast Asian countries such as Cambodia, China, Malaysia, Indonesia, Laos, Thailand, and Vietnam. Three species—Aquilaria crassna, Aquilaria malaccensis, and Aquilaria sinensis—are primarily being cultivated on thousands of hectares to produce agarwood, which comes from the resinous heartwood. ¹ In 1993, the Rainforest Project Foundation began to support Vietnamese farmers in cultivating Aquilaria plants in several provinces in southern Vietnam. That project supported plantations of more than one hectare; later the scheme was expanding into central Vietnam. The primary purpose of the planting was to produce the precious commercial product (used in perfumes and incense) known as agarwood, which is almost extinct, from 5- to 10-year-old plants. Aquilaria crassna has become the most cultivated plant in several farms in southern Vietnam, including in the provinces of An Giang, Kien Giang, and Binh Phuoc, alongside 2 other species—A. banaensae, and A baillonii. ¹ However, inoculation methods are not always successful from field to field. Some farmers have started to cut down the trees following their failure to produce agarwood. However, in other places, the farmers have changed their cultivation techniques in order to grow this plant to yield leaves by trimming the tree to keep it small, similar to a tea tree.^4,5

Apart from agarwood, the leaves of Aquilaria have traditionally been used as a herbal tea for laxative, anti-inflammatory, sedative, and pain-relieving purposes. ⁶ It has been widely reported that Aquilaria leaves are a very rich source of polyphenols, such as xanthones, flavonoids, and benzophenones. Mangiferin, iriflophenone 3-C-β-D-glucoside, iriflophenone 2-O-α-rhamnoside, genkwanin 5-O-β-primeveroside, genkwanin 4′-methyl ether 5-O-β-primeveroside, and genkwanin are the major components of A agallocha, A sinensis, and A crassna.^2,3 Recently, A crassna was reported to have activity on lipopolysaccharide-induced tumor necrosis factor-alpha production by human polymorphonuclear cells and the cell-signaling response. ⁶ The water extract of A crassna has been shown to exhibit antipyretic, analgesic, and anti-oxidative activities.^6,8 Iriflophenone 2-O-α-L-rhamnoside and iriflophenone 3-C-β-D-glucoside have had a positive effect on reducing blood glucose levels. Genkwanin 5-O-β-D-primeverosid and mangiferin have been proven to have laxative effects. In addition, genkwanin has been shown to have antibacterial and anti-inflammatory properties, in addition to anticancer activity against HT-29 and SW-480 human colorectal cancer lines in vitro, and on APC ^{Min /+} mice in vivo.^6,8,9

Several HPLC methods have been studied to determine the content of phenolic compounds in the leaves of A sinensis and A crassna. ¹⁰ In our preliminary screening of the Vietnamese natural and medicinal plants against diabetes, A crassna showed an inhibitory effect on α-glucosidase activity. Our preliminary study on the polyphenol content of young and old leaves also found that the content was much higher in the young leaves compared to the old leaves. Consequently, it became necessary to understand the dynamic changes in the significant phenolic compounds in the different ages of Aquilaria leaves. Hence, in this study, we isolated the major phenolic compounds as markers for establishing a quantification method to determine the major phenolic compound content in different-aged leaves of A crassna. This work provides valuable information that is essential for generating standardized materials for in vivo and in vitro studies, and for the authentication of the species as a food product.

Materials and Methods

Plant Materials and Chemicals

Branches of A crassna were collected from the Evergreen Forest Co. Ltd in Dong Phu District, Binh Phuoc Province, Vietnam (11°37′42.2″N 106°59′34.3″E) in March 2015 and identified by Dr Vo Van Chi, Faculty of Pharmacy, University of Medicine and Pharmacy at Ho Chi Minh city. The leaves were detached and sorted into sets of 10, from the youngest (first leaf) to oldest (tenth leaf) leaves that were then dried at 50 °C. The first 3 leaves were used for extraction and isolation. Voucher specimens (UMP-2015-AQ.001-010) were deposited at the Faculty of Pharmacy, University of Medicine and Pharmacy at Ho Chi Minh City, Vietnam. Each leaf (from the first to the tenth) was weighed, and then pulverized into powder of <355 µm. For extraction and isolation purposes, leaves were directly extracted without pulverization. The HPLC solvents were purchased from Merck and Labscan.

Apparatus

Preparative HPLC was performed on an LC-8A Shimadzu instrument, using a Discovery HS C18 column (25 cm × 21.2 mm, 10 µm). The injection volume was 10 mL, and the analysis was performed at room temperature. The flow rate was maintained at 10 mL/min, and the absorbance was measured at a wavelength of 330 nm. HPLC was performed on an ACQUITY UPLC H-Class system (Waters), equipped with a PDA detector (330 nm) and a Phenomenex Gemini C18 (150 mm × 4.6 mm, i.d., 3 µm) (Phenomenex), connected to Empower software. The separation was achieved using a mobile phase of 0.2% acetic acid/acetonitrile (A) and 0.2% acetic acid (B) in the following sequence: 0 to 1 min: 10% A; 1 to 5 min: 13% A; 5 to 10 min: 13% A; 10 to 13 min: 25% A; 13 to 14 min: 25% A; 14 to 15 min: 30% A; 15 to 18 min: 35% A; 18 to 19 min: 64% A; 19 to 22 min: 65% A; 22 to 23 min: 65% A; 23 to 28 min: 85% A; and 28 to 31 min: 10% A. The column temperature was set at 40 °C. The HPLC flow rate was 1 mL/min. A 4-μL sample solution was injected into the ultra-performance liquid chromatography (UPLC) system. TLC was performed using silica gel F254 (40-63 µm, Merck). Compounds were detected by UV light at 254 nm and 365 nm, and then were sprayed with 10% H₂SO₄ in EtOH, followed by heating. Column chromatography (CC) was performed using silica gel (200-300 mesh), along with liquid chromatography using LiChroprep RP-18 silica gel (40-63 µm) (Merck). Silica gel 60 (40-63 µm, 230-400 mesh ASTM) for the CC was purchased from Scharlau. The other chemicals were of the highest grade available. ¹H and ¹³C NMR spectra were recorded in DMSO-d6 using a Bruker AM-500 (500 MHz) spectrometer. ESI-MS were recorded on a MSQ_PLUS mass spectrometer (THERMO).

Extraction and Isolation of Standard Compounds

The isolation of polyphenolic compounds was targeted, with major polyphenol peaks from the total extract from UPLC chromatography occurring at 330 nm. The dried young leaves of A crassna (0.9 kg) were extracted under reflux using 100% MeOH (2 m 10 L) at 65 °C. The total MeOH extract was then cooled. A yellow precipitate resulted, which was filtered to obtain 70 g of residue. This was then crystallized several times using hot MeOH to obtain mangiferin (50 g, 3). The filtered extract was evaporated in vacuo to produce a MeOH extract (200 g), which was dissolved in water, and partitioned using CH2Cl2. A precipitate appeared in the CH2Cl2 extract, and was filtered and crystallized in CHCl3-MeOH (8:2) to obtain genkwanin (15 mg, 7). The CH2Cl2 fraction was evaporated in vacuo to obtain 25 g of extract. The water layer was extracted using EtOAc. The precipitate appeared as an EtOAc extract after removing 50% of the solvent. It was then filtered to yield 1.5 g of residue (Fraction E1). The EtOAc fraction was evaporated in vacuo to obtain 45 g of extract. The water layer was extracted using watersaturated n-butanol to give 15 g of n-butanol extract. The aqueous layer was then purified using Diaion HP-20, and eluted using 100% water and 100% methanol. The methanol eluate was evaporated in vacuo to obtain 12 g of dried extract. E1 (1.5 g) was passed through a silica gel column and eluted using CHCl3:MeOH (85:15) to obtain 7 fractions (E1-E7). Fraction E6 was continuously chromatographed on a silica-gel column and eluted with EtOAc-MeOH (9:1) to obtain genkwanin 4'methoxy 5-O-β-primeveroside (35 mg, 6). Fraction E7 was recrystallized in MeOH-H2O (5:5) to yield genkwanin 5-O-β-primeveroside (15 mg, 5). The EtOAc fraction (45 g) was subjected to a silica gel column, and eluted using EtOAc-MeOH (100:0, 100:5, 100:15) to obtain 9 fractions (EA1-EA9). Fraction EA3 (12.5 g) was crystallized in EtOAc to obtain iriflophenone 2-O-α-L-rhamnoside (5.13 g, 4). The n-BuOH fraction was then passed through a silica gel column and eluted using CHCl3-MeOH (10:1; 8:1; 5:1) to obtain 5 subfractions (Bu1-Bu5). Subfraction Bu5 (350 mg) was further separated by RP-18 and eluted using MeOH-H2O (4:6) to obtain 6 further fractions (Bu5.1-Bu5.6). Fraction Bu5.6 (55 mg) was further purified by prep-HPLC, using MeOH-H2O (24:74) as a mobile phase, to yield iriflophenone 3-C-β-D-glucoside (15 mg, 2). A 100% methanol extract (5 g), obtained from the Diaion HP-20 column, was subjected to the RP-18 column, and eluted with a gradient of water/methanol. The 100% MeOH fraction (200 mg) was then separated by prep-HPLC, eluted using MeOH-H2O (15:85), to yield iriflophenone 3,5-C-β-diglucoside (32 mg, 1). After confirming the purity (>95%) by HPLC-UV analysis and recorded 1H and 13C NMR spectra, each isolated compound was authorized as an analytical standard.

Figure 1.

Isolation procedure of primary polyphenols from Aquilaria crassna leaves.

Preparation of Standard Solution

A stock solution containing iriflophenone 3,5-C-β-diglucoside (1) (273.97 µg/mL), iriflophenone 3-C-β-glucoside (2) (614.11 µg/mL), mangiferin (3) (1462.80 µg/mL), iriflophenone 2-O-β-rhamnoside (4) (449.64 µg/mL), genkwanin 5-O-β-primeveroside (5) (134.94 µg/mL), genkwanin 4′-methoxy-5-O-β-primeveroside (6) (30.67 µg/mL) and genkwanin (7) (24.75 µg/mL) in methanol was prepared and diluted to 6 concentrations for the establishment of calibration curves. The limit of detection and quantitation under the chromatographic conditions used were determined based on the response at signal-to-noise ratios of approximately 2 to 10, respectively.

Optimization Extraction Solvent for HPLC

To obtain better extraction rates for the 7 polyphenols under investigation from A crassna leaves, different solvents and extraction methods were applied, including water and MeOH (30%, 60%, 80%, and 100%), with reflux and sonication also being employed, as described in Table 1. All extracts were filtered through a 0.2-µm membrane before injection into the UPLC.

Table 1.

Extraction Methods.

No.	Solvent	Methoda	Temperature (°C)	Duration (mins)
1	H2O	Reflux	100	60
2	MeOH	Reflux	60	60
3	MeOH	Sonication	55	30
4	80% MeOH	Reflux	65	60
5	80% MeOH	Sonication	55	30
6	60% MeOH	Reflux	75	60
7	60% MeOH	Sonication	55	30
8	30% MeOH	Reflux	80	60
9	30% MeOH	Sonication	55	30

^a

Each extraction was conducted in triplicate.

Preparation of HPLC Samples

A 100 mg aliquot of A crassna leaf powder for each different age was placed in a volumetric flask, to which 25 mL MeOH 100% was added. This was then shaken on a Vortex Genie (Scientific Industries, Inc) for 10 s before being subjected to ultrasonication (Branson 5510) for 30 min at 55 °C. The solution was filtered through a 0.22-µm membrane before UPLC analysis.

Method Validation

The precision of the analytical method was determined by intra- and inter-day variations. A 100 mg aliquot of A crassna leaves was extracted and analyzed, as described above. The intraday precision was performed by triplicate extraction and analysis on a single day. The inter-day precision was carried out on 3 different days. Variations were expressed as a percent relative standard deviation (% RSD). A recovery test was used to evaluate the accuracy of this quantification method. A precise amount of each polyphenol was added to a 100-mg A crassna leaf sample, and then extracted and analyzed, as described above. The following formula determined the average recovery: recovery (%) = (observed amount – original amount)/spiked amount × 100%, with % RSD = (SD/mean) × 100%. Using the validated method, samples from 10 groups of leaves were successfully analyzed. Each type of leaf was prepared as an extract, following the procedure described above, and was filtered before injection into the UPLC system. The contents of the 7 polyphenols obtained were calculated, using the formula:

$$\mathit{R}\mathit{i}={\displaystyle \frac{\mathit{S}\mathit{t}}{\mathit{S}\mathit{r}}\times \mathit{C}\mathit{r}\times {\displaystyle \frac{25}{{10}^6\times \mathit{m}}}}$$

where Ri is the polyphenol content (i) in the leaf (%), St is the peak area of polyphenol (i) in the sample solution, Sr is the peak area of polyphenol (i) in the standard solution, Cr is the polyphenol content in the standard solution, and m is the weight of the sample.

α-Glucosidase Inhibitory Assay

α-glucosidase (EC 3.2.1.20) isolated from Saccharomyces cerevisiae (750 UN) and p-nitrophenyl-α-d-glucopyranoside were obtained from Sigma Chemical Co., and acarbose and dimethylsulfoxide from Merck. The inhibitory activity of α-glucosidase was determined following a modified published method. ¹³

Aliquots of 1.5 mM p-nitrophenyl-α-d-glucopyranoside (50 μL) and 0.1 U/mL α-glucosidase (50 μL) in 0.01 M phosphate buffer (pH = 7.0) were added to the sample solution (525 μL) to start the reaction. Each reaction was allowed to occur at 37 °C for 30 min and was then stopped by adding 0.1 M Na₂CO₃ (375 μL). An inhibitory assay quantified the enzymatic activity by measuring the absorbance at 401 nm. One unit of α-glucosidase activity was defined as the amount of enzyme liberating p-nitrophenol at 1.0 μM/min. The IC₅₀ value was defined as the concentration of α-glucosidase inhibitor that inhibited 50% of the α-glucosidase activity. Acarbose, a known α-glucosidase inhibitor, was used as the positive control.

Protein and Ligand Preparation

The 3D structure of α-glucosidase (Protein Data Bank [PDB 2ZE0]) was obtained from the PDB (http://www.rcsb.org/pdb). The selected 2ZE0 was fixed using a CHARMM (Chemistry at Harvard Macromolecular Mechanics) force field in DS 2.5 (DS, Accelrys Software), which added up the hydrogen atoms, partial charges, and missing residues that are appropriately used for the processes of molecular docking. The 3D structures of 3 natural compounds—mangiferin (3), genkwanin-5-O-β-D-primeveroside (5), and genkwanin (7)—from the A crassna leaves were retrieved from the PubChem database or prepared by MarvinSketch (ChemAxon). In the final step, all ligands were converted into dockable pdbqt format, utilizing the Open Babel toolbox.

Molecular Docking

Virtual screening of the bioactive compounds on α-glucosidase [PDB 2ZE0] was carried out using AutoDock Vina software.^14,15 A grid box covering the active protein site was generated using the following parameters: center_x = 8.24; center_y = 26.90; center_z = 23.69; size_x = 25; size_y = 25; and size_z = 25. The docking scores are reported in kcal/mol, and the compounds were ranked by their docking scores. Finally, the molecular interactions between the proteins and selected ligands were visualized using Discovery Studio Visualizer software.

Statistical Analysis

The statistical significance was evaluated using an analysis of variance and SPSS software version 20. A P value of <.05 was considered to be statistically significant.

Results and Discussion

Isolation of Phenolic Compounds

Seven polyphenolic compounds were isolated from A crassna leaves using liquid–liquid partitioning, followed by normal-phase CC, RP-18 silica-gel CC, and/or semi-prep HPLC. The structures of the isolated compounds [iriflophenone 3,5-C-β-diglucoside (1), iriflophenone 3-C-β-glucoside (2), mangiferin (3), iriflophenone 2-O-β-rhamnoside (4), genkwanin 5-O-β-primeveroside (5), genkwanin 4’-methoxy-5-O-β-primeveroside (6), and genkwanin (7) (Figure 2)] were identified by comparing their spectroscopic data with literature data.^1,9,13 The compounds were all significant components in the UPLC chromatogram at 330 nm, and each indicated high purity of more than 95%. Mangiferin (3) was isolated from A crassna leaves in a high yield of ∼5.5% using a straightforward approach. Its poor solubility allowed it to be easily collected, as it precipitated after cooling the MeOH total extract. Using HPLC-guided isolation, iriflophenone 3,5-C-β-diglucoside (1) was found in the aqueous layer following liquid–liquid partitioning of n-BuOH, due to its high polarity, and could then be purified by RP-18 CC, followed by prep-HPLC.

Figure 2.

Structures of major polyphenols isolated from Aquilaria crassna leaves.

α-Glucosidase Inhibitory Activity of the Isolated Compounds and Extracts

The α-glucosidase inhibitory activity of compounds 1-7 was examined. Compound 7 (genkwanin) exhibited the most potent inhibitory effect, with an IC₅₀ value of 24.0 ± 3.6 µM, which was 4 times greater than that of acarbose, with an IC₅₀ value of 105.2 ± 11.7 µM. Genkwanin-5-O-β-D-primeveroside (5) presented significant α-glucosidase inhibitory activity, with an IC₅₀ value of 76.5 ± 6.8 µM, whereas the activity of mangiferin (3) was 217.6 ± 10.5 µM. Interestingly, genkwanin-4′-methoxy-5-O-β-D-primeveroside (6), a derivative of genkwanin-5-O-β-D-primeveroside (5), showed no inhibitory effect. Genkwanin-4′-methoxy-5-O-β-D-primeveroside (6) contains one methoxy group at C-4′ of the genkwanin skeleton. This linkage may reduce the α-glucosidase inhibitory activity of these flavone glycosides. Compounds 1, 2, and 4 showed no activity.

In the next study, the extracts from leaves 1 to 10 were tested for their α-glucosidase inhibitory activity using 2 concentrations of 1500 and 750 µg/mL. The results are presented in Table 2.

Table 2.

Screening for α-Glucosidase Inhibitory Effects of the Extracts of First—10th Leaves 1 to 10.

Samples	% inhibitiona
	1500 µg/mL	750 µg/mL
L1b	97.8 ± 0.4	89.5 ± 1.2
L2	102.6 ± 0.4	92.4 ± 1.4
L3	85.7 ± 0.3	73.7 ± 1.0
L4	90.1 ± 0.5	57.4 ± 1.1
L5	78.8 ± 0.8	57.3 ± 1.8
L6	70.5 ± 0.7	61.0 ± 2.8
L7	72.7 ± 0.4	56.2 ± 2.4
L8	55.1 ± 0.4	42.9 ± 3.6
L9	86.9 ± 1.4	43.6 ± 1.2
L10	43.1 ± 0.9	30.0 ± 3.2
Acarbosec	50.05 ± 2.3

^a

Experiments carried out in 3 replicates.

^b

L1: first leaves.

^c

Positive compound.

The α-glucosidase inhibition activity of the leaves indicates a gradual decrease in activity from leaf 1 to leaf 10 in both of the 2 concentrations (1500 and 750 µg/mL). These results are understandable because the highest polyphenol contents were found in leaves 1 and 2, and gradually decreased to leaf 10. The younger leaf samples (1-3) showed α-glucosidase inhibition of 89.5%, 92.4%, and 73.7%, respectively. This might indicate a correlation between the polyphenol content of the leaf samples and α-glucosidase inhibitory activity. Since leaves 1 to 3 showed the most potent α-glucosidase inhibitory activity, they were subsequently tested for their IC₅₀ values.

Leaf sample 1 exhibited the most potent α-glucosidase inhibiting effect, with an IC₅₀ value of 169.8 µg/mL, followed by leaf 2, with an IC₅₀ value of 248.0 µg/mL, and leaf 3, with an IC₅₀ value of 507.2 µg/ µg/ µg/mL (acarbose, positive control 50.0 ± 2.3 µg/ µg/mL) . These results are consistent with the screening results that indicated that the α-glucosidase inhibitory effect increased with the polyphenol content in the leaf samples. In terms of tea quality, this suggests that for a high polyphenol content and vigorous α-glucosidase inhibitory activity, the leaves should be collected from the ends of the branches.

Molecular Docking Results

α-Glucosidase inhibitors are common oral antidiabetic drugs used to control carbohydrates, typically by converting them into simple sugars, allowing them to be absorbed by the intestines. The 3D structures of the most potent compounds, such as mangiferin (3), genkwanin-5-O-β-D-primeveroside (5), and genkwanin (7), were constructed and used in the docking process with the α-glucosidase active site 2ZE0. In previous studies, acarbose has been shown to inhibit α-glucosidase. Therefore, we decided to choose acarbose as a reference inhibitor. The ligand bound with the α-glucosidase protein showed the crucial amino acid residue (Arg197) in the localized active site. Acarbose docking with α-glucosidase indicated that the model structure of the ligand–protein interaction between acarbose and α-glucosidase had formed 3 interactions of the ligand molecules with Asp199, Arg197, and Asn61, as illustrated in Figure 2A and B for the 2D and 3D structures, respectively. The key point is that, among these 3 compounds, genkwanin 5-O-β-D-primeveroside showed the highest binding affinity to the α-glucosidase active site 2ZE0, having a docking score of −10.5 Kcal/mol. Genkwanin 5-O-β-D-primeveroside showed interaction with α-glucosidase through hydrogen bonds at Tyr63, Asp199, Arg197, Glu256, and His203, and through van der Waals interactions at Val383, Asn61, Asp326, Asp98, His325, Asn324, Trp49, Phe282, Asn258, Phe225, Met229, ILE143, Phe163, Gln167, Asp60, and Val383. The interaction analysis revealed that genkwanin and mangiferin also formed several significant interactions with α-glucosidase. As shown in Figure 2 and Table 4, genkwanin formed 2 hydrogen bonds, with Glu256 and Arg411, forming Ala200, Asp199, Gln167, Asp60, Phe144, Asp382, Arg407, Asn61, Asp326, and Arg197 through van der Waals interactions. Meanwhile, mangiferin formed 2 hydrogen bonds, with Asp60 and Arg411, and had several hydrophobic interactions with Asp199, Ala200, His103, Gln167, Phe144, Ser384, Val383, Ala59, Asn58, Gly62, Arg407, Asp326, His325, Arg197, and Glu256. These interactions were somewhat similar to the interactions between the positive control of acarbose and 2ZE0 of this enzyme (Figure 3, Table 3).

Figure 3.

Molecular docking of α-glucosidase using 2ZEO active site. (A) 2D and (B) B 3D structures of acarbose. (C) 2D and (D) 3D structure of genkwanin. (E) 2D and (F) 3D structure of genkwanin 5-O-β-D-primeveroside. (G) 2D and (H) 3D structure of mangiferin.

Table 3.

Molecular Docking Simulation Results for Inhibitory Complexes Between the Ligands and Targeted Protein (PTP1P-2ZE0).

No	Liganda	Docking score (kcal/mol)	Hydrogen bond	Van der Waals interaction
1	Genkwanin	−8.9	Glu256, Arg411	Ala200, Asp199, Gln167, Asp60, Phe144, Asp382, Arg407, Asn61, Asp326, Arg197
2	Genkwanin 5-O-β-D-primeveroside	−10.5	Tyr63, Asp199, Arg197, Glu256, His203	Val383, Asn61, Asp326, Asp98, His325, Asn324, Trp49, Phe282, Asn258, Phe225, Met229, ILE143, Phe163, Gln167, Asp60, Val383
3	Mangiferin	−9.7	Asp60, Arg411	Asp199, Ala200, His103, Gln167, Phe144, Ser384, Val383, Ala59, Asn58, Gly62, Arg407, Asp326, His325, Arg197, Glu256

^a

Binding sites of tested compounds in α-glucosidase using 2ZEO and Autodock 4.0 molecular docking software.

Polyphenol Content of Different-Aged Leaves

Extraction Method Development

In order to obtain an effective extraction method, ultrasonication and reflux were used in tandem with the variables involved in the procedure, including methanol at different concentrations (30%, 60%, 80%, and 100%), temperature (55, 60, 80, and 100 °C), extraction duration (30-60 min), and extract repetition (3 times). The results showed that reflux using 60% MeOH gave a better extraction yield than ultrasonication with MeOH 100% and was suitable for extracting almost all the major polyphenolic compounds (data not shown). However, in the reflux method using 60% MeOH, mangiferin precipitated after the extract cooled. Therefore, 100% MeOH with ultrasonication is the most appropriate extraction method for the 7 primary polyphenols in A crassna leaves.

Optimization of HPLC Conditions

The optimal mobile phase composition for the analysis of iriflophenon 3,5-C-β-D-diglucoside (1), iriflophenon 3-C-β-D-glucoside (2), mangiferin (3), iriflophenon 2-O-α-D-rhamnoside (4), genkwanin 5-O-β-primeveroside (5), genkwanin-4’-methoxy-5-O-β-primeveroside (6), and genkwanin (7) from 100% MeOH with ultrasonication from A crassna leaves was selected by performing several HPLC runs using various concentrations of acetonitrile in acid-buffer solutions as the mobile phase. Among the buffer solutions, separation was achieved using a mobile phase of 0.2% acetic acid/acetonitrile (A) and 0.2% acetic acid (B) in the following sequence: 0 to 1 min: 10% A; 1 to 5 min: 13% A; 5 to 10 min: 13% A; 10 to 13 min: 25% A; 13 to 14 min: 25% A; 14 to 15 min: 30% A; 15 to 18 min: 35% A; 18 to 19 min: 64% A; 19 to 22 min: 65% A; 22 to 23 min: 65% A; 23 to 28 min: 85% A; and 28 to 31 min: 10% A. The column temperature was set at 40 °C. The HPLC peaks of the standard compounds (1-7) in each leaf were verified. The retention times of iriflophenon 3,5-C-β-D-diglucoside (1), iriflophenon 3-C-β-D-glucoside (2), mangiferin (3), iriflophenon 2-O-α-D-rhamnoside (4), genkwanin 5-O-β-primeveroside (5), genkwanin-4’-methoxy-5-O-β-primeveroside (6), and genkwanin (7) were 6.48, 6.94, 10.92, 13.94, 16.54, 19.55, and 22.18 min, respectively.

Validation of a Developed Method

The detection method calibration curves exhibited good linear regressions under the optimized chromatographic conditions, as shown in Table 4.

Table 4.

Validated Analytical Parameters for the UPLC/UV Quantification of 7 Major Polyphenols in Aquilaria crassna Leaves.

Compounds	Linear regression equation	R 2	Range (mg/mL)	Accuracy (recovery) (%)	RSD (%)
1	y = 1401.5x	0.9981	3.5-274	101.8	2.83
2	y = 4792.8x	0.9986	7.8-614	95.9	2.25
3	y = 4103x	0.9963	18.8-1462	103.4	1.57
4	y = 3845.6x	0.9999	5.8-450	97.4	2.27
5	y = 7770.3x	0.9998	1.4-135	100.6	2.81
6	y = 26737.7x	0.9999	0.4-30.7	99.4	1.91
7	y = 16546.9x	0.9999	0.3-24.7	102.7	1.15

Abbreviation: UPLC, ultra-performance liquid chromatography.

Polyphenol Content of A crassna Leaves

The major polyphenolic components of the leaves included compounds 1-3, which had the highest content in the youngest leaves (leaf number 1), gradually decreasing to 25% to 35% in leaf number 5, then not changing much from leaf number 5 to leaf number 10 (the oldest). Compound 4 was present in high amounts in leaf 1, decreasing gradually in leaves 2 to 4, being reduced to about 50% in leaf 5. However, the content of 4 was relatively stable in the remaining leaves. There was a low proportion of polyphenol components 5-7, with fluctuations in the leaf contents due to different ages. Compound 7, however, had a high content in leaves 1 and 2, showed a sharp decrease in leaf 3 (by ∼85%), and then, the content became relatively stable up to leaf 10. The polyphenolic components were mainly iriflophenone 3,5-C-β-D-diglucoside (1), iriflophenon 3-C-β-D-glucoside (2), mangiferin (3), and riflophenon 2-O-α-D-rhamnoside (4). These compounds had their highest contents in the young leaves 1 to 3; their contents decreased as the leaves got older. Some of the contents of the minor polyphenols in the leaves—genkwanin 5-O-β-primeveroside (5), genkwanin-4'-methoxy-5-O-β-primeveroside (6), and genkwanin (7)—fluctuated with increasing leaf age; for example, compound 7 was abundant in the young leaves 1 and 2 but decreased as the leaves aged. The others (5 and 6) had low concentrations in the young leaves 1 and 2, but increased in the middle-aged (3-7) leaves before slightly decreasing in the old (8-10) leaves (Table 5).

Table 5.

Polyphenol Contents (mg/g) in Leaves of Different Ages.

Compounds	Leaf’s agea
	1	2	3	4	5	6	7	8	9	10
1	24.12 ± 0.48	16.84 ± 0.2	11.96 ± 0.04	11.23 ± 0.2	6.95 ± 0.11	8.76 ± 0.16	8.39 ± 0.14	9.24 ± 0.18	7.65 ± 0.16	7.94 ± 0.05
2	33.29 ± 0.60	22.71 ± 0.44	19.51 ± 0.35	19.33 ± 0.29	11.11 ± 0.13	12.58 ± 0.17	12.77 ± 0.26	9.89 ± 0.16	10.89 ± 0.15	7.67 ± 0.13
3	150.07 ± 2.27	78.74 ± 1.3	52.73 ± 0.81	46.84 ± 0.44	36.09 ± 0.48	32.18 ± 0.29	33.68 ± 0.33	28.28 ± 0.21	23.16 ± 0.11	21.69 ± 0.43
4	11.87 ± 0.13	8.65 ± 0.15	7.37 ± 0.11	7.44 ± 0.03	6.69 ± 0.04	5.02 ± 0.05	5.55 ± 0.09	4.37 ± 0.05	4.35 ± 0.04	5.25 ± 0.05
5	1.05 ± 0.01	0.9 ± 0.01	2.79 ± 0.02	3.39 ± 0.03	3.06 ± 0.03	3.20 ± 0.02	2.76 ± 0.02	3.14 ± 0.03	3.25 ± 0.07	3.13 ± 0.03
6	0.13 ± 0.002	0.12 ± 0.003	0.36 ± 0.007	0.42 ± 0.006	0.31 ± 0.004	0.32 ± 0.003	0.25 ± 0.003	0.18 ± 0.004	0.16 ± 0.002	0.23 ± 0.003
7	1.94 ± 0.026	2.13 ± 0.031	0.88 ± 0.005	0.32 ± 0.002	0.26 ± 0.002	0.33 ± 0.002	0.28 ± 0.002	0.41 ± 0.009	0.1 ± 0.002	0.38 ± 0.005
Total content	222.48 ± 2.76	130.09 ± 1.74	95.61 ± 0.53	88.97 ± 0.32	64.47 ± 0.33	62.39 ± 0.17	63.68 ± 0.74	55.50 ± 0.06	49.54 ± 0.12	46.30 ± 0.51

^a

Data are presented as the mean ± SD of results from 3 independent experiments.

Conclusion

In this study, 7 phenolic compounds [iriflophenon 3,5-C-β-D-diglucoside (1), iriflophenon 3-C-β-D-glucoside (2), mangiferin (3), iriflophenon 2-O-α-D-rhamnoside (4), genkwanin 5-O-β-primeveroside (5), genkwanin-4’-methoxy-5-O-β-primeveroside (6), and genkwanin (7)] were isolated from the leaves of A crassna. They were subjected to HPLC to determine standards for quantification. A reliable and straightforward reversed-phase HPLC method was performed to determine the content of the secondary metabolites in different-aged A crassna leaves. The contents of 1-4 contributed to the total polyphenolic contents in all leaves. In addition, their contents significantly decreased from the youngest (leaf 1) to the oldest leaf (leaf 10). The total polyphenol contents were higher in the first 4 leaves. This study showed that the content of the main phenolic contents in the different leaves of A crassna changed with the age of the leaves. This result suggests that young agarwood leaves should be harvested to obtain the higher polyphenol contents. Genkwanin (7) was found to be the most active compound for glucosidase inhibition. A molecular docking analysis, using AutoDock 4.0 software, was performed in order to understand the binding interactions of the active compounds and locate the active sites of the enzymes. These results suggested that A crassna and its polyphenols may prevent the development and progression of diabetes through α-glucosidase inhibition and be a good source of anti-diabetes herbal materials. The method of analysis of polyphenols content in the leaves of A crassna helps orient the standardization of this plant material, as well as for the collection, classification, and processing of materials to ensure the development of products with high polyphenol content.