Soluble epoxide hydrolase inhibitory activity of phenolic components from the rhizomes and roots of Gentiana scabra

Abstract

Two new (1 and 2) and seven known phenolic compounds were isolated from a methanol extract of the rhizomes and roots of Gentiana scabra. Their structures were identified by spectroscopic analysis and comparing with reported values. The inhibitory effects of each compound on soluble epoxide hydrolase (sEH) were evaluated. Among them, compounds 2, 6, and 9 potently inhibited sEH activity.

Key words:

• Gentiana scabra
• Gentianaceae
• phenolic compound
• soluble epoxide hydrolase

The genus Gentiana, family Gentianaceae, comprises over 400 species that are widely distributed in alpine habitats within the temperate regions of Asia, Europe, and America.¹⁾ Gentiana scabra Bunge, a perennial herbaceous plant, grows mainly in temperate climates, such as China, Japan, South Korea, Russia, and North America. Rhizomes and roots of G. scabra are used in traditional Chinese medicine for the treatment of inflammation, anorexia, indigestion, and gastric infections.²⁾ Additionally, G. scabra has been shown to protect the liver, inhibit liver dysfunction, and promote gastric acid secretion, rendering it a popular ingredient in Chinese herbal medicine and health products.³⁾ Various secondary metabolites of G. scabra, such as secoiridoids, triterpenoids, flavonoids, xanthones, and alkaloids, have also been reported.^4–9) Current research on G. scabra focuses on its primary functional components, including gentiopicroside, swertiamarin, and sweroside.⁷⁾ These compounds exhibit smooth muscle relaxation, as well as antibacterial and free radical scavenging activities.^10,11) Although much research has focused on the biological activity of G. scabra, its soluble epoxide hydrolase (sEH) inhibitory activity has not yet been fully elucidated. In this study, the inhibitory effects of each isolated compound on sEH were evaluated.

sEH is a component of the arachidonic acid cascade, which plays an important role in the metabolism of eicosanoid epoxides.¹²⁾ sEH catalyzes the hydrolysis of epoxyeicosatrienoic acids (EETs) into the corresponding dihydroxyeicosatrienoic acids¹³⁾ and is expressed in various cells and tissues, including the liver, kidney, vascular endothelium, leukocytes, and adipocytes.^14–16) EETs have four major isomers: 5,6-EET, 8-9-EET, 11,12-EET, and 14,15-EET, all of which are produced from arachidonic acid by cytochrome P450s (CYPs) in the eicosanoid pathway.¹⁷⁾ These isomers are endothelium-derived hyperpolarizing factors with a variety of biological functions, such as anti-inflammatory, antihypertensive, cardioprotective, vasodilatory, and angiogenic effects.^18,19) Moreover, several studies have reported other potential anti-inflammatory effects of EETs, in animal models, including decreased production of nitric oxide, cytokines, and pro-inflammatory mediators.^18,20) Therefore, sEH inhibitors that decrease EETs levels could serve as novel agents for the treatment of hypertension and inflammation. Recent studies have developed several urea-type sEH inhibitors such as 12-(3-adamantan-1-yl-ureido)-dodecanoic acid butyl ester (AUDA), 1-trifluoromethoxyphenyl-3-(1-acetylpiperidin-4-yl) urea (TPAU), trans-4-[4-(3-adamantan-1-yl-ureido)-cyclohexyloxy]-benzoic acid (t-AUCB), and sorafenib.¹⁸⁾ However, several problems have arisen during clinical trials, such as poor solubility and rapid metabolism in microsomes. Therefore, novel sEH inhibitors from natural products that overcome these problems are urgently needed. In the study described herein, we report the isolation of nine phenolic compounds from the rhizomes and roots of G. scabra, as well as their sEH inhibitory activities.

Dried rhizomes and roots (2.5 kg) of G. scabra were extracted with MeOH (10 L × 3) under reflux. The MeOH extract (670.0 g) was suspended in water and partitioned with CHCl₃ and n-BuOH. The CHCl₃ fraction (90.0 g) was subjected to silica gel (8 × 30 cm) column chromatography with n-hexane–EtOAc (20:1, 5:1), CHCl₃–acetone (5:1), and CHCl₃–MeOH–H₂O (6:1:0, 3:1:0.1), resulting in five fractions (Fr. 1A–1E). Fraction 1D (28.6 g) was then subjected to silica gel (4 × 30 cm) column chromatography with CHCl₃–MeOH (20:1, 10:1, 7:1, 5:1, 3:1) elution solvent to yield eight subfractions (Fr. 1D-1–1D-8). Fraction 1D-1 was separated using YMC (1.5 × 80 cm) column chromatography with a MeOH–acetone–H₂O (1:1:1–4:4:1) elution solvent to yield compound 6 (23.7 mg; 0.0035%). Fraction 1D-2 was subjected to silica gel (1.0 × 80 cm) column chromatography with CHCl₃–MeOH (25:1 to 20:1) elution solvent to yield compounds 2 (39.6 mg; 0.0059%) and 9 (27.0 mg; 0.0040%). Fraction 1D-8 was separated using YMC (1.5 × 80 cm) column chromatography with a MeOH–acetone–H₂O (0.3:0.3:1–1:1:1) elution solvent to yield compounds 5 (187.0 mg; 0.028%) and 8 (87.6 mg; 0.013%). The n-BuOH fraction (190.0 g) was subjected to silica gel (8 × 30 cm) column chromatography with CHCl₃–MeOH–H₂O (12:1:0, 6.5:1:0, 4:1:0, 2.5:1:0.1) to yield four fractions (Fr. 2A–2D). Fraction 1B (15.0 g) was subjected to YMC (3 × 80 cm) column chromatography with a MeOH–H₂O (0.05:1–0.95:1) elution solvent to yield eight subfractions (Fr. 2B-1–2B-8). Fraction 1B-3 was subjected to silica gel (2.5 × 80 cm) column chromatography with CHCl₃–MeOH (15:1–8:1) elution solvent to yield compounds 1 (9.3 mg; 0.0014%), 3 (169.0 mg; 0.025%), 4 (105.0 mg; 0.016%), and 7 (39.1 mg; 0.0058%).

Compound 1 was obtained as a yellowish powder. The molecular structure of 1 was inferred to be C₁₈H₂₄O₉ from a pseudomolecular ion peak [M + Na]⁺ at m/z 407.1353 in the HR-ESI-MS spectrum (calcd for C₁₈H₂₄O₉Na, 407.1313). The ¹H-NMR spectrum of 1 (Table 1) revealed two aromatic proton signals at δ_H 7.55 (brs, H-2) and 7.88 (brs, H-6), an olefinic proton signal at δ_H 5.26 (t, J = 6.8 Hz, H-8), an anomeric proton signal at δ_H 5.57 (d, J = 6.5 Hz, H-1′), a methylene group signal at δ_H 3.20 (d, J = 6.8 Hz, H-7), two methyl group signals at δ_H 1.51 (s, H-10, 11), and among other signals (δ_H 3.97–4.43) belonging to a sugar moiety. The sugar moiety was identified as d-glucose by acid hydrolysis. The ¹³C NMR spectrum showed 18 signals, including a carboxyl group [δ_C 174.9 (COOH)], a benzene ring [δ_C 117.9 (C-5), 122.2 (C-2), 123.9 (C-6), 130.7 (C-1), 146.6 (C-3), 152.1 (C-4)], a glucosyl group [δ_C 61.7 (C-6′), 70.8 (C-4′), 74.7 (C-2′), 78.1 (C-3′), 78.5 (C-5′), 103.6 (C-1′)], and a prenyl group [δ_C 17.5 (C-11), 25.5 (C-10), 33.8 (C-7), 124.0 (C-8), 131.8 (C-9)]. Analyses of the ¹H and ¹³C NMR spectral data indicated that compound 1 was a tetra-substituted phenolic glycoside. The position of substituent groups was determined by HMBC spectrum. Key correlations between H-7 (δ_H 3.20)/C-2 (δ_C 122.2), C-6 (δ_C 123.9), and C-1 (δ_C 130.7) indicated that the prenyl group was located at C-1. Correlations between H-2 (δ_H 7.55)/C-3 (δ_C 146.6) and H-1′ (δ_H 5.57)/C-3 (δ_C 146.6) indicated that the glucosyl group was located at C-3. Both H-2 (δ_H 7.55) and H-6 (δ_H 7.88) exhibited correlations with C-4 (δ_C 152.1); however, only H-6 (δ_H 7.88) correlated with the carboxyl group (δ_C 174.9), indicating that the carboxyl group was located at C-5 and the hydroxy group was located at C-4 (Fig. 1(B)). Based on these data, compound 1 was characterized as 4-hydroxy-1-prenyl-5-(3-O-β-d-glucopyranosyl)-benzoic acid.

Table 1. The ¹H and ¹³C NMR spectroscopic data of compounds 1 and 2.

	1		2
	δC^a^,^b	δH^a^,^c (J in Hz)	δC^a^,^b	δH^a^,^c (J in Hz)
1	130.7		130.8
2	122.2	7.55 brs	160.4
3	146.6		116.1	7.11 d (8.0)
4	152.1		130.1	7.98 dd (8.0, 2.0)
5	117.9		123.0
6	123.9	7.88 brs	129.0	8.53 d (2.0)
7	33.8	3.20 d (6.8)	122.8	7.80 d (16.0)
8	124.0	5.26 t (6.8)	137.5	7.01 d (16.0)
9	131.8		73.8
10	25.5	1.51 s	70.9	3.96 m
11	17.5	1.51 s	25.4	1.67 s
1′	103.6	5.57 d (6.5)
2′	74.7	4.29 m
3′	78.1	4.28 m
4′	70.8	4.28 m
5′	78.5	3.97 m
6′	61.7	4.43 m
		4.33 m
C=O	174.9		166.9
OCH3			51.4	3.74 s

Note:

Assignments were done by HMQC and HMBC experiments; J values (Hz) are in parentheses.

^a Measured in pyridine-d₅.

^b 150 MHz.

^c 600 MHz.

Fig. 1. Structures of compounds 1–9 (A) and ¹H-¹H COSY and HMBC correlations (B) of compounds 1 and 2.

Compound 2 was obtained as a yellowish gum. The molecular formula was defined as C₁₃H₁₆O₅ from a pseudomolecular ion peak [M + H]⁺ at m/z 253.1072 (calcd. 253.1071) in the HR-ESI-MS spectrum. The ¹H NMR spectrum of 2 (Table 1) revealed three aromatic proton signals at δ_H 7.11 (d, J = 8.0 Hz, H-3), 7.98 (dd, J = 8.0, 2.0 Hz, H-4), and 8.53 (d, J = 2.0 Hz, H-6), belonging to an ABX system. Two olefinic proton signals at δ_H 7.01 (d, J = 16.0 Hz, H-8) and 7.80 (d, J = 16.0 Hz, H-7) indicated the presence of a trans double bond. Moreover, an oxymethylene group [δ_H 3.96 (m, H-10)], a methoxyl group [δ_H 3.74 (s, OCH₃)], and a methyl group [δ_H 1.67 (s, H-11)] were also observed. Correspondently, the ¹³C NMR spectrum (Table 1) exhibited 13 signals, including a carbonyl group signal at δ_C 166.9 (COCH₃), a benzene ring signal at δ_C 116.1 (C-3), 123.0 (C-5), 129.0 (C-6), 130.1 (C-4), 130.8 (C-1), and 160.4 (C-2), two olefinic carbon atoms at δ_C 122.8 (C-7), and 137.5 (C-8), an oxygenated quaternary carbon atom at δ_C 73.8 (C-9), an oxygenated methylene carbon atom at δ_C 70.9 (C-10), a methoxyl carbon atom at δ_C 51.4 (OCH₃), and a methyl carbon atom at δ_C 25.4 (C-11). The HMBC spectrum revealed correlations between H-7 (δ_H 7.80)/C-9 (δ_C 73.8), H-8 (δ_H 7.01)/C-9 (δ_C 73.8), C-10 (δ_C 70.9), and C-11 (δ_C 25.4), and H-11 (δ_H 1.67)/C-9 (δ_C 73.8) and C-10 (δ_C 70.9), which were indicative of a 1,2-dihydroxy-2-methyl-3-butenyl group in compound 2. The absolute configuration of 2 was determined as R by comparing the sign of the optical rotation (+62.8°) with the literature value.^21,22) The position of three substituent groups was determined by HMBC spectrum analysis. Key correlations between H-7 (δ_H 7.80)/C-6 (δ_C 129.0), C-1 (δ_C 130.8), and C-2 (δ_C 160.4) indicated that the 1,2-dihydroxy-2-methyl-3-butenyl group was located at C-1. Correlations between H-4 (δ_H 7.98) and H-6 (δ_H 8.53)/C = O (δ_C 166.9) indicated that the carboxyl methyl ester group was attached to C-5. In addition, correlations between H-4 (δ_H 7.98), H-6 (δ_H 8.53), and H-7 (δ_H 7.80)/C-2 (δ_C 160.4) revealed that the hydroxy group was located at C-2 (Fig. 1(B)). Accordingly, the structure of compound 2 was determined to be (R)-2-hydroxy-1(1,2-dihydroxy-2-methyl-3-butenyl)-5-benzoic acid methyl ester.

By comparing our physical and spectroscopic data with those reported in the literature, the other known compounds were identified as 1-​O-​caffeoyl glucose (3),²³⁾ siringin (4),²⁴⁾ 3-​(β-​d-​glucopyranosyloxy)​-​2-​hydroxy-​benzoic acid methyl ester (5),⁴⁾ berchemol (6),²⁵⁾ koaburaside (7),²⁶⁾ benzyl-​O-​β-​d-​glucopyranoside (8),²⁷⁾ and (2S)​-3-​(4-​hydroxy-​3-​methoxyphenyl)​-​1,​2-​propanediol (9).²⁸⁾ Here, we report for the first time the isolation of compounds 6–9 from G. scabra (Fig. 1(A)).

The inhibitory activity of isolated compounds against sEH was also evaluated. The amount of 6-methoxy-2-naphthaldehyde produced from the substrate (PHOME) was quantified in the presence or absence of compounds 1–9 using a fluorescence photometer at wavelengths of 330 and 465 nm, respectively.^29,30) AUDA served as a positive control (IC₅₀ = 14.2 ± 3.2 nM). First, the inhibitory activity of all nine compounds was evaluated against 50 μM sEH. Compounds 2, 6, and 9 exhibited the highest inhibitory activities, with inhibition effects of 73.2 ± 1.3, 114.3 ± 2.9, and 68.8 ± 4.5% of the control value, respectively. In contrast, compounds 1, 3–5, and 7 demonstrated moderate inhibitory activity, with inhibition effects ranging from 41.1 to 53.3% of the control value. Compound 8 lacked inhibitory activity in the bioassay system (Table 2). Compounds 2, 6, and 9 were further selected to examine IC₅₀ values using a dose-dependent evaluation. Results showed that compounds 2, 6, and 9 exerted sEH inhibitory activity in dose-dependent manner, with IC₅₀ values of 27.3 ± 0.8, 8.2 ± 2.1, and 26.9 ± 1.7, respectively (Table 2; Supplementary Material: Fig. S1). Interestingly, upon examination of the structure–activity relationship of isolated phenolic compounds, we found that compounds 2, 6, and 9, which exhibited highest inhibitory activities, contain more hydroxy groups than the other compounds. Correspondently, compounds 1, 3–5, and 7, which showed moderate inhibitory effects, contain only one or two hydroxy groups. In contrast to compound 8, which lacks a hydroxy group, does not exhibit inhibitory activity. Therefore, the number of hydroxy groups in these structures may represent a key functional element.

Table 2. The sEH inhibitory activities of compounds 1–9.

Compound	Inhibitory activity of compounds on sEH^c
	50 μM(%)	IC50 (μM)	Type (Ki^d/Kis^e, μM)
1	50.4 ± 10.8	N.T^b	N.T^b
2	73.2 ± 1.3	27.3 ± 0.8	Mixed (23.7 ± 2.1/2.1 ± 0.7)
3	53.3 ± 2.3	N.T^b	N.T^b
4	41.1 ± 2.0	N.T ^b	N.T^b
5	40.6 ± 1.6	N.T^b	N.T^b
6	114.3 ± 2.9	8.2 ± 2.1	Mixed (4.2 ± 1.1/0.5 ± 1.3)
7	43.9 ± 0.1	N.T^b	N.T^b
8	8.2 ± 3.6	N.T^b	N.T^b
9	68.8 ± 4.5	26.9 ± 1.7	Non-competitive (11.7 ± 2.7)
AUDA^a		14.2 ± 3.2 nM

^a Positive control.

^b N.T = Not tested.

^c Tested compounds were examined in a set of experiments repeated three times.

^d Values of inhibition constant.

^e Values of enzyme–substrate complex constant.

Lineweaver–Burk plots were used to determine the inhibitory mode of compounds 2, 6, and 9. Compounds 2 and 6 demonstrated changes in the apparent V_max and km values, indicating that these compounds induce a mixed type of inhibition (Supplementary Material: Fig. S2-A and B), whereas compound 9 only changed the apparent V_max value (Supplementary Material: Fig. S2-C). Therefore, we conclude that 9 is a noncompetitive inhibitor. Moreover, K_i (free enzyme) and K_is (enzyme–substrate complex) binding constants were obtained from the secondary replots of slope and y-intercept vs. [I], which were fitted to Equations (2) and (3) (Supplementary Material: Fig. S2). Table 2 indicates their values as follows: compound 3, K_i = 23.7 ± 2.1 μM and K_is = 2.1 ± 0.7 μM; compound 6, K_i = 4.2 ± 1.1 and K_is = 0.5 ± 1.3 μM; compound 9 K_i = 11.7 ± 2.7 μM).

In this study, nine phenolic compounds (1–9) were isolated from G. scabra. Their inhibitory effects on sEH were further evaluated. To our knowledge, this is the first report on the phenolic components of G. scabra and their inhibitory effects on sEH. Notably, our data suggest that phenolic components may represent new bioactive components of this plant and provide a scientific basis for the complement of the active components in G. scabra. In particular, compounds 2, 6, and 9 may have potential as natural sEH inhibitors or as lead molecules for developing new inhibitors sEH.

Supplemental material

The supplemental material for this paper is available at http://dx.doi.org/10.1080/09168451.2014.1002451.

Additional information

Funding

This study was supported by the Priority Research Center Program [2009–0093815] through the National Research [Foundation of Korea (NRF) and a grant [K14050] awarded to the Korean Institute of Oriental Medicine funded by the Ministry of Education, Science and Technology (MEST), Republic of Korea.