Chemical Structures and Hepatoprotective Effects of Constituents from Cassia auriculata Leaves

Abstract

An 80% aqueous acetone extract of Cassia auriculata leaves was found to show a protective effect on D-galactosamine-induced cytotoxicity in primary cultured mouse hepatocytes. From the 80% aqueous acetone extract, we isolated a new benzocoumarin glycoside, avaraoside I (1), and a new flavanol dimer, avaraol I (2), together with 29 known constituents. The structures of the new compounds were elucidated on the basis of chemical and physicochemical evidence. In addition, three isolated compounds, pseudosemiglabrin (15, 0.0011%), (2S)-7,4′-dihydroxyflavan(4β→8)-catechin (22, 0.00075%), and (2S)-7,4′-dihydroxyflavan(4β→8)-gallocatechin (23, 0.092%), displayed hepatoprotective effects equivalent to that of the hepatoprotective agent, silybin.

Cassia auriculata LINN. (Leguminosae) is a perennial shrub called Avarai in Tamil, and is distributed in India, Sri Lanka, etc. This plant has been used as a remedy for diabetes, rheumatism, asthma, and skin diseases in the Ayurvedic system of traditional Indian medicine. Extracts from the leaves of C. auriculata were reported to exert a protective effect on free radical mediated oxidative stress in experimental hepatotoxicity.¹⁾ In addition, extracts from the roots of C. auriculata were found to possess a protective effect on ethanol and antitubercular-drug-induced hepatotoxicity in rats.²⁾ During the course of our characterization studies on the bioactive compounds of traditional medicines in South or Southeast Asia,^3–9) we found that the 80% aqueous acetone extract of C. auriculata leaves exhibited a protective effect on D-galactosamine (D-GalN)-induced cytotoxicity in primary cultured mouse hepatocytes. From the 80% aqueous acetone extract, we have isolated a new benzocoumarin glycoside, avaraoside I (1), and a new flavanol dimer, avaraol I (2), together with 29 known constituents. Furthermore, we examined the protective effects of the principal constituents on D-GalN-induced cytotoxicity in primary cultured mouse hepatocytes. In this paper, we describe the isolation and structural elucidation of the new constituents and the hepatoprotective effects of the principal constituents from the leaves of C. auriculata.

C. auriculata leaves were extracted with 80% aqueous acetone. The 80% aqueous acetone extract (27.2% from the leaves) was partitioned with EtOAc–H₂O (1 : 1, v/v) to furnish an EtOAc-soluble fraction (11.8%) and an aqueous phase. The aqueous phase was further extracted with n-BuOH to give n-BuOH-(4.6%) and H₂O- (10.8%) soluble fractions. The 80% aqueous acetone extract [inhibition (%): 22.5±0.8 (p<0.01) at 100 µg/mL] and the EtOAc- and n-BuOH-soluble fractions [inhibition (%): 62.5±1.6 and 22.7±0.9, respectively (p<0.01) at 100 µg/mL] showed protective effects on D-GalN-induced cytotoxicity in primary cultured mouse hepatocytes. The H₂O-soluble fraction showed no detectable biological effect at 100 µg/mL. Then, the EtOAc-soluble fraction was subjected to normal-phase and reversed-phase silica gel column chromatography and finally HPLC to afford avaraol I (2, 0.00071%) together with 24 known compounds, luteolin (3, 0.010%),¹⁰⁾ kaempferol (4, 0.0026%), quercetin (5, 0.0012%), myricetin (6, 0.00070%),¹¹⁾ 3-methoxyluteolin (7, 0.0018%),¹²⁾ kaempferol 3-O-β-D-glucopyranoside (8, 0.0054%),^10,13) isoquercetin (9, 0.00098%),¹⁰⁾ myricetin-3-O-β-D-glucopyranoside (10, 0.0016%),¹⁴⁾ rutin (12, 0.87%),¹⁵⁾ lanceolatin B (14, 0.0015%),^16,17) pseudosemiglabrin (15, 0.0011%),¹⁸⁾ (+)-catechin (16, 0.0038%),^19–22) (+)-gallocatechin (17, 0.0086%),^19–22) (−)-epicatechin (18, 0.0060%),^19–22) (−)-epigallocatechin (19, 0.0028%),^11,19–22) 6-demethoxycapillarisin (20, 0.0011%),²³⁾ 6-demethoxy-7-methylcapillarisin (21, 0.0013%),²⁴⁾ (2S)-7,4′-dihydroxyflavan(4β→8)-catechin (22, 0.00075%),²⁵⁾ (2S)-7,4′-dihydroxyflavan(4β→8)-gallocatechin (23, 0.092%),²⁶⁾ (2S)-7,4′-dihydroxyflavan(4β→8)-epicatechin (24, 0.00070%),²⁶⁾ (2S)-7,4′-dihydroxyflavan(4β→8)-epigallocatechin (25, 0.0068%),²⁶⁾ chrysophanol (26, 0.0014%),²⁷⁾ emodin (27, 0.0019%),²⁸⁾ and physcion (28, 0.0036%).²⁷⁾ The n-BuOH soluble fraction was subjected to normal-phase and reversed-phase silica gel column chromatography and finally HPLC to afford avaraoside I (1, 0.0052%) together with six known compounds, kaempferol 3-O-rutinoside (11, 0.036%),^10,29) rutin (12, 0.091%),¹⁵⁾ myricetin 3-O-rutinoside (13, 0.0052%),³⁰⁾ roseoside (29, 0.0014%),^31,32) bridelionoside C (30, 0.0046%),^31,32) and benzyl O-β-D-apiofuranosyl(1→2)-β-D-glucopyranoside (31, 0.0015%).³³⁾

Structures of Avaraoside I (1) and Avaraol I (2)

Avaraoside I (1) was obtained as a yellow powder with a negative optical rotation ([α]_D²³ −88.2 in MeOH). Its IR spectrum showed absorption bands at 3423, 1717, 1619, 1509, and 1069 cm⁻¹, which were ascribable to hydroxy, ester, aromatic ring, and ether functionalities, respectively. Fast atom bombardment (FAB) MS in the positive-ion mode revealed a quasimolecular ion (M+Na)⁺ at m/z 567 from which the molecular formula C₂₆H₃₁O₁₄ was determined via high resolution-mass spectrum (HR-MS) and ¹³C-NMR data. Acid hydrolysis of 1 liberated D-glucose and L-rhamnose, which were identified by HPLC analysis using an optical rotation detector. The ¹H- (acetone-d₆) and ¹³C-NMR (Table 1) spectra of 1, which were assigned by various NMR experiments,³⁴⁾ showed signals ascribable to a methyl [δ 2.79 (3H, s, CH₃-5)], an olefinic proton [δ 6.03 (1H, s, H-3)], three aromatic protons [δ 6.64, 6.75 (1H each, both d, J=2.0 Hz, H-7, 9), 7.30 (1H, s, H-6)], an α-L-rhamnopyranosyl moiety [δ 4.79 (1H, br s, H-1″)], and a β-D-glucopyranosyl moiety [δ 5.38 (1H, d, J=7.3 Hz, H-1′)]. The proton and carbon signals of 1 in the ¹H- and ¹³C-NMR spectra resembled those of pannorin,³⁵⁾ a naphthopyrone, except for the signals due to the glycosyl moiety, which were similar to those of rutin (12).¹⁵⁾ In the heteronuclear multiple bond connectivity (HMBC) experiment, long-range correlations were observed between the following protons and carbons: H-3 and C-2, 4, 4a; H-6 and C-4a, 6a, 7, 10a; H-7 and C-6, 8, 9, 10a; H-9 and C-7, 8, 10, 10a; CH₃-5 and C-4a, 5, 6; H-1′ and C-4; and H-1″ and C-6′ (Fig. 2). Furthermore, in the nuclear Overhauser effect spectroscopy (NOESY) experiment, NOE correlations were observed between the following proton pairs: H-6 and H-7; CH₃-5 and H-6; and H-1′ and H-3 (Fig. 2). On the basis of all these pieces of evidence, the chemical structure of avaraoside I (1) was determined to be pannorin 4-O-α-L-rhamnopyranosyl(1→6)-β-D-glucopyranoside.

Table 1. ¹³C-NMR Spectroscopic Data for Compounds 1 and 2 in Acetone-d₆

Position	1	Position	1	Position	2	Position	2
Aglycon		β-D-Glucopyranosyl		Upper unit		Lower unit
2	163.0	1′	100.7	2	76.1	2″	77.9
3	91.2	2′	73.9	3	35.5	3″	30.3
4	169.4	3′	77.0	4	28.9	4″	20.3
4a	108.5	4′	70.6	5	130.4	5″	155.5
5	133.5	5′	77.6	6	109.3	6″	96.8
6	126.2	6′	67.2	7	157.5	7″	154.9
6a	139.2	α-L-Rhamnopyranosyl		8	104.1	8″	110.8
7	102.6	1″	101.5	9	156.7	9″	155.7
8	160.3	2″	71.8	10	115.9	10″	102.8
9	104.1	3″	71.4	1′	133.9	1‴	134.2
10	156.7	4″	73.2	2′,6′	127.9	2‴,6‴	106.1
10a	106.8	5″	69.4	3′,5′	115.9	3‴,5‴	146.5
10b	154.8	6″	17.9	4′	158.0	4‴	133.1

Avaraol I (2), obtained as a yellow amorphous powder with a positive optical rotation ([α]_D²³ +146.6 in MeOH), showed absorption bands ascribable to hydroxy and aromatic ring functionalities in the IR spectrum, respectively. FAB-MS in the positive-ion mode revealed a quasimolecular ion (M+Na)⁺ at m/z 553 from which the molecular formula C₃₀H₂₆O₉ was determined via HR-MS and ¹³C-NMR data. The ¹H-NMR (acetone-d₆) and ¹³C-NMR (Table 1) spectra³⁴⁾ of 2 showed signals assignable to two methines with an oxygen function [δ 4.75 (1H, dd, J=2.2, 10.1 Hz, H-2″), 5.38 (1H, dd, J=3.4, 6.4 Hz, H-2)], and 10 aromatic protons [δ 6.01 (1H, s, H-6″), 6.33 (1H, dd, J=2.5, 8.6 Hz, H-6), 6.43 (1H, d, J=2.5 Hz, H-8), 6.49 (2H, s, H-2‴, 6‴), 6.76 (1H, d, J=8.6 Hz, H-5), 6.81 (2H, d, J=8.6 Hz, H-3′, 5′), 7.22 (2H, d, J=8.6 Hz, H-2′, 6′)]. The structure of 2 was characterized by means of double quantum filter correlation spectroscopy (DQF COSY) and HMBC experiments (Fig. 1). The DQF COSY data of 2 indicated the presence of the partial structures (bold lines), and long-range correlations in the HMBC experiment were observed between the following protons and carbons: H-2 and C-1′, 2′, 6′; H-3 and C-10, 8″; H-4 and C-2, 5, 8″; H-5 and C-7, 9; H-6 and C-8; H-8 and C-6, 7, 9; H-2′, 6′ and C-4′; H-3′, 5′ and C-1′, 4′; H-2″ and C-1‴, 2‴, 6‴; H-3″ and C-10″; H-4″ and C-2,″ 5,″ 9,″ 10″; H-6″ and C-5,″ 7,″ 8,″ 10″; and H-2‴, 6‴ and C-1‴, 3‴, 4‴. The proton and carbon signals of 2 in the ¹H- and ¹³C-NMR spectra resembled those of 23 and 25, except for the signals around the 3″-position. On the basis of the above results and the comparison of the NMR data for 2 with those of known flavanol dimers such as (2S)-3′,4′,7-trihydroxyflavan(4β→8)-catechin,²⁵⁾ the planar structure of 2 was determined. The relative stereostructure between 2- and 4-positions in 2 was characterized by NOESY experiment, which showed NOE correlations between the following proton pairs: H-2 and H-3β; and H-3α and H-4. The configuration at the 4-position in 2 was determined by circular dichroism (CD) measurement. It was reported that the configuration at the 4-position of the flavanol dimer, in which a linkage is formed between the 4-position of upper flavanol and the 8-position of lower flavanol, was deduced from the Cotton effect around 240 nm in the CD spectrum.^36,37) The CD spectrum of 2 showed a positive Cotton effect at 239 nm (Δε+17.07), similar to that of a known flavanol dimer with an R-configuration at the 4-position, (−)-epiafzelechin-(4β→8)-4β-carboxymethyl-(−)-epicatechin methyl ester [234 nm (Δε+17.19)],^36,37) so that the 4-position of 2 was found to possess an R-configuration. On the basis of all these pieces of evidence, the chemical structure of avaraol I (2) was determined to be (2S)-7,4′-dihydroxyflavan(4β→8)-(2ξ)-5,7,3′,4′,5′-pentahydroxyflavan.

Fig. 1. Structures of Constituents Isolated from the Leaves of C. auriculata

Fig. 2. Important 2D-NMR Correlations of 1 and 2

Recently, we have reported the isolation of several constituents with hepatoprotective effects from medicinal plants, including Salacia chinensis,³⁸⁾ Cistanche tubulosa,³⁹⁾ Hedychium coronarium,⁴⁰⁾ Sinocrassula indica,⁴¹⁾ Piper chaba,⁴²⁾ Camellia sinensis,⁴³⁾ Sedum sarmentosum,^44,45) and Rhodiola sachalinensis.⁴⁶⁾ In addition, we have reported that flavonols and their glucosides, 3, 4, 5, 8, and 9, showed hepatoprotective effects.⁴¹⁾ As a continuing exploratory study of the hepatoprotective constituents from natural products, the protective effects of the principal constituents, 11–19, 22–28, on M-GalN-induced cytotoxicity in primary cultured mouse hepatocytes were examined. As shown in Table 2, compounds 15, 19, 22, and 23 exhibited significant protective effects [inhibition (%) 15: 37.0±3.7 (p<0.01), 19: 24.4±3.1 (p<0.01), 22: 33.5±2.9 (p<0.01), and 23: 28.2±4.7 (p<0.01) at 30 µM, respectively]. The effects were equivalent to that of the reference compound, silybin [45.2±8.8 (p<0.01) at 30 µM].⁴¹⁾ On the other hand, compounds 11–13 showed weak or no effects. We have reported that the effects of 3-O-monoglycosides 8 and 9 were stronger than those of corresponding aglycones 4 and 5.⁴¹⁾ In the present study, the effects of 8 and 9 were also found to be stronger than those of corresponding 3-O-rutinosides 11 and 12. Furthermore, among the catechin dimers, 22–25, the hepatoprotective effects of 22 and 23 having the S-configuration at 3″-position were stronger than those of 24 and 25 having the R-configuration at 3″-position.

Table 2. Effects of Constituents on D-GalN-Induced Cytotoxicity in Primary Cultured Mouse Hepatocytes^a)

Treatment conc. (µM)	Inhibition (%)
	0	3	10	30	100
11	0.0±2.4	−2.1±1.2	−5.3±1.1	−6.8±1.1	8.9±1.9**
13	0.0±1.0	−2.2±1.2	5.0±0.2	8.6±0.8*	35.0±2.3**
14	0.0±1.7	2.4±1.4	−0.1±0.6	0.6±1.0	11.7±1.2**
15	0.0±0.5	4.6±1.9	13.3±1.8	37.0±3.7**	73.7±1.4**
16	0.0±1.4	3.1±2.4	6.5±1.0	19.2±2.1**	49.9±3.1**
17	0.0±3.0	0.6±3.6	1.3±2.6	11.4±2.9	12.9±2.2
18	0.0±5.0	−2.4±3.2	−3.4±0.4	9.1±1.8	57.6±4.6**
19	0.0±2.8	16.0±5.3*	20.7±2.5**	24.4±3.1**	33.5±3.4**
22	0.0±1.2	5.2±2.5	10.1±1.6	33.5±2.9**	85.4±5.1**
23	0.0±3.9	16.0±4.0	21.8±2.5*	28.2±4.7**	79.6±6.5**
24	0.0±1.0	−3.2±3.3	−4.9±2.4	−2.2±2.6	31.6±3.0**
25	0.0±0.9	1.5±0.8	−3.4±1.1	−2.8±1.4	15.0±1.3**
28	0.0±5.6	2.0±4.7	−2.4±4.0	−1.2±2.5	20.4±3.6*
Silybin41,b)	0.0±0.3	4.8±1.1	7.7±0.7	45.2±8.8**	77.0±5.5**

a) Each value represents the mean±S.E.M. (N=4). Significantly different from control, * p<0.05, ** p<0.01. b) Reference compound.⁴¹⁾ Compounds 12, 26, and 27 did not show any effect at 100 µM. IC₅₀ values (µM) of compounds 15, 22, 23, and silybin⁴¹⁾ were 45.2, 44.6, 50.6, and 38.8, respectively.

Experimental

General Experimental Procedures

The following instruments were used to obtain physical data: specific rotations, a Horiba SEPA-300 digital polarimeter (l=5 cm); IR spectra, a Shimadzu FTIR-8100 spectrometer; CD spectra, a JASCO J-720WI spectrometer; electron ionization-mass spectra (EI-MS) and HR-EI-MS, JEOL JMS-GCMATE mass spectrometer; FAB-MS and HR-FAB-MS, a JEOL JMS-SX 102 A mass spectrometer; ¹H-NMR spectra, JEOL EX-270 (270 MHz), JNM-LA500 (500 MHz), and JEOL JNM-ECA600 (600 MHz) spectrometers; ¹³C-NMR spectra, JEOL EX-270 (68 MHz), JNM-LA500 (125 MHz), and JEOL JNM-ECA600 (150 MHz) spectrometers with tetramethylsilane as an internal standard; and HPLC, a Shimadzu RID-6A refractive index and SPD-10Avp UV-VIS detectors. YMC-Pack ODS-A (YMC) and COSMOSIL-5C₁₈-PAQ (Nacalai Tesque) {[250×4.6 mm i.d. (5 µm) for analytical purposes] and [250×20 mm i.d. (5 µm) for preparative purposes]} columns were used. The following experimental conditions were used for chromatography: ordinary-phase silica gel column chromatography (CC), Silica gel BW-200 (Fuji Silysia Chemical, Ltd., 150–350 mesh); reverse-phase silica gel CC, Chromatorex ODS DM1020T (Fuji Silysia Chemical, Ltd., 100–200 mesh); TLC, precoated TLC plates with Silica gel 60F₂₅₄ (Merck, 0.25 mm) (ordinary phase) and Silica gel RP-18 F_254S (Merck, 0.25 mm) (reverse phase); reversed-phase HPTLC, precoated TLC plates with Silica gel RP-18 WF_254S (Merck, 0.25 mm). Detection was achieved by spraying with 1% Ce(SO₄)₂−10% aqueous H₂SO₄ followed by heating.

Plant Material

The dried leaves of C. auriculata (1.5 kg), which were cultivated in India, were purchased from N.T.H. Co., Ltd. A voucher of the plant is on file in our laboratory.

Extraction and Isolation

The dried leaves of C. auriculata (1.5 kg) were finely cut and extracted three times with 80% aqueous acetone under room temperature for 12 h. Evaporation of the solvent under reduced pressure provided the 80% aqueous acetone extract (407.7 g, 27.2%). A part of the extract (377.7 g) was partitioned into an EtOAc–H₂O (1 : 1, v/v) mixture to furnish an EtOAc-soluble fraction (164.1 g, 11.8%) and aqueous layer. The aqueous layer was extracted with n-BuOH to give n-BuOH- (63.7 g, 4.6%) and H₂O- (149.9 g, 10.8%) soluble fractions.

A part of the EtOAc-soluble fraction (144.1 g) was subjected to normal-phase silica gel column chromatography [2.5 kg, n-hexane–EtOAc (10 : 1→5 : 1→2 : 1→1 : 1→1 : 2, v/v)→CHCl₃–MeOH–H₂O (10 : 3 : 1, lower layer→7 : 3 : 1, lower layer→6 : 4 : 1, v/v)→MeOH] to give 8 fractions [Fr. 1 (1.92 g), Fr. 2 (10.19 g), Fr. 3 (2.09 g), Fr. 4 (2.66 g), Fr. 5 (9.27 g), Fr. 6 (65.07 g), Fr. 7 (14.54 g), Fr. 8 (35.70 g)]. Fraction 1 (1.92 g) was separated by reversed-phase silica gel column chromatography [60 g, MeOH–H₂O (40 : 60→60 : 40→75 : 25, v/v)→MeOH] to give 7 fractions [Fr. 1-1 (18 mg), Fr. 1-2, Fr. 1-3, Fr. 1-4, Fr. 1-5, Fr. 1-6, Fr. 1-7]. Fraction 1-1 (18 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (80 : 20, v/v)] to give chrysophanol (26, 5.0 mg). A part of fraction 2 (4.45 g) was separated by reversed-phase silica gel column chromatography [60 g, MeOH–H₂O (40 : 60→60 : 40→75 : 25→90 : 10, v/v)→MeOH] to give 4 fractions [Fr. 2-1, Fr. 2-2 (43 mg), Fr. 2-3 (799 mg), Fr. 2-4]. Fraction 2-2 (43 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (80 : 20, v/v)] to give chrysophanol (26, 5.1 mg) and emodin (27, 10 mg). Fraction 2-3 (799 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (85 : 15, v/v)] to give physcion (28, 19 mg). Fraction 3 (2.09 g) was separated by reversed-phase silica gel column chromatography [60 g, MeOH–H₂O (30 : 70→50 : 50→70 : 30→90 : 10, v/v)→MeOH] to give 9 fractions [Fr. 3-1, Fr. 3-2, Fr. 3-3, Fr. 3-4 (61 mg), Fr. 3-5 (59 mg), Fr. 3-6, Fr. 3-7, Fr. 3-8, Fr. 3-9]. Fraction 3-4 (61 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (65 : 35, v/v)] to give pseudosemiglabrin (15, 13 mg) and 21 (16 mg). Fraction 3-5 (59 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (65 : 35, v/v)] to give lanceolatin B (14, 18 mg). Fraction 4 (2.66 g) was separated by reversed-phase silica gel column chromatography [80 g, MeOH–H₂O (15 : 85→30 : 70→45 : 55→60 : 40→75 : 25, v/v)→MeOH] to give 12 fractions [Fr. 4-1, Fr. 4-2, Fr. 4-3, Fr. 4-4 (114 mg), Fr. 4-5 (309 mg), Fr. 4-6 (152 mg), Fr. 4-7, Fr. 4-8, Fr. 4-9, Fr. 4-10, Fr. 4-11, Fr. 4-12]. Fraction 4-4 (114 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (40 : 60, v/v)] to give myricetin (6, 8.5 mg) and isoquercetin (9, 12 mg). Fraction 4-5 (309 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, [1] MeOH–H₂O (55 : 45, v/v), [2] CH₃CN–H₂O (30 : 70, v/v)] to give luteolin (3, 122 mg), quercetin (5, 15 mg), 3-methoxyluteolin (7, 22 mg), and 6-demethoxycapillarism (20, 13 mg). Fraction 4-6 (152 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (50 : 50, v/v)] to give kaempferol (4, 32 mg). Fraction 5 (9.27 g) was separated by reversed-phase silica gel column chromatography [280 g, MeOH–H₂O (15 : 85→30 : 70→45 : 55→60 : 40→75 : 25→90 : 10, v/v)→MeOH] to give 14 fractions [Fr. 5-1, Fr. 5-2 (260 mg), Fr. 5-3 (316 mg), Fr. 5-4, Fr. 5-5, Fr. 5-6, Fr. 5-7, Fr. 5-8, Fr. 5-9 (323 mg), Fr. 5-10, Fr. 5-11, Fr. 5-12, Fr. 5-13, Fr. 5-14]. Fraction 5-2 (260 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (20 : 80, v/v)] to give (+)-catechin (16, 46 mg). Fraction 5-3 (316 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (25 : 75, v/v)] to give (−)-epicatechin (18, 73 mg). Fraction 5-9 (323 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (50 : 50, v/v)] to give avaraol I (2, 8.7 mg), 22 (9.2 mg), and 24 (8.6 mg). A part of fraction 6 (50.0 g) was separated by reversed-phase silica gel column chromatography [1.5 kg, MeOH–H₂O (15 : 85→30 : 70→50 : 50→70 : 30, v/v)→MeOH] to give 14 fractions [Fr. 6-1, Fr. 6-2, Fr. 6-3 (1.23 g), Fr. 6-4, Fr. 6-5 (768 mg), Fr. 6-6, Fr. 6-7 (6.66 g), Fr. 6-8, Fr. 6-9 (2.67 g), Fr. 6-10, Fr. 6-11, Fr. 6-12 (2.25 g), Fr. 6-13 (8.21 g), Fr. 6-14]. Fraction 6-3 (1.23 g) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (15 : 85, v/v)] to give (+)-gallocatechin (17, 81 mg). Fraction 6-5 (768 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (20 : 80, v/v)] to give (−)-epigallocatechin (19, 26 mg). A part of Fr. 6-7 (500 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (35 : 65, v/v)] to give 23 (65 mg). A part of Fr. 6-9 (630 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (40 : 60, v/v)] to give 25 (15 mg). A part of fraction 6-12 (580 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (50 : 50, v/v)] to give 8 (13 mg). A part of fraction 6-13 (800 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (40 : 60, v/v)] to give rutin (12, 129 mg). Fraction 7 (14.54 g) was separated by reversed-phase silica gel column chromatography [420 g, MeOH–H₂O (30 : 70→45 : 55→60 : 40→75 : 25, v/v)→MeOH] to give 12 fractions [Fr. 7-1, Fr. 7-2 (846 mg), Fr. 7-3 (856 mg), Fr. 7-4, Fr. 7-5, Fr. 7-6, Fr. 7-7, Fr. 7-8, Fr. 7-9, Fr. 7-10, Fr. 7-11, Fr. 7-12]. Fraction 7-2 (846 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (40 : 60, v/v)] to give 10 (20 mg, 0.0016%) and rutin (12, 55 mg). A part of Fr. 7-3 (120 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (45 : 55, v/v)] to give rutin (12, 64 mg). A part of fraction 8 (30.0 g) was separated by reversed-phase silica gel column chromatography [600 g, MeOH–H₂O (30 : 70→45 : 55→60 : 40→75 : 25, v/v)→MeOH] to give 11 fractions [Fr. 8-1, Fr. 8-2, Fr. 8-3, Fr. 8-4, Fr. 8-5, Fr. 8-6 (7.08 g), Fr. 8-7 (5.16 g), Fr. 8-8, Fr. 8-9, Fr. 8-10, Fr. 8-11]. A part of Fr. 8-6 (100 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (40 : 60, v/v)] to give rutin (12, 64 mg). A part of Fr. 8-7 (100 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeOH–H₂O (40 : 60, v/v)] to give rutin (12, 49 mg). A part of the n-BuOH-soluble fraction (53.7 g) was subjected to reversed-phase silica gel column chromatography [1.2 kg, MeOH–H₂O (15 : 85→30 : 70→45 : 55→60 : 40→75 : 25, v/v)→MeOH] to give 10 fractions [Fr. 1, Fr. 2, Fr. 3 (3.09 g), Fr. 4, Fr. 5 (3.76 g), Fr. 6, Fr. 7, Fr. 8 (4.72 g), Fr. 9, Fr. 10]. A part of fraction 3 (800 mg) was purified by HPLC [YMC-Pack ODS-A, MeOH–H₂O (25 : 75, v/v)] to give roseoside (29, 4.2 mg), 30 (14 mg), and 31 (4.4 mg). A part of fraction 5 (800 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, [1] MeCN–H₂O (20 : 80, v/v), [2] MeOH–H₂O (35 : 65, v/v)] to give avaraoside I (1, 13 mg) and 13 (13 mg). A part of fraction 8 (200 mg) was purified by HPLC [COSMOSIL-5C₁₈-PAQ, MeCN–H₂O (20 : 80, v/v)] to give 11 (18 mg) and rutin (12, 45 mg).

Avaraoside I (1)

a yellow powder; [α]_D²³ −88.2° (c=0.6, MeOH); UV (MeOH) λ_max (log ε) 327 (3.76), 292 (4.02), 235 (4.24) nm; IR (KBr) v_max 3423, 1717, 1619, 1509, 1069 cm⁻¹; ¹H-NMR (acetone-d₆, 600 MHz) δ 2.79 (3H, s, CH₃-5), 4.79 (1H, br s, H-1″), 5.38 (1H, d, J=7.3 Hz, H-1′), 6.03 (1H, s, H-3), 6.64 (1H, d, J=2.0 Hz, H-9), 6.75 (1H, d, J=2.0 Hz, H-7), 7.30 (1H, s, H-6); ¹³C-NMR data see Table 1; positive-ion FAB-MS m/z 567 [M+H]⁺; HR-FAB-MS: m/z 567.1706 (Calcd for C₂₆H₃₁O₁₄ [M+H]⁺, 567.1713).

Avaraol I (2)

a yellow powder; [α]_D²³ +146.6° (c=0.7, MeOH); UV (MeOH) λ_max (log ε) 281 (3.90), 230 (4.67) nm; CD (MeOH) nm (Δε) 274 (−3.03), 239 (+17.07) nm; IR (KBr) v_max 3430, 1509 cm⁻¹; ¹H-NMR (acetone-d₆, 500 MHz) δ 1.89, 2.10 (1H each, both m, H₂-3″), 2.28 (1H, m, H-3α), 2.55 (1H, m, H-3β), 2.61 (2H, m, H₂-4″), 4.51 (1H, dd like, H-4α), 4.75 (1H, dd, J=2.2, 10.1, H-2″), 5.38 (1H, dd, J=3.4, 6.4, H-2), 6.01 (1H, s, H-6″), 6.33 (1H, dd, J=2.5, 8.6 Hz, H-6), 6.43 (1H, d, J=2.5 Hz, H-8), 6.49 (2H, s, H-2‴,6‴), 6.76 (1H, d, J=8.6 Hz, H-5), 6.81 (2H, d, J=8.6 Hz, H-3′,5′), 7.22 (2H, d, J=8.6 Hz, H-2′,6′); ¹³C-NMR data see Table 1; positive-ion FAB-MS m/z 553 [M+Na]⁺; HR-FAB-MS: m/z 553.1472 (Calcd for C₃₀H₂₆O₉Na [M+Na]⁺, 553.1475).

Acid Hydrolyses of 1

A solution of 1 (1.0 mg) in 1 M HCl–1,4-dioxane (1 : 1, v/v, 1.0 mL) was heated under reflux for 3 h. After cooling, the reaction mixture was extracted with EtOAc. The aqueous layer was subjected to HPLC analysis under the following conditions, respectively: HPLC column, Kaseisorb LC NH₂-60-5, 4.6 mm i.d.×250 mm (Tokyo Kasei Co., Ltd., Tokyo, Japan); detection, optical rotation [Shodex OR-2 (Showa Denko Co., Ltd., Tokyo, Japan); mobile phase, CH₃CN–H₂O (85 : 15, v/v); flow rate 0.8 mL/min]. Identifications of L-rhamnose and D-glucose present in the aqueous layer were carried out by comparison of their retention times and optical rotations with those of authentic samples [t_R: L-rhamnose, 7.4 min (negative optical rotation); D-glucose, 11.5 min (positive optical rotation)].

Protective Effect on Cytotoxicity Induced by M-GalN in Primary Cultured Mouse Hepatocytes

The hepatoprotective effects of the constituents were determined by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) colorimetric assay using primary cultured mouse hepatocytes. Hepatocytes were isolated from male ddY mice (30–35 g) by collagenase perfusion method.^47,48) The cell suspension at 4×10⁴ cells in 100 µL William’s E medium containing fetal calf serum (10%), penicillin G (100 units/mL), and streptomycin (100 µg/mL) was inoculated in a 96-well microplate, and precultured for 4 h at 37°C under a 5% CO₂ atmosphere. The fresh medium (100 µL) containing D-GalN (2 mM) and a test sample were added and the hepatocytes were cultured for 44 h. The medium was exchanged with 100 µL of the fresh medium, and 10 µL of MTT (5 mg/mL in phosphate buffered saline) solution was added to the medium. After 4 h culture, the medium was removed, 100 µL of isopropanol containing 0.04 M HCl was then added to dissolve the formazan produced in the cells. The optical density (O.D.) of the formazan solution was measured by microplate reader at 562 nm (reference: 660 nm). Inhibition (%) was obtained by following formula.

  

Statistics

Values were expressed as means±S.E.M. For statistical analysis, ANOVA followed by Dunnett’s test was used. Probability (p) values less than 0.05 were considered significant.

Acknowledgment

This research was supported in part by a Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan—Supported Program for the Strategic Research Foundation at Private Universities.

References

• 1) Kumar Rajagopal S., Manickam P., Periyasamy V., Namasivayam N., J. Nutr. Biochem., 14, 452–458 (2003).
• 2) Jaydeokar A. V., Bandawane D. D., Bibave K. H., Patil T. V., Pharm. Biol., 52, 344–355 (2014).
• 3) Matsumoto T., Nakamura S., Ohta T., Fujimoto K., Yoshikawa M., Ogawa K., Matsuda H., Org. Lett., 16, 3076–3078 (2014).
• 4) Matsumoto T., Nakamura S., Nakashima S., Fujimoto K., Yoshikawa M., Ohta T., Ogawa K., Matsuda H., J. Nat. Prod., 77, 990–999 (2014).
• 5) Yamaguchi I., Matsuda H., Zhang H., Hamao M., Yamashita C., Kogami Y., Kon’I H., Murata M., Nakamura S., Yoshikawa M., J. Nat. Med., 68, 74–82 (2014).
• 6) Nakamura S., Nakashima S., Tanabe G., Oda Y., Yokota N., Fujimoto K., Matsumoto T., Sakuma R., Ohta T., Ogawa K., Nishida S., Miki H., Matsuda H., Muraoka O., Yoshikawa M., Bioorg. Med. Chem., 21, 779–787 (2013).
• 7) Nakamura S., Nakashima S., Oda Y., Yokota N., Fujimoto K., Matsumoto T., Ohta T., Ogawa K., Maeda S., Nishida S., Matsuda H., Yoshikawa M., Bioorg. Med. Chem., 21, 1043–1049 (2013).
• 8) Mourad A. A., Nakamura S., Ueno T., Minami T., Yagi T., Yasue H., Komatsu R., Yoshikawa M., Taye A. M., El-Moselhy M. A., Khalifa M. M., Matsuda H., Bioorg. Med. Chem. Lett., 23, 4813–4816 (2013).
• 9) Matsumoto T., Nakamura S., Nakashima S., Yoshikawa M., Fujimoto K., Ohta T., Morita A., Yasui R., Kashiwazaki E., Matsuda H., Bioorg. Med. Chem. Lett., 23, 5178–5181 (2013).
• 10) Markham K. R., Ternai B., Stanley R., Geiger H., Mabry T. J., Tetrahedron, 34, 1389–1397 (1978).
• 11) Nonaka G., Muta M., Nishioka I., Phytochemistry, 22, 237–241 (1983).
• 12) Roitman J. N., James L. F., Phytochemistry, 24, 835–848 (1985).
• 13) Han J. T., Bang M. H., Chun O. K., Kim D. O., Lee C. Y., Baek N. I., Arch. Pharm. Res., 27, 390–395 (2004).
• 14) Cui B., Nakamura M., Kinjo J., Nohara T., Chem. Pharm. Bull., 41, 178–182 (1993).
• 15) Markham K. R., Ternai B., Tetrahedron, 32, 2607–2612 (1976).
• 16) Rao E. V., Raju N. R., Phytochemistry, 18, 1581–1582 (1979).
• 17) Talapatra S. K., Mallik A. K., Talapatra B., J. Indian Chem. Soc., 59, 534–535 (1982).
• 18) Ahmad S., Phytochemistry, 25, 955–958 (1986).
• 19) Coxon D. T., Holmes A., Ollis W. D., Vora V. C., Grant M. S., Tee J. L., Tetrahedron, 28, 2819–2826 (1972).
• 20) Nonaka G., Kawahara O., Nishioka I., Chem. Pharm. Bull., 31, 3906–3914 (1983).
• 21) Nishino C., Enoki N., Tawata S., Mori A., Kobayashi K., Fukushima M., Agric. Biol. Chem., 51, 139–143 (1987).
• 22) Davis A. L., Cai Y., Davies A. P., Lewis J. R., Magn. Reson. Chem., 34, 887–890 (1996).
• 23) Hashidoko Y., Tahara S., Mizutani J., Phytochemistry, 30, 3837–3838 (1991).
• 24) Huang Y. L., Ou J. C., Chen C. F., Chen C. C., J. Nat. Prod., 56, 275–278 (1993).
• 25) Hatano T., Yamashita A., Hashimoto T., Ito H., Kubo N., Yoshiyama M., Shimura S., Itoh Y., Okuda T., Yoshida T., Phytochemistry, 46, 893–900 (1997).
• 26) Coetzee J., Mciteka L., Malan E., Ferreira D., Phytochemistry, 53, 795–804 (2000).
• 27) Kitanaka S., Takido M., Chem. Pharm. Bull., 32, 860–864 (1984).
• 28) Varshney S. C., Rizvi S. A. I., Gupta P. C., Planta Med., 23, 363–369 (1973).
• 29) Chaurasia N., Wichtl M., Planta Med., 53, 432–434 (1987).
• 30) Parker W. H., Bohm B. A., Phytochemistry, 14, 553–555 (1975).
• 31) Otsuka H., Yao M., Kamada K., Takeda Y., Chem. Pharm. Bull., 43, 754–759 (1995).
• 32) Sueyoshi E., Liu H., Matsunami K., Otsuka H., Shinzato T., Aramoto M., Takeda Y., Phytochemistry, 67, 2483–2493 (2006).
• 33) Miyaichi Y., Tomimori T., Nat. Med., 49, 350–353 (1995).
• 34) The ¹H- and ¹³C-NMR spectra of 1 and 2 were assigned with the aid of distortionless enhancement by polarization transfer (DEPT), DQF COSY, heteronuclear multiple quantum coherence spectroscopy (HMQC), and HMBC experiments.
• 35) Ogawa H., Hasumi K., Sakai K., Murakawa S., Endo A., J Antibiot., 44, 762–767 (1991).
• 36) Liang Y.-H., Ye M., Yang W.-Z., Qiao X., Wang Q., Yang H.-J., Wang X.-L., Guo D.-A., Phytochemistry, 72, 1876–1882 (2011).
• 37) Coetzee J., Mciteka L., Malan E., Ferreira D., Phytochemistry, 53, 795–804 (2000).
• 38) Nakamura S., Zhang Y., Matsuda H., Ninomiya K., Muraoka O., Yoshikawa M., Chem. Pharm. Bull., 59, 1020–1028 (2011).
• 39) Morikawa T., Pan Y., Ninomiya K., Imura K., Matsuda H., Yoshikawa M., Yuan D., Muraoka O., Bioorg. Med. Chem., 18, 1882–1890 (2010).
• 40) Nakamura S., Okazaki Y., Ninomiya K., Morikawa T., Matsuda H., Yoshikawa M., Chem. Pharm. Bull., 56, 1704–1709 (2008).
• 41) Ninomiya K., Morikawa T., Xie H., Matsuda H., Yoshikawa M., Heterocycles, 75, 1983–1995 (2008).
• 42) Matsuda H., Ninomiya K., Morikawa T., Yasuda D., Yamaguchi I., Yoshikawa M., Bioorg. Med. Chem. Lett., 18, 2038–2042 (2008).
• 43) Li N., Morikawa T., Matsuda H., Ninomiya K., Li X., Yoshikawa M., Heterocycles, 71, 1193–1201 (2007).
• 44) Ninomiya K., Morikawa T., Zhang Y., Nakamura S., Matsuda H., Muraoka O., Yoshikawa M., Chem. Pharm. Bull., 55, 1185–1191 (2007).
• 45) Zhang Y., Morikawa T., Nakamura S., Ninomiya K., Matsuda H., Muraoka O., Yoshikawa M., Heterocycles, 71, 1565–1576 (2007).
• 46) Nakamura S., Li X., Matsuda H., Ninomiya K., Morikawa T., Yamaguti K., Yoshikawa M., Chem. Pharm. Bull., 55, 1505–1511 (2007).
• 47) Oka M., Maeda S., Koga N., Kato K., Saito T., Biosci. Biotechnol. Biochem., 56, 1472–1473 (1992).
• 48) Seglen P. O., Methods Cell Biol., 13, 29–83 (1976).