Clinoposides A–F: meroterpenoids with protective effects on H9c2 cardiomyocyte from Clinopodium chinense

Abstract

Six novel flavonoid–triterpene saponin meroterpenoids, clinoposides A–F (1–6), with two unusual skeletons were isolated from Clinopodium chinense. Clinoposides A–F represent a new class of isopentenyl flavonoid saponin. Their structures were determined based on spectroscopic data and chemical methods. The relative and absolute stereochemistries were assigned using a combination of NOESY and ECD. A possible biogenetic pathway for 1–6 is proposed. The protective effects of clinoposides A–F against H₂O₂-induced H9c2 cardiomyocyte injury were tested, and all compounds exhibited significantly dose-dependent effects, clinoposides B, D and F showed better protective effects as evidence by increased levels of SOD, CAT and GSH-Px but reduced MDA, LDH, caspase-3 and -9 levels.

---

Introduction

Meroterpenoids are natural products containing isopentenyl moieties in their molecules, along with carbon skeletons derived from other pathways. A common terpenoid moiety in such cases is a single C5 unit, and is usually a dimethylallyl substituent.¹ Within the flavonoid class of natural products, the prenylated sub-class is quite rich in structural diversity and biological activity.² A few meroterpenoids with monoterpenes attached to dihydrochalcones and flavanones have been isolated from Metrodorea stipularis,³ Goniothalamus macrophyllus⁴ and Lindera umbellate,⁵ some of which showed significant biological activities. However, flavonoid-higher terpene adducts such as flavonoid coupled with triterpene have been not reported so far.

Clinopodium chinense (Benth.) O. Kuntze (Labiatae), a perennial herb prevalently distributed in China, is popularly used in TCM for the treatment of heliosis, dysentery and hematuria.⁶ Previous phytochemical investigations of C. chinense disclosed the presence of flavonoids,⁷ triterpenoid saponins,⁸ and phenylpropanoids,⁹ which exhibited anti-inflammatory and anti-immunity,¹⁰ anti-hyperglycemic,¹¹ anti-tumor and anti-radiation pharmacological effects.¹² Our previous chemical studies led to the isolation of diterpenes and meroterpenes.^13,14 Further investigations of the extracts from the aerial parts of C. chinense resulted in the isolation of six new flavonoid–triterpene saponin meroterpenoids, clinoposides A–F (1–6), these compounds have been obtained in trace amounts (2.1–5.6 mg from 15 kg of dried plant material) after two hours extraction with 100 L of 70% ethanol at 80 °C, which possess an unique skeleton. Herein, we report the isolation and structural elucidation of compounds 1–6 (Fig. 1) and their protective effects against H₂O₂-induced H9c2 cardiomyocyte injury.

Fig. 1 Structures of compounds 1–6.

Results and discussion

Clinoposide A (1) was isolated as a yellow powder ([α]²⁰_D 6.7 (c 0.12, MeOH)). In conjunction with NMR analysis (Table S1†), its molecular formula was determined to be C₆₃H₉₀O₂₃ from the pseudomolecular [M − H]⁻ ion at m/z 1213.5745 (calcd 1213.5795, Δ – 0.005 mmu) in the HR-ESI-MS spectrum. The ¹H NMR spectrum of 1 (Table S1†) presented five aromatic protons, three anomeric protons, one double-bond proton and six three-proton singlets signals. The spin systems located at δ_H 5.21, 3.23, and 2.74 together with a carbonyl resonance at δ_C 196.9 in the ¹³C APT spectrum, indicated the presence of a flavanone skeleton with an OH at C-5. An aromatic singlet at δ_H 6.18 suggested a tri-substituted A-ring. Two pairs of aromatic protons at δ_H 7.19 and 7.49, appearing as an AA′XX′ spin–spin system in 1, indicated a para-substituted B-ring. Based on the above analyses of the spectroscopic data, the flavanone moiety was suggested to be 5,7,4′-trihydroxyflavanone.

Except the signals assigned to the flavonoid moiety, the ¹H NMR spectrum also showed six methyl singlets at δ_H 0.82, 0.92, 1.13, 1.42, 1.52, and 1.77, two pairs of oxymethylene protons at δ_H 3.74 and 4.35, and at δ_H 3.77 and 4.52, and one olefinic proton resonance at δ_H 5.28. Accordingly, the ¹³C APT spectrum exhibited six sp³ carbon signals at δ_C 13.9, 17.8, 18.8, 24.3, 25.7 and 33.8, and signals representing two oxygenated methylenes at δ_C 67.3 and 69.3. By comparison of the chemical shifts with those reported in the literature data,¹⁵ an oleanane-type triterpene moiety was established. The position of the double bond at Δ¹²⁽¹³⁾ was determined by the HMBC correlations of H-12 (δ_H 5.28, d, J = 1.8 Hz) with C-9 (δ_C 48.5), C-11 (δ_C 32.8) and C-14 (δ_C 44.4) (Fig. 2). For the sugar variety, the ¹H NMR spectrum presented three anomeric protons at δ_H 4.73, 5.25, and 5.57, which were assigned to fucose, glucose and glucose, respectively (Table S1†). The sugar units attached to C-3 were determined by the HMBC correlations between the proton signal of H-1′′′ (δ_H 4.73, d, J = 1.8) and the carbon resonance of C-3 (δ_C 83.6), one of the glucose linked to C-2′′′ was confirmed by the HMBC cross-peak of H-1′′′′/C-2′′′ and the COSY cross-peak of H-1′′′/H-2′′′, the other glucose linked to C-3′′′ due to the C-3′′′ shifted downfield at δ_C 85.5 and the HMBC cross-peak of H-1′′′′′/C-3′′′. The β-configuration for the fucose and glucose residues were determined through the vicinal coupling constants of the anomeric protons (J_1,2 = 7.8 Hz). The ¹H NMR spectrum showed that H-4′′′ at δ_H 4.15 (d, J = 2.4 Hz), which was different from H-4′′′′ (δ_H 4.18, br t, J = 9.0 Hz) and H-4′′′′′ (δ_H 4.21, br t, J = 9.6 Hz) due to axial–axial couplings with H-3′′′′ and H-5′′′′, H-3′′′′′ and H-5′′′′′. The H-4′′′ (δ_H 4.15, d, J = 2.4 Hz) indicated equatorial–axial couplings. Thus, 4′′′-OH should be β-oriented, and the 4′′′′-OH and 4′′′′′-OH were α-oriented. The NOE cross-peaks of H-1′′′/H-5′′′, H-1′′′′/H-5′′′′ and H-1′′′′′/H-5′′′′′ indicated that H-5′′′, H-5′′′′ and H-5′′′′′ were α-oriented. To further confirm the sugar residues, hydrolysis of 1 yielded D-fucopyranoside and D-glucopyranose, which were confirmed by subsequent GC analysis following derivatization with L-cysteine methyl ester and silylation. The complete assignment for all proton and carbon resonances of each sugar residue in the oligosaccharide chain was established through extensive 2D NMR experiments.

Fig. 2 Key COSY and HMBC correlations of 1.

The above-mentioned NMR spectroscopic data indicated that 1 was meroterpenoid consisting of a flavonoid and a triterpene saponin. The C-8′ was shifted downfield at δ_C 115.3, indicating that H-8′ was substituted by an alkyl moiety. The C-8′ linked to C-11 was confirmed by the HMBC cross-peaks of H-6′/C-4′, C-5′, C-7′, C-8′ and C-10′, and of H-11/C-8, C-12, C-13, C-7′, C-8′ and C-9′. Compared with compound 2 (Fig. 3), the absence of a correlation of H-11/H-26 in the NOESY spectrum indicated that the H-11 was in the α-face. According to the empirical rules established for flavanone,¹⁶ the ECD spectrum of 1, exhibiting a negative Cotton effect at 290 nm (Fig. 4), indicated the absolute configuration of C-2′ in 1 is S. ECD calculations were further conducted using the time-dependent density functional theory (TD-DFT) method at the B3LYP SCRF(PCM)/6-31+G(d,p) level in MeOH solution (the substituted triheteroglycan was simplified to a methyl group).

Fig. 3 Key NOESY correlations of 2.

Fig. 4 The experimental and calculated ECD spectra of 1–4.

It is observed that the calculated weighted ECD spectrum of the isomer of 1 with 11S, 2′S comparably conformed to the experimental spectrum (Fig. 4).

Thus, the structure of 1 was unanimously determined to be (11S,2′S)-3β,16β,23,28-tetrahydroxy-11-[5,7,4′-trihydroxyflavanone(8′→11)]-oleana-12-en-3-yl-[β-D-glucopyranosyl(1→2)]-[β-D-glucopyranosyl(1→3)]-β-D-fucopyranoside and named clinoposide A.

Clinoposide B (2) and clinoposide C (3) were assigned the same molecular formula as 1 based on the pseudomolecular [M − H]⁻ ion at m/z 1213. Their NMR spectra resembled each other and those of 1, except for the configuration of H-11 and H-2′. NOE experiments of 2 revealed correlations of H-11/H-25 and H-11/H-26. Therefore, the H-11 was in the β-face, whereas H-11 was α in 3. Detailed analyses of the ¹H NMR spectra between 1 and 2 disclosed that when H-11 was in the β-face, H-27 would downshift to 1.80, but when H-11 was in the α-face, the H-27 could shift upfield to 0.90, which could be due to the anisotropic effects on the A-ring.¹⁷ The ECD spectra of 2 and 3 displayed a positive Cotton effect at 290 nm (Fig. 4), and the absolute configuration of C-2′ in 2 and 3 was determined to be the R.¹⁶ Thus, the structure of 2 was elucidated to be (11R,2′R)-3β,16β,23,28-tetrahydroxy-11-[5,7,4′-trihydroxyflavanone(8′→11)]-oleana-12-en-3-yl-[β-D-glucopyranosyl(1→2)]-[β-D-glucopyranosyl(1→3)]-β-D-fucopyranoside and named clinoposide B. The structure of 3 was elucidated to be (11S, 2′R)-3β, 16β,23,28-tetrahydroxy-11-[5,7,4′-trihydroxyflavanone(8′→11)]-oleana-12-en-3-yl-[β-D-glucopyranosyl(1→2)]-[β-D-glucopyranosyl(1→3)]-β-D-fucopyranoside and named clinoposide C.

The molecular formula of clinoposide D (4) was determined to be C₆₄H₉₂O₂₃ based on a positive HR-ESI-MS at m/z 1227.5927 [M − H]⁻, which is 14 mass units larger than that of 1. A direct comparison of the NMR data (Table S2†) between 4 and 1 indicated that both structures shared the same skeleton except for the presence of an extra methoxy moiety in 4. The HMBC correlation between OCH₃ and C-4′′ confirmed the structural assignment. NOE experiments of 4 revealed correlations of H-11/H-25 and H-11/H-26; therefore, the H-11 was in the β-face. Again, a positive Cotton effect at 290 nm and negative effect at 340 nm were observed for 4 in the ECD spectrum, indicating that 4 and 2 possessed the same configuration of C-11 and 2′. Thus, the structure of 4 was determined to be (11R,2′R)-3β,16β,23,28-tetrahydroxy-11-[5,7-dihydroxy-4′-methoxyflavanone(8′→11)]-oleana-12-en-3-yl-[β-D-glucopyranosyl(1→2)]-[β-D-glucopyranosyl(1→3)]-β-D-fucopyranoside and named clinoposide D.

Clinoposide E (5) was assigned the same molecular formula as 4. A direct comparison of the NMR data between 5 and 4 indicated that compound 5 was also meroterpenoid with a flavonoid and a triterpene saponin moiety. The HMBC cross-peaks of H-8′/C-6′, C-7′, C-9′ and C-10′ and of H-11/C-12, C-13, C-6′, and C-7′ indicated that C-6′ was linked to C-11, which suggested that 5 possessed new framework completely different from those of compounds 1–4. The absence of a NOE correlation of H-11/H-26 indicated that the H-11 was in the α-face, and the H-27 downshift to δ_H 1.83 confirmed the structural assignment. The ECD spectrum of 5, exhibiting a negative Cotton effect at 290 nm, indicated the absolute configuration of C-2′ in 5 was in the S-configuration.¹⁶ Thus, the structure of 5 was determined to be (11S,2′S)-3β,16β,23,28-tetrahydroxy-11-[5,7-dihydroxy-4′-methoxyflavanone(6′→11)]-oleana-12-en-3-yl-[β-D- glucopyranosyl(1→2)]-[β-D-glucopyranosyl(1→3)]-β-D-fucopyranoside and named clinoposide E.

The molecular formula of clinoposide F (6) was determined to be C₆₄H₉₂O₂₃ based on a negative HR-ESI-MS at m/z 1227.5955 [M − H]⁻ (calcd 1227.5956). Inspection of the HSQC spectrum with ¹H and ¹³C APT data suggested that the skeleton was identical to that of 5. Through the detailed interpretation of the NOE experiment, the absence of a correlation of H-11/H-26 indicated that the H-11 was in the α-face, and the H-27 downshift to δ_H 1.82 provided evidence for this assignment. Thus, the structure of 6 was determined to be (11S,2′R)-3β,16β,23,28-tetrahydroxy-11-[5,7-dihydroxy-4′-methoxyflavanone(6′→11)]-oleana-12-en-3-yl-[β-D-glucopyranosyl(1→2)]-[β-D-glucopyranosyl(1→3)]-β-D-fucopyranoside and named clinoposide F.

To date, about 22 flavonoids and 46 triterpenoid saponins have been isolated from plants of the genus Clinopodium.^6,8 Although prenylated flavonoids and monoterpenes attached to flavonoids were isolated from plants, the flavonoid–triterpene adducts has not been reported. While the discovery of clinoposides A–F are of great interest, it also puts forward a question whether these compounds are natural product or handling artifacts produced during the extraction and isolation procedure. To prove this, the aerial parts of C. chinense were extracted methanol at 60 °C for 2 h, we analyzed the methanol extract with LC-ESIMS (Fig. S79†) which revealed two ion peaks at m/z 1213.3 ([M − H]⁻, t_R 14.773 min) and 1213.4 ([M − H]⁻, 15.157 min) in accord with that of 1–3 and confirmed the natural occurrence of flavonoid and triterpene saponin meroterpenoids. Structurally, 1–6 are derivatives of 5,7,4′-trihydroxyflavanone and buddlejasaponin IV. Their most intriguing feature is the formation of a C–C bond between flavonoid and triterpene saponin meroterpenoids as unusual. A possible biogenetic pathway for 1–6 is proposed in Scheme 1. The cleavage of the ether linkage between C-13 and the oxygen atom and the formation of double bonds at Δ¹²⁽¹³⁾ in buddlejasaponin IV are followed by radical coupling at 6′ or 8′, which involves the participation of involving neighboring groups participation.¹⁸

Scheme 1 Hypothetical biogenetic pathway of 1–6 (R = triheteroglycan).

The protective effects of 1–6 against H₂O₂-induced H9c2 cardiomyocyte injury were tested, and 1–6 all exhibited dose-dependent effects (Fig. 5A). Compounds 2, 4 and 6 exhibited significantly protective effects. Compared with the cell viability of 62.6 ± 5.3% in the model group, 2, 4 and 6 viabilities were 87.2 ± 7.7%, 82.7 ± 8.3% and 90.8 ± 6.5% at 25.0 μg ml⁻¹, respectively, using quercetin¹⁹ (cell viability of 83.6 ± 5.2%, 20 μg ml⁻¹) and ginsenoside Rb 1 (ref. 20) (cell viability of 78.1 ± 4.5%, 50 μg ml⁻¹) as positive controls.

Fig. 5 Protective effects of compounds 1–6 against H₂O₂-induced H9c2 cardiomyocyte injury. (A) Effects of 1–6 pretreatment on H₂O₂-induced H9c2 cell injury. (B) Effects of 1–6 pretreatment on caspase-3 activity. (C) Effects of 1–6 pretreatment on caspase-9 activity. The data are expressed as the means ± SD from three independent experiments. ^aP < 0.01 versus control; ^bP < 0.05 versus H₂O₂-treated cells; ^cP < 0.01 versus H₂O₂-treated cells.

Treatment with 150 mM H₂O₂ caused a significant increase in LDH release and intracellular MDA levels in H9c2 cardiomyocyte, whereas preincubation with 25 and 50 μg ml⁻¹ compounds 1–6 markedly decreased LDH release and intracellular MDA levels (Table 1). In addition, the activities of some endogenous anti-oxidative enzymes, such as SOD, CAT, and GSH-Px activities in the H₂O₂-treated cells were decreased compared with the control group, whereas pretreatment with 1–6, especially 2, 4 and 6, effectively increased SOD, CAT, and GSH-Px levels. These data suggest that 1–6 reduced oxidative injury by enhancing the endogenous anti-oxidative capacity.

Table 1 Effects of compounds 1–6 on lipid peroxidation and anti-oxidant enzyme activity (means ± SD, n = 3)

	Group	LDH (U L^−1)	MDA (nmol per mg protein)	SOD (U per mg protein)	GSH-Px (U per mg protein)	CAT (U per mg protein)
a P < 0.01 versus control.b P < 0.05.c P < 0.01 versus H2O2.
	Control	176.53 ± 15.68	1.21 ± 0.13	86.61 ± 12.47	3.33 ± 0.37	35.06 ± 5.22
	H2O2	1790.57 ± 84.12^a	3.55 ± 0.70^a	30.22 ± 2.38^a	1.36 ± 0.15^a	12.55 ± 2.35^a
1	(12.5 μg ml^−1)	1395.74 ± 53.75^b	3.12 ± 0.54	35.96 ± 1.96	1.42 ± 0.23	13.34 ± 1.72
	(25 μg ml^−1)	875.45 ± 41.97^c	2.23 ± 0.36^c	46.74 ± 3.57^c	1.53 ± 0.31	14.58 ± 1.29
	(50 μg ml^−1)	686.27 ± 33.62^c	1.58 ± 0.11^c	66.59 ± 3.38^c	1.95 ± 0.17^b	18.67 ± 0.94^b
2	(12.5 μg ml^−1)	549.58 ± 44.52^c	2.46 ± 0.42^b	36.14 ± 1.24	1.56 ± 0.22	26.94 ± 2.74^c
	(25 μg ml^−1)	445.13 ± 35.91^c	1.33 ± 0.22^c	47.26 ± 3.75^b	2.53 ± 0.31^c	31.75 ± 2.28^c
	(50 μg ml^−1)	431.53 ± 33.67^c	1.35 ± 0.28^c	62.47 ± 5.16^c	2.74 ± 0.19^c	29.56 ± 1.67^c
3	(12.5 μg ml^−1)	1266.12 ± 93.26^c	3.03 ± 0.33	43.52 ± 1.82	1.62 ± 0.22^b	22.53 ± 1.58^c
	(25 μg ml^−1)	717.56 ± 62.71^c	3.12 ± 0.24	56.30 ± 2.79^c	1.58 ± 0.24	28.68 ± 3.11^c
	(50 μg ml^−1)	560.49 ± 18.38^c	2.66 ± 0.41^b	72.95 ± 3.65^c	1.72 ± 0.20^b	29.92 ± 2.08^c
4	(12.5 μg ml^−1)	964.28 ± 71.11^c	2.85 ± 0.27^b	43.63 ± 3.43	1.77 ± 0.27^b	14.71 ± 0.76
	(25 μg ml^−1)	776.53 ± 31.89^c	2.02 ± 0.25^c	56.53 ± 2.86^c	2.55 ± 0.16^c	19.64 ± 1.37^c
	(50 μg ml^−1)	651.45 ± 26.71^c	1.66 ± 0.25^c	72.24 ± 4.98^c	3.11 ± 0.25^c	19.77 ± 1.90^c
5	(12.5 μg ml^−1)	1463.44 ± 99.16	2.83 ± 0.34^b	39.36 ± 2.08	1.59 ± 0.08	16.48 ± 2.00^b
	(25 μg ml^−1)	1109.02 ± 88.32^c	2.71 ± 0.27^b	52.37 ± 3.14^c	1.53 ± 0.13	18.99 ± 0.93^b
	(50 μg ml^−1)	716.37 ± 49.77^c	2.76 ± 0.35^b	59.44 ± 5.83^c	1.73 ± 0.22^b	16.43 ± 1.86^b
6	(12.5 μg ml^−1)	610.80 ± 41.08	1.63 ± 0.23^c	66.68 ± 5.71^c	2.14 ± 0.24^c	21.85 ± 1.69^c
	(25 μg ml^−1)	361.66 ± 12.55^c	1.33 ± 0.11^c	77.55 ± 8.20^c	3.25 ± 0.34^c	25.16 ± 2.09^c
	(50 μg ml^−1)	356.29 ± 28.95^c	1.28 ± 0.14^c	82.38 ± 7.31^c	3.31 ± 0.19^c	25.38 ± 1.66^c
Quercetin (20 μg ml^−1)		479.90 ± 21.36^c	1.33 ± 0.09^c	72.21 ± 4.55^c	3.10 ± 0.36^c	27.72 ± 1.79^c
Ginsenoside Rb 1 (50 μg ml^−1)		513.44 ± 23.01^c	1.38 ± 0.28^c	54.04 ± 2.63^c	1.76 ± 0.22^b	18.53 ± 1.08^c

---

Caspases are a family of aspartate-specific cysteine proteases that play key roles in regulating the cellular and biochemical changes associated with apoptosis. As shown in Fig. 5B and C, the level of activated caspase-3 and caspase-9 increased significantly in H₂O₂-treated cardiomyocytes. However, lower levels of activated caspase-3 and caspase-9 in compounds 1–6, especially 2, 4, and 6, pretreatment groups were observed.

In the present investigation, 2, 4 and 6 have powerful activities in protecting against H₂O₂ induced H9c2 cardiomyocyte injury. The mechanism may be associated with the inhibition of oxidative stress and subsequent apoptosis, as characterized by increasing SOD, GSH-Px and CAT activities, and inhibiting the activation of associated caspases. These results suggest that clinoposides B, D and F are powerful antioxidants. Of course, further studies will be necessary to determine the cardioprotective effects of clinoposides B, D and F in vivo assays.

Experimental

General experimental procedures

Sephadex LH-20 (Pharmacia, Uppsala, Sweden), MCI gel (CHP 20P, 75–150 μm, Mitsubishi Chemical Corporation, Tokyo, Japan), C-18 reversed-phase silica gel (50 μm, YMC CO., LTD., Kyoto, Japan), and silica gel (100–200 mesh, Qingdao Haiyang Chemical Co., Ltd, Qingdao, China) were used for column chromatography, and precoated silica gel GF254 plates (Yantai Chemical Industry Research Institute, Yantai, China) were used for TLC. HPLC separation was performed on a CXTH LC-3000 HPLC system equipped with a CXTH LC-3000 UV spectrophotometric detector and a YMC (250 × 10 mm) semi-preparative column packed with C18 (5 μm, YMC CO., LTD., Kyoto, Japan). Optical rotation data were recorded using a Perkin-Elmer 341 digital polarimeter in MeOH. UV data were obtained using a Shimadzu UV2550 spectrometer in MeOH. ECD spectra were measured in MeOH on a JASCO J-815 spectropolarimeter. IR data were recorded using an FTIR-8400S spectrometer. NMR spectra were obtained using a Bruker AV III 600 NMR spectrometer with the chemical shift values presented as δ values having TMS as an internal standard. HR-ESI-MS spectra were measured using a Synapt G2 MS system (Waters Corp., USA). Sugar analysis was performed on an Agilent 6890N GC equipped with a FID detector under the following conditions: injector temp. of 250 °C, detector temp. of 250 °C, and N₂ as carrier gas. D-Glucopyranose was purchased from Sinopharm Chemical Reagent Co., Ltd, and D-fucose was purchased from Alfa Aesar. TLC was conducted on precoated silica gel GF₂₅₄ plates (Qingdao Haiyang Chemical Co., Ltd). All organic solvents employed were of analytical grade (Beijing Chemical Works, China), solvents used for HPLC were of HPLC grade from Thermo Fisher Scientific Inc. (Waltham, MA, USA), and water for extraction was produced by Beijing University of Chinese Medicine, water for HPLC was purchased from Wow haha food co., ltd (Hangzhou, Zhejiang, China). Quercetin and ginsenoside Rb 1 (purity > 99%) were obtained from Shanghai Winherb Medical S & T Development (Shanghai, China). H₂O₂ was purchased from Beijing Chemical Works (Beijing, China). Cell culture products were purchased from Gibco BRL (Grand island, NY). The kits for determining lactate dehydrogenase (LDH) and malondialdehyde (MDA) contents and superoxide dismutase (SOD), catalase (CAT), glutathione peroxidase (GSH-Px) activity were obtained from Jiancheng Bioengineering Institute (Nanjing, China). Caspase-3 and -9 Fluorometric Assay Kits were acquired from BioVision (Palo Alto, CA).

Plant materials

C. chinense was purchased from Chinese Medicinal Material Markets (Bozhou, China) in 2013. Identification of plant overground parts was provided by Professor Shichun Yu of the School of Pharmacy, Anhui University of Chinese Medicine. A voucher specimen (20130312) was deposited in Institute of Medicinal Plant Development.

Extraction and isolation

The dried and powdered aerial parts (15 kg) of C. chinense were extracted twice with 100 L of 70% ethanol (v/v) at 80 °C for 2 h. The extract was filtered and evaporated to completely remove ethanol. The residue was suspended in water and successively extracted with petroleum ether, ethyl acetate and n-BuOH. Approximately 310 g of n-BuOH residue was loaded on a D101 column (700 mm × 250 mm) and then eluted with 10 L of 20% ethanol, 20 L of 50% ethanol (v/v), 25 L of 85% ethanol (v/v), and 15 L of 100% ethanol (v/v). The effluents of 50% and 85% ethanol were collected and evaporated at 55 °C under vacuum. Finally, we obtained 180 g of the 50% and 85% ethanol fractions of the total saponins. The total saponins (180 g) were subjected to SiO₂ column chromatography with CHCl₃/MeOH. The fraction eluted with CHCl₃/MeOH (100 : 15) (20 g) was purified by flash chromatography on RP-18 with MeOH/H₂O to afford fractions A1–A6. Fraction A4 was applied to an ODS column (MeOH/H₂O), 50 : 50 → 85 : 15 and then only MeOH to afford six fractions (frs. A4.1–6). Fraction A4.3 was separated by ODS HPLC (YMC ODS-A, 10 × 250 mm; flow rate 2.0 ml min⁻¹; UV detection at 210 nm; eluent MeOH/H₂O, 74 : 26) to isolate clinoposide D (4, 4.5 mg, 0.00003%, wet weight), and fraction A4.4 was separated by ODS HPLC (YMC ODS-A, 10 × 250 mm; flow rate 2.0 ml min⁻¹; UV detection at 210 nm; eluent MeOH/H₂O, 74 : 26) to isolate clinoposides E (5, 3.5 mg, 0.000023%, wet weight) and F (6, 2.1 mg, 0.000014%, wet weight). The fraction eluted with CHCl₃/MeOH (100 : 25) (30 g) was purified by flash chromatography on RP-18 with MeOH/H₂O to afford fractions B1–B9. The fraction was eluted with 75% MeOH (300 mg) by Sephadex LH-20 (MeOH) to afford 4 fractions (Frs B7.1–7.4). Fraction B7.3 was separated by ODS HPLC (YMC ODS-A, 10 × 250 mm; flow rate 2.0 ml min⁻¹; UV detection at 210 nm; eluent MeOH/H₂O, 70 : 30) to isolate clinoposides A (1, 5.6 mg, 0.000037%, wet weight) and B (2, 3.4 mg, 0.000023%, wet weight). The fraction was eluted with CHCl₃/MeOH (100 : 35) (25 g) was purified by flash chromatography on RP-18 with MeOH/H₂O (50 to 100%) to afford fractions C1–C7. The fraction eluted with 75% MeOH (500 mg) was purified by Sephadex LH-20 (MeOH) to give 4 fractions. Fraction B6.4 was separated by ODS HPLC (YMC ODS-A, 10 × 250 mm; flow rate 2.0 ml min⁻¹; UV detection at 210 nm; eluent MeOH/H₂O, 70 : 30) to isolate clinoposide C (3, 5.3 mg, 0.000035%, wet weight).

Clinoposide A (1). Yellow powder; [α]²⁰_D = 6.7 (c 0.12, MeOH); UV (MeOH) λ_max (log ε) 227 (4.33), 297 (4.10), 341 (3.27) nm; IR (film) v_max 3368, 2942, 1628, 1520, 1446, 1068, 833 cm⁻¹; ¹H and ¹³C NMR (Table S1†); HR-ESI-MS: m/z 1213.5745 [M − H]⁻ (calcd for C₆₃H₈₉O₂₃, 1213.5795).

Clinoposide B (2). Yellow powder; [α]²⁰_D = −10.0 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 298 (4.22), 344 (3.65) nm; IR (film) v_max 3360, 2943, 1634, 1520, 1436, 1074, 835 cm⁻¹; ¹H and ¹³C NMR (Table S1†); HR-ESI-MS: m/z 1213.5743 [M − H]⁻ (calcd for C₆₃H₈₉O₂₃, 1213.5795).

Clinoposide C (3). Yellow powder; [α]²⁰_D = 20.0 (c 0.12, MeOH); UV (MeOH) λ_max (log ε) 228 (4.19), 298 (3.98), 341 (3.13) nm; IR (film) v_max 3368, 2938, 1630, 1448, 1071, 833 cm⁻¹; ¹H and ¹³C NMR (Table S2†); HR-ESI-MS: m/z 1213.5751 [M − H]⁻ (calcd for C₆₃H₈₉O₂₃, 1213.5795).

Clinoposide D (4). Yellow powder; [α]²⁰_D = −6.7 (c 0.09, MeOH); UV (MeOH) λ_max (log ε) 297 (4.15), 342 (3.46) nm; IR (film) v_max 3366, 2943, 1635, 1517, 1074, 832 cm⁻¹; ¹H and ¹³C NMR (Table S2†); HR-ESI-MS: m/z 1227.5927 [M − H]⁻ (calcd for C₆₄H₉₁O₂₃, 1227.5956).

Clinoposide E (5). Yellow powder; [α]²⁰_D = −4.6 (c 0.13, MeOH); UV (MeOH) λ_max (log ε) 226 (4.26), 297 (3.99), 341 (3.30) nm; IR (film) v_max 3367, 2939, 1635, 1516, 1447, 1056, 832 cm⁻¹; ¹H and ¹³C NMR (Table S3†); HR-ESI-MS: m/z 1227.5916 [M − H]⁻ (calcd for C₆₄H₉₁O₂₃, 1227.5956).

Clinoposide F (6). Yellow powder; [α]²⁰_D = −20.0 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 227 (4.33), 296 (4.05), 338 (3.34) nm; IR (film) v_max 3369, 2934, 1635, 1516, 1444, 1070, 830 cm⁻¹; ¹H and ¹³C NMR (Table S3†); HR-ESI-MS: m/z 1227.5955 [M − H]⁻ (calcd for C₆₄H₉₁O₂₃, 1227.5956).

Acid hydrolysis

Clinoposides A–F (1 mg) were heated at 95 °C for 2 h in 1 ml of 2 N TFA. The solutions were extracted three times with 1 ml of CHCl₃. TLC analysis of the organic phase revealed decomposition of the aglycone. The aqueous phase was dried, and the residue was redissolved in anhydrous pyridine. The sugars were derivatized with L-cysteine methyl ester hydrochloride (1.5 mg, 60 °C, 1 h) and subsequently silylated with hexamethyldisilazane and chlorotrimethylsilane (Fluka) (2 : 1, 0.5 ml; 60 °C, 30 min). GC analysis was performed on a capillary HP-5 column (30 m × 0.25 mm i.d., 0.25 m; Agilent; column temp. 150 °C for 2 min, then 5 °C min⁻¹ to 210 °C) using the N₂ (1 ml min⁻¹) as carrier gas. A 3 μL aliquot of the supernatant was injected for analysis. The injection and detector temperature were set at 250 °C, and the splitting ratio was 1/50.

General procedure for ECD calculation

The absolute configurations of compounds 1–6 were computationally determined by comparing the experimental and calculated electronic circular dichroism (ECD) spectra using time-dependent density functional theory (TDDFT). Conformational analysis was initially conducted by using Discovery Studio 4.1 client conformational searching and MMFF94 molecular mechanics methods. The selected conformers were then optimized at the B3LYP/6-31G(d) level in the gas phase using Gaussian09.²¹ Further ECD calculations were performed at the (PCM-B3LYP)/6-31G(d,p)//B3LYP/6-31G(d,p) level in MeOH solution. The results suggested that the calculated weighted ECD spectra of 1 with 11S, 2′S configurations, 2 and 4 with 11R, 2′R configurations, 3 with 11S, 2′R, 5 with 11S, 2′S, and 6 with 11S, 2′R configurations are in accordance with the experimental spectra (see Fig. S12, S25, S38, S64 and S77†). Consequently, the absolute configurations of compounds 1–6 were unambiguously assigned.

Cell culture

H9c2 cells were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai). The cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% (v/v) fetal bovine serum, 2 mM L-glutamine, 100 U ml⁻¹ of penicillin, 100 mg ml⁻¹ of streptomycin and maintained in a humidified incubator of 95% air/5% CO₂ at 37 °C. For all experiments, the cells were plated at an appropriate density according to the experimental design and were grown for 36 h before experiment.

Analysis of cell viability

Cell viability was determined by MTT assay. Briefly, H9c2 cells at a density of 5 × 10⁴ cells/well in 96-well plates were cultured at 37 °C under 5% CO₂ in a humidified incubator for 24 h. After 24 h of treatment with different concentrations of drugs, followed by incubation with 150 mM H₂O₂ for 2 h, 20 μL of 5 mg ml⁻¹ MTT solution was added to each well (0.1 mg per well), and the plates were incubated for 4 h. The supernatants were discarded, and the metabolized MTT in each well was dissolved in 150 μL of DMSO. Optical density was measured at 570 nm on a microplate reader (BioTek, Vermont). The survival rate of H9c2 cells was evaluated and the inhibition (%) was expressed as the percentage of control.

Measurement of LDH and MDA levels and the activity of SOD, CAT, and GSH-Px

H9c2 cells were cultured in six-well plates at 3 × 10⁵ cells per well. After the experimental procedures, the supernatant was used to measure the level of LDH release using an LDH assay kit. The cells were collected, ultrasonicated, and centrifuged at 1000 rpm for 5 min at 4 °C. The supernatant was used to assess activities of MDA, SOD, CAT, and GSH-Px according to the corresponding detection kits.

Analysis of caspase-3 activation and caspase-9 activity

Caspase-3 activation was measured using a fluorescein active caspase-3 staining kit (BioVision). About 300 ml (1 × 10⁶ cells per ml) of the cultures were incubated with 1 ml substrate FITC-DEVD-FMK for 1 h at 37 °C. The cells were centrifuged at 3000 rpm for 5 min and the supernate was removed, then the cells were washed twice with PBS, resuspended in 300 ml of wash buffer, and kept on ice. The samples were analyzed by flow cytometry using the FL-1 channel. Caspase-9 activity was measured using a Fluorometric Assay Kit (BioVision) according to the manufacturer's instructions. The fold-increases in caspase activity were determined by comparing the results with the level of the control group.

Statistical analysis

The bioactivity data are expressed as means ± SD from three independent experiments. The differences were analyzed by ANOVA, followed by post hoc analysis with Student–Newman–Keuls test. Statistical significance was considered at P < 0.05.

Conclusions

In conclusion, six novel flavonoid–triterpene saponin meroterpenoids with two unique skeletons were isolated from C. chinense, representing a new family of isopentenyl flavonoids. Some meroterpenoid exhibited strong protective effects against H₂O₂-induced H9c2 cells injury. Further investigation of action mechanism of these compounds in vivo will be of great significance to drug discovery.

Acknowledgements

This work was supported by grants (No. 81173511 and 81374010) from the National Natural Sciences Foundation of China. The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Notes and references

1. P. M. Dewick, Medicinal Natural Products: A Biosynthetic Approach, John Wiley & sons, Baffins Lane, Chichester, West Sussex, UK, 3rd edn, 2009, p. 187.
2. B. Botta, A. Vitali, P. Menendez, D. Misiti and G. Delle Monache, Curr. Med. Chem., 2005, 12, 717–739.
3. M. C. Burger, J. B. Fernandes, M. F. da Silva, A. Escalante, J. Prudhomme, K. G. Le Roch, M. A. Izidoro and P. C. Vieira, J. Nat. Prod., 2014, 77, 2418–2422.
4. Y. J. Sun, Z. Y. Hao, J. G. Si, Y. Wang, Y. L. Zhang, J. M. Wang, M. L. Gao and H. Chen, RSC Adv., 2015, 5, 82736–82742.
5. K. Ichino, H. Tanaka and K. Ito, Chem. Pharm. Bull., 1989, 37, 944–947.
6. M. L. Zhong, X. D. Xu, S. C. Yu and G. L. Sun, Chin. Tradit. Herb. Drugs, 2012, 43, 820.
7. H. Aoshima, T. Miyase and T. Warashina, Chem. Pharm. Bull., 2012, 60, 499–507.
8. S. N. Wang, G. X. Ma, M. L. Zhong, S. C. Yu, X. D. Xu, Y. X. Hu, Y. Z. Zhang, H. Wei and J. S. Yang, Fitoterapia, 2013, 90, 14–19.
9. T. Murata, K. Sasaki, K. Sato, F. Yoshizaki, H. Yamada, H. Mutoh, K. Umehara, T. Miyase, T. Warashina, H. Aoshima, H. Tabata and K. Matsubara, J. Nat. Prod., 2009, 72, 1379–1384.
10. G. X. Li, J. Biol., 1989, 3, 1381.
11. D. N. Tian, F. H. Wu, S. C. Ma, D. Li and Y. Dai, China J. Chin. Mater. Med., 2008, 33, 1313–1316.
12. B. Dzhambazov, S. Daskalova, A. Monteva and N. Popov, Biol. Pharm. Bull., 2002, 25, 499–504.
13. M. L. Zhong, H. F. Wu, X. P. Zhang, G. L. Sun, G. B. Sun, S. C. Yu and X. D. Xu, Nat. Prod. Res., 2014, 28, 467–472.
14. M. L. Zhong, G. B. Sun, X. P. Zhang, G. L. Sun, X. D. Xu and S. C. Yu, Molecules, 2012, 17, 13910–13916.
15. S. B. Mahato and A. P. Kundu, Phytochemistry, 1994, 37, 1517–1575.
16. D. Slade, D. Ferreira and J. P. Marais, Phytochemistry, 2005, 66, 2177–2215.
17. J. B. Harborne, The Flavonoids: Advances in Research since 1980, Chapman & Hall, London, UK, 1988, p. 223.
18. K. Yazaki, K. Sasaki and Y. Tsurumaru, Phytochemistry, 2009, 70, 1739–1745.
19. X. Sun, G. B. Sun, M. Wang, J. Xiao and X. B. Sun, J. Cell. Biochem., 2011, 112, 2019–2029.
20. Z. Liu, J. Chen, W. Huang, Z. Zeng, Y. Yang and B. Zhu, Mol. Med. Rep., 2013, 8, 1397–1403.
21. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and D. J. Fox, 09 Gaussian, Revision C.01, Gaussian, Inc., Wallingford, CT, 2010.

---

Footnotes

† Electronic supplementary information (ESI) available: Supporting information as well as MS, NMR, UV, CD, and IR spectra of compounds 1–6. See DOI: 10.1039/c5ra27485k

‡ These authors contributed equally to this work.

---