Skip to content
Publicly Available Published by De Gruyter July 30, 2015

Two new secondary metabolites from the fruits of mangrove Avicennia marina

Dong-Mei Yan , Cheng-Hai Gao , Xiang-Xi Yi , Wen-Pei Xie , Ming-Ben Xu and Ri-Ming Huang EMAIL logo

Abstract

Further chemical investigation of the fruit of the mangrove Avicennia marina led to the isolation of a new caffeic acid derivative, maricaffeolylide A (1), and a new megastigmane derivative, maricyclohexene A (2). The structures of the isolates were elucidated on the basis of extensive spectroscopic analysis. The antioxidant activity of 1 and 2 was also evaluated using a cellular antioxidant activity assay, and catechol 1 showed antioxidant activity with an EC50 value of 24 ± 0.3 μm.

1 Introduction

Previous chemical investigations on the genus Avicennia have shown the presence of iridoid glucosides [1–4], naphthoquinone derivatives [5, 6], flavonoids [3, 7], and diterpenoids [8]. In our study of bioactive compounds from A. marina, we have reported some secondary metabolites [9, 10]. Further searching for bioactive metabolites from the fruits of this specimen has led to the isolation of a new caffeoyl derivative, maricaffeolylide A (1) and a new megastigmane derivative, diendiol maricyclohexene A (2) (Fig. 1). In this paper, we describe the isolation and structure elucidation of these two new compounds and their antioxidant activity.

Fig. 1: Compounds 1 and 2.
Fig. 1:

Compounds 1 and 2.

2 Results and discussion

Compound 1, a yellow oil, was established as C17H20O6 from the HR-ESI-MS at m/z = 321.1336 [M+H]+ and the NMR data, indicating eight degrees of unsaturation. Analysis of the NMR data (Table 1) displayed the presence of one methyl, four sp3 methylenes, two sp3 methines, five sp2 methines, and five sp2 quaternary carbons. 1H NMR resonances at δH = 7.02 (d, J = 1.9 Hz, H-2), 6.93 (dd, J = 8.2, 1.9 Hz, H-6) and 6.77 ppm (d, J = 8.2 Hz, H-5) showed the typical coupling pattern of a caffeoyl moiety with two olefinic protons at δH = 7.51 (d, J = 15.9 Hz, H-7) and 6.19 ppm (d, J = 15.9 Hz, H-8), exhibiting the expected coupling constant of 15.9 Hz for an E-configuration of the caffeoyl residue in compound 1. The protons resonating at δH = 5.28–5.24 (m, H-6′), 4.04–4.00 (m, H-7′), 2.24–2.20 (m, H-2′a), 2.01–1.97 (m, H-2′b), 1.68–1.64 (m, H-3′), 1.66–1.62 (m, H-5′), and 1.51–1.49 ppm (m, H-4′), were determined to be spin systems of H2-2′/H2-3′/H-4′/H2-5′/H-6′/H2-7′ by COSY experiments. These data, together with 13C NMR resonances at δC = 175.0 (C-1′), 72.2 (C-6′), 66.4 (C-7′), 38.0 (C-2′), 31.6 (C-4′), 31.4 (C-5′), and 20.1 ppm (C-3′) indicated an eight-membered lactone in 1, which was fully supported by the HMBC correlation H-7′ with C-1′. The gross structure was further established by the aid of COSY and HMBC experiments (Fig. 2). In the HMBC spectra, the methyl group (4′-Me) attached to C-4′ was evident from the key HMBC correlations of 4′-Me with C-3′ and C-5′. The caffeoyl moiety attached to C-6′ was confirmed by the key HMBC correlation of H-6′ with C-9. The gross structure of 1 containing an unusual saturated 8-membered ring lactone with two chiral centers was determined by mass spectra and NMR studies. Though the relative stereochemistries at C-4′/C-6′ with H-4′ and H-6′ in the α-equatorial position were determined by the strong NOESY correlation between the H-4′ and H-6′ (Fig. 3). Although the similarly featured characteristic structure was also found in cephalosporolide D whose configuration was established independently by synthesis [11], it is hard for us to have an explanation for the difference in the absolute stereochemistry of the both compound 1 and cephalosporolide D through the NMR, CD spectrum and optical rotation. In this communication, the determination of the exact absolute stereochemistry by the crystallization is further developed. Taking all of the spectroscopic data into consideration, the structure of 1 was determined as shown in Fig. 1.

Table 1

1H and 13C NMR spectroscopic data of 1 (in CD3OD).

PositionδCδH (J)
1127.6
2114.87.02 (d, 1.9)
3146.9
4149.7
5116.26.77 (d, 8.2)
6122.76.93 (dd, 8.2, 1.9)
7147.27.51 (d, 15.9)
8114.76.19 (d, 15.9)
9168.2
1′175.0
2′a38.02.24–2.20 (m)
 b2.01–1.97 (m)
3′20.11.68–1.64 (m)
4′31.61.53–1.49 (m)
5′31.41.66–1.62 (m)
6′72.25.28–5.24 (m)
7′66.44.04–4.00 (m)
4′-Me13.70.98 (d, 6.5)

Chemical shifts δ in ppm, multiplicities and J values (in Hz) in parentheses.

Fig. 2: Key 1H-1H COSY and HMBC correlations of 1 and 2.
Fig. 2:

Key 1H-1H COSY and HMBC correlations of 1 and 2.

Fig. 3: Key NOESY correlations of 1 and 2.
Fig. 3:

Key NOESY correlations of 1 and 2.

Compound 2, a yellow oil, was established as C12H20O2 from the HR-ESI-MS peak at m/z = 197.1540 [M+H]+ and the NMR data, indicating three degrees of unsaturation. In the 1H NMR spectrum (Table 2) three methyl singlet were present at δH = 1.06 (s, H-10), 1.09 (s, H-11) and 2.28 ppm (s, H-12) appeared, one methyl doublet at δH = 1.85 ppm (d, J = 6.5 Hz, H-9), one methylene proton as two double doublets at δH = 1.80 (dd, J = 12.5, 12.0 Hz, H-2a) and 1.43 ppm (dd, J = 12.5, 3.0 Hz, H-2b), two methine protons geminal to hydroxyls resonated as a double triplet at δH = 3.78 ppm (dt, J = 12.0, 3.0 Hz, H-3) and a doublet at δH = 3.85 ppm (d, J = 3.0 Hz, H-4), and the resonances of two olefinic protons were found as a double doublet at δH = 6.53 ppm (d, J = 16.0 Hz, H-7) and a doublet at δH = 6.03 ppm (dd, J = 16.0, 6.5 Hz, H-8). The 13C NMR spectrum (Table 2) showed 12 carbon signals, identified by a DEPT experiment as four methyl groups, one methylene, four methines and three quaternary carbons. All the carbons were correlated to the corresponding protons on the basis of the HMQC experiment.

Table 2

1H and 13C NMR spectroscopic data of 2 (in CD3OD).

PositionδCδH (J)
137.8
2a41.71.80 (dd, 12.5, 12.0)
b1.43 (dd, 12.5, 3.0)
367.83.78 (dt, 12.0, 3.0)
472.63.85 (d, 3.0)
5132.9
6142.9
7130.36.53 (d, 16.0)
8139.26.03 (dd, 16.0, 6.5)
9-Me20.01.85 (d, 6.5)
10-Me30.41.06 (s)
11-Me27.91.09 (s)
12-Me13.72.28 (s)

Chemical shifts δ in ppm, multiplicities and J values (in Hz) in parentheses.

The 1H-1H COSY experiment showed a correlation beginning with the doublet methyl H3-9 which was coupled with the H-8. This latter proton was correlated with the H-7. H-3 was correlated with H-4 and H2-2. Two hydroxyl groups were positioned at the C-3 and C-4 on the basis of the HMBC and NOESY experiments. The correlations of the H-4, H-8 and H-12 protons with the C-6 carbon and those of the H-12 protons with C-4 and C-5 confirmed the hypothesis. The analysis of the NOESY spectrum evidenced an NOE between the H-12 methyl and the H-4 protons. The cis-relative configuration at C-3 and C-4 was predicted on the basis of the small coupling constant 3JH-3,H-4 (J = 3.0 Hz). Finally, the absolute configurations of C-3 and C-4, was determined by its CD spectrum, in which a positive Cotton effect by cyclohexene group was show at 205 nm (Δε + 10.2), whereas the reported value for ixero B whose stereostructures of C-3 and C-4 were determined by CD spectrum, was negative [12]. Although we do not have an explanation for the difference in the absolute values of the Cotton effect, appplicaton of the octant rule to the compound depicted in the formula of 2 found that the expected sign of the Cotton effect should be positive. On the basis of this cumulative analysis, the structure of diendiol 2 was thus determined as shown in Fig. 1.

The cellular antioxidant assay (CAA) is a new approach to quantify antioxidants under physiological conditions when compared to the chemical antioxidant activity assays, and the CAA assay has been used in the marine natural product field recently [13]. The EC50 of compounds 1 and 2 were 24 ± 0.3 and 339 ± 3 μm, respectively. The EC50 value of catechol 1 is of the same order of the positive control, quercetin (EC50 = 11 ± 0.2 μm).

3 Experimental section

3.1 General

NMR spectra were recorded on a Bruker AC 600 NMR spectrometer (Bruker, Bremen, Germany) with TMS as an internal standard. HR-ESI-MS (Bruker, Bremen, Germany) data were obtained from a Bruker Maxis mass spectrometer. Waters-2695 HPLC (Waters, Milford, MA, USA) system was used with a SunfireTM C18 column (250 × 10 mm i.d., 10 μm) coupled to a Waters 2998 Photodiode Array detector (210 nm, 254 nm, 278 nm, 300 nm, 350 nm). Optical rotation data were measured on a Perkin-Elmer Model 341 polarimeter (Wellesley, MA, USA). CD spectra were recorded on a spectropolarimeter (MODEL J-810-150S; Tokyo, Japan). The silica gel GF254 used for TLC was supplied by the Qingdao Marine Chemical Factory (Qingdao, China). Spots were detected on TLC under UV light or by heating after spraying with 5 % H2SO4 in EtOH.

3.2 Plant material

The fruits of A. marina were collected from Beihai City, Guangxi province, China, in September 2011. The specimen was identified by Professor Hangqing Fan, Guangxi Mangrove Research Center, Guangxi Academy of Sciences. A voucher specimen (2011-GXAS-008) has been deposited in Guangxi Key Laboratory of Marine Environmental Sciences, Guangxi Academy of Sciences, China.

3.3 Extraction and isolation

The fruits of A. marina (35.4 kg, wet wt) were exhaustively extracted in a large metal bowl (diameter 80 cm, volume 50 L) with EtOH-CH2Cl2 (2:1, 3 × 30 L) at 25 °C for 3 × 4 d. The solvent was evaporated in vacuo [0.09 Mpa, 35 °C, rotary evaporator (Shanghai Biochemical Equipment Co., Ltd.)] to afford a syrupy residue (935 g) that was suspended in distilled water (1.5 L) and fractionated successively with petroleum ether (3 × 2 L), ethyl acetate (3 × 2 L), and n-butanol (3 × 2 L). The n-butanol-soluble portion (269 g) was subjected to column chromatography on silica gel, using CHCl3-MeOH (from 10:0 to 0:10) as eluent, giving eleven fractions (A–K). Fraction B was subjected to column chromatography to afford two sub-fractions (B1 and B2). Fraction B2 was separated by HPLC, using MeOH-H2O (45:55) to yield 1 (5.9 mg, Rt = 24.4 min, λmax = 360 nm). Fraction F was subjected to column chromatography on silica gel, eluted with CHCl3-MeOH (from 9:1 to 7:3), to yield five sub-fractions (F1–F5). Sub-fraction F3 was then further purified by HPLC, using MeOH-H2O (35:65) to yield 2 (1.5 mg, Rt = 16.1 min, λmax = 254 nm).

Maricaffeolylide A (1): Yellow oil (5.9 mg from 35.4 kg sample). – [α]20D = −26.2° (c = 0.13, MeOH). – CD (MeOH) Δε205 nm = −4.7. – IR (KBr): νmax = 3525 (OH, s), 2923 (CH2, m), 1740 (C=O, s), 1625 (C=C, m) cm–1. – 1H and 13C NMR data: see Table 1. – HRMS [(+)-ESI]: m/z (%) = 321.1336 (100) (calcd. 321.1338 for C17H21O6, [M+H]+).

Maricyclohexene A (2): Yellow oil (1.5 mg from 35.4 kg sample). – [α]20D = −18.1° (c = 0.11, MeOH). – CD (MeOH) Δε205 nm = +10.2. – IR (KBr): νmax = 3425 (OH, s), 1700 (C=O, s), 1635 (C=C, m) cm–1. – 1H and 13C NMR data: see Table 2. – HRMS [(+)-ESI]: m/z (%) = 197.1540 (100) (calcd. 197.1542 for C12H21O2, [M+H]+).

3.4 Antioxidant activity

The cellular antioxidant activity assay was performed following the literature method [13].


Corresponding author: Ri-Ming Huang, Key Laboratory of Plant Resources Conservation and Sustainable Utilization, South China Botanical Garden, Chinese Academy of Sciences, Guangzhou 510650, P. R. China; and Department of Pharmacy and Pharmacology, University of Bath, Bath BA2 7AY, UK, Fax: +86-20-37252958, E-mail:

Acknowledgments

This study was supported by grants from National Natural Science Foundation of China (nos. 31100260, 81260480), the Foundation of Key Laboratory of Plant Resources Conservation and Sustainable Utilization, South China Botanical Garden, Chinese Academy of Sciences (no. 201210ZS) and the Natural Science Foundation of Guangxi (2014GXNSFAA118048). Ri-Ming Huang acknowledges a CAS Academic Visitor Fellowship and thanks Dr. Ian S. Blagbrough, University of Bath, for many helpful discussions.

References

[1] Y. Sun, O. Y. Jie, Z. W. Deng, Q. S. Li, W. H. Lin, Magn. Reson. Chem.2008, 46, 638.Search in Google Scholar

[2] G. Konig, H. Rimpler, Phytochemistry1995, 24, 1245.10.1016/S0031-9422(00)81110-3Search in Google Scholar

[3] Y. Feng, X. M. Li, X. J. Duan, B. G. Wang, Chem. Biodivers. 2006, 3, 799.Search in Google Scholar

[4] M. T. Fauvel, A. Bousquetmelou, C. Moulis, J. Gleye, S. R. Jensen, Phytochemistry 1995, 38, 893.10.1016/0031-9422(94)00750-NSearch in Google Scholar

[5] C. Ito, S. Katsuno, Y. Kondo, H. T. W. Tan, H. Furukawa, Chem. Pharm. Bull.2000, 48, 339.Search in Google Scholar

[6] L. Han, X. S. Huang, H. M. Dahse, U. Moellmann, H. Z. Fu, S. Grabley, I. Sattler, W. H. Lin, J. Nat. Prod.2007, 70, 923.Search in Google Scholar

[7] M. Sharaf, M. A. El-Ansari, N. A. M. Saleh, Fitoterapia2000, 71, 274.10.1016/S0367-326X(99)00169-0Search in Google Scholar

[8] L. Han, X. S. Huang, H. M. Dahse, U. Moellmann, S. Grabley, W. H. Lin, I. Sattler, Planta Med. 2008, 74, 432.Search in Google Scholar

[9] X. X. Yi, Y. Chen, W. P. Xie, M. B. Xu, Y. N. Chen, C. H. Gao, R. M. Huang, Mar. Drugs2014, 12, 2515.10.3390/md12052515Search in Google Scholar PubMed PubMed Central

[10] C. H. Gao, X. X. Yi, W. P. Xie, Y. N. Chen, M. B. Xu, Z. W. Su, L. Yu, R. M. Huang, Mar. Drugs2014, 12, 4353.10.3390/md12084353Search in Google Scholar PubMed PubMed Central

[11] I. Shiina, Y. Fukuda, T. Ishii, H. Fujisawa, T. Mukaiyama, Chem. Lett.1998, 8, 831.Search in Google Scholar

[12] Y. F. Han, K. Gao, Z. J. Jia, Chin. Chem. Lett.2006, 17, 913.Search in Google Scholar

[13] A. L. K. Faller, E. Fialho, R. H. Liu, J. Agr. Food Chem.2012, 60, 4826.Search in Google Scholar

Received: 2014-5-19
Accepted: 2015-5-6
Published Online: 2015-7-30
Published in Print: 2015-9-1

©2015 by De Gruyter

Downloaded on 9.12.2022 from https://www.degruyter.com/document/doi/10.1515/znb-2014-0111/html?lang=en
Scroll Up Arrow