Briaexcavatins C–F, four new briarane-related diterpenoids from
the Formosan octocoral Briareum excavatum (Briareidae)
Ping-Jyun Sung,
a,b,*
Yu-Pei Chen,
a,cTsong-Long Hwang,
dWan-Ping Hu,
eLee-Shing Fang,
a,fYang-Chang Wu,
gJan-Jung Li
aand Jyh-Horng Sheu
c,*
a
National Museum of Marine Biology and Aquarium, Pingtung 944, Taiwan, ROC
b
Institute of Marine Biotechnology, National Dong Hwa University, Pingtung 944, Taiwan, ROC
c
Department of Marine Biotechnology and Resources, National Sun Yat-sen University, Kaohsiung 804, Taiwan, ROC
d
Graduate Institute of Natural Products, Chang Gung University, Taoyuan 333, Taiwan, ROC
e
Faculty of Biotechnology, Kaohsiung Medical University, Kaohsiung 807, Taiwan, ROC
f
Institute of Marine Biodiversity and Evolution, National Dong Hwa University, Pingtung 944, Taiwan, ROC
g
Graduate Institute of Natural Products, Kaohsiung Medical University, Kaohsiung 807, Taiwan, ROC
Received 18 January 2006; revised 23 March 2006; accepted 24 March 2006 Available online 13 April 2006
Abstract—Four new briarane-related diterpenoids, designated as briaexcavatins C–F (1–4), were isolated from the Formosan octocoral Briar-eum excavatum, collected off southern Taiwan coast. The structures of these new metabolites were elucidated by the interpretation of spec-troscopic and chemical methods. The configuration of 1 was further supported by molecular mechanics calculations. Briarane 1 has been shown to exhibit mild cytotoxicity toward MDA-MB-231 human breast tumor cells and briarane 3 was found to show activity to inhibit neutrophil elastase release in humans.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
Octocorals belonging to the genus Briareum (¼Solenopo-dium) (Briareidae) are organisms taxonomically placed within the order Alcyonacea and Gorgonacea.1,2 These organisms inhabit abundantly in the coral reefs of tropical and subtropical Indo-Pacific Ocean and Caribbean Sea and have been recognized as rich sources for marine natural products with novel structural features.3–5 Since the first briarane-type natural product, briarein A, was obtained from the Caribbean octocoral Briareum asbestinum in 1977,6 a number of briarane derivatives have been isolated from various marine organisms.3,5Briarane-related natural prod-ucts continue to attract the attentions of investigators because of the structural complexity and interesting biolog-ical activity associated with numerous compounds of this type.3,5In previous studies, 64 new briaranes, including ste-cholides I–N and 16-hydroxy-stecholide C acetate,7 excava-tolides A–Z,8 briaexcavatolides A–Z,9 briantheins A–C,10 and briaexcavatins A and B,11have been isolated from the octocoral Briareum excavatum (Nutting, 1911). During our
continuing studies on the chemical constituents of B. exca-vatum, four new diterpenoids, briaexcavatins C–F (1–4), were isolated. We describe herein the isolation, structure elu-cidation, and biological activity of these new metabolites.
2. Results and discussion
Specimens of the octocoral B. excavatum, collected at south-ern Taiwan coast, were minced and extracted with EtOAc. The extract was separated on silica gel column chromato-graphy to afford briaranes 1–4. Briaexcavatin C (1) was obtained as a white powder. The HRESIMS data recorded at m/z 549.2340 established the molecular formula of 1 as C28H36O11 (calcd for C28H36O11+H, 549.2336). Thus, 11 degrees of unsaturation were determined for 1. The IR spec-trum showed bands at 1792, 1744, and 1688 cm1, consis-tent with the presence of g-lactone, ester, and conjugated ketone groups in 1. The conjugated ketone group was further confirmed by13C NMR signals at d 200.7 (s), 154.5 (d), and 126.4 (d) (Table 1). The FABMS of 1 exhibited peaks at m/z 549 (M+H)+, 489 (M+HAcOH)+, 429 (M+H2AcOH)+, and 341 (M+HC3H7CO2H2AcOH)+, which suggested the presence of a butyryloxy and two acetoxy groups. In the13C NMR spectrum, five carbonyl resonances appeared at d 200.7 (s), 172.4 (s), 170.3 (2s), and 169.0 (s),
Keywords: Briarane; Briaexcavatin; Briareum excavatum; Octocoral; Cyto-toxicity; Human neutrophil elastase.
* Corresponding authors. Tel.: +886 8 8825037; fax: +886 8 8825087; e-mail:[email protected]
0040–4020/$ - see front matterÓ 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.tet.2006.03.086
supporting the presence of a ketone, a lactone, and three esters. In the1H NMR spectrum (Table 2), an n-butyryloxy group (d 2.24, 2H, t, J¼7.2 Hz; 1.66, 2H, m; 0.95, 3H, t, J¼7.2 Hz) and two acetate methyls (d 2.25, 3H, s; 2.23, 3H, s) were further observed. Thus, the13C NMR data ac-counted for six degrees of unsaturation and required 1 to be pentacyclic. The presence of a tetrasubstituted epoxide containing a methyl substituent was elucidated from the sig-nals of two oxygen-bearing quaternary carbons at d 68.8 (s, C-8) and 59.9 (s, C-17), and further confirmed from the proton signal of a methyl singlet resonating at d 1.64 (3H, s, H3-18). In addition, a trisubstituted epoxide containing a methyl substituent was deduced from the signals of an oxy-methine (dH3.06, 1H, d, J¼8.0 Hz, H-6; dC62.8, d, C-6), an oxygen-bearing quaternary carbon (d 59.5, s, C-5), and a methyl singlet resonating at d 1.34 (3H, s, H3-16). More-over, a methyl doublet (d 1.27, 3H, d, J¼7.2 Hz, H3-20), a methyl singlet (d 1.08, 3H, s, H3-15), two conjugated olefin protons (d 6.62, 1H, d, J¼10.4 Hz, H-14; 6.19, 1H, d,
J¼10.4 Hz, H-13), two aliphatic methine protons (d 2.94, 1H, dd, J¼9.6, 4.8 Hz, H-10; 2.93, 1H, m, H-11), a pair of methylene protons (d 2.48, 1H, dd, J¼16.0, 6.4 Hz; 2.11, 1H, d, J¼16.0 Hz, H2-4), and five oxygenated methine pro-tons (d 5.27, 1H, br s, H-2; 5.24, 1H, d, J¼9.6 Hz, H-9; 5.08, 1H, d, J¼6.4 Hz, H-3; 4.43, 1H, d, J¼8.0 Hz, H-7; 3.06, 1H, d, J¼8.0 Hz, H-6) were further assigned by the assistance of 1H–1H COSY and HSQC spectrum. From the1H–1H COSY and HMBC correlations, the epoxy groups positioned at C-5/ C-6 and C-8/C-17, acetoxy groups positioned at C-2 and C-9, and an n-butyryloxy group positioned at C-3, were established (Fig. 1).
The relative stereochemistry of 1 was elucidated from the NOE interactions observed in an NOESYexperiment (Fig. 2) and by the vicinal1H–1H coupling constants. As per conven-tion while analyzing the stereochemistry of briarane-type diterpenoids, H-10 and H3-15 were assigned to the a and b face, anchoring the stereochemical analysis because no NOE correlation was found between H-10 and H3-15. In the NOESY experiment of 1, H-10 gives NOE correlations to H-3 and H-11, and H-3 was found to show responses with H-2 and one proton of the C-4 methylene (d 2.11), indicating that these protons (H-2, H-3, H-4a, H-10, and H-11) are lo-cated on the same face of the molecule and therefore are as-signed as a protons, as the C-15 methyl is the b-substituent at C-1. Furthermore, H3-16 exhibited NOE correlations with H-4a and H-6, but not with H-7, suggesting that H3-16 and H-6 were positioned on the a face in the epoxy group and H-7 was b-oriented in the 10-membered ring. H-9 was found to show NOE correlations with H-10, H-11, and H3-18. From the detailed consideration of molecular models, H-9 was found to be reasonably close to H-10, H-11, and H3-18, while H-9 should be placed on the a face in 1, and H3-18 was b-oriented in the g-lactone ring. The cis geometry of the C-13/C-14 double bond was indicated by a 10.4 Hz coupling constant between H-13 (d 6.19) and H-14 (d 6.62) and by the NOE correlation between H-13 and H-14. Based on above observations, the configurations of all chiral centers of 1 are assigned as 1R*,2R*,3S*,5R*, 6S*,7S*,8R*,9S*,10S*,11R*,13Z,17R*. Geometrical opti-mization of 1 was performed with DISCOVER utilizing the consistent valence force field (CVFF) calculations for energy minimization. The calculated results were visualized using INSIGHT II, running on a Silicon Graphics IRIS (SGI) Indigo XS24/4000 workstation. The conformation search suggested that the most stable conformation and the calculated dis-tances of selected key protons of 1 are shown inTable 3. The new briarane diterpene, briaexcavatin D (2), had the mo-lecular formula of C33H44O13, as determined by HRESIMS. It was found that the13C and1H NMR spectra of 2 in CDCl
3 revealed mostly broad peaks when measured at room tem-perature (25C). In order to make more reliable assignments of the NMR signals of briarane 2, the13C and1H NMR spec-tra of 2 were measured at20C in CDCl
3(Tables 1 and 2). It was found that at this temperature the signals for each pro-ton and carbon of the molecule were sharpened and could be assigned by the assistance of 1D and 2D NMR experiments. By detailed analysis, the NMR data of 2 were very similar to those of a known metabolite, excavatolide C (5).8aHowever, the13C and1H NMR spectra revealed that the signals corre-sponding to the hydroxy group in 5 (dC 65.8, d, C-12; dH
Table 1.13C NMR data for diterpenoids 1–4
Carbon 1a 2b 3c 4b 1 43.3 (s)d 43.7 (s) 44.1 (s) 43.6 (s) 2 80.7 (d) 81.0 (d) 81.4 (d) 81.3 (d) 3 70.5 (d) 73.2 (d) 73.5 (d) 72.9 (d) 4 33.3 (t) 33.6 (t) 33.9 (t) 33.9 (t) 5 59.5 (s) 138.9 (s) 139.2 (s) 139.1 (s) 6 62.8 (d) 121.6 (d) 122.0 (d) 121.6 (d) 7 76.5 (d) 73.7 (d) 74.0 (d) 73.7 (d) 8 68.8 (s) 68.6 (s) 69.0 (s) 68.7 (s) 9 66.2 (d) 65.8 (d) 66.2 (d) 65.9 (d) 10 41.8 (d) 39.7 (d) 40.1 (d) 39.7 (d) 11 40.0 (d) 32.2 (d) 32.6 (d) 32.1 (d) 12 200.7 (s) 69.2 (d) 69.2 (d) 69.2 (d) 13 126.4 (d) 26.9 (t) 27.2 (t) 26.9 (t) 14 154.5 (d) 80.6 (d) 80.9 (d) 80.4 (d) 15 17.5 (q) 18.0 (q) 18.3 (q) 17.9 (q) 16 21.5 (q) 22.2 (q) 22.5 (q) 22.3 (q) 17 59.9 (s) 60.0 (s) 60.3 (s) 60.0 (s) 18 10.4 (q) 10.0 (q) 10.3 (q) 10.0 (q) 19 170.3 (s) 171.7 (s) 172.0 (s) 171.6 (s) 20 14.5 (q) 10.2 (q) 10.5 (q) 10.2 (q) Acetate methyls 21.8 (q) 22.2 (q) 22.6 (q) 22.2 (q) 21.7 (q) 22.2 (q) 22.5 (q) 22.1 (q) 21.6 (q) 21.9 (q) 21.6 (q) 21.0 (q) 21.3 (q) Acetate carbonyls 170.3 (s) 171.0 (s) 172.7 (s) 171.3 (s) 169.0 (s) 170.3 (s) 170.7 (s) 170.9 (s) 169.5 (s) 169.8 (s) 169.4 (s) 169.4 (s) 169.7 (s) n-Butyrate 13.6 (q) 13.6 (q) 18.1 (t) 18.1 (t) 36.0 (t) 35.8 (t) 172.4 (s) 171.7 (s) 3-Vinylpropionate 115.6 (t) 115.6 (t) 136.4 (d) 136.4 (d) 28.5 (t) 28.5 (t) 33.2 (t) 33.3 (t) 172.3 (s) 172.3 (s) Isovalerate 22.7 (2q) 25.8 (d) 43.7 (t) 171.3 (s) a Spectra recorded at 100 MHz in CDCl3at 25C. b Spectra recorded at 150 MHz in CDCl 3at20C. c Spectra recorded at 100 MHz in CDCl3at20C.
d Multiplicity deduced by DEPT and indicated by usual symbols. The
3.89, 1H, m, H-12) were replaced by those of a 3-vinyl-propionyloxy group (dC172.3, s; 136.4, d; 115.6, t; 33.2, t; 28.5, t; dH 5.77, 1H, m; 5.02, 2H, m; 2.35, 2H, m; 2.30, 2H, t, J¼7.2 Hz) in 2. In the HMBC experiment of 2, the carbon signal at d 172.3 (s), which showed a correlation with H-12 (d 5.02) was found to be correlated with the signal of the methylene protons at d 2.30, and was consequently assigned as the carbon atom of the 3-vinylpropionate car-bonyl. Thus, the 3-vinylpropionate ester should be posi-tioned at C-12 in 2. Furthermore, vinylpropionylation of 5 gave a less polar product, which was found to be identical with natural product 2 by comparison of the physical and spectral data, and confirmed the structure of diterpenoid 2. Briaexcavatin E (3) had the molecular formula of C33H46O13 as determined by HRESIMS, with 11 degrees of unsatura-tion thereby being determined for the molecule. The IR absorptions of 3 showed the presence of g-lactone (nmax 1787 cm1) and ester carbonyl groups (nmax 1742 cm1). Like as those of 2, the sharpened13C and1H NMR signals of 3 (Tables 1 and 2) were also obtained in CDCl3 at 20C. Carbonyl resonances in the13C NMR spectrum of 3 at d 172.7 (s), 172.0 (s), 171.3 (s), 170.7 (s), 169.8 (s), and 169.7 (s) confirmed the presence of a g-lactone and five other esters in the molecule. In the1H NMR spectrum, four acetate methyls were observed at d 2.37 (3H, s), 2.24 (3H, s), 2.16 (3H, s), and 1.93 (3H, s). The additional acyl group was found to be an isovaleryl group, which showed nine contiguous protons (d 2.12, 2H, t, J¼7.2 Hz; 2.03,
1H, m; 0.89, 23H, d, J¼6.8 Hz) were observed. The 13C NMR signal appeared at d 171.3 (s) correlated with the sig-nal of the methylene protons at dH2.12 in the HMBC spec-trum and was consequently assigned as the carbon atom of the isovalerate carbonyl. The 1H–13C long-range correla-tions observed in an HMBC experiment of 3 further revealed the connectivity between H-12 (d 5.04) and the carbonyl car-bon (d 171.3) of the isovalerate unit and demonstrated the location of the isovalerate to be at C-12. The positions of the other four acetoxy groups at C-2, C-3, C-9, and C-14 were also confirmed by the connectivities between the four oxymethine protons at d 5.17 (H-2), 5.72 (H-3), 5.32 (H-9), 4.82 (H-14) and the four ester carbonyls (d 172.7, s; 170.7, s; 169.8, s; 169.7, s) in the HMBC spectrum of 3. Moreover, isovalerylation of excavatolide C (5)8a yielded a compound that was found to be identical with diterpenoid 3by physical and spectral data comparison.
Our present study has also led to the isolation of the new briarane, briaexcavatin F (4). The molecular formula of C35H48O13was deduced from HRESIMS with m/z 699.2996 (calcd for C35H48O13+Na, 699.2993). This showed that briar-ane 4 contained 12 degrees of unsaturation. From detailed analysis, the NMR data of 4 (measured in CDCl3 at 20C) (Tables 1 and 2) were found to be close to those
of 2 and the known metabolite, excavatolide B (6),8a and showed the presence of a g-lactone, an n-butyryloxy, a 3-vinylpropionyloxy, and three acetoxy groups. Compari-son of the1H and13C NMR spectral data of 4 with those of
Table 2.1H NMR data for diterpenoids 1–4
Proton 1a 2b 3b 4b 2 5.27 br s 5.13 br s 5.17 br s 5.16 br s 3 5.08 d (6.4)c 5.69 m 5.72 d (6.8) 5.74 m 4a 2.11 d (16.0) 1.96 m 1.96 m 1.97 m 4b 2.48 dd (16.0, 6.4) 3.67 dd (16.0, 7.2) 3.72 dd (16.0, 6.8) 3.73 dd (15.2, 6.8) 6 3.06 d (8.0) 5.33 d (6.8) 5.37 d (6.4) 5.37 d (6.8) 7 4.43 d (8.0) 5.25 d (6.8) 5.30 d (6.4) 5.29 d (6.8) 9 5.24 d (9.6) 5.28 d (10.4) 5.32 d (10.4) 5.31 d (10.0) 10 2.94 dd (9.6, 4.8) 2.97 dd (10.4, 4.8) 3.01 dd (10.4, 5.2) 3.00 dd (10.0, 6.8) 11 2.93 m 2.57 m 2.59 m 2.60 m 12 5.02 m 5.04 m 5.03 m 13/130 6.19 d (10.4) 2.29 m; 1.86 m 2.01 m; 1.83 m 2.32 m; 1.87 m 14 6.62 d (10.4) 4.80 br s 4.82 br s 4.85 br s 15 1.08 s 0.78 s 0.82 s 0.83 s 16 1.34 s 1.88 s 1.92 s 1.93 s 18 1.64 s 1.55 s 1.58 s 1.59 s 20 1.27 d (7.2) 1.00 d (6.8) 1.04 d (7.2) 1.95 d (7.2) Acetate methyls 2.25 s 2.43 s 2.37 s 2.37 s 2.23 s 2.21 s 2.24 s 2.25 s 2.13 s 2.16 s 2.17 s 1.90 s 1.93 s n-Butyrate 2.24 t (7.2) 2.33 t (7.2) 1.66 m 1.57 m 0.95 t (7.2) 0.88 t (7.2) 3-Vinylpropionate 5.77 m 5.78 m 5.02 m 5.03 m 2.35 m 2.36 m 2.30 t (7.2) 2.36 t (7.2) Isovalerate 0.89 d (23H, 6.8) 2.03 m 2.12 d (7.2) a Spectra recorded at 400 MHz in CDCl 3at 25C. b Spectra recorded at 400 MHz in CDCl3at20C.
2showed that the acetoxy group attached to C-3 in 2 could be replaced by an n-butyryloxy group in 4. These observations were further confirmed by the correlations observed in the 2D NMR experiments of 4 including1H–1H COSY, HSQC, and HMBC spectra. Similar to that of briarane 2, vinylpropio-nylation of 6 gave a less polar product, which was found to be identical with natural product 4 by comparison of the physi-cal and spectral data. Since the absolute configuration of the known briarane, excavatolide B (6), had been determined by modified Mosher’s method,11we were able to assign the ab-solute configurations of all the chiral centers of 4 as 1R,2R, 3S,5Z,7S,8S,9S,10S,11R,12S,14S,17R. Based on above find-ings, the structure of 4 was established unambiguously. In a previous study, we reported the isolation and structure elucidation of two 5,6-dihydroxybriarane metabolites, briaexcavatolides X (7) and Y (8).9fHowever, based on the detailed spectral data analysis and by comparing the 1H and13C NMR chemical shifts of C-5 and C-6 in briaranes 7(dC59.3, s, C-5; 62.7, d, C-6; dH3.05, 1H, d, J¼8.5 Hz, H-6) and 8 (dC 59.4, s, C-5; 62.7, d, C-6; dH 3.06, 1H, d, J¼8.5 Hz, H-6) with those of briarane 1 (dC 59.5, s, C-5; 62.8, d, C-6; dH3.06, 1H, d, J¼8.0 Hz, H-6) and a known 5,6-dihydroxybriarane, junceellolide L (9) (dC89.6, s, C-5; 82.7, d, C-6; dH 4.22, 1H, br s, H-6),12the 5,6-dihydroxy groups in briaranes 7 and 8 should be revised as 5b,6b-epoxy groups as presented in briaranes 10 and 11, respectively. The
spectral data (IR and MS) for briaranes 10 (briaexcavatolide X) and 11 (briaexcavatolide Y) are reassigned in this study (see Section3).
In the biological activity testing, briaexcavatin C (1) ex-hibited mild cytotoxicity toward MDA-MB-231 human breast tumor cells (IC50¼17.50 mg/mL) and briaexcavatin O O O O O AcO H AcO OR1 H 1: R1= CO(CH2)2CH3, R2= H 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 R2 O O O O AcO H AcO OAc 7: R = OH 8: R = H R OH OH 10: R1= Ac, R2= OH (revised) 11: R1= Ac, R2= H (revised) O O CH3CO O OCCH2CH2CH3 CH3CO O O O :1H_1H COSY : HMBC (H O C) O
Figure 1.The1H–1H COSY and selective HMBC correlations of 1.
2: R1= Ac, R2= CO(CH2)2CH CH2 4: R1= CO(CH2)2CH3, R2= CO(CH2)2CH CH2 O O H AcO AcO AcO OR1 O R2O 5: R1= Ac, R2= H 6: R1= CO(CH2)2CH3, R2= H 3: R1= Ac, R2= COCH2CH(CH3)2 : NOE OCCH2CH2CH3 O H H H OAc H O H H O O O H H H H H O OAc S* S* S* S* S* R* R* R* R* R* R* O O OAc H AcO 9 OH OH O AcO OH
Figure 2.Selective NOESY correlations of 1.
Table 3. The stereoview of 1 (generated from computer modeling) and the calculated distances (A˚ ) between selected protons having key NOE correlationsa Briaexcavatin C (1) H/H (A˚ ) H-2/H-3 2.44 H-3/H-10 3.09 H-3/H-4a 2.24 H-4a/H3-16 2.57 H-6/H3-16 2.40 H-9/H-10 2.73 H-9/H-11 2.41 H-9/H3-18 2.37 H-10/H-11 2.28 H-13/H-14 2.38 a
E (3) was found to show the activity to inhibit human neutro-phil elastase (HNE) release but not superoxide anion gener-ation (Table 4). To the best of our knowledge, briaexcavatin E (3) is the first briarane-type natural product reported to possess the activity to inhibit HNE release.
3. Experimental 3.1. General experimental procedures
Optical rotation values were measured with a JASCO P-1010 digital polarimeter at 25C. Infrared spectra were obtained on a VARIAN DIGLAB FTS 1000 FTIR spectrometer. EIMS and FABMS were obtained with a VG QUATTRO GC/MS spectrometer. HRMS data were recorded by ESI FT-MS on a BRUKER APEX II mass spectrometer or by EI on a JEOL JMS-700 mass spectrometer.1H NMR spectra were recorded on a VARIAN MERCURY PLUS 400 FT-NMR at 400 MHz and13C NMR spectra were recorded on a VAR-IAN MERCURY PLUS 400 FT-NMR at 100 MHz or on a VARIAN UNITY INOVA 600 FT-NMR at 150 MHz, in CDCl3using TMS as an internal standard. Column chroma-tography was performed on silica gel 60 (230–400 mesh) (Merck, Darmstadt, Germany). TLC spots (silica gel 60 F254, Merck) were detected with an UV254 lamp and by 10% H2SO4followed by heating at 120C for 5 min. All solvents and reagents used were analytical grade.
3.2. Animal material
Specimen of the octocoral B. excavatum was collected by hand using scuba off the coast of southern Taiwan in October 2003, at a depth of 10 m. Live reference specimens are being maintained in the authors’ marine organism cultivating tanks and a voucher specimen was deposited in the National Museum of Marine Biology and Aquarium (NMMBA). 3.3. Extraction and isolation
The organism (wet weight 1.0 kg) was collected and freeze-dried. The freeze-dried material (0.57 kg) was minced and extracted with EtOAc. The extract was separated by silica gel column chromatography, using n-hexane and n-hexane– EtOAc mixtures of increased polarity. Briarane 3 was eluted with n-hexane–EtOAc (6:1/5:1), 4 with n-hexane–EtOAc (5:1), 2 with n-hexane–EtOAc (4:1), and 1 with n-hexane– EtOAc (5:2).
3.3.1. Briaexcavatin C (1). White powder (2.1 mg); 79– 81C; [a]D25 25 (c 0.40, CHCl3); IR (neat) nmax 1792, 1744, 1688 cm1;13C (CDCl3, 100 MHz) and1H (CDCl3,
400 MHz) NMR data, seeTables 1 and 2; FABMS m/z 549 (M+H)+, 489, 429, 341; HRESIMS m/z 549.2340 (calcd for C28H36O11+H, 549.2336).
3.3.2. Briaexcavatin D (2).Colorless gum (2.4 mg); [a]D25 +32 (c 0.12, CH2Cl2); IR (neat) nmax 1792, 1737 cm1; 13C (CDCl
3, 150 MHz) and 1H (CDCl3, 400 MHz) NMR data, see Tables 1 and 2; FABMS m/z 671 (M+Na)+, 649, 589, 529, 489, 469, 429, 369, 309; HRESIMS m/z 671.2684 (calcd for C33H44O13+Na, 671.2680).
3.3.3. Briaexcavatin E (3). White powder (2.2 mg); 251– 252C; [a]D29 +42 (c 0.40, CHCl3); IR (neat) nmax 1787, 1742 cm1; 13C (CDCl
3, 100 MHz) and 1H (CDCl3, 400 MHz) NMR data, seeTables 1 and 2; FABMS m/z 673 (M+Na)+, 651, 591, 549, 489, 471, 429, 369, 309; HRESIMS m/z 673.2834 (calcd for C33H46O13+Na, 673.2836). 3.3.4. Briaexcavatin F (4). Colorless gum (1.7 mg); [a]D25 +46 (c 0.09, CH2Cl2); IR (neat) nmax 1787, 1735 cm1; 13C (CDCl
3, 150 MHz) and 1H (CDCl3, 400 MHz) NMR data, see Tables 1 and 2; FABMS m/z 699 (M+Na)+, 677 (M+H)+, 617, 589, 557, 529, 517, 469, 409, 369, 309; HRE-SIMS m/z 699.2996 (calcd for C35H48O13+Na, 699.2993). 3.3.5. Briaexcavatolide X (10). IR (neat) nmax 3472, 1792, 1748, 1703 cm1; EIMS m/z 537 (M+H)+, 459 (M+HH2OAcOH)+, 417 (M+H2AcOH)+, 399 (M+H H2O2AcOH)+, 357 (M+H3AcOH)+, 339 (M+HH2O 3AcOH)+; HREIMS m/z 536.1894 (calcd for C
26H32O12, 536.1894).
3.3.6. Briaexcavatolide Y (11).IR (neat) nmax1792, 1744, 1686 cm1; FABMS m/z 521 (M+H)+, 461 (M+HAcOH)+,
401 (M+H2AcOH)+, 341 (M+H3AcOH)+; HRESIMS
m/z 543.1844 (calcd for C26H32O11+Na, 543.1842). 3.4. Vinylpropionylation of excavatolide C (5)
Excavatolide C (5) (5.0 mg) was stirred with 2 mL of 3-vinylpropionic anhydride (pent-4-enoic anhydride) in 2 mL of pyridine for 96 h at room temperature. After evaporation of excess reagent, the residue was separated by column chro-matography on silica gel to give pure briaexcavatin D (2) (n-hexane–EtOAc, 4:1, 4.0 mg, 70%); physical (Rfand opti-cal rotation values) and NMR data were in full agreement with those of natural product 2.
3.5. Isovalerylation of excavatolide C (5)
Excavatolide C (5) (5.0 mg) was stirred with 2 mL of iso-valeric anhydride in 2 mL of pyridine for 96 h at room tem-perature. After evaporation of excess reagent, the residue was separated by column chromatography on silica gel to give pure briaexcavatin E (3) (n-hexane–EtOAc, 6:1/5:1, 3.9 mg, 68%); physical (mp, Rf, and optical rotation values) and NMR data were in full agreement with those of natural product 3.
3.6. Vinylpropionylation of excavatolide B (6)
Excavatolide B (6) (8.0 mg) was stirred with 2.5 mL of 3-vinylpropionic anhydride (pent-4-enoic anhydride) in
Table 4. Inhibitory effects of briarane 3 on superoxide generation and elastase release by human neutrophil in response to fMet-Leu-Phe/cyto-chalasin Ba Compound Concn (mM) Superoxide generation (%) Elastase release (%) 3 3 — 87.775.86 5 — 65.969.94 10 101.194.15 37.8913.53
a Data obtained without any drugs were set to 100%. MeansSEM of three
2.5 mL of pyridine for 96 h at room temperature. After evap-oration of excess reagent, the residue was separated by column chromatography on silica gel to give pure briaexca-vatin F (4) (n-hexane–EtOAc, 5:1, 6.6 mg, 72%); physical (Rfand optical rotation values) and NMR data were in full agreement with those of natural product 4.
3.7. Molecular mechanics calculations
The minimum energy conformation of briaexcavatin C (1) was determined using the MSI Insight II/DISCOVER ver-sion 95 molecular modeling package incorporating an empirical force field, the consistent valence force field (CVFF),13 on Silicon Graphic IRIS (SGI) Indigo XS24/ 4000 workstation. All force field calculations were carried out in vacuo (dielectric constant¼1). The conformational space of 1 was scanned using the high-temperature molec-ular dynamics simulation technique followed by energy minimization. A 100 ps molecular dynamics simulation at 1000 K provided a set of 500 conformations of 1. Each con-formation was used as a starting structure for the subsequent energy minimization (1000 steps, conjugated gradient algo-rithm). In the subsequent analysis, only 10 conformations with a reasonably low energy (at most 5 kcal/mol higher with respect to the lowest energy conformer) were used. The conformational search suggested that the most stable conformation of briarane 1 shown inTable 3is the lowest. 3.8. Cytotoxicity assays
Compounds were assayed for cytotoxicity against MDA-MB-231 cells using the MTT method. Freshly trypsinized cell suspensions were seeded in 96-well microtitre plates at densities of 5000–10,000 cells per well with tested com-pounds added from DMSO-diluted stock. After 3 days in culture, attached cells were incubated with MTT (0.5 mg/ mL, 1 h) and subsequently solubilized in DMSO. The absor-bency at 550 nm was then measured using a microplate reader. The IC50is the concentration of agent that reduced cell growth by 50% under the experimental conditions. 3.9. Human neutrophil superoxide generation and elas-tase release
Human neutrophils were obtained by means of dextran sedi-mentation and Ficoll centrifugation. Superoxide generation and elastase release were carried out according to the proce-dures described previously.14,15 Briefly, superoxide anion production was assayed by monitoring the superoxide dis-mutase-inhibitable reduction of ferricytochrome c. Elastase release experiments were performed using MeO-Suc-Ala-Ala-Pro-Valp-nitroanilide as the elastase substrate.
Acknowledgements
This research work was supported by grants from the Na-tional Science Council, Taiwan, ROC, and by the intramural funding from the NMMBA awarded to P.-J. Sung.
References and notes
1. Bayer, F. M. Proc. Biol. Soc. Wash. 1981, 94, 902–947. 2. Benayahu, Y.; Jeng, M.-S.; Perkol-Finkel, S.; Dai, C.-F. Zool.
Stud. 2004, 43, 548–560.
3. Sung, P.-J.; Sheu, J.-H.; Xu, J.-P. Heterocycles 2002, 57, 535–579. 4. Sung, P.-J.; Chen, M.-C. Heterocycles 2002, 57, 1705–1715. 5. Sung, P.-J.; Chang, P.-C.; Fang, L.-S.; Sheu, J.-H.; Chen, W.-C.;
Chen, Y.-P.; Lin, M.-R. Heterocycles 2005, 65, 195–204. 6. Burks, J. E.; van der Helm, D.; Chang, C. Y.; Ciereszko, L. S.
Acta Crystallogr. 1977, B33, 704–709.
7. Schmitz, F. J.; Schulz, M. M.; Siripitayananon, J.; Hossain, M. B.; van der Helm, D. J. Nat. Prod. 1993, 56, 1339–1349. 8. (a) Sheu, J.-H.; Sung, P.-J.; Cheng, M.-C.; Liu, H.-Y.; Fang,
L.-S.; Duh, C.-Y.; Chiang, M. Y. J. Nat. Prod. 1998, 61, 602–608; (b) Sung, P.-J.; Su, J.-H.; Wang, G.-H.; Lin, S.-F.; Duh, C.-Y.; Sheu, J.-H. J. Nat. Prod. 1999, 62, 457–463; (c) Neve, J. E.; McCool, B. J.; Bowden, B. F. Aust. J. Chem. 1999, 52, 359–366; (d) Sheu, J.-H.; Sung, P.-J.; Su, J.-H.; Wang, G.-H.; Duh, C.-Y.; Shen, Y.-C.; Chiang, M. Y.; Chen, I.-T. J. Nat. Prod. 1999, 62, 1415–1420.
9. (a) Sheu, J.-H.; Sung, P.-J.; Su, J.-H.; Liu, H.-Y.; Duh, C.-Y.; Chiang, M. Y. Tetrahedron 1999, 55, 14555–14564; (b) Sung, P.-J.; Su, J.-H.; Duh, C.-Y.; Chiang, M. Y.; Sheu, J.-H. J. Nat. Prod. 2001, 64, 318–323; (c) Wu, S.-L.; Sung, P.-J.; Chiang, M. Y.; Wu, J.-Y.; Sheu, J.-H. J. Nat. Prod. 2001, 64, 1415– 1420; (d) Wu, S.-L.; Sung, P.-J.; Su, J.-H.; Sheu, J.-H. J. Nat. Prod. 2003, 66, 1252–1256; (e) Wu, S.-L.; Sung, P.-J.; Su, J.-H.; Wang, G.-H.; Sheu, J.-H. Heterocycles 2004, 63, 895– 898; (f) Sung, P.-J.; Hu, W.-P.; Wu, S.-L.; Su, J.-H.; Fang, L.-S.; Wang, J.-J.; Sheu, J.-H. Tetrahedron 2004, 60, 8975– 8979.
10. Aoki, S.; Okano, M.; Matsui, K.; Itoh, T.; Satari, R.; Akiyama, S.; Kobayashi, M. Tetrahedron 2001, 57, 8951–8957. 11. Sung, P.-J.; Chao, C.-H.; Chen, Y.-P.; Su, J.-H.; Hu, W.-P.;
Sheu, J.-H. Tetrahedron Lett. 2006, 47, 167–170.
12. Sheu, J.-H.; Chen, Y.-P.; Hwang, T.-L.; Chiang, M. Y.; Fang, L.-S.; Sung, P.-J. J. Nat. Prod. 2006, 69, 269–273.
13. MSI INSIGHT II/DISCOVER (version 95.0/2.97) is a molec-ular modeling software package of MSI Technologies, Barnes Canyon Road, San Diego, CA, 92121, USA.
14. Hwang, T.-L.; Hung, H.-W.; Kao, S.-H.; Teng, C.-M.; Wu, C.-C.; Cheng, S.-J. Mol. Pharmacol. 2003, 64, 1419–1427. 15. Yeh, S.-H.; Chang, F.-R.; Wu, Y.-C.; Yang, Y.-L.; Zhuo, S.-K.;
