Anti-inflammatory cembranolides from the soft coral Lobophytum durum

Article abstract of DOI:10.1016/j.bmc.2009.04.053

Chemical investigation of the soft coral Lobophytum durum resulted in the isolation of seven new cembranolides, durumolides F-L (1-7), as well as one previously characterized cembranolides, sinularolide D (8). The molecular structures of these isolated me

Full text of DOI:10.1016/j.bmc.2009.04.053

Bioorganic & Medicinal Chemistry 17 (2009) 3763–3769
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Anti-inflammatory cembranolides from the soft coral Lobophytum durum
Shi-Yie Cheng a, Zhi-Hong Wen a, Shang-Kwei Wang b, Shu-Fen Chiou a, Chi-Hsin Hsu a,d
,
Chang-Feng Dai c, Chang-Yih Duh a,d,
*
a Department of Marine Biotechnology and Resources, National Sun Yat-sen University, Kaohsiung 804, Taiwan
b Department of Microbiology, Kaohsiung Medical University, Kaohsiung 807, Taiwan
c Institute of Oceanography, National Taiwan University, Taipei, Taiwan
d Center of Asia-Pacific Marine Researches, National Sun Yat-sen University, Kaohsiung, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history:
Chemical investigation of the soft coral Lobophytum durum resulted in the isolation of seven new cembr-
anolides, durumolides F–L (1–7), as well as one previously characterized cembranolides, sinularolide D
(8). The molecular structures of these isolated metabolites were determined mainly through NMR tech-
niques and HRESIMS analysis. Moreover, the absolute configurations of 1 and 5 were established by
application of modified Mosher’s method. The antibacterial activities, anti-inflammatory effects, and
anti-HCMV (Human cytomegalovirus) endonuclease activity of metabolites 1–8 were also evaluated
Received 13 March 2009
Revised 23 April 2009
Accepted 24 April 2009
Available online 3 May 2009
Keywords:
Lobophytum durum
Cembranolides
Antibacterial activities
Anti-inflammatory effects
HCMV endonuclease activity
in vitro. Anti-inflammatory activity of metabolites 1 and 6 (10
lM) significantly reduced the levels of
the iNOS protein to 0.8 0.6% and 5.7 2.2%, respectively, and COX-2 protein to 47.8 9.0% and
71.6 5.8%, respectively. Metabolites 1–8 (100
monella enteritidis.
l
g/disk) exhibited weak antibacterial activity against Sal-
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
was then subjected to column chromatography on silica gel. The
fractions containing cembranolides were selected for further puri-
fication by C-18 HPLC to obtain metabolites 1–8 (see Section 3).
Durumolide F (1) was isolated as a colorless oil. Its HRESIMS (m/
z 415.2097, [M+Na]+) and NMR spectroscopic data (Tables 1 and 2)
established the molecular formula C22H32O6, implying seven de-
Soft corals belonging to the genus Lobophytum (Alcyoniidae)
have been well recognized as a rich source of macrocyclic cem-
brane-type diterpeniods and their cyclized derivatives.1–19 Previ-
ous bioassay results of some cembranoid analogues have been
shown to exhibit diverse biological properties such as cyto-
toxic,2–6,19 anti-inflammatory,18 and antibacterial activities.18 The
chemical investigations for bioactive constituents prompted us to
explore the secondary metabolites of the soft coral Lobophytum
durum (Tixier-Durivault, 1956). We have previously isolated five
new cembranolides, named as durumolides A–E, from the ace-
tone-soluble of the organism.18 Our continuous chemical examina-
tion of the secondary of this soft coral led to the isolation of seven
new cembranolides, durumolides F–L (1–7), and sinularolide D
(8).20 Herein, we describe the isolation, structural elucidation,
anti-inflammatory effects, antibacterial activities, and inhibition
assay of HCMV endonuclease activity of these metabolites.
grees of unsaturation. The IR spectrum of
1 at 1767 and
1665 cmꢀ1 demonstrated absorption bands diagnostic of an
a-
methylene-c-lactone functionality. This was further indicated from
the 1H NMR signals at dH 6.29 (1H, d, J = 1.3 Hz) and 5.68 (1H, d,
J = 1.9 Hz) and 13C NMR signals at dC 170.2 (qC, C-16), 139.7 (qC,
C-15), 123.2 (CH2, C-17), 81.8 (CH, C-14), and 42.5 (CH, C-1).18
A
strong IR spectrum absorption at 1735 cmꢀ1 indicated the pres-
ence of an acetoxy group. A secondary hydroxyl and a tertiary hy-
droxyl were recognized as being present in 1 from its 1H NMR
signals at dH 3.49 (1H, br t, J = 8.3 Hz) and 13C NMR signals at dC
70.9 (CH, C-3) and 75.2 (qC, C-4), as well as from a broad IR absorp-
tion at 3462 cmꢀ1. The 13C NMR signals at dC 124.2 (CH, C-7), 136.7
(qC, C-8), 128.4 (CH, C-11), and 130.3 (qC, C-12) were assigned two
trisubstituted double bonds in 1. The above functionalities account
for five of the seven degrees of unsaturation, suggesting a bicyclic
structure in 1. By interpretation of 1H–1H COSY correlations, it was
possible to establish three partial structures of consecutive proton
systems extending from H-3 to H2-13 through H2-2, H-1 and H-14,
from H2-5 to H-7, and from H-9 to H-11, as well as long-range
COSY correlations between H-1/H2-17, H3-19/H-7, and H3-20/H-
11. Moreover, the connectivities of these partial structures were
2. Results and discussion
The acetone extract of the soft coral L. durum was partitioned
between EtOAc and H2O to afford the EtOAc-soluble portion, which
* Corresponding author. Tel.: +886 7 5252000x5036; fax: +886 7 5255020.
E-mail address: yihduh@mail.nsysu.edu.tw (C.-Y. Duh).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.04.053

3764
S.-Y. Cheng et al. / Bioorg. Med. Chem. 17 (2009) 3763–3769
Table 1
Table 2
1H NMR spectroscopic data of metabolites 1–3
13C NMR spectroscopic data of metabolites 1–7
1a
2b
3a
C#
1a
2b
3a
4a
5a
6c
7c
H#
1
2
3
4
5
6
7
8
42.5 (CH)d 39.3 (CH)d 39.9 (CH)d 41.9 (CH)d 42.3 (CH)d 42.6 (CH)d 42.3 (CH)d
36.4 (CH2) 33.2 (CH2) 33.6 (CH2) 31.3 (CH2) 31.9 (CH2) 34.9 (CH2) 34.8 (CH2)
70.9 (CH) 73.3 (CH) 73.3 (CH) 62.2 (CH) 63.4 (CH) 58.5 (CH) 59.1 (CH)
1
2
2.97 m
2.63 m
a: 2.07 m
2.75 m
1.96 m
a: 1.93 m
b: 1.64 m
3.49 br t (8.3)c
a: 1.86 m
b: 1.68 m
2.20 m
b: 2.00 m
75.2 (qC)
75.7 (qC)
75.5 (qC)
60.3 (qC)
60.2 (qC)
63.3 (qC)
65.6 (qC)
3
5
5.22 dd (8.0, 5.5)c
a: 1.76 m
5.17 dd (7.4, 5.1)c
a: 1.83 m
b: 1.70 m
34.2 (CH2) 33.4 (CH2) 33.6(CH2) 32.3 (CH2) 38.0 (CH2) 30.5 (CH2) 26.0 (CH2)
21.7 (CH2) 22.7 (CH2) 22.5 (CH2) 23.9 (CH2) 24.4 (CH2) 21.4 (CH2) 21.2 (CH2)
124.2 (CH) 122.3 (CH) 122.9 (CH) 128.7 (CH) 124.5 (CH) 127.1 (CH) 126.3 (CH)
136.7 (qC) 136.8 (qC) 136.3 (qC) 137.7 (qC) 134.6 (qC) 133.9 (qC) 134.7 (qC)
37.6 (CH2) 37.4 (CH2) 37.5 (CH2) 77.4 (CH) 38.7 (CH2) 37.5 (CH2) 37.6 (CH2)
24.9 (CH2) 24.7 (CH2) 24.8 (CH2) 33.2 (CH2) 24.8 (CH2) 25.9 (CH2) 25.6 (CH2)
128.4 (CH) 132.1 (CH) 133.7 (CH) 128.1 (CH) 132.4 (CH) 145.4 (CH) 145.5 (CH)
130.3 (qC) 131.7 (qC) 129.6 (qC) 130.7 (qC) 131.7 (qC) 135.8 (qC) 136.2 (qC)
46.1 (CH2) 80.4 (CH) 80.1 (CH) 79.9 (CH) 81.3 (CH) 195.3 (qC) 194.8 (qC)
81.8 (CH) 84.4 (CH) 82.2 (CH) 79.4 (CH) 82.3 (CH) 77.2 (CH) 76.4 (CH)
139.7 (qC) 139.8 (qC) 138.9 (qC) 137.9 (qC) 138.7 (qC) 137.0 (qC) 136.3 (qC)
170.2 (qC) 168.6 (qC) 168.5 (qC) 168.8 (qC) 169.1 (qC) 169.8 (qC) 169.1 (qC)
123.2 (CH2) 124.0 (CH2) 123.8 (CH2) 124.3 (CH2) 124.3 (CH2) 122.9 (CH2) 123.6 (CH2)
67.7 (CH2) 66.8 (CH2) 66.9 (CH2) 63.6 (CH2) 16.7 (CH3) 63.8 (CH2) 199.3 (CH)
17.0 (CH3) 17.2 (CH3) 17.1 (CH3) 11.2 (CH3) 15.5 (CH3) 15.7 (CH3) 15.4 (CH3)
17.5 (CH3) 12.6 (CH3) 13.2 (CH3) 13.3 (CH3) 12.2 (CH3) 11.6 (CH3) 11.7 (CH3)
b: 1.64 m
a: 2.25 m
b: 2.20 m
5.08 br t (6.5)
2.20 m
a: 2.46 m
b: 2.25 m
5.44 dd (7.5, 3.0)
3.96 d (8.5)
6
2.20 m
9
10
11
12
13
14
15
16
17
18
19
20
7
9
10
5.26 br t (6.9)
2.18 m
a: 2.39 m
5.06 br t (6.8)
2.20 m
a: 2.40 m
b: 2.25 m
5.54 br d (6.6)
5.10 d (8.5)
b: 2.17 m
11
13
5.09 br d (4.9)
a: 2.57 br d (12.4)
b: 2.16 m
4.32 dt (9.9, 2.8)
a: 6.29 d (1.3)
b: 5.68 d (1.9)
a: 4.20 d (11.8)
b: 4.13 d (11.8)
1.64 s
14
17
4.24 dd (8.5, 6.0)
a: 6.29 d (2.5)
b: 5.85 d (2.5)
a: 4.14 d (12.0)
b: 4.05 d (12.0)
1.67 s
4.33 dd (8.5, 4.8)
a: 6.29 d (1.9)
b: 5.78 d (1.3)
a: 4.13 d (11.8)
b: 4.05 d (11.8)
1.67 s
18-OAc 20.9 (CH3) 20.9 (CH3) 21.0 (CH3) 20.7 (CH3)
171.7 (qC) 171.2 (qC) 171.2 (qC) 170.7 (qC)
3-OAc
18
21.1 (CH3) 21.0 (CH3)
170.3 (qC) 170.3 (qC)
19
20
1.76 s
2.11 s
1.74 s
2.12 s
2.06 s
1.74 s
2.12 s
2.07 s
2.08 s
13-OAc
20.9 (CH3) 21.0 (CH3)
169.9 (qC) 169.5 (qC)
18-OAc
3-OAc
13-OAc
a
Spectra were measured in CDCl3 (75 MHz).
b
c
Spectra were measured in CDCl3 (125 MHz).
a
Spectra were measured in CDCl3 (100 MHz).
Spectra were measured in CDCl3 (300 MHz).
Spectra were measured in CDCl3 (500 MHz).
J values (in Hz) are in parentheses.
d
b
c
Multiplicities are deduced by HSQC and DEPT experiments.
are oriented toward the opposite side. The absolute stereochemis-
try of 1 was determined by spectroscopic method for secondary
further established by the HMBC correlations (Fig. 1). The long-
range 1H–13C correlations observed from H2-18 to C-3, C-4, C-5,
and the carbonyl carbon of 18-OAc indicated the position of the
acetoxy group at C-18. In addition, the above HMBC correlations
also led the assignment of the two hydroxyls at C-3 and C-4,
respectively. The locations of the two double bonds at C-7/C-8
and C-11/C-12 were clarified by analysis of the HMBC correlations
from H3-19 to C-7, C-8, and C-9 and from H3-20 to C-11, C-12, and
C-13. Thus, the gross structure of durumolide F was unmistakably
alcohols.22 Analysis of the
DdS–R values (Fig. 3) according to the
Mosher’s model pointed to an S configuration for C-3 of 1, because
H-1, H2-2, and H-14 of (S)-MTPA ester 1a were more shielded by
the phenyl ring of MTPA products. From the above observations,
1 was elucidated as (1R,3S,4S,14S,7E,11E)-18-acetoxy-3,4-dihydr-
oxycembra-7,11,15(17)-trien-16,14-olide.
Durumolide G (2), appeared as a colorless oil, was isolated as
the most polar metabolite. Its HRESIMS (m/z 473.2153, [M+Na]+)
and NMR spectroscopic data (Tables 1 and 2) established the
molecular formula C24H34O8, implying the existence of eight de-
grees of unsaturation. The NMR spectroscopic data of metabolite
assigned as 1, possessing an
14-membered ring at C-1 and C-14.
The determination of the configurations of an
a-methylene-c-lactone ring fused to a
a-methylene-c-
2 were consistent with a cembranolide skeleton having an
a-meth-
lactone ring fused to a 14-membered ring based on the coupling
constant between the lactonic methine protons (3J1,14) is dubi-
ous.19 The relative stereochemistry of 1 assigned by NOESY spec-
trum was compatible with those of 1 offered by computer
modeling (Fig. 2), in which the close contacts of atoms calculated
in space were consistent with the NOESY correlations. The calcu-
ylene- -lactone moiety, corresponding to exomethylene protons at
c
dH 6.29 (1H, d, J = 2.5 Hz) and 5.85 (1H, d, J = 2.5 Hz) that correlated
to the methine carbon at dC 39.3 (C-1), the quaternary olefinic car-
bon at dC 139.8 (C-15), and the lactone carbonyl signal at dC 168.8
(C-16) in the HMBC spectrum. Two trisubstituted double bonds at
C-7 and C-8 [dH 5.08 (br t, J = 6.5 Hz, 1H); dC 136.8 (qC) and 122.3
(CH)], and C-11 and C-12 [dH 5.44 (dd, J = 7.5, 3.0 Hz, 1H); dC 131.7
(qC) and 132.1 (CH)] were evidenced by HMBC correlations of H3-
19 to C-7, C-8, and C-9 and of H3-20 to C-11, C-12, and C-13. In
addition to a lactone carbonyl, the 13C NMR data exhibited two car-
bonyl signals. The carbonyl signal at dC 170.3 was attributed to the
acetate moiety linked to C-3, as indicated by HMBC correlations of
H-3 to the carbonyl carbon of 3-OAc. The other acyl (dC 171.2) was
assigned to C-18, as indicated by the HMBC correlations of two iso-
lated oxymethylene protons (dH 4.14 and 4.05) to the carbonyl car-
bon of 18-OAc. Furthermore, the HMBC spectrum was used to
locate two hydroxyl substitutions at C-4 and C-13. The stereo-
chemistry of 2 assigned by NOESY spectrum was compatible with
that of 1 offered by computer-generated perspective model using
MM2 force field calculations, in which the close contacts of atoms
calculated space were consistent with NOESY correlations. The
NOESY spectrum and the conformer generated by Chem3D Pro
9.0 indicated that 2 possessed the same configuration as 1 at the
C-1, C-3, C-4, and C-14 stereocenters. From the NOESY spectrum
lated torsion angles for a trans-fused
a-methylene-c-lactone of 1
are identical to those of an analogue, deacetyl-13-hydroxy lobo-
lide,13 which were determined by an X-ray crystallographic analy-
sis.18 This suggests that the preferred conformation for a trans-
fused
a-methylene-c-lactone of 1 is generally in agreement with
its crystal structure. Also, it is noted that all the calculated dis-
tances between the protons in this computer-generated perspec-
tive model that show NOE correlations are less than 3.0 Å. The
geometry of the trisubstituted olefins was assigned as E based on
the c-effect of the olefinic methyl signals for C-19 and C-20 (less
than 20 ppm)21 and the NOESY correlations between H-7 and H2-
9, H2-6 and H3-19, H-11 and H2-13, and H2-10 and H3-20. The
NOESY correlations (Fig. 2) between H-1/H-2b (dH 1.64), H-1/H-3,
H-1/H-11, H-1/H-13b (dH 2.16), H-18a (dH 4.20)/H-5b (dH 1.68),
H-3/H-7, H-7/H-6, H-7/H-9, H-11/H-10b (dH 2.17), H-3/H-2b, H-
14/H-2a (dH 1.93), H-14/H3-20, and H-2b/H-17b (dH 5.68) con-
firmed the geometry of the trans-fused lactone ring and indicated
that H-1, H-2b, H-3, H-11, and H-13b are located on the same side
of the ring system, whereas H-2a, H-13a, H-14, H2-18, and H3-20

S.-Y. Cheng et al. / Bioorg. Med. Chem. 17 (2009) 3763–3769
3765
O
O
O
OH
O
O
OH
O
O
O
HO
O
HO
O
O
O
O
O
O
1
4
6
Figure 1. 1H–1H COSY (
) and key HMBC (?) correlations of 1, 4, and 6.
5b
5a
5b
6
17b
18a
5a
17b
7
6
17a
17a
3
2b
2a
7
18b
19
2
18a
10
3
18b
1
19
1
14
9
9
14
11
10
11
13b
13a
13
20
20
1
4
5b
5a
17b
6
5a
17b
2
5b
7
3
3
17a
17a
18
1
6
7
2
9
18a
19
18b
14
1
10
9
11
10
11
20
14
13
19
20
5
6
Figure 2. Key NOE correlations and computer-generated perspective model using MM2 force field calculations for 1 and 4–6.
of 2, H-13 was found to show the NOE correlations with both H-1
and H-11, indicating the R -configuration of the secondary hydro-
Durumolide H (3) was obtained as a colorless oil, which ana-
lyzed for the molecular formula C26H36O9 by HRESIMS coupled
with the DEPT and 13C NMR spectroscopic data (Table 2). Analysis
of the COSY and HMBC correlations were diagnostic in determining
that the gross framework of durumolide H, containing a 14-mem-
*
xyl attached at C-13. The NOE correlations between H2-18/H2-2,
H3-19/H2-6, H3-20/H2-10, H-3/H2-5, H2-9/H-7, and H-11/H-13,
respectively, reflected the E geometry of the two trisubstituted
double bonds in the molecule. According to the aforementioned
observations and other detailed NOESY correlations, the structure
bered ring fused to an a-methylene-c-lactone ring, was assigned as
3. The structure of 3 was identical to that of 2 with the exception
that the hydroxy group attached to C-13 was replaced by an acet-
oxy group. This was confirmed through HMBC correlations from H-
13 to C-1, C-14, C-20, and the carbonyl carbon of 13-OAc. Metabo-
of metabolite
2
was unambiguously determined as
*
*
*
*
*
(1R ,3S ,4S ,13R ,14S ,7E,11E)-3,18-diacetoxy-4,13-dihydroxycem-
bra-7,11,15(17)-trien-16,14-olide.

3766
S.-Y. Cheng et al. / Bioorg. Med. Chem. 17 (2009) 3763–3769
+0.03
Table 3
+0.01
+0.01
1H NMR spectroscopic data of metabolites 4–7
—0.00
OAc
+0.04
+0.01
OH
H#
4a
5a
6b
7b
O
—0.04
—0.02
+0.01
+0.03
—0.01
—0.00
3.12 dt (12.4, 2.4)c
a: 2.12 m
b: 1.77 m
3.35 dd (8.6, 2.4)
a: 2.33 m
b: 2.01 ddd (12.8, 6.4, 4.0)
a: 2.17 m
—0.03
+0.01
—0.01
1
2
2.82 m
2.77 m
3.06 m
+0.02
—0.18
+0.01
—0.01
a: 1.78 dt (14.4, 2.9)c a: 1.87 dt (14.6, 2.4)c a: 1.95 m
RO
RO
H
—0.01
b: 1.56 m
2.82 m
a: 2.46 m
b: 1.23 td (13.0, 2.6) b: 1.24 m
a: 2.35 m
b: 2.19 m
b: 1.40 m
2.70 dd (7.3, 2.3)
a: 2.08 m
b: 1.78 m
3.04 m
+0.39
—0.11
—0.05
+0.02
+0.11
3
5
O
O
O
O
a: 2.13 m
b: 1.65 m
a: 2.18 m
b: 2.10 m
+0.02
+0.02
5
6
a: 2.23 m
b: 2.10 m
1
b: 2.15 m
7
9
5.37 br dd (9.6, 3.9) 5.03 br d (4.3)
4.11 dd (10.4, 2.7)
5.18 br t (5.6)c 5.15 br t (6.0)
R= MTPA
a: 2.35 m
b: 2.12 m
a: 2.58 m
b: 2.52 m
2.27 m
a: 2.32 m
b: 2.27 m
a: 2.64 m
b: 2.42 m
Figure 3. Selected 1H NMR
and 5 in CDCl3.
DdS–R values in ppm for the S- and R-MTPA esters of 1
10
a: 2.67 m
b: 2.41 m
a: 2.58 m
b: 2.44 m
11
13
14
17
5.52 br dd (8.3, 5.6) 5.47 br dd (9.7, 3.1) 6.68 t (6.8)
6.63 dd (6.8, 4.8)
5.28 d (8.5)
4.17 d (8.5)
a: 6.33 d (2.9)
b: 6.03 d (2.5)
a: 4.41 d (12.2)
b: 3.85 d (12.2)
1.71 s
4.05 br s
4.06 d (6.0)
a: 6.32 d (3.2)
b: 6.08 d (2.8)
1.24 s
5.57 d (2.0)
5.51 d (2.0)
a: 6.28 d (2.0) a: 6.34 d (2.0)
b: 5.62 d (1.6) b: 5.67 d (1.6)
a: 3.72 d (11.6) 9.42 s
b: 3.67 d (11.6)
1.62 s
1.84 s
lite 3 is simply an acetylation product of 2. All the NMR spectro-
scopic data of 3, assigned by COSY, HMBC, and NOESY correlations,
were satisfactorily consistent with the structure shown as
(1R ,3S ,4S ,13R ,14S ,7E,11E)-3,13,18-triacetoxy-4-hydroxycem-
bra-7,11,15(17)-trien-16,14-olide.
The positive HRESIMS of 4 exhibited a pseudo molecular ion
peak at m/z 471.1993 [M+Na]+, corresponding to the molecular for-
mula C24H32O8 and nine degrees of unsaturation. Both the 1H and
13C NMR spectroscopic data (Tables 2 and 3) demonstrated the
18
19
20
1.64 s
1.73 s
1.61 s
1.86 s
*
*
*
*
*
1.72 s
18-OAc 2.15 s
13-OAc 2.12 s
a
Spectra were measured in CDCl3 (300 MHz).
Spectra were measured in CDCl3 (400 MHz).
J values (in Hz) are in parentheses.
b
c
presences of an
a-methylene-c-lactone ring, two acetate units,
an epoxide, and two trisubstituted double bonds, which accounted
for eight degrees of nine unsaturations and were suggestive of a
*
*
*
*
*
*
metabolite 4 was deduced as (1R ,3R ,4S ,9S ,13R ,14R ,7E,11E)-
13,18-diacetoxy-9-hydroxy-3,4-epoxycembra-7,11,15(17)-trien-
16,14-olide, which has been named durumolide I.
tricyclic cembranolide bearing an
and an epoxide ring. The lactone carbonyl signal at dC 168.8 (C-
16) was assigned to an -methylene- -lactone ring together with
a-methylene-c-lactone moiety
(1R,3R,4R,13S,14R,7E,11E)-13-hydroxy-3,4-epoxycembra-
a
c
7,11,15(17)-trien-16,14-olide (5) was assigned a molecular for-
mula of C20H28O4, according to its HRESIMS and NMR spectro-
scopic data (Tables 2 and 3). A careful analysis of the NMR data
coupled with COSY, HSQC, and HMBC correlations proved that
the structure of 5 was identical to that of 13-acetoxy isolobophyto-
lide14 except for the presence of a hydroxyl instead of an acetoxy
group at C-13. The NOE correlations (Fig. 2) between H-1/H-3, H-
1/H-11, H-11/H-9, H-3/H-5a (dH 2.08), H3-18/H-5b (dH 1.24), H-3/
H-7, H-7/H-9, and H-14/H3-20 indicated the relative configurations
for the 14-membered ring carbons, which were identical to the
same configuration as durumolide E18 at the C-1, C-3, C-4, C-13,
and C-14 stereocenters. In order to resolve the absolute stereo-
chemistry of 5, we determined the absolute configuration at C-13
the oxymethine at dC 79.4 (C-14) and the methine carbon at dC
41.9 (C-1). The NMR signals at dY 6.33 and 6.03 as well as at dC
124.3 were ascribed to an exocyclic methylene group and corre-
lated to C-1, C-15, and C-16, thereby proving the presence of a
14,16-lactone ring (Fig. 1). Two acetates resonated at dY 2.15 and
2.12 (each 3H, s) and dC 170.7 (C), 169.5 (C), 20.7 (CH3), and 21.0
(CH3). Meanwhile, the oxymethine observed at dY 5.28 (H-13),
and the oxymethylene at dY 4.41 (H-18a) and 3.85 (H-18b) exhib-
ited HMBC correlations to two acetoxy carbonyls, proving acetyl-
oxy substitution at these positions, and was confirmed by the
COSY correlations (Fig. 1). The long-range 1H–13C correlations ob-
served from H2-18 to C-3, C-4, C-5, and the carbonyl carbon of
18-OAc further indicated the position of the epoxide at C-3 and
C-4. The HMBC correlations from the vinylic methyl signal (Y3-
19) correlated with those of an olefinic methine carbon (C-7), a
methine carbon (C-9), and a quaternary olefinic carbon (C-8), led
the location of the double bond at C-7 and C-8. The assignment
of the secondary hydroxyl at C-9 was clarified by analysis of the
above HMBC correlations (Fig. 1). In addition, the vinylic methyl
signal (Y3-20) correlated with those of an olefinic methine carbon
(C-11), a methylene carbon (C-13), and a quaternary olefinic car-
bon (C-12). The oxymethine proton at dY 5.28 was coupled to the
olefinic CH at dY 5.52 (H-11) and correlated to C-20 and C-12, sup-
porting C-11/C-12 unsaturation. The structure of 4 was generated
by CS CHEM 3D version 9.0 using MM2 force field calculations for en-
ergy minimization, as shown in Figure 2. The result was consistent
with the stereochemistry of 4 as established by the NOESY evi-
dences and the computer-generated perspective modeling. NOE
correlations between H-3/H-7, H-3/H-5a (d 2.46), H-18a (d 4.41)/
H-5b (d 1.23), H-7/H-5a, H-3/H-1, H-1/H-13, H-9/H-7, H-9/H-11,
H-11/H-13, and H-14/H3-20 confirmed the geometry of the trans-
fused lactone ring and indicated that H-1, H-3, H-7, H-9, and H-
11 are located on the same side of the ring system, whereas H-
14, H2-18, and H3-20 are oriented toward the opposite side. The
using a modified Mosher’s acylation.22 The determination of
DdS–R
values (Fig. 3) for the protons neighboring C-13 led to the assign-
ment of the R-configuration at C-13 in 5. Consequently, the struc-
ture of durumolide J (5) was established definitively.
The molecular formula of C20H26O5 was assigned to metabolite
6 from its HRESIMS and 13C NMR spectroscopic data (Table 2), indi-
cating eight degrees of unsaturation. The NMR features (Tables 2
and 3) of 6 were analogous to those of 819 except for the reso-
nances in the vicinity of C-13. The exception was that the reso-
nances for the methylene at C-13 were replaced by those of a
keto group, consistent with the HMBC correlations (Fig. 1) from
H3-20 to C-11, C-12, and C-13. The computer-modeled structure
of 6 was generated by CS CHEM 3D version 9.0 using MM2 force field
calculations for energy minimization, as shown in Figure 2. The
NOESY correlations (Fig. 2) between H-1/H-3, H-1/H-14, H-3/H-7,
H-18a (dH 3.72)/H-5b (dH 1.65), H-7/H-9, H-14/H3-20, and H-7/H-
5a (dH 2.13) confirmed that H-1, H-3, and H-7 are located on the
same side of the ring system, whereas H-14, H2-18 and H3-20 are
oriented toward the opposite side. The above findings indicated
*
*
*
*
the 1R , 3R , 4S , and 14R configurations as depicted in Figure 2.
The trans-fused conformation of bicyclic ring junction was further
supported by the coupling constant between the lactonic methine
proton (3J1,14 = 2.0 Hz). In addition, the dihedral angle formed by H-
*
*
*
*
*
above findings also indicated the 1R , 3R , 4S , 9S , 13R , and
*
14S configurations as depicted in Figure 2. Thus, the structure of

S.-Y. Cheng et al. / Bioorg. Med. Chem. 17 (2009) 3763–3769
3767
1, C-1, C-14, and H-14 is 109.9°, also indicating that H-1 and H-14
to be pseudoaxially and pseudoequatorially oriented, respectively.
On the basis of the above observations, the structure of durumolide
K (6) was assigned as (1R ,3R ,4S ,14R ,7E,11E)-18-hydroxy-13-
oxo-3,4-epoxycembra-7,11,15(17)-trien-16,14-olide.
the in vitro anti-inflammatory activity of metabolites 1–8 was
tested using LPS-stimulated cells. Metabolites 1 and 6 significantly
reduced the levels of the iNOS protein to 0.8 0.6% and 5.7 2.2%,
respectively, and COX-2 protein to 47.8 9.0% and 71.6 5.8%,
respectively, compare with the control cells (LPS alone) at a con-
*
*
*
*
centration of 10 lM. Under the same concentration, metabolites
2–5, 7, and 8 did not inhibit the COX-2 protein expression, but sig-
nificantly inhibited iNOS protein expression to 7.3 0.9%,
22.6 4.9%, 6.2 0.7%, 0.1 0.1%, 1.2 0.1%, and 2.2 0.3%, respec-
tively, by LPS stimulation. Moreover, the housekeeping protein b-
actin was not changed by the presence of metabolites 1–8 at the
same concentration. Under the same experimental conditions,
OAc
R1
O
18
OH
6
5
3
4
7
9
17
2
1
14
R1
R2
8
15
19
R2
R3
H
O16
O
13
11
O
O
12
20
10
10 lM CAPE (caffeic acid phenylthyl ester) reduced the levels of
the iNOS and COX-2 protein to 1.5 2.1% and 70.2 11.5%, respec-
tively, relative to the control cells stimulated with LPS. The primary
4 R1= OAc R2= OAc R3= OH
5 R1= H R2= OH R3= H
8 R1= OH R2= H R3=H
1 R1= OH R2= H
2 R1= OAc R2= OH
3 R1= OAc R2= OAc
anti-inflammatory results suggested that the presence of
a-meth-
ylene- -lactone functionality is required for the significantly activ-
c
ity against LPS-stimulated RAW 264.7 cells.
OH
O
H
O
(S)
(R)
O
3. Experimental
(E)
(E)
(R)
(R)
H
O
(R)
(R)
H
O
3.1. General experimental procedures
(R)
(R)
O
O
O
O
(E)
(E)
Optical rotations were determined with a JASCO P1020 digital
polarimeter. IR spectra were recorded on a JASCO FT/IR-4100 infra-
red spectrophotometer. UV spectra were obtained on a JASCO V-
650 spectrophotometer. The NMR spectra were recorded on Bruker
Avance 300 NMR/Varian MR 400 NMR/Varian Unity INOVA 500 FT-
NMR spectrometers (300/400/500 MHz for 1H and 75/100/
125 MHz for 13C), using CDCl3 with TMS as internal standard.
Chemical shifts are given in d (ppm) and coupling constants in
hertz. ESIMS were recorded by ESI FT-MS on a Bruker APEX II mass
spectrometer. Si Gel 60 (Merck, 230–400 mesh) and LiChroprep
6
7
Durumolide L (7) analyzed for C20H24O5 from HRESIMS and 13C
NMR spectroscopic data (Table 2), indicating nine degrees of unsat-
uration. The IR spectrum of 7 showed the presence of aldehyde
(2846 and 1716 cmꢀ1), and
a-methylene-c-lactone (1776 and
1660 cmꢀ1) functionalities. The NMR features (Tables 2 and 3) of
7 were similar to those of 6, except for the replacement of the
hydroxymethyl by an aldehyde group [dH 9.42 (1H, s) and dC
199.3 (CH)] at C-4. This was supported by HMBC correlations from
H-18 to C-3, C-4, and C-5, in a manner similar to that of 6. The rel-
ative stereochemistry and the detailed 1H NMR spectroscopic data
assignment of 7 were determined mainly by the assistance of the
NOESY experiment. The NOESY spectrum and the selected geomet-
ric parameters generated by CHEM3D PRO 9.0 and the coupling con-
RP-18 (Merck, 40–63 lm) were used for column chromatography.
Precoated Si gel plates (Merck, Kieselgel 60 F254, 0.25 mm) and pre-
coated RP-18 F254s plates (Merck) were used for analytical TLC
analyses. High-performance liquid chromatography (HPLC) was
carried out using a Hitachi L-7100 pump equipped with a Hitachi
L-7400 UV detector at 220 nm and a semi-preparative reversed-
stant (3J1,14 = 2.0 Hz) indicated that
7 possessed the similar
phase column (Merck, Hibar Purospher RP-18e, 5
l
m, 250 mm ꢁ
configurations for each ring junction and chiral center as 6 at C-
1, C-3, C-4, and C-14. Therefore, metabolite 7 was definitely pro-
10 mm). S-(+)- and R-(ꢀ)-
a
-Methoxy- -trifluoromethylphenylace-
a
tyl chloride were obtained from ACROS Organics (Geel, Belgium).
*
*
*
*
posed as (1R ,3R ,4R ,14R ,7E,11E)-13,18-dioxo-3,4-epoxycembra-
7,11,15(17)-trien-16,14-olide.
3.2. Animal materials
Geranylgeranyl pyrophosphate23 is considered to be the precur-
sor that, through a series of acid-catalyzed hydrolysis, cyclization,
dehydrogenation, double bond rearrangement, epoxidation,
hydroxylation, oxidation, and lactonization, would yield the cem-
branoids and their analogues isolated from marine soft corals. So
far, it has been mentioned that almost all cembrane diterpenes of
known absolute configuration at C-1 reported from the order Alcy-
onacea belong to the R-configuration (pointing downward), while
the analogues isolated from the order Gorgonacea belong to the
S-configuration.24,25 Our results presented here have added further
support to this conclusion.
The soft coral L. durum was collected by hand using scuba at the
Dongsha Islands located in northeastern South China Sea in April
2007, at a depth of 8 m, and was stored in a freezer for two months
until extraction. This soft coral was identified by one of the authors
(C.-F.D.). A voucher specimen (TS-13) was deposited in the Depart-
ment of Marine Biotechnology and Resources, National Sun Yat-sen
University.
3.3. Extraction and isolation
The absolute stereochemistries of the cembranolides other than
1 and 5 remain to be determined because of the scarcity of material
or the absence of secondary hydroxyl groups.
The frozen soft coral (1.2 kg) was chopped into small pieces and
extracted with acetone for 24 h at room temperature. The quantity
of solvent used for each extraction (2.0 L) was at least three times
the amount of the soft coral material used (1.2 kg). The combined
extracts were concentrated in vacuo (under 30 °C) to obtain 35.0 g
of dry extract, which was suspended in water and extracted with
EtOAc and n-BuOH (saturated with H2O) sequentially. The EtOAc
phase was evaporated to dryness in vacuo to give a brown residue
(30.0 g). The resulting EtOAc residue was subjected to silica gel
chromatography using a stepwise gradient mixture of n-hexane–
EtOAc–MeOH as elution and separated into 40 fractions on the ba-
Metabolites1–8 ataconcentrationof100 lg/diskexhibitedweak
antibacterial activity against the bacterial strain, Salmonella enteriti-
dis (ATCC13076). The results for inhibition of HCMV endonuclease
activity assay are all negative at a concentration of 1 mg/mL.
Our previous study has reported that cembranolides possess
iNOS (inducible nitric oxide synthetase) and COX-2 (cyclooxygen-
ase-2) inhibition,18 which prompted us to evaluate the anti-inflam-
matory effect of these isolated metabolites. As shown in Figure 4,

3768
S.-Y. Cheng et al. / Bioorg. Med. Chem. 17 (2009) 3763–3769
Figure 4. Effect of metabolites 1–8 at 10
lM on the LPS-induced pro-inflammatory iNOS and on COX-2 protein expression of RAW 264.7 macrophage cells by immunoblot
analysis. (A) Immunoblot of iNOS; (B) Immunoblot of COX-2; (C) Immunoblot of b-actin. The values are mean S.E.M. (n = 5). The relative intensity of the LPS alone stimulated
*
*
group was taken as 100%. Significantly different from LPS-stimulated (control) group ( P < 0.05).
sis of TLC and 1H NMR analysis. Fraction 19 (1.18 g) derived from
the n-hexane–EtOAc (1:10) elution was subjected to a silica gel col-
umn using n-hexane–EtOAc mixtures of increasing polarity for elu-
tion, to afford 10 subfractions. A subfraction 19-3 (24 mg) eluted
with n-hexane–EtOAc (2:1) was fractionated over Sephadex LH-
20 (100% MeOH) to produce a mixture (16 mg) that was further
purified by HPLC (RP-18) using 75% MeOH in H2O to give 6
(1 mg) and 7 (1 mg). In turn, a subfraction 19-4 (60 mg) was ap-
plied to column chromatography on reversed-phase C18 gel column
eluting with 70% MeOH in H2O to afford a mixture (30 mg) that
was further separated by HPLC (RP-18) using 65% MeOH in H2O
to provide 5 (4 mg) and 8 (18 mg). Similarly, a subfraction 19-5
(336 mg) eluted with n-hexane–EtOAc (1:2) was subjected to a
RP-18 column eluting with 65% MeOH in H2O to afford a mixture
(214 mg) that was further purified by HPLC (RP-18) using 70%
MeOH in H2O to give 4 (2 mg). Metabolite 2 (1 mg) was obtained
by separation of a subfraction 19-8 (48 mg) eluted with 100%
EtOAc on a HPLC (RP-18) column eluting with 65% MeOH in H2O.
A subfraction 19-6 (202 mg) eluted with n-hexane–EtOAc (1:3)
was subjected to column chromatography on reversed-phase C18
gel eluting with 65% MeOH in H2O to yield a mixture (85 mg) that
was further purified by HPLC (RP-18) using 70% MeOH in H2O to
give 1 (3 mg) and 3 (2 mg).
3.3.5. Durumolide J (5)
Colorless, viscous oil; ½a D24
ꢂ
+6.7 (c 0.4, CHCl3); UV (MeOH) kmax
(log e) 215 (3.78) nm; IR (KBr) vmax 3438, 2957, 1776, 1665, 1447,
1378, 1272, 1239, 1105, 1049, 943, 739 cmꢀ1 1H NMR and 13C
;
NMR data, see Tables 2 and 3; ESIMS m/z 355 [M+Na]+; HRESIMS
m/z 355.1886 [M+Na]+ (calcd for C20H28O4Na, 355.1885).
3.3.6. Durumolide K (6)
Colorless, viscous oil; ½a D24
ꢀ118.0 (c 0.1, CHCl3); UV (MeOH)
ꢂ
kmax (log e) 236 (3.88), 210 (3.64) nm; IR (KBr) vmax 3314, 2962,
1776, 1665, 1447, 1378, 1276, 1239, 1110, 1049, 947, 739 cmꢀ1
;
1H NMR and 13C NMR data, see Tables 2 and 3; ESIMS m/z 369
[M+Na]+; HRESIMS m/z 369.1680 [M+Na]+ (calcd for C20H26O5Na,
369.1678).
3.3.7. Durumolide L (7)
Colorless, viscous oil; ½a D24
ꢀ160.4 (c 0.1, CHCl3); UV (MeOH)
ꢂ
kmax (log e) 235 (3.94), 210 (3.77) nm; IR (KBr) vmax 2952, 2846,
1776, 1716, 1660, 1447, 1378, 1244, 1044, 952, 735 cmꢀ1 1H
;
NMR and 13C NMR data, see Tables 2 and 3; ESIMS m/z 367
[M+Na]+; HRESIMS m/z 367.1361 [M+Na]+ (calcd for C20H24O5Na,
367.1362).
3.4. Preparation of (R)- and (S)-MTPA esters of 1 and 5
3.3.1. Durumolide F (1)
Colorless, viscous oil; ½a D24
ꢂ
ꢀ61.3 (c 0.3, CHCl3); UV (MeOH) kmax
Duplicate (1.0 mg) samples of 1 were prepared for both (R)- and
(S)-MTPA chloride acylation reactions. In separate vials, the samples
were dissolved in 0.5 mL of dry pyridine and allowed to react over-
night at room temperature with (R)- and (S)-MTPA chloride (one
drop), respectively. The reaction was quenched by the addition of
1.0 mL of H2O, followed by extraction with EtOAc (3 ꢁ 1.0 mL). The
EtOAc-soluble layers were combined, dried over anhydrous MgSO4
andevaporated. Theresidueof(R)-MTPAchlorideacylationwassub-
jectedto a short silicagel column elutingwith n-hexane–EtOAc(5:1)
to yield (S)-MTPA ester 1a (0.4 mg). The (R)-MTPA ester 1b (0.3 mg)
was prepared with (S)-MTPA chloride according to the same proce-
dure as described above. In the same manner as shown above, two
aliquots of 5 (1.0 mg) were dissolved in dry pyridine (0.5 mL) and al-
lowed to react overnight with (R)- and (S)-MTPA chloride (one drop),
affording the (S)-MTPA ester 5a (0.4 mg) and (R)-MTPA ester 5b
(0.3 mg), respectively. Selected 1H NMR(CDCl3, 400 MHz) of 1a: dH
7.40–7.52 (5H, m, Ph), 6.21 (1H, d, J = 2.0 Hz, H-17a), 5.68 (1H, d,
J = 2.0 Hz, H-17b), 5.22 (1H, t, J = 5.6 Hz, H-7), 5.16 (1H, t, J = 8.0 Hz,
H-3), 5.07 (1H, br d, J = 5.2 Hz, H-11), 4.33 (1H, dt, J = 9.2, 3.2 Hz,
H-14), 4.15 (1H, d, J = 11.6 Hz, H-18a), 4.08 (1H, d, J = 11.6 Hz, H-
18b), 3.54 (3H, s, OCH3), 2.74 (1H, m, H-1), 2.46 (1H, br d,
J = 11.6 Hz, H-13a), 2.08 (3H, s, 18-OAc), 2.04 (1H, m, H-5a), 2.01
(1H, m, H-2a), 1.73 (1H, m, H-5b), 1.67 (1H, m, H-2b), 1.66 (3H, s,
Me-20), 1.64 (3H, s, Me-19); Selected 1H NMR (CDCl3, 400 MHz) of
1b: dH 7.35–7.59 (5H, m, Ph), 6.20 (1H, d, J = 2.0 Hz, H-17a), 5.65
(1H, d, J = 2.0 Hz, H-17b), 5.25 (1H, t, J = 5.6 Hz, H-7), 5.18 (1H, t,
(log e) 216 (3.87) nm; IR (KBr) vmax 3462, 2948, 1767, 1735, 1665,
1442, 1378, 1267, 1244, 1133, 1049, 952, 735 cmꢀ1 1H NMR and
;
13C NMR data, see Tables 1 and 2; ESIMS m/z 415 [M+Na]+; HRE-
SIMS m/z 415.2097 [M+Na]+ (calcd for C22H32O6Na, 415.2096).
3.3.2. Durumolide G (2)
Colorless, viscous oil; ½a D24
ꢀ32.6 (c 0.1, CHCl3); UV (MeOH) kmax
ꢂ
(log e) 214 (3.68) nm; IR (KBr) vmax 3485, 2952, 1763, 1739, 1670,
1442, 1378, 1244, 1128, 1049, 952, 735 cmꢀ1 1H NMR and 13C
;
NMR data, see Tables 1 and 2; ESIMS m/z 473 [M+Na]+; HRESIMS
m/z 473.2153 [M+Na]+ (calcd for C24H34O8Na, 473.2152).
3.3.3. Durumolide H (3)
Colorless, viscous oil; ½a D24
ꢀ36.5 (c 0.2, CHCl3); UV (MeOH) kmax
ꢂ
(log e) 215 (3.85) nm; IR (KBr) vmax 3453, 2952, 1767, 1739, 1730,
1665, 1442, 1378, 1239, 1049, 952, 739 cmꢀ1 1H NMR and 13C
;
NMR data, see Tables 1 and 2; ESIMS m/z 515 [M+Na]+; HRESIMS
m/z 515.2259 [M+Na]+ (calcd for C26H36O9Na, 515.2257).
3.3.4. Durumolide I (4)
Colorless, viscous oil; ½a D24
ꢂ
+48.3 (c 0.2, CHCl3); UV (MeOH) kmax
(log e) 217 (3.97) nm; IR (KBr) vmax 3443, 2957, 1767, 1744, 1735,
1665, 1442, 1378, 1267, 1239, 1105, 1049, 957, 735 cmꢀ1; 1H NMR
and 13C NMR data, see Tables 2 and 3; ESIMS m/z 471 [M+Na]+;
HRESIMS m/z 471.1993 [M+Na]+ (calcd for C24H32O8Na, 471.1995).

S.-Y. Cheng et al. / Bioorg. Med. Chem. 17 (2009) 3763–3769
3769
J = 8.0 Hz, H-3), 5.08 (1H, br d, J = 5.2 Hz, H-11), 4.44 (1H, dt, J = 9.2,
3.2 Hz, H-14), 4.14 (1H, d, J = 11.6 Hz, H-18a), 4.07 (1H, d,
J = 11.6 Hz, H-18b), 3.45 (3H, s, OCH3), 2.92 (1H, m, H-1), 2.44 (1H,
br d, J = 11.6 Hz, H-13a), 2.05 (3H, s, 18-OAc), 2.05 (1H, m, H-2a),
2.00 (1H, m, H-5a), 1.72 (1H, m, H-5b), 1.69 (1H, m, H-2b), 1.64
(3H, s, Me-20), 1.62 (3H, s, Me-19). Selected 1H NMR (CDCl3,
400 MHz) of 5a: dH 7.41–7.57 (5H, m, Ph), 6.33 (1H, d, J = 3.2 Hz, H-
17a), 6.03 (1H, d, J = 3.2 Hz, H-17b), 5.63 (1H, br t, J = 6.0 Hz, H-11),
5.38 (1H, br d, J = 8.8 Hz, H-7), 5.28 (1H, d, J = 8.8 Hz, H-13), 4.11
(1H, dd, J = 10.8, 2.8 Hz, H-14), 3.55 (3H, s, OCH3), 2.94 (1H, m, H-
1), 2.82 (1H, m, H-3), 1.74 (3H, s, Me-20), 1.72 (3H, s, Me-19), 1.26
(3H, s, Me-18); Selected 1H NMR(CDCl3, 400 MHz) of 5b: dH 7.40–
7.58 (5H, m, Ph), 6.34 (1H, d, J = 3.2 Hz, H-17a), 6.03 (1H, d,
J = 3.2 Hz, H-17b), 5.52 (1H, br t, J = 6.0 Hz, H-11), 5.37 (1H, br d,
J = 8.8 Hz, H-7), 5.17 (1H, d, J = 8.8 Hz, H-13), 4.16 (1H, dd, J = 10.8,
2.8 Hz, H-14), 3.55 (3H, s, OCH3), 2.95 (1H, m, H-1), 2.83 (1H, m, H-
3), 1.72 (3H, s, Me-20), 1.71 (3H, s, Me-19), 1.26 (3H, s, Me-18).
Plus) according to the manufacturer, Sigma Chemical Company, St.
Louis, MO, mark an inner of CAPE in var. paragraph instructions.
3.8. Inhibition of endonuclease activity assay
The HCMV UL76 encodes an endonuclease32 was purified from
E. coli BL21-codon DE3-RIL strain, which was transformed with cal-
modulin-tagged UL76 plasmid pCBP-UL76. The double-stranded
supercoiled DNA uX174 was purified by CsCl gradient and used
as substrate.33 Tested compounds were series diluted and incu-
bated with 1 lg DNA substrate, reaction buffer and UL76 endonu-
clease at 37 °C for 1 h. After cleavage reaction, DNA was resolved in
1% agarose gel. Concentrations of tested compounds that exert 50%
of inhibition of cleaved substrate (IC50) were recorded.
Acknowledgments
This work was supported by grants from the Ministry of Educa-
tion (97C031703) and National Science Council (NSC96-2320-B-
110-003-MY3) of the Republic of China awarded to C.-Y.D.
3.5. Molecular mechanics calculations
Implementation of the MM2 force field26 in CHEM3D PRO software
from CambridgeSoft Corporation, Cambridge, MA, USA (ver. 9.0,
2005), was used to calculate molecular models.
References and notes
1. Blunt, J. W.; Copp, B. R.; Hu, W. P.; Munro, M. H. G.; Northcote, P. T.; Prinsep, M.
R. Nat. Prod. Rep. 2008, 25, 35.
3.6. Antimicrobial activity
2. Higuchi, R.; Miyamoto, T.; Yamada, K.; Komori, T. Toxicon 1998, 36, 1703.
3. Matthee, G. F.; Konig, G. M.; Wright, A. D. J. Nat. Prod. 1998, 61, 237.
4. Wang, S. K.; Duh, C. Y.; Wu, Y. C.; Wang, Y.; Cheng, M. C.; Soong, K.; Fang, L. S. J.
Nat. Prod. 1992, 55, 1430.
5. Coval, S. J.; Patton, R. W.; Petrin, J. M.; James, L.; Rothofsky, M. L.; Lin, S. L.; Patel,
M.; Reed, J. K.; McPhil, A. T.; Bishop, W. R. Bioorg. Med. Chem. Lett. 1996, 6, 909.
6. Duh, C. Y.; Wang, S. K.; Huang, B. T.; Dai, C. F. J. Nat. Prod. 2000, 63, 884.
7. Yamada, K.; Ryu, K.; Miyamoto, T.; Higuchi, R. J. Nat. Prod. 1997, 60, 798.
8. Subrahmanyam, C.; Rao, C. V.; Anjaneyulu, V.; Satyanarayana, P.; Rao, P. V. S.
Tetrahedron 1992, 48, 3111.
Bacterial strains were grown in LB (Luria-Bertani) broth med-
ium for 24 h at 37 °C. Then, 17 mL LB hard agar (1.5% agar) was
poured into sterile Petri dishes (9 cm) and allowed to set. Next,
2.7 mL molten LB soft agar (0.7% agar, 45 °C) was inoculated with
0.3 mL broth culture of the test organism and poured over the base
hard agar plates forming a homogenous top layer. Sterile paper
disks (Advantec, 8 mm) were placed onto the top layer of the LB
9. Bowdon, B. F.; Coll, J. C.; Heaton, A.; Konig, G.; Bruck, M. A.; Cramer, R. E.; Klein,
D. M.; Scheuer, P. J. J. Nat. Prod. 1987, 50, 650.
agar plates. Ten milliliters (2
l
g/lL) of the tested compounds were
10. Uchio, Y.; Eguchi, S.; Kuramoto, J.; Nakayama, M.; Hase, T. Tetrahedron Lett.
1985, 26, 4487.
11. Tursch, B.; Braekman, J. C.; Daloze, D.; Herin, M.; Karlsson, R. Tetrahedron Lett.
1974, 43, 3769.
applied on to each the filter paper disks. Ampicillin (5
l
g/ L) and
l
the same solvents were served as positive and negative controls.
All plates were incubated at 37 °C, 24 h prior to antibacterial activ-
ity evaluation. The antimicrobial activity of all isolated metabolites
12. Kashman, Y.; Groweiss, A. Tetrahedron Lett. 1977, 13, 1159.
13. Kinamoni, Z.; Groweiss, A.; Carmely, S.; Kashman, Y. Tetrahedron 1983, 39,
1643.
14. Kashman, Y.; Carmely, S.; Groweiss, A. J. Org. Chem. 1981, 46, 3592.
15. Rashid, M. A.; Gustafson, K. R.; Boyd, M. R. J. Nat. Prod. 2000, 63, 531.
16. Bowden, B. F.; Coll, J. C.; Tapiolas, D. M. Aust. J. Chem. 1983, 36, 2289.
17. Bowden, B. F.; Coll, J. C.; Decosta, M. S. L.; Desilva, E. D.; Mackay, M. F.;
Mahendran, M.; Willis, R. H. Aust. J. Chem. 1984, 34, 545.
18. Cheng, S. Y.; Wen, Z. H.; Chiou, S. F.; Wang, S. K.; Hsu, C. H.; Dai, C. F.; Chiang, M.
Y.; Duh, C. Y. Tetrahedron 2008, 64, 9698.
1–8 was tested up to 100 lg/disk on S. enteritidis (ATCC13076). The
bacterial strain was obtained from the American Type Culture Col-
lection. The antibiotic activity evaluation method was conducted
based on previously reports.27,28
3.7. Anti-inflammatory assay
19. Zhang, W.; Krohn, K.; Ding, J.; Miao, Z. H.; Zhou, X. H.; Chen, S. H.; Pescitelli, G.;
Salvadori, P.; Kurtan, T.; Guo, Y. W. J. Nat. Prod. 2008, 71, 961.
20. Li, G.; Zhang, Y.; Deng, Z.; van Ofwegen, L.; Proksch, P.; Lin, W. J. Nat. Prod. 2005,
68, 649.
21. Iwashima, M.; Matsumoto, Y.; Takenaka, Y.; Iguchi, K.; Yamori, T. J. Nat. Prod.
2002, 65, 1441.
22. Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem. Soc. 1991, 113,
4092.
23. Kinamoni, Z.; Groweiss, A.; Carmely, S.; Kashman, Y.; Loya, Y. Tetrahedron 1983,
39, 1643.
24. Rodriguez, A. D.; Li, Y.; Dhasmana, H.; Barnes, C. J. Nat. Prod. 1993, 56, 1101.
25. Pham, N. B.; Butler, M. S.; Quinn, R. J. J. Nat. Prod. 2002, 65, 1147.
26. Allinger, N. L. J. Am. Chem. Soc. 1977, 99, 8127.
27. Hou, L.; Shi, Y.; Zhai, P.; Le, G. J. Ethnopharmacol. 2007, 111, 227.
28. Arias, M. E.; Gomez, J. D.; Cudmani, N. M.; Vattuone, M. A.; Isla, M. I. Life Sci.
2004, 75, 191.
29. Ho, F. M.; Lai, C. C.; Huang, L. J.; Kuo, T. C.; Chao, C. M.; Lin, W. W. Br. J.
Pharmacol. 2004, 141, 1037.
30. Park, E. K.; Shin, Y. W.; Lee, H. U.; Kim, S. S.; Lee, Y. C.; Lee, B. Y.; Kim, D. H. Biol.
Pharm. Bull. 2005, 28, 652.
The anti-inflammatory assay was modified from Ho et al.29 and
Park et al.30 Murine RAW 264.7 macrophages were obtained from
the American Type Culture Collection (ATCC, No. TIB-71). The cells
were activated by incubation in medium containing Escherichia coli
LPS (0.01 lg/mL; Sigma) for 16 h in the presence or absence of vari-
ous compounds. Then, cells were washed with ice-cold PBS, lysed in
ice cold lysis buffer, and centrifuged at 20,000g for 30 min at 4 °C.
The supernatant was decanted from the pellet and retained for Wes-
tern blot analysis. Protein concentrations were determined by the
DC protein assay kit (Bio-Rad) modified by the method of Lowry
et al.31 Samples containing equal quantities of protein were sub-
jected to SDS–polyacrylamide gel electrophoresis, and the separated
proteins were electrophoretically transferred to polyvinylidene
difluoride membranes (PVDF; Immobilon-P, Millipore, 0.45 lm pore
size). The resultant PVDF membranes were incubated with blocking
solution and incubated for 180 min with antibody against inducible
nitric oxide synthase (iNOS; 1:1000 dilution; Transduction Labora-
tories) and cyclooxygenase-2 (COX-2; 1:1000 dilution; Cayman
Chemical) protein. The blots were detected using ECL detection re-
agents (Perkin–Elmer, Western Blot Chemiluminescence Reagent
31. Lowry, D. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J. J. Biol. Chem. 1951, 193,
265.
32. Knizewski, L.; Kinch, L.; Grishin, N. V.; Rychlewski, L.; Ginalski, K. J. Virol. 2006,
80, 2575.
33. Tachiwana, H.; Shimura, M.; Nakai-Murakami, C.; Tokunaga, K.; Takizawa, Y.;
Sata, T.; Kurumizaka, H.; Ishizaka, Y. Cancer Res. 2006, 66, 627.