Rare casbane diterpenoids from the hainan soft coral sinularia depressa

Article abstract of DOI:10.1021/np900484k

A series of nine casbane diterpenes, compounds 5-13, exhibiting either cis or trans ring junctions were isolated from the Hainan soft coral Sinularia depressa. The structures of this group of compounds, the basic member of which was named depressin (5), were established by detailed spectroscopic analysis. In addition, the absolute configuration of the main metabolite, 10-hydroxydepressin (7), and of its epimer, 1-epi-10-hydroxydepressin (8), was determined by a combination of conformational analysis and the modified Mosher's method. A stereochemical relationship between all isolated molecules was investigated by analyzing their circular dichroism profiles. Antiproliferative and antibacterial activities of the depressins were also evaluated.

Full text of DOI:10.1021/np900484k

J. Nat. Prod. 2010, 73, 133–138
133
Rare Casbane Diterpenoids from the Hainan Soft Coral Sinularia depressa
Yan Li,† Marianna Carbone,‡ Rosa Maria Vitale,‡ Pietro Amodeo,‡ Francesco Castelluccio,‡ Giovanna Sicilia,‡ Ernesto Mollo,‡
Michela Nappo,‡ Guido Cimino,‡ Yue-Wei Guo,*,† and Margherita Gavagnin*,‡
State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Chinese Academy of Sciences, Shanghai 201203, People’s
Republic of China, and Istituto di Chimica Biomolecolare (ICB), CNR, Via Campi Flegrei 34, 80078 Pozzuoli, Naples, Italy
ReceiVed August 5, 2009
A series of nine casbane diterpenes, compounds 5-13, exhibiting either cis or trans ring junctions were isolated from
the Hainan soft coral Sinularia depressa. The structures of this group of compounds, the basic member of which was
named depressin (5), were established by detailed spectroscopic analysis. In addition, the absolute configuration of the
main metabolite, 10-hydroxydepressin (7), and of its epimer, 1-epi-10-hydroxydepressin (8), was determined by a
combination of conformational analysis and the modified Mosher’s method. A stereochemical relationship between all
isolated molecules was investigated by analyzing their circular dichroism profiles. Antiproliferative and antibacterial
activities of the depressins were also evaluated.
Soft corals of the genus Sinularia (Alcyoniidae) are a rich source
of diterpene metabolites with intriguing structural features and
various promising bioactivities.1 In the course of our ongoing
research focused toward the isolation of biologically active
compounds from Chinese marine organisms,2-6 we carried out a
chemical investigation on the soft coral Sinularia depressa, collected
off Lingshui Bay, Hainan Province, China. The soft coral was found
to possess a secondary metabolite pattern characterized by the
presence of casbane diterpenes. The casbane framework, which was
found for the first time in casbene (1), isolated from an enzymatic
preparation of castor bean seedlings,7 is strictly related to the
cembrane skeleton and differs from this in the presence of a
dimethyl-cyclopropyl moiety rather than the isopropyl residue fused
to the 14-membered ring. Casbane diterpenes are extremely rare
in nature, being described so far only from some species of plants
of the family Euphorbiaceae (i.e., agrostistachin, 2,8 and yuexian-
dajisu A, 39) as well as from the soft coral Sinularia microclaVata
(microclavatin, 410). The latter report represents the only example
from the sea. In the majority of casbanes described in the literature
the two rings forming the macrocyclic structure are cis-fused (1,
2, 4), whereas a very few molecules exhibit a corresponding trans
junction (3).
petroleum ether/acetone gradient). Selected casbane-containing
fractions were subsequently purified on silica gel, Sephadex LH-
20, and reversed-phase HPLC to afford nine pure metabolites,
compounds 5-13.
A preliminary 1H NMR analysis showed that all molecules shared
the same carbon skeleton, differing from each other in either the
degree of oxidation or the configuration of one or more chiral
centers. On the basis of these considerations and with the aim at
simplifying the structure description, the compounds isolated were
grouped and named according to both the extent of oxidation and
the stereochemical relationship. Three cis/trans pairs of molecules,
which were epimeric at C-1, were recognized: depressin (5) and
1-epi-depressin (6), 10-hydroxydepressin (7) and 1-epi-10-hydroxy-
depressin (8), and 10-oxo-11,12-dihydrodepressin (10) and 1-epi-
10-oxo-11,12-dihydrodepressin (11). The remaining three co-
occurring compounds were 1-epi-10-oxodepressin (9), 2-epi-10-
oxo-11,12-dihydrodepressin (12), and 8,10-dihydroxy-iso-depressin
(13). In addition, the epimer at C-1 of compound 9, 10-oxodepressin
(14), was obtained by oxidation of 7. The elucidation of the structure
of 5-13 is here reported according to this grouping.
A total of nine unreported casbanes, compounds 5-13, have been
characterized in the course of this work. Five of these compounds
show a trans-fused bicyclic system, whereas the remaining four
molecules exhibit the cis-junction. This paper describes the isolation,
the structure elucidation including the stereochemical analysis, and
the bioactivity evaluation of compounds 5-13.
Results and Discussion
The Et2O-soluble portion of the acetone extract of the soft coral
S. depressa was subjected to silica gel chromatography (light
The molecular formula C20H30O of depressin (5) was deduced
by the HREIMS molecular peak at m/z 286.2291 and implied six
1
* To whom correspondence should be addressed. (M.G.) Tel: + 39 081
8675094. Fax: + 39 081 8675340. E-mail: margherita.gavagnin@icb.cnr.it.
(Y.W.G.) Tel: + 86 21 50805813. Fax: + 86 21 50807088. E-mail:
ywguo@mail.shcnc.ac.cn.
degrees of unsaturation. The H NMR spectrum of 5 contained
methyl singlets at δH 1.09 (s, 3H), 1.16 (s, 3H), 1.56 (s, 6H), and
1.87 (s, 3H), which were assigned to two tertiary and three vinyl
methyl groups, immediately suggesting a diterpenoid framework.
The presence of an R,ꢀ-unsaturated ketone group was inferred by
† Shanghai Institute of Materia Medica-CAS.
‡ Istituto di Chimica Biomolecolare-CNR.
10.1021/np900484k  2010 American Chemical Society and American Society of Pharmacognosy
Published on Web 02/02/2010

134 Journal of Natural Products, 2010, Vol. 73, No. 2
Li et al.
Table 1. 1H NMR Data (400 MHz, CDCl3) for Compounds 5-14a
proton
5
6
7
8
9
10
11
12
13
14
1
2
1.15, m
1.50, dd
(10.2, 8.7)
6.37, d
(10.2)
3.55, dd
0.71, m
1.08, m
1.15, m
1.50, dd
(10.2, 8.4)
6.30, d
(10.5)
3.50, dd
0.57, m
1.03, m
0.65, m
1.10, m
0.97, m
1.42, m
0.75, m
1.15, m
0.65, m
1.14, m
1.24, m
1.52, m
1.22, m
1.55, dd
(10.2, 9.0)
6.33, d
3
6.11, d
(10.2)
3.71, dd
6.04, d
(10.2)
3.63, dd
6.12, d
(9.9)
3.80, dd
6.35, d
(10.8)
3.85, dd
6.05, d
(10.5)
3.72, dd
6.08, d
(10.5)
3.63, dd
6.39, d
(10.5)
6.55, d
(16.2)
(10.2)
6a
6b
7
3.78, dd
(13.5, 10.2)
2.95, br d
(13.5)
5.35, br d
(10.2)
3.02, d
(13.2)
2.85, d
(13.2)
(13.8, 8.4) (13.8, 11.1) (13.8, 8.4)b (14.3, 11.0)b (13.5, 10.8) (13.5, 10.8) (12.0, 14.7) (14.7, 10.5)
2.97, dd
(13.8, 5.7)
5.08, t
(6.6)
2.15, m
2.83, br d
(13.8)
5.21, br d
(11.1)
3.02, dd
(13.8, 5.7)c
5.11, t
2.89, br d
2.91, d
(13.5)
5.35, br d
(10.8)
3.10, d
(13.2)
2.91, d
(13.2)
2.90, br d
(13.5)
5.42, dd
(10.8, 1.2)
3.07, d
(11.1)
2.95,
3.05, dd
(14.3)c
(overlapped) (14.7, 3.0)
5.27,
5.42, dd
(12.0, 1.8)
3.03,
(overlapped)
2.93,
(overlapped)
5.42, dd
(10.5, 3.0)
3.03, s
6.10, d
(16.2)
2.40, dd
(15.0, 3.6)
1.91, m
(6.6)
(overlapped)
2.48, br d
(11.4)b
9a
9b
10a
2.18, m
2.39, mb
2.00, m
2.17, m
2.10, m
2.28, m
2.07, mc
2.06, mc
2.80, d
(11.1)
3.03, s
4.38, td
4.31, td
4.85, m
(9.3, 3.9)
(9.6, 3.8)
10b
11
1.96, m
4.84, t
(5.4)
2.07, m
4.88, br d
(8.4)
5.06, d
(9.3)
5.26,
(overlapped)
6.03, s
2.60, dd
(18.0, 9.0)
2.12, d
(18.0)
2.70, d
(17.1)
2.05, d
(17.1)
2.35, m
2.35, m
5.67, d
(10.2)
6.17, s
12
2.08, m
1.40, m
0.80, m
1.50, m
0.60, m
1.16, s
1.15, s
1.83, s
1.79, s
0.95, d
(7.2)
2.02, m
1.45, m
1.25, m
1.83, m
1.20, m
1.12, s
1.13, s
1.82, s
1.61, s
0.90, d
(6.6)
2.10, m
1.65, m
1.30, m
2.00, m
1.07, m
1.13, s
1.15, s
1.83, s
1.64, s
0.88, d
(6.6)
13a
13b
14a
14b
16
17
18
19
20
2.20, m
1.75, m
2.05, m
0.80, m
1.16, s
1.09, s
1.87, s
1.56, s
1.56, s
2.17, m
1.93, m
1.88, m
1.03, m
1.16, s
1.11, s
1.80, s
1.59, s
1.59, s
2.38, mb
1.80, mc
2.07, mb
1.19, mc
1.18, s
1.06, s
1.83, s
1.66, s
1.63, s
2.33, mc
2.13, mb
1.99, mc
1.27, mb
1.15, s
1.11, s
1.77, s
1.71, s
1.59, s
2.35, m
1.98, m
2.02, m
1.18, m
1.17, s
1.11, s
1.84, s
1.61, s
2.17, s
2.20, m
1.78, m
1.91, m
1.40, m
1.27, s
1.15, s
1.87, s
1.45, s
1.69, s
2.30, m
1.88, m
2.24, m
0.80, m
1.16, s
1.08, s
1.92, s
1.56, s
2.07, s
a
The assignments were based on DEPT, H-1H COSY, HMQC, and HMBC experiments. b ProR. c ProS.
1
Table 2. 13C NMR Data (100 MHz, CDCl3) for Compounds 5-14a
carbon
5
6
7
8
9
10
11
12
13
14
1
2
3
4
5
6
7
8
35.2, CH
27.6, CH
143.1, CH
136.6, qC
199.9, qC
39.4, CH2
119.4, CH
137.1, qC
39.0, CH2
23.9, CH2
124.4, CH
135.9, qC
39.9, CH2
26.3, CH2
25.4, qC
37.5, CH
31.9, CH
149.4, CH
132.8, qC
201.2, qC
40.4, CH2
121.7, CH
135.1, qC
38.4, CH2
24.0, CH2
125.4, CH
133.5, qC
38.8, CH2
24.4, CH2
28.1, qC
35.2, CH
28.6, CH
142.9, CH
136.2, qC
200.0, qC
39.8, CH2
120.5, CH
134.5, qC
46.7, CH2
67.5, CH
127.8, CH
140.8, qC
39.2, CH2
25.8, CH2
25.3, qC
29.0, CH3
15.7, CH3
11.5, CH3
18.5, CH3
18.3, CH3
1
37.0, CH
35.2, CH
148.1, CH
133.0, qC
200.4, qC
40.6, CH2
122.9, CH
133.0, qC
46.4, CH2
68.7, CH
129.2, CH
140.1, qC
40.1, CH2
25.4, CH2
29.7, qC
37.2, CH
31.6, CH
148.1, CH
133.9, qC
199.6, qC
40.6, CH2
125.6, CH
130.1, qC
56.6, CH2
198.3, qC
122.9, CH
158.2, qC
41.1, CH2
24.0, CH2
28.3, qC
35.9, CH
27.8, CH
144.4, CH
136.0, qC
199.3, qC
40.7, CH2
125.1, CH
130.4, qC
53.6, CH2
207.1, qC
50.1, CH2
27.2, CH
39.9, CH2
24.5, CH2
25.2, qC
38.8, CH
32.6, CH
148.3, CH
134.6, qC
200.2, qC
41.0, CH2
126.1, CH
131.4, qC
56.0, CH2
209.1, qC
49.5, CH2
27.9, CH
35.7, CH2
27.5, CH2
29.5, qC
39.7, CH
33.6, CH
147.5, CH
134.2, qC
200.9, qC
34.6, CH
27.9, CH
145.9, CH
137.7, qC
195.1, qC
34.5, CH
27.8, CH
141.9, CH
137.7, qC
199.3, qC
39.3, CH2
124.2, CH
132.8, qC
55.9, CH2
200.1, qC
122.8, CH2
158.4, CH
41.1, CH2
26.8, CH2
26.1, qC
40.7, CH2 128.4, CH
124.7, CH
131.7, qC
55.8, CH2
208.7, qC
143.6, CH
83.0, qC
43.4, CH2
67.6, CH
9
10
11
12
13
14
15
16
17
18
19
20
a
50.8, CH2 124.5, CH
27.6, CH
35.6, CH2
24.5, CH2
27.9, qC
21.6, CH3
23.5, CH3
11.6, CH3
16.7, CH3
18.7, CH3
139.5, qC
37.1, CH2
22.9, CH2
27.2, qC
29.1, CH3
16.3, CH3
11.5, CH3
26.0, CH3
18.8, CH3
29.0, CH3
15.8, CH3
11.6, CH3
15.6, CH3
15.3, CH3
22.0, CH3
23.7, CH3
11.2, CH3
14.7, CH3
14.8, CH3
22.0, CH3
23.0, CH3
11.1 CH3
19.0, CH3
14.8, CH3
21.6, CH3
23.3, CH3
11.1, CH3
15.6, CH3
17.8, CH3
29.0, CH3
16.0, CH3
11.7, CH3
16.8, CH3
23.1, CH3
21.7, CH3
23.8, CH3
11.6, CH3
16.4, CH3
22.4, CH3
29.0, CH3
15.8, CH3
11.9, CH3
16.6, CH3
18.1, CH3
The assignments were based on DEPT, H-1H COSY, HMQC, and HMBC experiments.
both the 13C NMR signal at δC 199.9 (C) and the IR band at 1655
cm-1. Three trisubstituted double bonds were also evident by six
sp2 carbon signals in the 13C NMR spectrum at δ 143.1 (CH), 137.1
(C), 136.6 (C), 135.9 (C), 124.4 (CH), and 119.4 (CH) and by three
cyclopropyl ring. Consequently, the remaining ring was deduced
to be a 14-membered macrocycle, according to a casbane carbon
arrangement. A detailed analysis of 2D NMR spectra of 5 allowed
the assignment of all proton and carbon atoms (Tables 1 and 2).
The E geometry of all three double bonds ∆3(4), ∆7(8), and ∆11(12)
was deduced by the δC values of CH3-18, CH3-19, and CH3-20
(<20 ppm).11 The junction of the two rings at carbons C-1/C-2
was suggested to be cis on the basis of 13C NMR chemical shifts
of the geminal methyls C-16 and C-17 (δC 29.0 and 15.8,
respectively), analogously with literature cis-fused casbane diter-
penes, i.e., agrostistachin (2)8 and microclavatin (4).10 The proposed
configuration was further confirmed by a diagnostic NOE interaction
observed between H-1 and H-2.
1
olefinic multiplets in the H NMR spectrum at δ 4.84 (1H, t, J )
5.4 Hz, H-11), 5.08 (1H, t, J ) 6.6 Hz, H-7), and 6.37 (1H, d, J )
10.2, H-3). The remaining two unsaturation degrees required by
the molecular formula of 5 were thus attributed to two rings. The
presence of both a quaternary carbon (δ 25.4, C-15) and a gem-
dimethyl functionality [δΗ 1.16 (s, H3-16), δΗ 1.09 (s, H3-17); δC
29.0 (C-16), δC 15.8 (C-17)], together with the COSY correlation
between two cyclopropyl protons [δΗ 1.15 (m, H-1), δΗ 1.50 (dd,
J ) 10.2 and 8.7 Hz, H-2)], indicated the presence of a substituted

Rare Casbane Diterpenoids from Sinularia depressa
Journal of Natural Products, 2010, Vol. 73, No. 2 135
Compound 6 was an isomer of 5 with the same molecular
formula (C20H30O) as deduced by HREIMS. A detailed 2D NMR
analysis revealed that the planar structure of 6 was the same as 5,
thus implying that the difference between the two compounds
concerned the configuration. As the geometry of all three double
bonds of 6 was inferred to be E, analogously with 5, by both
NOESY and 13C NMR data (see Table 2), a different configuration
at one of the two chiral carbons was thus expected. Analysis of
the NOESY experiment of 6 showed the presence of interactions
between H-1 and H3-17, and between H-2 and H3-16, indicating
that H-1 and H-2 were oriented in opposite directions, consistent
with a trans ring junction. This assignment was confirmed by the
diagnostic 13C NMR chemical shifts of the C-16 and C-17 methyl
groups (δC 22.0 and 23.7, respectively), in accordance with those
reported for the trans-fused yuexiandajisu A (3).9 Due to the
different ring junctions, the 1H NMR values of H-1, H-2, and H-3
and the 13C NMR chemical shifts of C-1, C-2, C-3, C-4, and C-15
were also significantly shifted compared to 5 (Tables 1 and 2).
Detailed spectroscopic NMR analysis of compound 6 led us to
assign all proton and carbon values. Surprisingly, the CD profiles
of 5 and 6 were almost identical suggesting that C-2, near the
chromophore, had the same configuration in both molecules.
Consequently, the configuration of C-1 was assumed to be opposite,
and thus 6 was named 1-epi-depressin.
10-Hydroxydepressin (7), the main component of the depressin
fraction, and 1-epi-10-hydroxydepressin (8) had the molecular
formula C20H30O2 and were the 10-hydroxy derivatives of 5 and 6,
respectively, as evidenced by careful analysis of their spectroscopic
data (Tables 1 and 2). The geometry of the three double bonds in
7 and 8 was E, the same as in 5 and 6. Compound 7 exhibited the
cis junction at C-1/C-2, whereas in 8 the two rings were trans-
fused, as indicated by the diagnostic 13C NMR values of the geminal
methyl groups C-16 (δC 29.0 in 7, δC 22.0 in 8) and C-17 (δC 15.7
in 7, δC 23.0 in 8). The 1H-1H COSY spectra showed the diagnostic
correlation of a carbinolic signal (δH 4.38 in 7, δH 4.31 in 8) with
two multiplets (δH 2.39 and 2.07 in 7, δH 2.48 and 2.06 in 8) and
a doublet (δH 5.06 in 7, δH 5.26 in 8) that were attributed to an
allylic methylene (H2-9) and a vinylic proton (H-11), respectively.
This result led us to locate unambiguously the secondary hydroxy
function in the macrocyclic fragment of the molecule at C-10 in
both molecules. Analysis of 2D NMR spectra confirmed the
proposed planar structures and allowed the complete assignment
as listed in Tables 1 and 2.
Figure 1. ∆δ (δS - δR) values (in ppm) for the MTPA esters of
7 and 8.
integrated as arising from a single conformer. This, in turn, prevents
traditional structural refinement approaches based on deriving
distance restraints from NOE intensities to be applied in confor-
mational sampling and a refinement strategy.
In view of this, we performed an exhaustive conformational
search by generation of a large number of conformers (N ) 200)
for each model, by employing a simulated annealing (SA) protocol,
followed by a derivation of ensemble 1/r6-averaged interprotonic
distances. The SA protocol has been successfully applied in many
restrained and unrestrained calculations on systems varying from
small organic molecules up to peptides and proteins.14
For each molecule, predicted average distances of the two
diastereomers were compared to distances derived from experi-
mental peak volumes to check if all predicted short distances have
a counterpart in the NOESY spectrum and Vice Versa. In particular,
potentially diagnostic contacts, i.e., those giving rise to substantially
different distance patterns between the two diasteromers, were
monitored with special care.
In the case of 7, only the (1S,2S,10S)-diastereomer was able to
account for the whole experimental contact pattern, while the
(1R,2R,10S)-diastereomer exhibited systematic violations in both
the full ensemble containing all 200 calculated structures and the
subsets only including the representative structures of lowest energy
conformer clusters. For the (1S,2S,10S)-diastereomer, a subset
including the representative structures of the two lowest energy
conformer clusters, i.e., the two most two populated clusters (71
and 52 molecules, respectively) with an energy difference of 0.1
kcal/mol, fully accounts for the whole experimental NOE-derived
contact pattern (see Supporting Information for details) (Figure 2).
On these bases the absolute configuration of 7 was assigned as
(1S,2S,10S).
A similar approach was also used for the two possible diaster-
eomers of 8. In this case, there is a pseudoenantiomeric relationship
between the lowest energy conformers of the two diastereomers,
where only the OH and H substituent groups on the asymmetric
C-10 atom were swapped. Consequently, potentially diagnostic
interactions are located mainly around either the hydroxy group,
whose signals are not detected in the NOESY spectra due to fast
exchange in chloroform, or the H-10 atom. In spite of this, a limited
but clear-cut pattern of interactions differentiating the two possible
diastereomers can be safely identified. In fact, in the case of
(1S,2R,10S)-8 neither the lowest energy cluster nor the most
populated ones, singularly or in combination, fully accounted for
the experimental NOE cross-peaks (see Supporting Information for
details). On the contrary, in the case of the (1R,2S,10S)-8 diaste-
reomer, the observed pattern of 40 nontrivial NOE effects is fully
justified by a simple mixture of the two most stable and populated
conformer clusters (Figure 2), similar to that seen for (1S,2S,10S)-
7. Even the structural relationship between the two conformers is
similar to that detected in (1S,2S,10S)-7, their main difference
consisting in a rotation around the C-7/C-8 bond. On these bases,
the absolute configuration of 8 was determined as 1R,2S,10S.
Thus, compounds 7 and 8 were epimers at C-1. Due to the
similarity of the CD curves of 5 and 6 with those of 7 and 8, the
absolute configurations of depressin (5) and 1-epi-depressin (6) were
also defined as 1S,2S and 1R,2S, respectively.
The CD curves of 7 and 8 were similar to those of 5 and 6, thus
suggesting a stereochemical correspondence between the two pairs
of molecules. Due to the flexible nature of the 14-membered
macrocyclic ring, the relative configuration at C-10 could not be
deduced by 2D NOESY experiments. Furthermore, efforts to obtain
crystals suitable for X-ray crystallographic analysis were unsuc-
cessful. Thus, we decided first to determine the absolute configu-
ration at C-10 in 7 and 8 by applying the modified Mosher
method12,13 and then to undertake a conformational analysis with
the aim of relating this configuration to that of the ring junction
and establishing the absolute configuration of both molecules.
The (S)- and (R)-MTPA esters of 7 and 8 were prepared by
treatment with (R)- and (S)-MTPA chloride, respectively. The ∆δ
values observed for the signals of protons near the hydroxy group
at C-10 for both esters of each compound are reported in Figure 1
and indicated the S configuration for the carbinol asymmetric center
in 7 and 8.
Models of the four possible diastereomers, (1S,2S,10S) and
(1R,2R,10S) for compound 7 and (1S,2R,10S) and (1R,2S,10S) for
compound 8, were built in order to conduct a conformational study.
A preliminary analysis showed a potential, albeit limited, flexibility
for all these models, with the possible occurrence of a small number
of main conformers endowed with similar conformational energy.
In this case, NOESY spectra can not be safely assigned and

136 Journal of Natural Products, 2010, Vol. 73, No. 2
Li et al.
functional groups at C-5 and C-10 and only two E-double bonds
(∆3(4) and ∆7(8)), with the third one having been reduced, in contrast
to the other co-occurring depressins. The secondary methyl group
was located at C-12 in all three compounds by analysis of their
1H-1H COSY spectra. The correlations observed in the HSQC and
HMBC experiments confirmed the same planar structure and
implied that they had to differ from each other in the relative
configuration of one or two chiral centers. Analysis of their 13C
NMR spectra (Table 2) revealed that 10 had the 1,2-cis ring
junction, whereas the other two isomers, 11 and 12, were trans-
fused. Comparison of their CD curves disclosed interesting stere-
ochemical features. Compounds 10 and 11 had the same CD profile
as depressins 5-8, whereas compound 12 displayed an opposite
CD curve. This implied that in compounds 10 and 11 the absolute
configuration at the ring junction carbons was 1S,2S and 1R,2S,
respectively, the same as the other co-occurring cis and trans
molecules, whereas in 12 this configuration had to be opposite
(1S,2R). At this point, having assessed the configuration of the ring
junction, it remained to be defined the configuration of C-12. Due
to the flexibility of the casbane macrocycle, the relative orientation
of H3-20 in compounds 10-12 could not be confidently assigned
by NOE analysis. However, some considerations could be made
by comparing the spectroscopic data of 10-oxo-11,12-dihydrode-
pressin (10) with literature models. In particular, the 13C NMR value
of C-20 (δ 23.1) of 10 was almost comparable with that of the
structurally related microclavatin (4) (δ 21.8), the relative config-
uration of which has been secured by X-ray diffraction analysis.10
This led us to tentatively assign to C-12 the same relative
configuration as 4. By biogenetic considerations, the same config-
uration at C-12 was assumed for 1-epi-10-oxo-11,12-dihydrode-
pressin (11). According to this assumption, due to the fact that 11
and 12 had opposite configurations in the trans ring junction moiety
but are not enantiomers, the configuration of C-12 also had to be
the same in 12, 2-epi-10-oxo-11,12-dihydrodepressin.
Figure 2. The two representative clusters for the (1S,2S,10S)-7 (top)
and (1R,2S,10S)-8 (bottom) diastereomers are shown in ball-and-
stick representations and colored in light (71 and 69 overlapped
structures for (1S,2S,10S)-7 and (1R,2S,10S)-8, respectively) or dark
gray (52 and 58 overlapped structures for (1S,2S,10S)-7 and
(1R,2S,10S)-8, respectively). The oxygen atoms are in black with
a transparent contour. For clarity, only hydrogen atoms on C-1 and
C-2 atoms are evidenced to show the configuration of the cyclo-
propane ring.
The molecular formula (C20H30O3) of 8,10-dihydroxy-iso-de-
pressin (13) implied an additional oxygen atom with respect to 7
and 8. The spectroscopic data of 13 revealed structural features
similar to those of 7 including the secondary hydroxy group at C-10
and the 1,2-cis-junction, whereas the C-6/C-8 carbon fragment was
different. The presence of an isolated disubstituted double bond
The molecular formula (C20H28O2) of 1-epi-10-oxodepressin (9)
was established by HREIMS and contained an additional unsat-
uration degree with respect to 7 and 8. Analysis of the spectroscopic
data indicated the presence at C-10 in the macrocyclic ring of 9 of
a second ketone group, which was R,ꢀ-unsaturated, as indicated
by the 13C NMR signal at δC 198.3, rather than the secondary
hydroxy function present in 7 and 8. Diagnostic HMBC correlations
were observed between C-10 and both the isolated methylene H2-9
(δH 2.91 and 3.10) and the olefinic proton H-11 (δH 6.03), according
to the proposed arrangement. Compound 9 had the 1,2-trans-
6(7)
1
∆
was suggested by both H NMR signals at δH 6.55 (1H, d, J
) 16.2, H-6) and 6.10 (1H, d, J ) 16.2, H-7) and the corresponding
13C NMR signals at δC 128.4 (CH, C-6) and 143.6 (CH, C-7). The
carbon spectrum of 13 also contained a signal at δC 83.0 (qC, C-8)
attributable to a tertiary carbon linked to oxygen, thus indicating
the presence a second hydroxy group located at C-8. Analysis of
1H-1H COSY, HSQC, and HMBC experiments confirmed this
structural hypothesis. The geometry of the ∆6(7) double bond was
suggested to be E by the coupling constant value,11 whereas the
configurations at C-8 and C-10 were not determined. The CD profile
was almost comparable with those of depressins 5-11, and thus
the configuration of the cis ring junction was assumed to be 1S,2S,
the same as that of other cis-depressins.
junction analogously with 6 and 8, as indicated by the 1H and 13
C
NMR values (Tables 1 and 2). The CD curve of 9 was more
complex with respect to those of 6 and 8 due to the presence of
the second R,ꢀ-unsaturated ketone. However, a portion similar to
that observed in the CD profiles of compounds 6 and 8 was
recognizable in the curve of 9, suggesting the same 1R,2S absolute
configuration at the junction asymmetric centers. This was con-
firmed by oxidizing a sample of 1-epi-10-hydroxydepressin (8) by
Dess-Martin reagent. The product obtained had spectroscopic and
optical properties15 identical with those of 9 (see Experimental
Section and Supporting Information).
With the aim of confirming the stereochemical relationship
between Sinularia metabolites, a sample of 10-hydroxydepressin
(7) was oxidized under the same conditions as compound 8 to give
the corresponding oxidized derivative 14, which was fully char-
acterized (Tables 1 and 2). The CD curve of non-natural compound
14 was identical with that of 9. The two compounds were in the
same cis-trans relationship as the corresponding hydroxy deriva-
tives 7 and 8, respectively.
Compounds 5-13 were tested for cytotoxicity against human
tumor cell lines HepG2 and SW-1990. The activity against
Staphylococcus aureus and Escherichia coli was also evaluated for
5-8. 10-Hydroxydepressin (7) showed cytotoxic activity against
tumor cell lines HepG2 and SW-1990 with IC50 values of 61 and
37 µM, respectively, and antimicrobial activity against S. aureus
and E. coli at 17 µM.
Experimental Section
General Experimental Procedures. Optical rotations were measured
on a Perkin-Elmer 241MC polarimeter. UV spectra were recorded on
a Varian Cary 300 Bio spectrometer. IR spectra were recorded on a
Nicolet-Magna FT-IR 750 spectrometer. NMR spectra were measured
on Bruker DRX-400, Bruker DRX-600, and Varian Mercury 300
spectrometers with the residual CHCl3 (δH 7.26 ppm; δC 77.0 ppm) as
Compounds 10, 11, and 12 exhibited the same molecular formula,
1
C20H30O2, as deduced by HREIMS. Analysis of their H and 13C
NMR spectra (Tables 1 and 2) clearly revealed the presence in the
three molecules of an identical oxidation pattern with two keto

Rare Casbane Diterpenoids from Sinularia depressa
Journal of Natural Products, 2010, Vol. 73, No. 2 137
an internal standard. NOESY spectra were obtained with different
mixing times (0.180, 0.250, and 0.400 s). EIMS and HREIMS data
were obtained on a Finnigan-MAT-95 mass spectrometer. ESIMS and
HRESIMS spectra were recorded on a Q-TOF Micro LC-MS-MS mass
spectrometer. Reversed-phase HPLC (Agilent 1100 series liquid
chromatography using a VWD G1314A detector at 210 nm and a
semipreparative ODS-HG-5 [5 µm, 10 mm (i.d.) × 25 cm] column
was also employed. Commercial Si gel (Qing Dao Hai Yang Chemical
Group Co., 200-300 and 400-600 mesh) was used for column
chromatography (CC), and precoated Si gel plates (Yan Tai Zi Fu
Chemical Group Co., G60 F-254) were used for analytical and
preparative TLC.
8,10-Dihydroxy-iso-depressin (13): colorless oil; [R]2D5 -39.0
(c 0.20, CHCl3); CD (c 1.42 × 10-3 M, n-hexane) λ (∆ε) 203 (0.06),
217 (0.8), 262 (-2.3); IR (KBr) νmax 1668 cm-1; 1H and 13C NMR in
Tables 1 and 2; HRESIMS m/z 341.2096 (calcd for C20H30O3Na,
341.2093).
Preparation of MTPA Esters. (R)- and (S)-MTPA-Cl (10 µL) and
a catalytic amount of DMAP were separately added to two different
aliquots of alcohol 7 (1.0 mg each) in dry CH2Cl2 (0.5 mL), and the
resulting mixtures were allowed to stand at room temperature for 12 h.
After the evaporation of the solvent, the mixtures were purified on
preparative TLC (SiO2, light petroleum ether/Et2O, 8:2), affording pure
(S)- and (R)-MTPA esters of 7, respectively. The MTPA esters of 8
were prepared following the same procedure.
Collection and Extraction of the Animal Material. Specimens of
Sinularia depressa were collected at Lingshui Bay, Hainan Province,
China, in July 2004, at a depth of 20 m and were frozen immediately
after collection. A voucher specimen of S. depressa (LS-338) is
available for inspection at the Shanghai Institute of Materia Medica,
CAS. The frozen material (510 g dry weight) was cut into small pieces
and extracted exhaustively with acetone at room temperature (3 × 1.5
L). The organic extract was evaporated to give a residue, which was
partitioned between Et2O and H2O. The Et2O solution was concentrated
under reduced pressure to give a dark brown residue (9.6 g).
Purification of Compounds 5-13. The Et2O extract was fraction-
ated by gradient Si gel column chromatography eluting with a step
gradient (0-100% acetone in light petroleum ether) to yield three
casbane-containing fractions [A (900 mg), B (1.2 g), C (500 mg)]. An
aliquot of fraction A (50.0 mg) was further purified by Si gel CC (light
petroleum ether/Et2O, 95:5) to yield 5 (5.6 mg) and 6 (6.2 mg). An
aliquot of fraction B (61.0 mg) was subjected to a RP-HPLC purification
[semipreparative ODS-HG-5 (5 µm, 250 × 10 mm), MeCN/H2O, 70:
30, 2.0 mL/min], obtaining 9 (4.7 mg; tR 16.7 min), 11 (5.5 mg; tR
19.2 min), 12 (4.3 mg; tR 22.3 min), and 10 (5.0 mg; tR 25.5 min).
Fraction C was fractionated by Si gel CC (light petroleum ether/Et2O,
6:4 and 5:5) to give a mixture, which was further purified on a Sephadex
LH20 column (light petroleum ether, CHCl3, CH3OH, 2:1:1), yielding
the main component, 7 (134.5 mg), along with 8 (1.6 mg) and 13 (1.1
mg).
(S)-MTPA ester of 7: selected 1H NMR values (CDCl3, 400 MHz)
δ 6.27 (1Η, d, J ) 10.5 Hz, H-3), 5.65 (1H, ddd, J ) 14.0, 10.0, 4.0
Hz, H-10), 5.17 (1H, t, J ) 6.6 Hz, H-7), 5.06 (1H, d, J ) 9.3 Hz,
H-11), 3.53 (3H, s, OMe), 3.50 (1H, m, H-6a), 3.02 (1H, dd, J ) 13.8,
5.7, H-6b), 2.41 (1H, m, H-13a), 2.35 (1H, m, H-9a), 2.18 (1H, m,
H-9b), 1.85 (1H, m, H-13b), 1.85 (3H, s, H3-18), 1.72 (3H, s, H3-20),
1.64 (3H, s, H3-19), 1.18 (3H, s, H3-16), 1.05 (3H, s, H3-17).
(R)-MTPA ester of 7: selected 1H NMR values (CDCl3, 400 MHz)
δ 6.26 (1Η, d, J ) 10.5 Hz, H-3), 5.62 (1H, ddd, J ) 14.0, 10.0, 4.0
Hz, H-10), 5.19 (1H, t, J ) 6.6 Hz, H-7), 4.94 (1H, d, J ) 9.3 Hz,
H-11), 3.54 (3H, s, OMe), 3.52 (1H, m, H-6a), 3.03 (1H, dd, J ) 13.8,
5.7 Hz, H-6b), 2.45 (1H, m, H-9a), 2.35 (1H, m H-13a,), 2.28 (1H, m,
H-9b), 1.85 (1H, m, H-13b), 1.85 (3H, s, H3-18), 1.71 (3H, s, H3-20),
1.66 (3H, s, H3-19), 1.18 (3H, s, H3-16), 1.05 (3H, s, H3-17).
(S)-MTPA ester of 8: selected 1H NMR values (CDCl3, 400 MHz)
δ 6.02 (1Η, d, J ) 10.2 Hz, H-3), 5.66 (1H, ddd, J ) 13.4, 9.8, 3.8
Hz, H-10), 5.35 (1H, br d, J ) 8.3 Hz, H-7), 5.28 (1H, d, J ) 9.3 Hz,
H-11), 3.61 (1H, dd, J ) 14.3, 10.9 Hz, H-6a), 3.53 (3H, s, OMe),
2.92 (1H, br d, J ) 14.3 Hz, H-6b), 2.44 (1H, br d, J ) 11.4 Hz,
H-9a), 2.20 (1H, dd, J ) 11.4, 10.5 Hz, H-9b,), 2.32 (1H, m, H-13a),
2.16 (1H, m, H-13b), 1.99 (1H, m, H-14b), 1.26 (1H, m, H-14a), 1.78
(3H, s, H3-18), 1.66 (3H, s, H3-19), 1.62 (3H, s, H3-20), 1.17 (3H, s,
H3-16), 1.10 (3H, s, H3-17), 1.04 (1H, m, H-2), 0.56 (1H, m, H-1).
(R)-MTPA ester of 8: selected 1H NMR values (CDCl3, 400 MHz)
δ 6.01 (1Η, d, J ) 10.2 Hz, H-3), 5.65 (1H, ddd, J ) 13.2, 9.7, 3.7
Hz, H-10), 5.36 (1H, br d, J ) 8.3 Hz, H-7), 5.17 (1H, d, J ) 9.3 Hz,
H-11), 3.60 (1H, dd, J ) 14.3, 11.0 Hz, H-6a), 3.53 (3H, s, OMe),
2.93 (1H, br d, J ) 14.3 Hz, H-6b), 2.52 (1H, br d, J ) 11.4 Hz,
H-9a), 2.35 (1H, dd, J ) 11.4, 10.4 Hz, H-9b,), 2.24 (1H, m, H-13a),
2.13 (1H, m, H-13b), 2.01 (1H, m, H-14a), 1.25 (1H, m, H-14b), 1.78
(3H, s, H3-18), 1.68 (3H, s, H3-19), 1.58 (3H, s, H3-20), 1.15 (3H, s,
H3-16), 1.09 (3H, s, H3-17), 1.04 (1H, m, H-2), 0.55 (1H, m, H-1).
Oxidation of Compounds 7 and 8. Compounds 7 (6.0 mg) and 8
(1.1 mg) were separately treated with Dess-Martin reagent in CH2Cl2
(1 mL). The reaction mixtures were stirred at room temperature for
18 h. The excess Dess-Martin reagent was destroyed by adding
isopropyl alcohol, and the mixtures were washed with NaHCO3 solution.
The organic phases were separated and concentrated under vacuum,
and the residues were purified by TLC chromatography on Si gel CC
(light petroleum ether/diethyl ether, 7:3) to give 14 (4.6 mg, from 7)
and a compound with spectroscopic data identical with those of natural
9 (0.4 mg, from 8).
Depressin (5): colorless oil; [R]2D5 -80.0 (c 0.26, CHCl3); CD
(c 2.10 × 10-3 M, n-hexane) λ (∆ε) 204 (7.4), 249 (-2.1); UV (EtOH)
λ
max (log ε) 271 (3.56); IR (KBr) νmax 1655 cm-1; 1H and 13C NMR in
Tables 1 and 2; HREIMS m/z 286.2291 (calcd for C20H30O, 286.2297).
1-epi-Depressin (6): colorless oil; [R]2D5 +34.0 (c 0.25, CHCl3); CD
(c 1.05 × 10-3 M, n-hexane) λ (∆ε) 205 (5.1), 236 (0.2); UV (MeOH)
λ
max (log ε) 274 (5.0) nm; IR (KBr) νmax 1653 cm-1; 1H and 13C NMR
in Tables 1 and 2; HREIMS m/z 286.2306 (calcd for C20H30O,
286.2297).
10-Hydroxydepressin (7): colorless oil; [R]2D5 -218 (c 0.55, CHCl3);
CD (c 9.93 × 10-4 M, n-hexane) λ (∆ε) 202 (27.6), 252 (-17.7); UV
(EtOH) λmax (log ε) 270 (4.76); IR (KBr) νmax 3388, 1650 cm-1; H
1
and 13C NMR in Tables 1 and 2; HREIMS m/z 302.2240 (calcd for
C20H30O2, 302.2246).
1-epi-10-Hydroxydepressin (8): colorless oil; [R]2D5 +4.5 (c 0.54,
CHCl3); CD (c 9.93 × 10-4 M, n-hexane) λ (∆ε) 206 (31.9), 232
(-10.5); UV (EtOH) λmax (log ε) 270 (4.82); IR (KBr) νmax 3399, 1651
cm-1; H and 13C NMR in Tables 1 and 2; HRESIMS m/z 325.2140
1
1
(calcd for C20H30O2Na, 325.2144).
10-Oxodepressin (14): colorless oil; [R]2D5 -76 (c 0.4, CHCl3); H
1-epi-10-Oxodepressin (9): colorless oil; [R]2D5 +11.0 (c 0.27,
CHCl3);15 CD (c 1.00 × 10-3 M, n-hexane) λ (∆ε) 201 (2.1), 241 (2.1),
and 13C NMR in Tables 1 and 2; ESIMS [M + Na]+ m/z 323.
Synthetic 1-epi-10-oxodepressin: colorless oil; [R]2D5 +210 (c 0.02,
CHCl3);15 1H and 13C NMR identical to those of 9 in Tables 1 and 2;
ESIMS [M + Na]+ m/z 323.
1
265 (-1.1); IR (KBr) νmax 1668 cm-1; H and 13C NMR in Tables 1
and 2; HREIMS m/z 300.2089 (calcd for C20H28O2, 300.2089).
10-Oxo-11,12-dihydrodepressin (10): colorless oil; [R]2D5 +155
(c 0.13, CHCl3); CD (c 8.28 × 10-4 M, n-hexane) λ (∆ε) 202 (43.0),
Conformational Analysis. Structures were obtained by unrestrained
SA/EM.16 Calculations were performed with Sander module of
AMBER10,17 using AMBER GAFF parametrization18 and RESP
charges,19 fitted from ab initio 6-31G* calculations performed with
GAMESS.20 Starting structures for all molecules were built with
Ghemical 2.95.21 Conformational sampling was obtained by unre-
strained simulated annealing. The starting structure of each molecule
underwent 200 SA cycles of 100 000 MD steps, where system
temperature was linearly raised from 10 to 1200 K (steps 1 to 5000),
then kept constant at 1200 K (steps 5001 to 50 000), and finally, linearly
decreased down to 10 K (steps 50 001 to 100 000). A time step of 1
fs, with no constraints or restraints on bond lengths, a nonbonded cutoff
of 16 Å, and a 0.05 fs time constant for heat bath coupling were used,
with all other parameters set at their default values. trans-Alkene bonds
were forced into a trans orientation (ω ) 180°) by torsional constraints
1
234 (-26.6); 268 (9.9); IR (KBr) νmax 1709, 1647 cm-1; H and 13C
NMR in Tables 1 and 2; HREIMS m/z 302.2247 (calcd for C20H30O2,
302.2246).
1-epi-10-Oxo-11,12-dihydrodepressin (11): colorless oil; [R]D25
-74.0 (c 0.24, CHCl3); CD (c 1.32 × 10-3 M, n-hexane) λ (∆ε) 204
1
(7.1), 232 (-0.7), 268 (3.4); IR (KBr) νmax 1709, 1653 cm-1; H and
13C NMR in Tables 1 and 2; HREIMS m/z 302.2250 (calcd for
C20H30O2, 302.2246).
2-epi-10-Oxo-11,12-dihydrodepressin (12): colorless oil; [R]D25
+133 (c 0.28, CHCl3); CD (c 1.16 × 10-3 M, n-hexane) λ (∆ε) 204
(-7.6), 235 (2.6), 267 (-1.7); IR (KBr) νmax 1709, 1655 cm-1; 1H and
13C NMR in Tables 1 and 2; HREIMS m/z 302.2257 (calcd for
C20H30O2, 302.2246).

138 Journal of Natural Products, 2010, Vol. 73, No. 2
Li et al.
with a force constant of 30 kcal mol-1 operating for deviations higher
than 20° from the trans form. Final structures were energy minimized
(EM) with 2000 steps of steepest descent followed by a conjugate
gradient method, down to a gradient norm value less than 10-3 kcal
mol-1 Å-1 (thus leading to the final SA/EM structure set), both with
the same energy setup used for SA. The MOLMOL22 program was
used for structural analysis and to obtain clusters of conformers by
superposition of heavy atoms, using a 0.5 Å cutoff to include structures
in the same cluster.
computational analysis details are available free of charge via the
Internet at http://pubs.acs.org.
References and Notes
(1) Kamel, H. N.; Slattery, M. Curr. Pharm. Biol. 2005, 43, 253–269.
(2) Zhang, W.; Guo, Y.-W.; Gu, Y.-C. Curr. Med. Chem. 2006, 13, 2041–
2090.
(3) Yan, X.-H.; Gavagnin, M.; Cimino, G.; Guo, Y.-W. Tetrahedron Lett.
2007, 48, 5313–5316.
(4) Yan, X.-H.; Lin, L.-P.; Ding, J.; Guo, Y.-W. Bioorg. Med. Chem. Lett.
2007, 17, 2661–2663.
6
The inverse of the sixth power (1/rij ) of each possible interprotonic
distance was computed either on single structures, or as an average
over the whole ensemble of 200 structures, or as a Boltzmann-weighted
average, at 300 K, of the representative structures of the selected
conformer clusters (see also Supporting Information). Distances involv-
ing one or more methyl groups were computed and stored as single
distances, by averaging all possible distances involving methyl H atoms
for each bundle structure. The inverse of the sixth root of these average
values provided the ensemble-averaged distances, directly related to
observed peak volumes. Experimental NMR-NOESY peaks were
integrated with the CCPNMR 1.0 rel. 15 program.23
By using reference peak volumes corresponding to fixed-distance
geminal interactions (-3.06 × 10-7 for the H-9a/H-9b geminal peak
of 7, distance 1.75 Å; -4.42 × 10-7 for H-14a/H-14b in 8, distance
1.76 Å), conversion factors were calculated for all peaks, to be used in
the translation of peak volumes into average experimental distances.
Antiproliferative Assays. The inhibition of cancer cell proliferation
by the compounds 5-12 was measured using the MTT assay.24 Cell
suspensions were plated in 96-well plates at a density of 1 × 104 cells
mL-1. The test compound solutions (10 µL in DMSO) at various
concentrations were added to each well, and the culture was incubated
for 72 h at 37 °C in a 5% CO2 incubator. After the treatment, 20 µL
of the MTT solution (5 mg/mL) was added to each well and cells were
incubated at 37 °C in 5% CO2. After incubation for 4 h the medium
was removed and 150 µL of DMSO was added. The absorbance was
measured at 570 nm. Dose-response curves were generated, and the
concentration of each compound required to inhibit cell proliferation
by 50% (IC50) was calculated from the linear portion of the log dose
response curves.
Antibacterial Assays. Antibacterial assays were performed by the
broth macrodilution method following the guidelines of the National
Committee for Clinical Laboratory Standards (NCCLS) document
M27-P (NCCLS, 1992)25 and of the Clinical and Laboratory Standard
Institute (CLSI) document M7-A7 (CLSI, 2007).26 The medium used
to prepare the 10× drug dilutions, and the inoculum suspension was
liquid LB (Luria-Bertani medium: 10 g/L Bactotryptone, 5 g/L
Bactoyeast, and 10 g/L NaCl, pH 7.5).27,28 The bacteria suspensions
were adjusted with the aid of a spectrophotometer to a cell density of
0.5 McFarland (2 × 108 cfu/mL) standard at 530 nm and diluted to
1:4000 (50 000 cfu/mL) in LB medium. The bacteria suspension (0.9
mL) was added to each test tube that contained 0.1 mL of eleven 2-fold
dilutions (512-0.05 µg/mL final) of each tested compound.
(5) Zhang, W.; Gavagnin, M.; Guo, Y.-W.; Mollo, E.; Ghiselin, M. T.;
Cimino, G. Tetrahedron 2007, 63, 4725–4729.
(6) Jia, R.; Guo, Y.-W.; Mollo, E.; Gavagnin, M.; Cimino, G. J. Nat.
Prod. 2006, 69, 819–822.
(7) Robinson, D. R.; West, C. A. Biochemistry 1970, 9, 70–79.
(8) Chio, Y.-H.; Kim, J.; Pezzuto, J. M.; Kinghorn, A. D.; Farnsworth,
N. R. Tetrahedron Lett. 1986, 48, 5795–5798.
(9) Xu, Z.-H.; Sun, J.; Xu, R.-S.; Qin, G.-W. Phytochemistry 1998, 49,
149–151.
(10) Zhang, C.-X.; Yan, S.-J.; Zhang, G.-W.; Lu, W.-G.; Su, J.-Y.; Zeng,
L.-M.; Gu, L.-Q.; Yang, X.-P.; Lian, Y.-J. J. Nat. Prod. 2005, 68,
1087–1089.
(11) Bretmaier, E.; Voelter, W. Carbon-13 NMR Spectroscopy; VCH: New
York, 1987; pp 192-196.
(12) Sullivan, G. R.; Dale, J. A.; Mosher, H. S. J. Org. Chem. 1973, 38,
2143–2147.
(13) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem.
Soc. 1991, 113, 4092–4096.
(14) Gavagnin, M.; Carbone, M.; Amodeo, P.; Mollo, E.; Vitale, R. M.;
Roussis, V.; Cimino, G. J. Org. Chem. 2007, 72, 5625–5630.
(15) The difference observed in the [R]D absolute values of natural and
synthetic 9 could be due to an impurity present in the natural sample.
(16) Gippert, G. P.; Yip, P. F.; Wright, P. E.; Case, D. A. Biochem.
Pharmacol. 1990, 40, 15–22.
(17) Case, D. A.; Darden, T. A.; Cheatham, T. E., III; Simmerling, C. L.;
Wang, J.; Duke, R. E.; Luo, R.; Crowley, M.; Ross, C.; Walker, W.;
Zhang, K. M.; Merz, B.; Wang, S.; Hayik, A.; Roitberg, G.; Seabra,
I.; Kolossva´ry, K. F.; Wong, F.; Paesani, J.; Vanicek, X. W.; Brozell,
S. R.; Steinbrecher, T.; Gohlke, H.; Yang, L.; Tan, C.; Mongan, J.;
Hornak, V.; Cui, G.; Mathews, D. H.; Seetin, M. G.; Sagui, C.; Babin,
V.; Kollman, P. A. AMBER 10; University of California, San
Francisco, 2008.
(18) Wang, J.; Wolf, R. M.; Caldwell, J. W.; Kollman, P. A.; Case, D. A.
J. Comput. Chem. 2004, 25, 1157–1174.
(19) Bayly, C. I.; Cieplak, P.; Cornell, W. D.; Kollman, P. A. J. Phys.
Chem. 1993, 97, 10269–10280.
(20) Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S. T.; Gordon,
M. S.; Jensen, J. H.; Koseki, S.; Matsunaga, N.; Nguyen, K. A.; Su,
S. J.; Windus, T. L.; Dupuis, M.; Montgomery, J. A. J. Comput. Chem.
1993, 14, 1347–1363.
(21) Hassinen, T.; Pera¨kyla¨, M. J. Comput. Chem. 2001, 22, 1229-1242.
(22) Koradi, R.; Billeter, M.; Wu¨thrich, K. J. Mol. Graph. 1996, 14, 51–
55.
(23) Vranken, W. F.; Boucher, W.; Stevens, T. J.; Fogh, R. H.; Pajon, A.;
Llinas, M.; Ulrich, E. L.; Markley, J. L.; Ionides, J.; Laue, E. D.
Proteins 2005, 59, 687–696.
(24) Cho, J. Y.; Park, J.; Yoo, E. S.; Baik, K. U.; Lee, J.; Park, M. H.
Planta Med. 1998, 64, 594–597.
(25) National Committee for Clinical Laboratory Standards (NCCLS). In
NCCLS document M27-P, NCCLS, Villanova, PA, 1992.
(26) Clinical and Laboratory Standard Institute (CLSI). In CLSI document
M7-A7, 7th ed.; Wayne, PA, 2007.
(27) Rodriguez-Tudela, J. L.; Berenguer, J.; Martinez-Suarez, L. V.;
Sanchez, R. Antimicrob. Agents Chemother. 1996, 40, 1998–2003.
(28) Hong, S. Y.; Oh, J. E.; Kwon, M. Y.; Choi, M. J.; Lee, B. L.; Moon,
H. M.; Lee, K. H. Antimicrob. Agents Chemother. 1998, 42,
2534–2541.
Acknowledgment. This research work was financially supported by
national S & T major project (20092X09301-001), national marine 863
project (2007AA09Z447), NSFC grants (Nos. 40976048, 30730108,
and 20721003), CAS key projects (grant KSX2-YW-R-18 and
SIMM0907KF-09), and PRIN-MIUR 2007 Project “Antitumor natural
products and synthetic analogues”. The authors acknowledge Prof. H.
Huang for identifying the material, Prof. C.-G. Huang for the cytotoxic
bioassay and Mrs. D. Melck for antibacterial assays.
Supporting Information Available: 1H and 13C NMR, 1H-1H
COSY, HSQC, HMBC, and CD spectra of compounds 5-14 and
NP900484K