Dolabellane-type diterpenoids with antiprotozoan activity from a southwestern caribbean gorgonian octocoral of the genus Eunicea

Article abstract of DOI:10.1021/np100074r

Ten new diterpenes, 1-10, having a dolabellane skeleton were isolated from a Colombian gorgonian coral of the genus Eunicea. Their structures, as well as those of known compounds 11-18, were determined on the basis of spectroscopic analysis and, in some instances, by chemical conversion and X-ray crystallographic analysis. The absolute structure of 7 was established by chemical conversion from 11, a co-occurring dolabellane congener of known absolute structure. Most of these diterpenoids showed antimalarial activity against the protozoan parasite Plasmodium falciparum.

Full text of DOI:10.1021/np100074r

J. Nat. Prod. 2010, 73, 925–934
925
Dolabellane-Type Diterpenoids with Antiprotozoan Activity from a Southwestern Caribbean
Gorgonian Octocoral of the Genus Eunicea
Xiaomei Wei, Abimael D. Rodr´ıguez,* Peter Baran, and Raphael G. Raptis
Department of Chemistry, UniVersity of Puerto Rico, P.O. Box 23346, U. P. R. Station, San Juan, Puerto Rico 00931-3346
ReceiVed February 4, 2010
Ten new diterpenes, 1-10, having a dolabellane skeleton were isolated from a Colombian gorgonian coral of the genus
Eunicea. Their structures, as well as those of known compounds 11-18, were determined on the basis of spectroscopic
analysis and, in some instances, by chemical conversion and X-ray crystallographic analysis. The absolute structure of
7 was established by chemical conversion from 11, a co-occurring dolabellane congener of known absolute structure.
Most of these diterpenoids showed antimalarial activity against the protozoan parasite Plasmodium falciparum.
Gorgonian octocorals belonging to the genus Eunicea, which can
be found in great abundance in waters of the Caribbean Sea, have
been recognized as rich sources for marine natural products, most
of which possess unique structural features and remarkable biologi-
cal activities.1,2 A variety of chemically complex diterpenoids, such
as cembrane lactones, dolabellanes, dolastanes, cubitanes, dilophols,
and fuscols, have been isolated since the early 1960s from several
species of Eunicea, including E. succinea, E. mammosa, E.
laciniata, E. palmeri, E. calyculata, E. asperula, E. fusca, and E.
tourneforti.1–10 As part of a study on the chemical constituents of
Caribbean gorgonians of this genus, we reported in 2004 the
isolation and structure characterization of five new representatives
of the cembrane class of diterpenes from an unknown species of
Eunicea collected near the Colombian Southwestern Caribbean
Sea.11 During our continuing studies on the chemical constituents
of this gorgonian species, 10 new dolabellane-type diterpenoids
(1-10) were isolated, as well as eight known dolabellane and
dolastane congeners (11-18). This paper describes the isolation
and structure determination of these compounds.12 The structures
of 1-18 were established by detailed spectroscopic analysis,
including extensive examination of 2D NMR (1H-1H COSY,
HMQC, HMBC, and NOESY) correlations, chemical conversion,
and, in the case of metabolites 1, 2, 8, and 12, X-ray crystallographic
analysis. Identification of known compounds was based on com-
parison of their physical properties (mp, [R]D, MS, and NMR) with
those reported in the literature, by direct comparison with authentic
samples, and, in the case of 12, by X-ray crystallography. It is to
be noted that, except for 7, only the relative configuration of the
new compounds has been determined. However, in light of the
established absolute structures of the co-occurring 11 and 16,
the configurations shown are considered more probable than the
enantiomeric ones.13,14 The antiprotozoan activity of 1-17 against
the Plasmodium falciparum W2 strain (chloroquine resistant) was
also studied. The results displayed that these compounds are gener-
ally toxic toward P. falciparum, with 9 being the most potent.
The HREIMS (m/z 302.2252) of 1 established the molecular
formula C20H30O2, appropriate for six degrees of unsaturation, and
more, the chemical shift for the C-13 carbonyl carbon in the 13C
NMR spectrum of 1 (206.6 ppm), the IR absorption at 1700 cm-1
,
and the UV absorption data were virtually identical with those
reported for known compounds 11-14, thus allowing the further
assignment of this chromophore as R-isopropylidenylcyclopen-
tanone. Signals appearing at δC 135.7 (C, C-4), δC 125.1 (CH, C-3),
δH 5.27 (dd, 1H, J ) 3.2, 11.8 Hz, H-3), and δH 1.63 (s, 3H, H3-
16) revealed the presence of one trisubstituted carbon-carbon
double bond. One R-methylene-R′-methyl-substituted ketone was
also identified from the NMR signals at δC 215.2 (C, C-7), 47.2
(CH, C-8), 40.3 (CH2, C-6), and 18.0 (CH3, C-17); δH 2.89 (dt,
1H, J ) 2.7, 14.0 Hz, H-6ꢀ), 2.66 (m, 1H, H-8), 2.38 (dq, 1H, J
) 2.1, 14.0 Hz, H-6R), and 1.05 (d, 3H, J ) 7.1 Hz, H3-17). In the
1H-1H COSY spectrum, it was possible to identify three different
structural units, which were assembled with the assistance of an
HMBC experiment (Figure 1). Key HMBC correlations of H3-15
to C-1, C-2, C-11, and C-14; H3-16 to C-5; H2-6 to C-7; H3-17 to
C-7, C-8, and C-9; and H-11 to C-1, C-10, C-12, C-13, C-14, and
C-18 permitted connection of the carbon skeleton. Furthermore,
the geminal methyl groups positioned at C-18 were confirmed from
the simultaneous HMBC correlations of H3-19 and H3-20 (δH 1.88
and 2.20, respectively) to the sp2 quaternary carbons C-18 (δC 147.9)
and C-12 (δC 137.6). On the basis of the above analysis, the planar
structure of 1 was established unambiguously.
The relative configurations of three asymmetric centers (C-1, C-8,
and C-11), as well as the configuration of the trisubstituted double
bond in 1, were determined by X-ray crystallographic analysis on
a single crystal of 1. The results of the X-ray analysis is shown in
Figure 2, disclosing the 1R*, 8R*, and 11S* relative configurations
for the asymmetric centers as well as the E configuration for the
double bond. Compound 1 was thus elucidated to be 7,13-diketo-
1R*,8R*,11S*-dolabell-3E,12(18)-diene, a derivative most likely
stemming from the acid-promoted epoxide rearrangement of 12,
the main compound isolated during this investigation.3
The molecular formula of C22H32O4 for 2 was demonstrated by
HREIMS and the 13C NMR spectrum. All 22 carbons appeared in
the 13C NMR spectrum, and DEPT experiments indicated the
presence of five methyls (including one acetate methyl), six sp3
methylenes, one sp2 methylene, three sp3 methines (including one
oxymethine), and one sp3 and six sp2 quaternary carbons (including
one ester and two ketone carbonyls). The 1H and 13C NMR spectra
(Tables 1 and 3) showed the presence of an R-isopropylidenecy-
clopentanone system [δC 205.9 (C, C-13), 150.4 (C, C18), 136.6
(C, C-12), 54.6 (CH2, C-14), 25.0 (CH3, C-19), 21.3 (CH3, C-20);
δH 2.37 (d, 1H, J ) 18.0 Hz, H-14R), 1.92 (d, 1H, J ) 18.0 Hz,
H-14ꢀ), 2.17 (s, 3H, H3-20), 1.69 (s, 3H, H3-19)], which was
demonstrated to be the same as that in 1 by the UV [λmax (log ε)
the IR spectrum revealed the presence of carbonyl (1700 cm-1
)
and olefin (1621 cm-1) groups. The 13C NMR (Table 3) and DEPT
spectroscopic data showed signals of five methyls, six sp3 meth-
ylenes, two sp3 methines, one sp2 methine, and one sp3 and five
sp2 quaternary carbons (including two ketone carbonyls). The NMR
signals (Tables 1 and 3) observed at δC 206.6 (C, C-13), 147.9 (C,
C-18), 137.6 (C, C-12), 24.3 (CH3, C-19), and 21.5 (CH3, C-20)
and δH 2.20 (s, 3H, H3-20) and 1.88 (s, 3H, H3-19) revealed the
presence of the R-isopropylidenyl ketone functionality.3 Further-
* To whom correspondence should be addressed. Tel: + 787-764-0000,
ext. 4799. Fax: + 787-756-8242. E-mail: abrodriguez@uprrp.edu.
1
203 (3.4), 255 (3.2) nm] and IR (1698, 1622 cm-1) spectra. H
10.1021/np100074r  2010 American Chemical Society and American Society of Pharmacognosy
Published on Web 04/12/2010

926 Journal of Natural Products, 2010, Vol. 73, No. 5
Wei et al.
Chart 1
and 13C NMR spectra also showed signals due to a terminal olefin
[δC 145.8 (C, C-4), 118.7 (CH2, C-16); δH 5.32 (br s, 1H, H-16R),
5.18 (br s, 1H, H-16ꢀ)] and due to an R-methylene-R′-methyl-
substituted ketone [δC 211.7 (C, C-7), 45.8 (CH, C-8), 37.2 (CH2,
C-6), 14.5 (CH3, C-17); δH 3.28 and 2.40 (each, m, 1H, H2-6), 2.67
(m, 1H, H-8), 1.08 (d, 3H, J ) 6.8 Hz, H3-17)]. The presence of
a secondary acetate group was demonstrated by the IR absorption
Compound 3 analyzed for C22H32O3, as revealed by the HREIMS
(m/z 344.2346) and 13C NMR data, a molecular formula possessing
one oxygen atom less in comparison to that of compound 2. Of
the seven degrees of unsaturation implied by the molecular formula,
five were accounted for by multiple bonds, three carbon-carbon
double bonds [δC 148.7 (C, C-4), 148.1 (C, C-18), 137.9 (C, C-12),
134.6 (C, C-8), 127.7 (CH, C-7), 115.0 (CH2, C-16)], a keto group
[δC 206.9 (C, C-13)], and an acetate function [δC 170.6 (C, C-21),
21.4 (CH3, C-22)], indicating 3 to be bicyclic. Comparison of all
spectroscopic data of 3 with those of 2, and the results of a 2D
1H-1H COSY experiment, revealed that the two molecules were
identical, except for the fact that the keto group at C-7 in 2 was
replaced by a carbon-carbon double bond in 3. On the basis of
the overall 2D NMR data it was apparent that the new olefin
functionality in compound 3 had to reside between carbons C-7
and C-8. From the NOESY spectrum, it was found that H3-15 (δ
1.30, s, 3H) showed an NOE interaction with H-3 (δ 5.39, dd, 1H,
J ) 2.2, 10.6 Hz), but not with H-11 (δ 3.28, br d, 1H, J ) 9.4
Hz), revealing the ꢀ-orientation of H-3. Furthermore, we concluded
that the acetate function had to be R, based on a 10.6 Hz coupling
between H-2R (δ 2.17) and H-3, as well as a 2.2 Hz coupling
1
(1735 cm-1), the low-field H signal at δ 5.25 (m, 1H) due to the
oxymethine proton H-3, and the 1H and 13C NMR data [δH 1.99 (s,
3H, H3-22); δC 170.7 (C, C-21), 70.9 (CH, C-3), 21.4 (CH3, C-22)].
These spectroscopic findings coupled with 1H-1H correlations
observed in the 1H-1H COSY spectrum gave several partial
structures, as shown in Figure 1. After assignments of the direct
13C-1H correlations were made on the basis of the HMQC analysis,
the HMBC spectrum of 2 was measured and analyzed to ob-
tain the gross structure of 2 (Figure 1). The relative configuration
of the four asymmetric centers (C-1, C-3, C-8, and C-11) was
determined by X-ray crystallographic analysis on a single crystal
of 2. The result of the X-ray analysis is shown in Figure 2,
demonstrating 1R*, 3R*, 8S*, and 11S* relative configurations for
the asymmetric centers.

Dolabellane-Type Diterpenoids from Eunicea
Journal of Natural Products, 2010, Vol. 73, No. 5 927
Table 1. 1H NMR (300 MHz) Assignments for Compounds 1-5 in CDCl3 [δH, mult (J in Hz)]a
H
1
2
3
4
5
2R
2ꢀ
3
5R
5ꢀ
6R
6ꢀ
7
2.06, m
2.06, m
1.35, m
5.25, m
2.35, m
3.04, m
2.40, m
3.28, m
2.17, m
1.28, m
5.39, dd (10.6, 2.2)
2.49, m
2.15, m
2.38, m
2.19, m
1.85, m
4.01, dd (5.1, 3.9)
2.13, m
2.03, m
2.28, m
2.20, m
1.30, m
5.48, dd (11.4, 1.5)
2.63, m
2.22, m
2.10, m
1.72, dd (13.1, 4.2)
5.27, dd (11.8, 3.2)
2.07, m
2.72, m
2.38, dq (14.0, 2.1)
2.89, dt (14.0, 2.7)
2.01, m
5.18, m
1.94, m
5.34, br s
1.82, m
3.00, d (8.1)
8
2.66, m
2.67, m
9R
9ꢀ
10R
10ꢀ
11
14R
14ꢀ
15
2.24, m
1.38, m
1.54, m
1.26, m
2.28, m
1.76, m
1.58, m
1.32, m
2.28, m
2.12, m
1.68, m
1.35, m
3.28, br d (9.4)
2.36, d (18.2)
2.00, d (18.2)
1.30, s
2.18, m
2.18, m
1.72, m
1.56, m
2.48, m
2.25, br s
2.25, br s
1.16, s
2.08, m
1.40, m
1.75, m
1.41, m
3.23, br d (12.2)
2.45, br d (18.2)
2.00, br d (18.2)
1.34, s
2.30, m
2.29, m
2.28, d (17.6)
2.11, d (17.6)
1.06, s
2.37, d (18.0)
1.92, d (18.0)
1.42, s
16R
16ꢀ
17
1.63, s
5.32, br s
5.18, br s
1.08, d (6.8)
1.69, s
5.18, br s
5.14, br s
1.62, s
5.27, br s
5.02, br s
1.54, s
5.34, br s
5.28, br s
1.31, s
1.05, d (7.1)
1.88, s
19
1.76, s
1.83, s
1.82, s
20
2.20, br s
2.17, s
2.21, s
2.17, s
2.24, s
22
1.99, s
2.00, s
2.01, s
a Chemical shift values are in ppm relative to TMS. Spectra were recorded at 25 °C. Proton assignments were aided by 1Hs1H COSY, HMQC,
HMBC, and NOESY experiments.
Table 2. 1H NMR (300 MHz) Assignments for Compounds 6-10 in CDCl3 [δH, mult (J in Hz)]a
H
6
7
8
9
10
2R
2ꢀ
3
5R
5ꢀ
6R
6ꢀ
7
9R
9ꢀ
10R
10ꢀ
11
13
14R
14ꢀ
15
16
17
18
19
20
2.20, m
1.97, m
5.09, dd (11.6, 4.5)
2.33, m
2.22, m
1.92, m
1.59, m
4.13, dd (10.7, 2.6)
5.41, t (7.8)
1.77, m
1.37, m
3.06, dd (11.2, 2.9)
2.27, m
1.42, m
1.76, m
1.47, m
2.96, br d (7.7)
2.12, m
1.43, m
1.78, m
2.51, m
2.34, m
4.68, br d (11.7)
1.89, m
2.53, m
1.48, m
2.03, m
2.93, br d (10.3)
2.15, m
1.42, m
1.94, m
2.47, br d (14.5)
2.18, dd (14.6, 12.3)
4.66, br d (12.2)
1.78, m
2.30, m
1.83, m
1.83, m
4.02, dd (10.0, 5.2)
5.68, t (7.4)
2.42, dd (11.8, 11.6)
1.85, m
5.08, br d (11.3)
1.79, m
2.27, m
1.83, m
1.83, m
4.03, dd (10.0, 5.1)
5.66, t (7.4)
2.40, m
1.78, m
2.77, br d (11.8)
2.62, m
1.95, m
2.98, dt (10.2, 2.6)
5.83, br s
2.80, m
1.82, m
3.01, br d (11.1)
1.57, m
2.81, br d (11.4)
1.53, m
2.98, br d (11.0)
5.97, br s
2.37, d (16.3)
2.10, d (16.3)
1.08, s
1.62, s
1.68, s
2.44, d (18.4)
2.13, d (18.4)
1.33, s
1.21, s
1.43, s
2.53, d (15.7)
1.81, d (15.7)
1.01, s
1.75, s
1.62, s
1.04, s
1.64, s
1.30, s
2.62, m
1.13, d (6.8)
1.20, d (6.8)
1.06, s
1.64, s
1.58, s
2.62, m
1.18, d (6.3)
1.20, d (6.7)
1.96, s
2.14, s
1.94, s
2.24, s
1.98, br s
2.20, br s
a Chemical shift values are in ppm relative to TMS. Spectra were recorded at 25 °C. Proton assignments were aided by 1Hs1H COSY, HMQC,
HMBC, and NOESY experiments.
between H-2ꢀ (δ 1.28) and H-3. The E geometry of the trisubsti-
tuted double bond was also assigned from the NOE correlation of
H3-17 (δ 1.62, s, 3H) with H-6R (δ 2.38), but not with olefinic
proton H-7 (δ 5.18, m, 1H), and also the upper field chemical shift
of C-17 (δ 16.8). On the basis of the above findings and detailed
examination of other NOE correlations (Figure 3), we concluded
that the relative structure of compound 3 is 3R*-acetoxy-13-keto-
1R*,11S*-dolabell-4(16),7E,12(18)-triene.
Compound 4 was shown by HREIMS to possess the molecular
formula C20H30O2 (m/z 302.2242). The IR spectrum of 4 revealed
the presence of hydroxy (3434 cm-1), ketone carbonyl (1701 cm-1),
and olefin (1620 cm-1) groups. Comparison of the 1H and 13C NMR
data (Tables 1 and 3) of compounds 3 and 4 showed that the
structure of 4 should be very close to that of 3, with the exception
of signals assigned to C-3, where an acetate methyl (δ 2.00) and
acetoxymethine (δ 5.39, dd, 1H, J ) 2.2, 10.6 Hz) in 3 are replaced
by a hydroxymethine (δ 4.01, dd, 1H, J ) 3.9, 5.1 Hz) in 4. The
overall planar structure of 4 was fully established by analyzing the
1H-1H COSY and HMBC correlations (Figure 1). The relative
configuration of 4 was confirmed to be 1R*, 3R*, and 11S* from
the following NOESY correlations: both H-3 (δ 4.01) and H3-15
(δ 1.16) with H-7 (δ 5.34), and H3-15 with H-3. The ∆7,8 double
bond was assigned with the E configuration on the basis of the 13
C
NMR chemical shift of C-17 (δC 15.9), which is in the same range
as that of compound 3. These results, together with other detailed
2D NMR correlations of 4 (Figure 1), unambiguously established
the structure of this secondary metabolite as shown in formula 4.
The structural relationship of compounds 3 and 4 was decisively
demonstrated when 4, upon acetylation at room temperature, yielded
acetate 3 quantitatively. Thus, 4 is the 3-deacetyl derivative of 3.
Compound 5 possessed the same molecular formula (C22H32O4)
as that of compound 2, as revealed from HREIMS, suggesting 5 to
be an isomer of 2. Furthermore, it was found that the NMR
spectroscopic data of 5 were very similar to those of 2. The four
oxygen atoms in 5 were readily assigned to an acetate, a trisub-
stituted epoxide, and an R,ꢀ-unsaturated ketone constellation by

928 Journal of Natural Products, 2010, Vol. 73, No. 5
Wei et al.
Table 3. 13C NMR (75 MHz) Assignments for Compounds 1-10 in CDCl3 [δC, mult]a
C
1
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
7
8
40.5, C
38.1, C
38.0, C
40.4, C
38.3, C
43.5, CH2
75.0, CH
148.3, C
33.6, CH2
29.3, CH2
64.4, CH
60.7, C
36.3, CH2
28.0, CH2
42.1, CH
137.4, C
206.2, C
54.8, CH2
23.2, CH3
117.4, CH2
17.3, CH3
148.6, C
25.1, CH3
21.3, CH3
170.5, C
21.4, CH3
39.3, C
37.9, C
40.3, CH2
63.0, CH
60.3, C
37.3, CH2
23.4, CH2
63.8, CH
60.6, C
36.7, CH2
27.9, CH2
43.0, CH
137.1, C
205.4, C
54.8, CH2
23.6, CH3
15.6, CH3
17.5, CH3
149.5, C
25.2, CH3
21.7, CH3
50.8, C
53.5, C
39.0, C
40.3, CH2
125.1, CH
135.7, C
35.6, CH2
40.3, CH2
215.2, C
47.2, CH
30.9, CH2
28.6, CH2
44.4, CH
137.6, C
45.3, CH2
70.9, CH
145.8, C
30.7, CH2
37.2, CH2
211.7, C
45.8, CH
31.9, CH2
27.0, CH2
43.2, CH
136.6, C
43.8, CH2
74.4, CH
148.7, C
35.1, CH2
28.5, CH2
127.7, CH
134.6, C
37.9, CH2
27.5, CH2
42.4, CH
137.9, C
41.0, CH2
74.8, CH
153.9, C
34.5, CH2
34.4, CH2
123.5, CH
136.7, C
37.9, CH2
30.2, CH2
44.0, CH
137.6, C
42.8, CH2
125.1, CH
133.8, C
38.5, CH2
28.0, CH2
80.6, CH
137.5, C
129.3, CH
30.0, CH2
47.9, CH
137.0, C
36.4, CH2
122.4, CH
136.7, C
28.9, CH2
26.1, CH2
67.8, CH
60.6, C
35.9, CH2
23.6, CH2
45.5, CH
190.4, C
125.6, CH
213.4, C
21.1, CH3
23.9, CH3
17.5, CH3
29.0, CH
22.5, CH3
21.2, CH3
33.8, CH2
122.9, CH
134.8, C
27.1, CH2
30.6, CH2
80.6, CH
138.8, C
124.0, CH
23.6, CH2
46.7, CH
187.2, C
123.6, CH
212.4, C
24.2, CH3
22.9, CH3
10.1, CH3
29.4, CH
21.7, CH3
20.7, CH3
37.4, CH2
121.6, CH
135.0, C
27.4, CH2
27.6, CH2
80.6, CH
138.2, C
125.4, CH
30.4, CH2
43.2, CH
134.5, C
9
10
11
12
13
14
15
16
17
18
19
20
21
22
206.6, C
205.9, C
206.9, C
206.3, C
205.5, C
205.5, C
55.3, CH2
21.3, CH3
15.6, CH3
18.0, CH3
147.9, C
54.6, CH2
23.0, CH3
118.7, CH2
14.5, CH3
150.4, C
25.0, CH3
21.3, CH3
170.7, C
55.7, CH2
23.0, CH3
115.0, CH2
16.8, CH3
148.1, C
24.7, CH3
21.3, CH3
170.6, C
56.0, CH2
22.1, CH3
110.7, CH2
15.9, CH3
147.8, C
57.5, CH2
23.1, CH3
15.5, CH3
11.2, CH3
145.9, C
52.8, CH2
25.0, CH3
23.3, CH3
10.6, CH3
148.4, C
24.3, CH3
21.5, CH3
24.1, CH3
21.6, CH3
23.8, CH3
21.4, CH3
24.7, CH3
22.5, CH3
21.4, CH3
21.4, CH3
a Chemical shift values are in ppm relative to TMS. Spectra were recorded at 25 °C. Atom multiplicities were obtained from DEPT NMR
1
experiments. Assignments were aided by H-1H COSY, HMQC, HMBC, and NOESY experiments.
C-8 (δC 60.7, C) and H2-9 (δH 2.08 and 1.40), as well as from C-7
(δC 64.4, CH) to H2-6 (δH 2.10 and 1.82), which allowed the
epoxide group to be unambiguously positioned across C-7 and C-8.
The relative configuration of 5 at C-1, C-3, and C-11 was identical
to that of 2 on the basis of comparable NOE interactions (Figure
3) and interproton coupling patterns. For C-7 and C-8 it was
concluded that the epoxide function had to possess the 7R*, 8R*
relative configuration, on the basis of a 8.1 Hz coupling between
H-7 (δ 3.00) and H-6R (δ 2.10), as well as strong NOE effects
observed between H3-15 (δ 1.34), H-3 (δ 5.48), and H-7 (δ 3.00)
and between H-11 (δ 3.23) and H3-17 (δ 1.31). Thus, compound
5, a likely precursor to 2 upon acid-promoted epoxide rearrange-
ment, is 3R*-acetoxy-7R*,8R*-epoxy-13-keto-1R*,11S*-dolabell-
4(16),12(18)-diene.
Mass spectrometry and NMR data revealed compound 6 to have
a molecular formula of C20H30O3. The presence of seven sp2-
hybridized carbon atoms in the molecule, as deduced from the 13
C
and DEPT NMR spectra, being for three carbon-carbon double
bonds and one carbon-oxygen bond as the only multiple bonds,
indicated compound 6 to be bicyclic. The IR spectrum showed the
presence of hydroxy and R,ꢀ-unsaturated ketone functionalities (νmax
3397, 1704 cm-1). The 1H and 13C NMR spectra (Tables 2 and 3)
resembled those of known dolabellatrienone (11).3 However, a
secondary hydroperoxy [δH 4.13 (dd, 1H, J ) 2.6, 10.7 Hz, H-7);
δC 80.6 (CH, C-7)] adjacent to a methyl bearing an E trisubstituted
double bond [δH 5.41 (t, 1H, J ) 7.8 Hz, H-9), 1.68 (s, 3H, H3-
17); δC 137.5 (C, C-8), 129.3 (CH, C-9), 11.2 (CH3, C-17)] in 6
replaced the E trisubstituted ∆7,8 double bond in 11. The presence
of a secondary hydroperoxy group in 6 was confirmed by the EIMS
data, which showed pronounced fragment ions at m/z 301 and 284
through elimination of HO · and H2O2, respectively, from the
molecular ion of 6 at m/z 318. 1H-1H COSY cross-peaks between
H-7 (δ 4.13) and H2-6 (δ 1.92 and 1.59) and between H-9 (δ 5.41)
and H2-10 (δ 2.40 and 1.78), as well as HMBC correlations between
H-7 and C-6, C-8, C-9, C-17; H-9 and C-7, C-8, C-10, C-11, and
C-17 (Figure 1), positioned the secondary hydroperoxy group at
C-7 and the nearby trisubstituted double bond between C-8 and
C-9. In the NOESY spectrum (Figure 4), NOEs between H-7 and
H-9, between H-7 and H3-15, between H-9 and H3-15, and between
H-11 and H3-17, which indicated that the 11-membered ring in 6
adopts a crown conformation, allowed the configuration of the
hydroperoxy and ∆8,9 moieties to be assigned as 7R* and E,
respectively. Correspondingly, the triene 6 was fully characterized
1
Figure 1. H-1H COSY and HMBC correlations for compounds
1-10.
evaluation of spectral information. 13C NMR bands (Table 3) at δ
64.4 (CH, C-7) and 60.7 (C, C-8) in conjunction with H NMR
1
resonances at δ 3.00 (d, 1H, J ) 8.1 Hz, H-7) and 1.31 (s, 3H,
H3-17) illustrated 5 to possess a methyl-substituted epoxide group.
By 2D NMR spectroscopic data, including 1H-1H COSY, HMQC,
and HMBC, compound 5 was shown to possess many of the same
structural features as those of 2. Comparison of the 13C NMR data
for the two compounds revealed that the major differences between
them lay in the vicinity of the C-7,8 junction (see Table 3). On the
basis of H-1H couplings observed in the H NMR spectrum and
overall HMBC spectral data of 5 (Figure 1) it was apparent that
the trisubstituted epoxide group had to reside at the C-7,8 position.
The HMBC spectrum of 5 showed heteronuclear couplings between
1
1

Dolabellane-Type Diterpenoids from Eunicea
Journal of Natural Products, 2010, Vol. 73, No. 5 929
Figure 2. Molecular structures of compounds 1, 2, 8, and 12, which define only the relative configuration, with atom-labeling scheme. The
carbon and oxygen atoms are drawn as 30% thermal ellipsoids.
Figure 3. Key NOESY correlations for 3 and 5.
as 7R*-hydroperoxy-13-keto-1R*,11S*-dolabell-3E,8E,12(18)-
triene. The co-occurrence of 6 and 11 within the same Eunicea sp.
extract suggests that 6 might be formed from 11 during the isolation
process upon air oxidation of the double bond at C-7 of 11.
Compound 7 was isolated as a colorless oil. HREIMS, 13C NMR,
and DEPT spectra established the molecular formula of 7 as
C20H30O3. The IR and UV absorptions clearly defined the analogous
R-isopropylidenylcyclopentanone group as in 1-6. The NMR data
of compound 7 were strongly reminiscent of those of known
epoxide 12, apart from the replacement of the NMR bands of the
trisubstituted olefin [δH 5.43 (dd, 1H, J ) 4.7, 11.6 Hz, H-3), 1.57
(br s, 3H, H3-16); δC 124.7 (CH, C-3), 135.9 (C, C-4), 15.7 (CH3,
C-16)]3 by those of an additional epoxide [δH 3.06 (dd, 1H, J )
2.9, 11.2 Hz, H-3), 1.21 (s, 3H, H3-16); δC 63.0 (CH, C-3), 60.3
(C, C-4), 15.6 (CH3, C-16)]. On the basis of this very favorable
comparison, dolabellanone 7 was formulated as the C-3,4 epoxy
derivative of 12. The proposed structural type and substitution
pattern of 7 were confirmed by means of 2D NMR correlated
spectroscopy including HMQC, HMBC, and 1H-1H COSY (Figure
1). The relative structure of diepoxide 7 was elucidated by the
analysis of NOE correlations, as shown in Figure 4. It was found
that epoxymethine protons H-3 (δ 3.06) and H-7 (δ 2.96) each
showed NOE interactions with H3-15 (δ 1.33); therefore, assuming
a ꢀ-orientation of H3-15, H-3 and H-7 should also be positioned
on the ꢀ-face. NOE correlations observed between H-11 (δ 2.81)
and H3-16 (δ 1.21), and between H-11 and H3-17 (δ 1.43), reflected
the R-orientations of H3-16 and H3-17. Furthermore, H-3 exhibited
an interaction with H-7, and H3-19 (δ 1.94) exhibited a NOE

930 Journal of Natural Products, 2010, Vol. 73, No. 5
Wei et al.
Figure 4. Key NOESY correlations for 6 and 7.
Figure 5. Key NOESY correlations for 9 and 10.
interaction with H3-17, all of which pointed toward identical R*
relative configurations at C-3, C-4, C-7, and C-8. To confirm this
contention, a solution of known dolabellatrienone (11) in CH2Cl2
was treated with m-CPBA at 25 °C to find a 20% conversion of 11
to 7, whose NMR data and the [R]D value were identical to those
of 7 obtained by the above-mentioned isolation. Because the
absolute configuration of 11 was previously established by syn-
thesis,13 the absolute configuration of 7 was determined by
correlation. Therefore, 7 was found to be 3R,4R,7R,8R-diepoxy-
13-keto-1R,11S-dolabell-12(18)-ene.
must be stereochemical. The relative configuration, 1R*, 7R*, 8R*,
11S*, as well as the Z configuration of the ∆3,4 trisubstituted double
bond in 8 were determined by X-ray crystallographic analysis on
a single crystal of 8 (Figure 2). Compound 8 was thus elucidated
to have a structure corresponding to the ∆3,4 Z-isomer of 15.
The molecular formula of C20H30O3 for 9 was determined by
HRESIMS [found m/z 319.2259 (M + 1)+; calcd for C20H31O3
319.2273]. All 20 carbons appeared in the 13C NMR spectrum, and
DEPT experiments indicated the presence of five methyls, four sp3
methylenes, three sp3 methines, three sp2 methines, one sp3
quaternary carbon, and four sp2 quaternary carbons (Table 3). The
presence of a ꢀ-isopropyl-R,ꢀ-cyclopentenone system, the same as
that in compound 8, was demonstrated by UV, IR, 13C NMR (Table
The molecular formula of compound 8 was established as
1
C20H30O2 from its HREIMS data. The H and 13C NMR spectra
suggested the same dolabellane diterpenoid skeleton of 1-7 with
signals for an epoxy functionality, five methyl groups, two
trisubstituted double bonds, and one carbonyl group [δ 213.4 (C,
3), and H NMR (Table 2) data. The H and 13C NMR spectra
showed signals due to two methyl-bearing trisubstituted olefins [δC
134.8 (C, C-4), 122.9 (CH, C-3), 22.9 (CH3, C-16); δH 4.66 (br d,
1H, J ) 12.2 Hz, H-3), 1.64 (s, 3H, H3-16)] and [δC 138.8 (C,
C-8), 124.0 (CH, C-9), 10.1 (CH3, C-17); δH 5.68 (t, 1H, J ) 7.4
Hz, H-9), 1.58 (s, 3H, H3-17)] and due to a secondary hydroperoxide
group [δC 80.6 (CH, C-7); δH 4.02 (dd, 1H, J ) 5.2, 10.0 Hz, H-7)].
These spectroscopic findings coupled with proton-proton and long-
range 13C-1H correlations observed in the 1H-1H COSY and
HMBC spectra gave a series of partial structures, as shown in Figure
1, to give a gross structure for 9. In the NOESY experiment (Figure
5), the secondary hydroperoxide at C-7 was assigned as R on the
basis of NOEs between H-7 and H-9, H-5ꢀ; between H-11 and
H-3, H3-17; between H3-15 and H-9, H-10ꢀ; and between H-2ꢀ
and H-9, H3-15. The geometries of the trisubstituted ∆3,4 and ∆8,9
double bonds were certified to be Z and E, respectively, on the
basis of NOE experiments and the 13C chemical shift values of the
corresponding vinyl methyls. Compound 9 is thus 7R*-hydroperoxy-
14-keto-1S*,11S*-dolabell-3Z,8E,12Z-triene.
1
1
1
C-14)]. While the H NMR spectrum of 8 (Table 2) displayed an
unusually shielded broad doublet at δ 4.68 (1H, J ) 11.7 Hz) for
H-3 (cf. δ 5.27 for H-3 of 1), it contained an additional one-proton
vinyl singlet at δ 5.97 and an additional one-proton multiplet at δ
2.62. The usual methyl signals near δ 2.20 and 1.90, ascribable to
the vinyl methyl groups of the R-isopropylidenylcyclopentanone
moiety of 1-7 (H3-19 and H3-20), were conspicuously absent in
the 1H NMR spectrum of 8 and showed in their place two doublets
at δ 1.13 and 1.20 (each, 3H, J ) 6.8 Hz). Coupling was observed
1
in the H-1H COSY spectrum between these methyl signals and
the multiplet at δ 2.62, indicating the presence of an isopropyl group
at C-12 of the dolabellane skeleton. The 13C NMR spectrum (Table
3) showed, in addition to the carbonyl absorption at δ 213.4, sp2
carbon signals at δ 190.4 (C, C-12), 136.7 (C, C-4), 125.6 (CH,
C-13), and 122.4 (CH, C-3). All of these features were suggestive
of a structure isomeric with that of known compound 15. Further
analysis of the 2D NMR (1H-1H COSY, HMQC, and HMBC)
correlations showed compound 8 to possess the same planar
structure as that of 15 (Figure 1); thus the differences between them
Compound 10 was shown by HREIMS to possess the same
molecular formula (C20H30O3) as those of 6 and 9. Furthermore, it

Dolabellane-Type Diterpenoids from Eunicea
Journal of Natural Products, 2010, Vol. 73, No. 5 931
Providence Island (March 2002), Colombia, located off the Nicaraguan
shelf in the southwestern Caribbean Sea. A voucher specimen (No.
Eunicea sp. 2) is on deposit at the Chemistry Department of the
University of Puerto Rico. The gorgonian specimens were partially air-
dried, freeze-dried, and then kept frozen prior to extraction. The dried
animal (1.7 kg) was blended with a mixture of 1:1 MeOH/CH2Cl2 (5
× 4 L), and after filtration, the combined extracts were concentrated
in Vacuo to afford a gummy, green residue (74 g). The extract was
suspended in H2O (1 L) and extracted successively with hexane (4 ×
2 L), CH2Cl2 (4 × 2 L), and EtOAc (3 × 2 L). The hexane extract (46
g) was chromatographed on a large Si gel column by stepwise elution
with 100% hexane, hexane/EtOAc mixtures (100-0%), and then 100%
MeOH. Fractions were pooled on the basis of their TLC and NMR
profiles to yield 15 primary fractions, denoted I-XV. Purification of
fraction III (1.5 g) by Si gel CC with 100:1 hexane/EtOAc led to the
isolation of known compounds dolabellatrienone (11, 7.6 mg, 0.02%
yield), epoxide 12 (800 mg, 1.8% yield), and palominol (16, 5.3 mg,
0.01% yield).3–5 A solution of fraction VII (420 mg) in toluene was
filtered under vacuum and purified by size-exclusion chromatography
on a Bio-Beads SX-3 (90 g) column with toluene as eluant. The
combined portions (NMR and TLC guided) were concentrated to yield
four subfractions, designated as A (80 mg), B (139 mg), C (155 mg),
and D (40 mg). Subfraction C was purified successively by Si gel (18
g) CC with 80:1 hexane/acetone and reversed-phase HPLC (Ultrasphere
ODS column with 65:35 MeOH/H2O) to afford pure compound 1 (33.0
mg, 0.07% yield) as a white crystalline solid. Plate crystals of
dolabellanone 1 suitable for X-ray crystallographic analysis were
obtained upon recrystallization from a 90:10 MeOH/H2O mixture.
Purification of fraction IX (0.6 g) by size-exclusion chromatography
on a Bio-Beads SX-3 column (toluene) led to four subfractions. The
penultimate subfraction was eluted through a Si gel column (100:1
hexane/acetone), and then the mixture obtained was separated by
successive HPLC analyses (Ultrasphere Si gel column with 95:5 hexane/
2-propanol followed by a Cyano column with 97:3 hexane/2-propanol),
affording pure compounds 3 (5 mg, 0.01% yield) and 10 (4 mg, 0.009%
yield). Fraction X (0.6 g) was chromatographed directly over Si gel
(30 g) and separated by stepwise elution with hexane/acetone (2-3%)
to yield three fractions (A, B, and C). Fraction B (0.1 g) was purified
successively by Si gel CC with 125:1 hexane/2-propanol and then by
HPLC (Cyano column with 95:5 hexane/2-propanol) to afford pure
compound 6 (15 mg, 0.03% yield) along with the known compounds
12 (10 mg, 0.02% yield), 17 (1 mg, 0.002% yield), and 18 (1 mg,
0.002% yield).3 As crystals of suitable quality were obtained for
previously reported compound 12, a single-crystal X-ray analysis was
undertaken (Figure 2).17 Fraction XI (1.2 g) was purified by size
exclusion chromatography (Bio-Beads SX-3, toluene) to yield five
fractions, denoted A-E. Fraction D (0.5 g) was fractionated succes-
sively by normal-phase HPLC (Si gel column with 95:5 hexane/2-
propanol), polar-bonded HPLC (Cyano column with 95:5 hexane/2-
propanol), and reversed-phase HPLC (Ultrasphere ODS column with
75:25 MeOH/H2O). The main fraction obtained (100 mg) was loaded
onto a Si gel column and separated by stepwise elution with hexane/
acetone (2-100%) to afford pure dolabellanones 8 (14 mg, 0.03% yield)
and 9 (28 mg, 0.06% yield). Needle crystals of dolabellanone 8 suitable
for X-ray crystallographic analysis were subsequently obtained upon
recrystallization from a 80:20 MeOH/CHCl3 mixture. Size exclusion
CC (Bio-Beads SX-3 with toluene) of fraction XII (614 mg) followed
by successive reversed-phase HPLC (Ultrasphere ODS column with
75:25 MeOH/H2O) and Si gel CC (with 50:1 hexane/acetone) yielded
known dolabellane 15 (3.4 mg, 0.007% yield).8 Fraction XIII (900 mg)
was purified upon elution through a Bio-Beads SX-3 column with
toluene. The penultimate subfraction (564 mg) was subsequently
purified by Si gel CC and normal-phase HPLC (Cyano column with
97:3 hexane/2-propanol) to yield pure compound 2 (8 mg, 0.02% yield)
and known dolabellane 14 (3.6 mg, 0.008% yield).8 Recrystallization
of dolabellanone 2 from 90:10 MeOH/CHCl3 afforded white cubic
crystals. After fraction XIV (1.5 g) was fractionated by size-exclusion
chromatography on a Bio-Beads SX-3 column (toluene), the penultimate
fraction eluted (0.5 g) was purified by Si gel CC and separated by
stepwise elution with hexane/acetone mixtures (5-100%). The second
subfraction was purified by HPLC (Cyano column with 97:3 hexane/
2-propanol) to yield pure dolabellanone 5 (13 mg, 0.03% yield). The
third subfraction was purified by HPLC (Cyano column with 95:5
hexane/2-propanol) to afford dolabellanone 7 (15 mg, 0.03% yield)
along with known compound 13 (10 mg, 0.02% yield),8 and the last
was found that the NMR spectroscopic data of 10 were very similar
to those of 6, suggesting that these compounds possess the same
planar structure. Comparison of the 1H and 13C NMR data (Tables
2 and 3) of compounds 6 and 10 showed that the structure of 10
should be very close to that of 6, with the exception of signals
assigned to the methyl bearing E-trisubstituted olefin ∆3,4 in 6,
which were replaced by resonances for a Z-trisubstituted olefin [δH
5.08 (br d, 1H, J ) 11.3 Hz, H-3), 1.75 (s, 3H, H3-16); δC 135.0
(C, C-4), 121.6 (CH, C-3), 23.3 (CH3, C-16)] in 10. The planar
structure of 10 was fully established by analyzing the 1H-1H COSY
and HMBC correlations (Figure 1), and its relative configuration
was confirmed to be 1R*, 7R*, and 11S* from the following
NOESY correlations: both H-5ꢀ (δ 2.27) and H-9 (δ 5.66) with
H-7 (δ 4.03), both H-11 (δ 3.01) and H-14R (δ 2.53) with H-3 (δ
5.08), and H3-16 (δ 1.75) with H-3. These results, together with
other detailed NOE correlations of 10 (Figure 5), unambiguously
established that compound 10 is 7R*-hydroperoxy-13-keto-1R*,11S*-
dolabell-3Z,8E,12(18)-triene.
The spread of drug resistance in Plasmodium falciparum has
made it essential to look into new effective chemotherapeutic agents
that possess antimalarial activity with favorable pharmacokinetic
properties.15 Thus, Eunicea sp. dolabellanes 1-16 as well as
dolastane 17 were tested for their inhibitory activity toward the
growth of P. falciparum W2 (chloroquine-resistant). Interestingly,
most of the compounds were active with IC50 values ranging from
9.4–59.6 µM. The most active metabolite was 9 (IC50 ) 9.4 µM),
and the least active compound was epoxide 8 (IC50 g 100 µM).16
Some of the present isolates (1, 3, 6, 8, 9, 11, 12, 16, 17, and 18)
were also tested for their inhibitory activity toward the growth of
Mycobacterium tuberculosis, but were deemed inactive (% of
inhibition at 6.25 µg/mL e 50%). Furthermore, dolabellanes 1, 7,
11, and 12 were evaluated in a three-cell-line panel consisting of
MCF-7 breast cancer, NCI-H460 non-small-cell lung cancer, and
SF-268 (CNS). However, results from the one-dose primary
anticancer assay showed a lack of significant cytotoxicity when
compared to the untreated control cells. Likewise, at a concentration
of 30 mM, compounds 1, 6, 9, 11, and 16 were not active against
the hepatitis B virus (HBV).
Experimental Section
General Experimental Procedures. Optical rotations were recorded
with a Rudolph Autopol IV polarimeter and melting point determina-
tions with an electrothermal IA 900 digital apparatus. The UV spectra
were recorded with a Shimadzu UV-2401PC spectrometer, and the IR
analyses were performed with a Nicolet Magna IR 750FT-IR spec-
trometer. 1D and 2D NMR data were recorded on a Bruker DRX-300
spectrometer. 1H NMR and 13C NMR chemical shifts were recorded at
25 °C with respect to TMS. Mass spectrometric measurements were
generated at the Mass Spectrometry Laboratory of the University of
Illinois at Urbana-Champaign. Column chromatography was performed
using Si gel (35-75 mesh); TLC analysis was carried out using glass
precoated Si gel plates, and the spots were visualized using a UV lamp
at λ ) 254 nm or by exposure to I2 vapor. HPLC was performed using
either an Ultrasphere polar-bonded Cyano semipreparative column (5
µm, 10 mm × 25 cm), an Ultrasphere normal-phase Si gel semi-
preparative column (5 µm, 10 mm × 25 cm), or an Ultrasphere ODS
reversed-phase Si gel semipreparative column (5 µm, 10 mm × 25
cm). All HPLC separations were monitored simultaneously with a
refractive index detector and a UV detector set at λ ) 220 nm using a
flow rate of 2 mL/min with isocratic elution of the mobile phase. All
solvents were distilled from glass prior to use. The percentage yield of
each compound is based on the weight of the hexane extract.
Animal Material. A detailed taxonomical description of the
biological specimens used in this investigation has been disclosed
previously.11 A photograph of the gorgonian coral voucher specimen
is available as Supporting Information.
Collection, Extraction, and Isolation. Medium to large colonies
(0.5-1.3 m) of the gorgonian coral Eunicea sp. (undescribed species;
order: Gorgonacea; family: Gorgoniidae, phylum: Cnidaria) were
collected by scuba at depths of 23-26 m from the coral reefs off Old

932 Journal of Natural Products, 2010, Vol. 73, No. 5
Wei et al.
subfraction was purified by Si gel CC with 20:1 hexane/acetone
followed by HPLC (Cyano column with 93:7 hexane/2-propanol),
yielding dolabellanone 4 (8 mg, 0.02% yield).
respectively; EIMS m/z [M]+ 318 (1), 302 (2), 301(1), 284 (1), 269
(2), 241 (2), 203 (2), 175 (3), 161 (5), 150 (37), 135 (11), 121 (6), 85
(72), 83 (100); HRESIMS m/z [M + 1]+ 319.2259 (calcd for C20H31O3,
319.2273).
Dolabellanone 1: white crystalline solid; mp 128.5-128.9 °C; [R]25
D
+41.2 (c 1.0, CHCl3); UV (MeOH) λmax (log ε) 203 (3.1), 254 (3.7)
nm; IR (film) νmax 2975, 2957, 2929, 2873, 1700, 1621, 1459, 1406,
Dolabellanone 10: colorless oil, [R]25 -15.0 (c 1.0, CHCl3); UV
D
(MeOH) λmax (log ε) 202 (3.3), 238 (3.1) nm; IR (film) νmax 3426, 2965,
2935, 2874, 1712, 1621, 1452, 1377, 1242, 1038, 756 cm-1; 1H NMR
(300 MHz, CDCl3) and 13C NMR (75 MHz, CDCl3), see Tables 2 and
3, respectively; EIMS m/z [M]+ 318 (4), 302 (18), 301 (5), 284 (12),
269 (7), 189 (22), 150 (84), 136 (45), 121 (35), 107 (46), 95 (39), 83
(100), 69 (44), 55 (63); HREIMS m/z [M - H2O2]+ 284.2131 (calcd
for C20H28O, 284.2140).
Acetylation of Dolabellanone 4. Compound 4 (3.0 mg, 0.01 mmol)
was dissolved in a mixture of dry pyridine (500 µL) and acetic
anhydride (500 µL) and stirred at rt for 12 h. The reaction mixture
was concentrated in Vacuo, and the oily residue obtained was purified
by CC over Si gel (1 g) using 5% EtOAc in hexane to yield a
homogeneous oil (3.3 mg, quantitative yield) whose NMR (1H and 13C)
data, TLC retention time, and [R]D value were identical to those of
dolabellanone 3.
1
1369, 1273, 1182 cm-1; H NMR (300 MHz, CDCl3) and 13C NMR
(75 MHz, CDCl3), see Tables 1 and 3, respectively; EIMS m/z [M]+
302 (100), 287 (7), 269 (3), 231 (7), 219 (14), 163 (53), 151 (42), 137
(36), 121 (23), 109 (24), 93 (33); HREIMS m/z [M]+ 302.2252 (calcd
for C20H30O2, 302.2245).
Dolabellanone 2: white crystalline solid; mp 169-170 °C; [R]25
D
+26.3 (c 1.3, CHCl3); UV (MeOH) λmax (log ε) 203 (3.4), 255 (3.2)
nm; IR (film) νmax 2981, 2969, 2930, 1735, 1698, 1622, 1441, 1412,
1
1371, 1242, 1224, 1186, 1014, 958, 912 cm-1; H NMR (300 MHz,
CDCl3) and 13C NMR (75 MHz, CDCl3), see Tables 1 and 3,
respectively; EIMS m/z [M]+ 360 (21), 318 (29), 300 (95), 285 (14),
189 (23), 163 (58), 151 (100), 136 (50), 121 (34), 107 (49); HREIMS
m/z [M]+ 360.2306 (calcd for C22H32O4, 360.2301).
Dolabellanone 3: colorless oil, [R]25 +5.7 (c 1.3, CHCl3); UV
D
(MeOH) λmax (log ε) 203 (3.5), 247 (3.2) nm; IR (film) νmax 3077, 2963,
2931, 2869, 1737, 1702, 1624, 1447, 1370, 1280, 1243, 1180, 1016,
966, 906 cm-1; 1H NMR (300 MHz, CDCl3) and 13C NMR (75 MHz,
CDCl3), see Tables 1 and 3, respectively; EIMS m/z [M]+ 344 (9),
302 (18), 284 (87), 269 (21), 241 (26), 201 (38), 187 (30), 150 (96),
135 (74), 121 (82), 93 (100), 81 (76), 69 (55); HREIMS m/z [M]+
344.2346 (calcd for C22H32O3, 344.2351).
Epoxidation of Dolabellatrienone (11). We followed the general
procedure reported by Look and Fenical.3 A mixture of dolabella-
trienone (11, 0.33 g, 1.2 mmol), anhydrous Na2HPO4 (0.17 g, 1.2
mmol), and m-chloroperbenzoic acid (0.20 g, 1.2 mmol) in dry CH2Cl2
(54 mL) was stirred at 25 °C for 1 h. The reaction mixture was diluted
and washed with 10% Na2SO3 (2 × 25 mL), 5% NaHCO3 (2 × 25
mL), and brine (2 × 25 mL). The organic layer was dried over MgSO4,
filtered, and concentrated. The product mixture (305 mg) was chro-
matographed on a Si gel (15 g) column by elution with benzene/EtOAc
(95:5) to afford pure diepoxide 7 (73.7 mg, 20%), C-7R,8R monoep-
oxide 12 (71.2 mg, 21%), and claenone (19, 145 mg, 42%);18,19 partial
data for claenone (19) [3R,4R-epoxy-13-keto-1R,11S-dolabell-7E,12(18)-
diene]: [R]25D -64 (c 0.2, CHCl3); 13C NMR (75 MHz, CDCl3) δ 205.9
(C, C-13), 148.5 (C, C-18), 137.0 (C, C-12), 132.8 (C, C-8), 128.2
(CH, C-7), 63.6 (CH, C-3), 61.2 (C, C-4), 55.4 (CH2, C-14), 42.2 (CH,
C-11), 40.7 (C, C-1), 38.4 (CH2, C-9), 37.4 (CH2, C-2), 37.1 (CH2,
C-5), 27.2 (CH2, C-6), 24.4 (CH3, C-19), 24.3 (CH2, C-10), 23.2 (CH3,
C-15), 21.2 (CH3, C-20), 16.5 (CH3, C-17), 15.4 (CH3, C-16). The [R]D,
IR, UV, HREIMS, and 1H NMR (CDCl3) data for 19 have been
previously described.3,18
Single-Crystal X-ray Analysis of Dolabellanones 1, 2, 8, and 12.
The X-ray data were collected at 298 K with a Bruker SMART 1 K
CCD diffractometer equipped with a graphite monochromator and Mo
KR radiation (λ ) 0.71073 Å) using the SMART software. Final values
of the cell parameters were obtained from least-squares refinement. The
frames were processed using the SAINT software to give the hkl file
corrected for Lorentz and polarization effects. No absorption correction
was applied. The structures were solved by direct methods with the
SHELX-90 program and refined by least-squares methods on F2,
SHELXTL-93, incorporated in SHELXTL, Version 5.1.20 The initial
E-maps yielded all non-hydrogen atom positions. All non-hydrogen
atoms were refined anisotropically, and the H atoms were positioned
geometrically and treated as riding, with C-H distances in the range
0.93-0.98 Å and with Uiso(H) ) 1.2 or 1.5 × Ueq(C). The crystal-
lographic data for dolabellanones 1, 2, 8, and 12 reported in this article
(Table 4) have been deposited at the Cambridge Crystallographic Data
Centre, under the reference numbers CCDC 709008-709011. Copies
of the data can be obtained, free of charge, on application to the
Director, CCDC, 12 Union Road, Cambridge CB2 1EZ, UK (fax: +44
1223 336033 or e-mail: deposit@ccdc.cam.ac.uk). ORTEP representa-
tions of 1, 2, 8, and 12 (Figure 2) show the relative configuration for
each of these molecules.
Dolabellanone 4: colorless oil, [R]25 +11.8 (c 0.83, CHCl3); UV
D
(MeOH) λmax (log ε) 202 (3.9), 235 (3.7) nm; IR (film) νmax 3434, 2933,
2873, 1701, 1620, 1453, 1373, 1242, 1188, 1037, 903 cm-1; 1H NMR
(300 MHz, CDCl3) and 13C NMR (75 MHz, CDCl3), see Tables 1 and
3, respectively; EIMS m/z [M]+ 302 (16), 284 (6), 189 (19), 175 (25),
161 (40), 150 (54), 135 (47), 121 (54), 107 (61), 93 (87), 81 (70), 61
(100); HREIMS m/z [M]+ 302.2242 (calcd for C20H30O2, 302.2246).
Dolabellanone 5: colorless oil, [R]25 -7.4 (c 1.3, CHCl3); UV
D
(MeOH) λmax (log ε) 204 (3.9), 235 (3.4) nm; IR (film) νmax 3079, 2971,
2933, 2864, 1735, 1706, 1626, 1454, 1372, 1282, 1247, 1190, 1018,
1
978, 931, 914, 756 cm-1; H NMR (300 MHz, CDCl3) and 13C NMR
(75 MHz, CDCl3), see Tables 1 and 3, respectively; EIMS m/z [M]+
360 (2), 318 (1), 300 (3), 257 (1), 215 (2), 189 (4), 163 (5), 149 (11),
135 (7), 107 (11), 91 (11), 61 (100); HREIMS m/z [M]+ 360.2298
(calcd for C22H32O4, 360.2301).
Dolabellanone 6: colorless oil, [R]25 -29.5 (c 1.0, CHCl3); UV
D
(MeOH) λmax (log ε) 206 (3.3), 234 (3.7), 235 (3.6) nm; IR (film) νmax
3397, 2969, 2935, 2874, 1704, 1619, 1457, 1376, 1040 cm-1; 1H NMR
(300 MHz, CDCl3) and 13C NMR (75 MHz, CDCl3), see Tables 2 and
3, respectively; EIMS m/z [M]+ 318 (21), 302 (36), 301 (22), 284 (64),
189 (24), 149 (43), 136 (100), 121 (26), 109 (30), 83 (36); HRESIMS
m/z [M - 1]- 317.2102 (calcd for C20H29O3, 317.2117).
Dolabellanone 7: colorless oil, [R]25 -26 (c 0.9, CHCl3); UV
D
(MeOH) λmax (log ε) 203 (3.9), 254 (3.8) nm; IR (film) νmax 2968, 2932,
2859, 1701, 1621, 1454, 1385, 1272, 1256, 1188, 1086, 1057, 945,
888 cm-1 1H NMR (300 MHz, CDCl3) and 13C NMR (75 MHz,
;
CDCl3), see Tables 2 and 3, respectively; EIMS m/z [M]+ 318 (54),
163 (37), 149 (79), 135 (73), 121 (59), 109 (79), 93 (72), 81 (70), 69
(80), 55 (100); HREIMS m/z [M]+ 318.2197 (calcd for C20H30O3,
318.2195). The physical data of dolabellanone 7 were found to be
identical with those of a diepoxide that was chemically derived from
dolabellatrienone (11) isolated from the Caribbean gorgonian Eunicea
calyculata and reported with incomplete relative configuration about
the epoxides.3
Dolabellanone 8: white crystalline solid, mp 102.2-102.7 °C, [R]20
D
Biological Screening Assays. For a general description of the
approach used by the NIAID’s Antimicrobial Acquisition and Coor-
dinating Facility (AACF) for determining antiviral activity and toxicity
for hepatitis B virus, visit http://niaid-aacf.org/protocols/HBV.htm.
Anticancer activity screening by the Developmental Therapeutics
Program (DTP) of the National Cancer Institute is conducted following
this general protocol: most of the compounds screened have no
antiproliferative activity (up to 85%). In order to avoid screening
inactive compounds across all the cell lines, a prescreen is done using
three highly sensitive cell lines (breast MCF-7, lung NCI-H640, CNS
SF-268). Antiproliferative activity must be seen in these cell lines in
order to continue to the 60-cell-line panel. The 60 different human
tumor lines are incubated with five different doses of compound, and
+ 93.0 (c 1.0, CHCl3); UV (MeOH) λmax (log ε) 229 (3.7) nm; IR
(film) νmax 3077, 2986, 2957, 2919, 2862, 1686, 1601, 1447, 1376,
1
1266, 1217, 1190, 1162, 1095, 1083, 1024, 868 cm-1; H NMR (300
MHz, CDCl3) and 13C NMR (75 MHz, CDCl3), see Tables 2 and 3,
respectively; EIMS m/z [M]+ 302 (19), 287 (8), 269 (5), 241 (8), 217
(9), 203 (20), 189 (15), 161 (33), 150 (100), 135 (54), 121 (41), 107
(41), 95 (45), 81 (51), 55 (69); HREIMS m/z [M]+ 302.2252 (calcd
for C20H30O2, 302.2246).
Dolabellanone 9: colorless oil, [R]25 +43.1 (c 1.0, CHCl3); UV
D
(MeOH) λmax (log ε) 203 (4.2), 229 (4.1) nm; IR (film) νmax 3419, 3075,
2969, 2932, 2875, 1697, 1600, 1458, 1381, 1341, 1264, 1214, 1190,
1110, 1090, 1040, 1014, 864, 825, 810, 758 cm-1; 1H NMR (300 MHz,
CDCl3) and 13C NMR (75 MHz, CDCl3), see Tables 2 and 3,

Dolabellane-Type Diterpenoids from Eunicea
Journal of Natural Products, 2010, Vol. 73, No. 5 933
Table 4. Single-Crystal X-ray Diffraction Data for Compounds 1, 2, 8, and 12a
parameter
1
2
8
12
empirical formula
fw
C20H30O2
302.44
C22H32O4
360.48
C20H30O2
302.44
C20H30O2
302.44
temperature (K)
wavelength (Å)
cryst syst
space group
unit cell dimens
a (Å)
299 (2)
0.71073
monoclinic
P21 (No.4)
298 (2)
0.71073
orthorhombic
299 (2)
0.71073
orthorhombic
298 (2)
0.71073
monoclinic
P212121 (No. 19)
P21212 (No. 18)
P21 (No. 4)
9.777(2)
10.719(2)
17.521(3)
90
90.174(3)
90
1836.2(5)
4
1.094
8.994(2)
12.140(2)
18.745(3)
90
90.174(3)
90
2046.8(6)
4
1.170
11.149(2)
21.716(3)
7.590(2)
90
90
90
1837.5(4)
4
8.267(2)
8.670(2)
12.790(3)
90
90.451(4)
90
904.3(3)
2
1.111
b (Å)
c (Å)
R (deg)
ꢀ (deg)
γ (deg)
volume (Å3)
Z
density (calcd) (Mg m-3
)
1.093
0.068
664
absorp coeff (mm-1
F(000)
)
0.068
664
0.079
784
0.069
332
cryst size (mm)
θ range for data
collection (deg)
index ranges
0.15 × 0.14 × 0.14
0.46 × 0.09 × 0.05
0.34 × 0.29 × 0.24
0.19 × 0.16 × 0.08
1.16 to 23.25
2.00 to 25.52
1.88 to 27.99
1.61 to 23.33
-10 e h e 10,
-11 e k e 9,
-19 e l e 18
8096
4306 [R(int) ) 0.0235]
3121 [R(int) ) 0.0372]
-10 e h e 10,
-12 e k e 14,
-22 e l e 21
-14 e h e 14,
-26 e k e 28,
-10 e l e 8
12 536
4358 [R(int) ) 0.0217]
3596 [R(int) ) 0.0206]
-7 e h e 9,
-9 e k e 9,
-14 e l e 14
4050
2282 [R(int) ) 0.0379]
1913 [R(int) ) 0.0397]
reflns collected
indep reflns
indep reflns [I > 2σ(I)]
max. and min. transmn
refinement method
11 514
3795 [R(int) ) 0.0688]
2439 [R(int) ) 0.0630]
0.0233, 0.0075
full-matrix least-squares on F2
data/restraints/params
4306/1/407
0.976
3795/0/240
1.263
4358/0/205
1.036
2282/1/204
1.096
goodness-of-fit on F2
final R indices [I > 2σ(I)]
R indices (all data)
largest diff peak
R1 ) 0.0418,
wR2 ) 0.0942
R1 ) 0.0675,
wR2 ) 0.1049
0.175 and -0.119
R1 ) 0.0657,
wR2 ) 0.1278
R1 ) 0.1405,
wR2 ) 0.1435
0.139 and -0.114
R1 ) 0.0416,
wR2 ) 0.0973
R1 ) 0.0541,
wR2 ) 0.1035
0.175 and -0.119
R1 ) 0.0479,
wR2 ) 0.1168
R1 ) 0.0618,
wR2 ) 0.1356
0.148 and -0.131
and hole (e Å-3
)
a Atomic coordinates for all X-ray structures can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
a sulforhodamine blue (SRB) assay is performed after 48 h to determine
cytotoxicity. From the five-point curve, the following concentrations
are extrapolated: GI50 (inhibits growth by 50%), TGI (totally inhibits
growth), LC50 (kills 50% of cells). For the specific screening methods
from the DTP Web site, visit http://www.dtp.nci.nih.gov/branches/btb/
ivclsp.html. Compounds shown to have anticancer activity in cell lines
within the NCI 60 panel may then move on to animal trials and, if
successful, may eventually move on to be tested in clinical trials.
Additional experimental details for our primary in Vitro antimicrobial
assays against Mycobacterium tuberculosis and Plasmodium falciparum
have been previously described.21,22
References and Notes
(1) Rodr´ıguez, A. D. Tetrahedron 1995, 51, 4571–4618.
(2) Berrue, F.; Kerr, R. G. Nat. Prod. Rep. 2009, 26, 681–710.
(3) Look, S. A.; Fenical, W. J. Org. Chem. 1982, 47, 4129–4134.
(4) Ca´ceres, J.; Rivera, M. E.; Rodr´ıguez, A. D. Tetrahedron 1990, 46,
341–348.
(5) Shin, J.; Fenical, W. J. Org. Chem. 1991, 56, 3392–3398.
(6) Rodr´ıguez, A. D.; Acosta, A. L.; Dhasmana, H. J. Nat. Prod. 1993,
56, 1843–1849.
(7) Rodr´ıguez, A. D.; Gonza´lez, E.; Gonza´lez, C. J. Nat. Prod. 1995, 58,
226–232.
(8) Govindan, M.; Govindan, G. N.; Kingston, D. G. I. J. Nat. Prod. 1995,
58, 1174–1184.
Acknowledgment. This work was supported by a grant from the
National Institutes of Health SCORE Program (Grant S06GM08102)
awarded to A.D.R. We thank J. A. Sa´nchez for assistance during the
collection and identification of the gorgonian coral, E. Gonza´lez for
the chemical conversion of 11 to 7, and K. Nieves for measuring the
specific rotation of 8. Mass spectrometry determinations were provided
by the Mass Spectrometry Laboratory of the University of Illinois at
Urbana-Champaign. We gratefully acknowledge the assistance of E.
Ortega-Barria and J. Gonza´lez from the Instituto de Investigaciones
Avanzadas y Servicios de Alta Tecnolog´ıa in Panama for the antima-
larial bioassays. The National Cancer Institute (NCI), the National
Institute of Allergy and Infectious Diseases (NIAID), and the TAACF
provided in Vitro cytotoxicity, antiviral, and antituberculosis activity
data, respectively.
(9) Bashyal, B.; Desai, P.; Rao, K. V.; Hamann, M. T.; Avery, B. A.;
Reed, J. K.; Avery, M. A. J. Chem. Res. 2006, 165–167.
(10) Xiang, W.; Chang, L. C. Planta Med. 2006, 72, 735–739.
(11) Wei, X.; Rodr´ıguez, A. D.; Baran, P.; Raptis, R. G.; Sa´nchez, J. A.;
Ortega-Barria, E.; Gonza´lez, J. Tetrahedron 2004, 60, 11813–11819.
(12) For a review on dolabellane diterpenes, see: Rodr´ıguez, A. D.;
Gonza´lez, E.; Ram´ırez, C. Tetrahedron 1998, 54, 11683–11729.
(13) Corey, E. J.; Kania, R. S. J. Am. Chem. Soc. 1996, 118, 1229–1230.
(14) Snyder, S. A.; Corey, E. J. J. Am. Chem. Soc. 2006, 128, 740–742.
(15) (a) Enserink, M. Science 2008, 322, 1776. (b) Noedl, H.; Socheat,
D.; Satimai, W. N. Engl. J. Med. 2009, 361, 540–541.
(16) The antimalarial activity of the compounds tested was as follows: 1
(IC50 13.2 µM); 2 (IC50 50.0 µM); 3 (IC50 20.3 µM); 4 (IC50 36.4
µM); 5 (IC50 36.1 µM); 6 (IC50 50.3 µM); 7 (IC50 37.7 µM); 8 (IC50
g 100 µM); 9 (IC50 9.4 µM); 10 (IC50 40.9 µM); 11 (IC50 45.5 µM);
12 (IC50 23.2 µM); 13 (IC50 30.6 µM); 14 (IC50 33.3 µM); 15 (IC50
43.0 µM); 16 (IC50 31.2 µM); 17 (IC50 59.6 µM).
(17) Although the same X-ray structure of 12 has already been described
in the literature by Xiang and Chang (see ref 10), it was incorrectly
referred to in the text as the enantiomer 1S,7S,8S,11R.
(18) Mori, K.; Iguchi, K.; Yamada, N.; Yamada, Y.; Inouye, Y. Chem.
Pharm. Bull. 1988, 36, 2840–2852.
1
Supporting Information Available: H and 13C NMR spectra in
CDCl3, including representative 2D NMR spectra, for compounds 1-10,
as well as a photograph of the air-dried voucher specimen Eunicea sp.
2. This material is available free of charge via the Internet at http://
pubs.acs.org.

934 Journal of Natural Products, 2010, Vol. 73, No. 5
Wei et al.
(19) Miyaoka, H.; Isaji, Y.; Kajiwara, Y.; Kunimune, I.; Yamada, Y.
Tetrahedron Lett. 1998, 39, 6503–6506.
(20) Data Collection: (a) SMART-NT Software Reference Manual, version
5.0; Bruker AXS, Inc.: Madison, WI, 1998. Data Reduction: (b)
SAINT-NT Software Reference Manual, verion 4.0; Bruker AXS, Inc.:
Madison, WI, 1996. (c) Sheldrick, G. M. SHELXTL-NT, version 5.1;
Bruker AXS, Inc.: Madison, WI, 1999.
(21) Collins, L. A.; Franzblau, S. G. Antimicrob. Agents Chemother. 1997,
41, 1004–1009.
(22) Corbett, Y.; Herrera, L.; Gonza´lez, J.; Cubilla, L.; Capson, T.; Colley,
P. D.; Kursar, T. A.; Romero, L. I.; Ortega-Barria, E. J. Trop. Med.
Hyg. 2004, 70, 119–124.
NP100074R