Bioactive metabolites from the South African marine mollusk Trimusculus costatus

Article abstract of DOI:10.1021/np070612y

A reinvestigation of extracts of the endemic South African intertidal limpet Trimusculus costatus yielded the known labdane diterpenes 6β,7α-diacetoxylabda-8,13E-dien-15-ol (1) and 2α,6β, 7α-triacetoxylabda-8,13E-dien-15-ol (2) and three new metabolites, 6β,7α,15-triacetoxylabda-8,13E-diene (3), 3α,11-dihydroxy-9,11- seco-cholest-4,7-dien-6,9-dione (4), and cholest-7-en-3,5,7-triol (5). Chiral derivatization and X-ray analysis were used to confirm the labdane absolute configuration of 2. Compounds 1, 2 and 4 exhibited moderate activity (3-25 μM) against the WHCO1 human esophageal cancer cell line.

Full text of DOI:10.1021/np070612y

420
J. Nat. Prod. 2008, 71, 420–425
Bioactive Metabolites from the South African Marine Mollusk Trimusculus costatus
Albert W. W. van Wyk,† Christopher A. Gray,† Catherine E. Whibley,‡ Omolaja Osoniyi,‡ Denver T. Hendricks,‡
Mino R. Caira,§ and Michael T. Davies-Coleman*,†
Department of Chemistry, Rhodes UniVersity, Grahamstown, South Africa, DiVision of Medical Biochemistry, UniVersity of Cape Town,
Rondebosch, South Africa, and Department of Chemistry, UniVersity of Cape Town, Rondebosch, South Africa.
ReceiVed October 31, 2007
A reinvestigation of extracts of the endemic South African intertidal limpet Trimusculus costatus yielded the known
labdane diterpenes 6ꢀ,7R-diacetoxylabda-8,13E-dien-15-ol (1) and 2R,6ꢀ,7R-triacetoxylabda-8,13E-dien-15-ol (2) and
three new metabolites, 6ꢀ,7R,15-triacetoxylabda-8,13E-diene (3), 3R,11-dihydroxy-9,11-seco-cholest-4,7-dien-6,9-dione
(4), and cholest-7-en-3,5,7-triol (5). Chiral derivatization and X-ray analysis were used to confirm the labdane absolute
configuration of 2. Compounds 1, 2 and 4 exhibited moderate activity (3–25 µM) against the WHCO1 human esophageal
cancer cell line.
Shelled, marine pulmonate mollusks of the genus Trimusculus
are well-adapted to the challenges of living in the intertidal zone
of rocky shores and are truly amphibious, possessing a mantle cavity
that has evolved to serve as both a lung and a gill.1 Trimusculids
are prolific where they occur and congregate in large colonies under
shady rocky overhangs.1 In this habitat algal growth is slow, thus
reducing the significance of algae (typically exploited by other
intertidal limpets) as a primary food source for trimusculids.1–3
Trimusculus species have, therefore, evolved a unique feeding
method and use a mucus net, secreted by the mantle glands, to
filter phytoplankton out of the water column.1 The mucus net and
trapped food particles are ingested once the presence of food in
the mucous net is detected by oral lobes. A secondary mucous,
distinct from that used for filter feeding, is produced by numerous
subepithelial glands situated in the mantle and sides of the foot.1
This secondary mucus, containing diterpene metabolites, has
chemical defense properties and has been shown to deter predation
of trimusculid limpets by both seastars and fish.2–5
3,5,7-triol (5), from selected fractions generated from diol flash
chromatography.
Diterpenes routinely dominate the natural product investigations
of Trimusculus species and are often characterized by a 6ꢀ,7R,15-
oxygenation pattern as exemplified by 1 from T. costatus and 6
from the Chilean species T. peruVianus.4,9 Occasionally, additional
oxygenation is observed at C-2, e.g., 2 from T. costatus and 7 from
the Californian species T. reticulatus, or at C-3 and C-19, e.g., 8
also from T. peruVianus.2,4,6 Isovalerate and acetate ester func-
tionalities are the only two ester groups reported thus far from
Trimusculus species and, with the exception of four ∆13 Z geometric
isomers, e.g., 9 isolated from T. peruVianus,8 the ∆13 E olefin is
predominant. A labdane stereochemistry was randomly assigned
by Manker and Faulkner to the first two metabolites, 7 and 10,
isolated from a trimusculid mollusk, and this stereochemical
assignment was replicated without justification in subsequent
publications describing trimusculid diterpene metabolites.2 Recently,
the application of the modified Mosher’s method to 8 and a
stereoselective synthesis of 10 from (+)-larixol confirmed a labdane
stereochemistry in these two compounds and appeared to justify
the original tenuous assignment of a labdane stereochemistry to
the diterpenes isolated from Trimusculus species.6,10 Confirmation
that the diterpenes isolated from T. costatus could also be assigned
to the labdane, as opposed to the ent-labdane series, regularly
encountered in diterpenes isolated from marine mollusks,11 was
provided by esterification of 2 with (-)-camphanic chloride and
crystallization of the ester product (11) from ethyl acetate/hexane
to provide crystals suitable for X-ray analysis (Figure 1).12 During
structural elucidation of 11, 2-fold disorder of the COO moiety of
the camphanate ester was detected and modeled accordingly,
resulting in refined site-occupancy factors of 0.6 and 0.4 for the
major and minor components, respectively. That this disorder was
not peculiar to the crystal specimen selected was subsequently
verified by X-ray analysis of a second crystal specimen, which
revealed the same phenomenon. The presence of this disorder did
not, however, compromise the unequivocal assignment of a labdane
absolute configuration to 11, as indicated in Figure 1.
Our earlier investigation of the diterpene metabolites of the
endemic South African trimusculid species Trimusculus costatus,
the only African representative of the genus Trimusculus, yielded
two major diterpene metabolites, 6ꢀ,7R-diacetoxylabda-8,13E-dien-
15-ol (1) and 2R,6ꢀ,7R-triacetoxylabda-8,13E-dien-15-ol (2).4
Prompted by the successful reinvestigation of other Trimusculus
species that have afforded new minor bioactive metabolites, missed
during the initial natural product studies,6–8 we returned to extracts
of T. costatus with the objectives of, first, isolating minor
metabolites for screening against the WHCO1 esophageal cancer
cell line and, second, confirming the labdane configuration of the
diterpene metabolites.
Trimusculus costatus is common along the rocky shore near the
coastal resort of Cintsa situated on the warm temperate southeast
coast of South Africa. An acetone extract of specimens of T.
costatus collected during the autumn of 2007 was subjected to initial
polymeric reversed-phase separation followed by flash chromatog-
raphy using a diol solid support. Finally, an exhaustive combination
of normal-phase and diol HPLC afforded 1 and 2 and the minor
metabolites 6ꢀ,7R,15-triacetoxylabda-8,13E-diene (3) 3R,11-dihy-
droxy-9,11-seco-cholest-4,7-dien-6,9-dione (4), and cholest-7-en-
Results and Discussion
A molecular formula of C26H40O6 was established for 3 from
1
HRFABMS data. The congruency between the H, 13C NMR and
mass data of 1, 2, and 3 and the presence of three ester carbonyl
resonances (δC 169.7, 169.9, and 171.1) in the 13C NMR spectrum
of 3 suggested that 3 was a triacetate, isomeric with 2 and differing
by one acetate unit from 1. A three-bond gHMBC correlation from
the oxymethylene protons (δH 4.50) in 3 to the carbonyl carbon of
an acetate functionality secured the assignment of the ester moiety
Dedicated to Dr. G. Robert Pettit of Arizona State University for his
pioneering work on bioactive natural products.
* To whom correspondence should be addressed. Tel: +27 46 603 8924.
Fax: +27 46 6225109. E-mail: M.Davies-Coleman@ru.ac.za.
† Department of Chemistry, Rhodes University.
‡ Division of Medical Biochemistry, University of Cape Town.
§ Department of Chemistry, University of Cape Town.
10.1021/np070612y CCC: $40.75
2008 American Chemical Society and American Society of Pharmacognosy
Published on Web 02/21/2008

BioactiVe Metabolites from the Mollusk Trimusculus costatus
Journal of Natural Products, 2008, Vol. 71, No. 3 421
of the carbonyl at C-9 was supported by gHMBC correlations from
H3-19 (δH 1.37) to C-9 (δC 201.8). Further 1H and 13C NMR
assignments were made from gHMBC correlations from H3-19 to
quaternary carbons C-10 (δC 49.2) and C-5 (δC 140.4), as well as
to methylene carbon C-1 (δC 25.0). A two-bond gHMBC correlation
from the olefinic methine proton H-7 (δH 6.69) to the remaining
carbonyl carbon (δC 187.5), in addition to a three-bond gHMBC
correlation from the olefinic proton H-4 (δH 6.70) to the same
carbon placed the second ketone at C-6. A gCOSY correlation
between H-4 and oxymethine proton H-3 (δH 4.34) provided the
assignment of C-3 (δC 63.0) from gHSQC data. gCOSY correlations
between oxymethine H-3 (δH 4.34) and H2-2 (δH 1.89) facilitated
the assignment of C-2 (δC 27.2) from gHSQC data. Two- and three-
bond gHMBC correlations from H-3 to both C-2 and C-4 (δC 135.2)
and both C-1 and C-5 confirmed the structure of ring A.
The relative configuration of the substituents around ring A of 4
was determined through NOE correlations. A correlation between
H3-19 (δH 1.37) and both H-1ꢀ (δH 1.77) and H-3 (δH 4.34) implied
that these methyl and methine protons are situated on the same
face of the ring. The configuration of the substituents around the
cyclopentane ring and at C-20 was proposed from biogenetic
considerations. Attempted selective acetylation of the primary
alcohol at C-11 of 4 afforded both the monoacetate (13) and the
diacetate (14). Subsequent application of the modified Mosher’s
method15 to 13 and calculation of the ∆δ values from the 1H NMR
spectra of the respective (R)- and (S)-MTPA esters (15 and 16) in
the usual manner (Figure 2) suggested an S configuration at C-3 in
13, and hence 4.
Figure 1. Perspective view of the structure of 11. For clarity, only
the major disordered component of the camphanate ester COO-
group is shown (key: C black; H blue; O red).
The molecular formula of 5, established as C27H46O3 by
HRFABMS data, implied five degrees of unsaturation. A disub-
stituted olefin [δC 130.7 (d) and 135.4 (d)] was evident from the
13C and DEPT135 NMR data acquired for 5 and accounted for one
of the degrees of unsaturation intimated by the molecular formula.
The absence of any spectroscopic evidence to suggest the presence
of other unsaturated functionalities in 5 led us to deduce a tetracyclic
skeleton for this compound. A closer inspection of the 13C and
DEPT NMR spectra confirmed the presence of 27 carbons compris-
ing one trisubstituted olefin, four quaternary, five methyl, six
methine (one oxygenated), and 10 methylene carbons. The 27
carbons and tetracyclic nature of 5 were suggestive of a steroid
nucleus. Rings C and D and the side chain of 5 were assigned
through analogy with the 1H and 13C NMR data of 5 with those of
4, and the individual chemical shift assignments confirmed by
exhaustive gHMBC and gCOSY correlations. A two-bond gHMBC
correlation from the methyl protons H3-19 (δH 0.86) enabled the
assignment of quaternary carbon C-10 (δC 36.92). Further three-
bond gHMBC correlations from H3-19 tentatively allowed us to
assign methylene carbon C-1 (δC 34.7) and methine carbons C-9
(δC 51.6) and C-5 (δC 82.1). gHMBC correlations from the two
vicinal olefinic methine protons (δH 6.49 and 6.22), constituting
an isolated spin system, to C-5 (δC 82.1) and another oxygenated
quaternary carbon (δC 79.4) placed the disubstituted olefin between
these two quaternary carbons. Although the 1H and 13C NMR
chemical shifts for C-6 and C-7 were reminiscent of epidioxy sterols
(e.g., 17),16,17 the HRFABMS data unequivocally supported a
molecular formula of C27H46O3 for 5 (the epidioxy analogue of 5
would require C27H44O3). In addition, a fragmentation peak corre-
sponding to the loss of molecular oxygen, which is diagnostic for
an epidioxy sterol, was absent from the mass spectrum of 5.18
Another three-bond gHMBC correlation from H-9 (δH 1.54) to δC
79.4 provided the assignment of C-8. A three-bond gHMBC
correlation from the more deshielded of the olefinic resonances (H-
7, δ6.49) to C-14 (δC 51.0) enabled us to assign C-7 (δC 130.7)
and suggested that the remaining olefinic carbon resonance was
C-6 (δC 135.4). A three-bond gHMBC correlation between H-6
and C-4 provided the assignment of C-4 (δC 36.89), which was
further confirmed by a gCOSY correlation between H-3 (δH 3.95)
1
Figure 2. Calculated ∆δ values obtained from the H NMR data
(600 MHz, CDCl3) of the MPTA esters (17 and 18) of 13.
to C-15. Confirmation of the ∆13 E olefin in 3 was provided by the
13C chemical shifts of the vinylic methyl (C-16) and allylic
methylene (C-12) carbon resonances (δC 19.0 and 39.2, respec-
tively).13 Acetylation of a portion of 1 in the usual manner afforded
3, which was identical in all respects with the natural product and
unequivocally confirmed the structure of this compound as 6ꢀ,7R,15-
triacetoxylabda-8,13E-diene.
The molecular formula of 4, established as C27H42O4 from
HRFABMS data, implied seven degrees of unsaturation. The
presence of two R,ꢀ-unsaturated ketones (δC 187.5 and 201.8) and
their corresponding two trisubstituted olefins [δC 135.2 (d), 140.4
(s), 138.7 (d), and 153.5 (s)] and one secondary hydroxyl
functionality (δC 63.0) was further corroborated by the IR absorption
bands at 1683 and 1624 cm-1 and a broad absorbance band at 3420
cm-1, respectively. The two R,ꢀ-unsaturated ketones accounted for
four of the seven degrees of unsaturation required by the molecular
formula of 4, and the absence of any further carbonyl or olefinic
functionalities suggested a tricyclic skeleton for this compound.
Analysis of the 13C and DEPT NMR spectra of 4 confirmed the
presence of 27 carbons, comprising two trisubstituted olefins, one
oxymethylene, two conjugated carbonyl, one oxymethine, two
quaternary, four methyl, four methine, and eight methylene carbons.
The presence of 27 carbons suggested a steroidal structure, but the
tricyclic nature of 4 implied rearrangement of the typical tetracyclic
steroid skeleton, such as in a 9,11-secosterol, e.g., 12.14 Comparison
of the 1H and 13C NMR spectra of 4 and 12 limited the differences
to the substitution pattern around the A and B rings. The position

422 Journal of Natural Products, 2008, Vol. 71, No. 3
Van Wyk et al.
Chart 1
and both H-4ꢀ (δH 1.88) and H-4R (δH 2.09). A two-bond gHMBC
correlation from H-3, as well as gCOSY correlations between H-3
and both H-2ꢀ (δH 1.51) and H-2R (δH 1.82), allowed us to assign
C-2 (δC 30.1).
cancer,20 with studies in Soweto revealing that residents there had
a 5-fold higher risk of developing the cancer than the world
average.21 Although a wide range of factors have been linked to
the onset of the disease, incidences of esophageal cancer in South
Africa have been associated traditionally with low socioeconomic
status, poor nutrition, the consumption of maize contaminated with
a Fusarium fungus, and the use of alcohol and tobacco.21–24
NOESY correlations between H3-19 (δH 0.86) and all three H-1ꢀ
(δH 1.67), H-2ꢀ (δH 1.51), and H-4ꢀ (δH 1.88) suggested that these
protons were all on the ꢀ-face of 5. Similarly, NOESY correlations
between H-3 (δH 3.95) and H-1R (δH 1.95), H-2R (δH 1.82), and
H-4R (δH 2.09) suggested a ꢀ-equatorial orientation of the hydroxyl
functionality at C-3. The configurations in ring C and ring D of 5
were proposed from standard steroidal biogenetic considerations,
and the configuration at C-5 was proposed from an expectation of
a trans fusion of rings A and B. The congruency between the
chemical shift data of 5 and 17 would suggest that the configuration
at C-8 was consistent with that observed at this position in epidioxy
sterols.
As part of an ongoing collaborative esophageal cancer research
project we screened four of the metabolites (1-4) isolated from T.
costatus for their cytotoxicity toward WHCO1 esophageal cancer
cells. Globally, esophageal cancer is the eighth most common form
of cancer, with remission rates rarely exceeding 10% after diag-
nosis.19 In South Africa esophageal cancer is the fifth most common
Compounds 1–3 exhibited weak activity (IC50 25, 24, and 84
µM, respectively) against the WHCO1 esophageal cancer cell line
when compared to the commonly used chemotherapeutic agent
cisplatin (IC50 13 µM).25 Compound 4, however, exhibited reason-
able activity (IC50 3 µM) against this cell line. A recent study of
the cellular mechanism of the antiesophageal cancer activity of the
triprenylated toluquinone (18) and related analogues (originally
isolated from the endemic South African opisthobranch mollusk
Leminda millecra) has revealed that these compounds mediate cell
death by triggering the production of reactive oxygen species (ROS),
leading to the activation of signaling pathways (cJun and p38) that
ultimately induce apoptosis.26 Since ring B of 4 is structurally
similar to a quinone, it is possible that this secosterol may exhibit
a similar mode of action against esophageal cancer cells.

BioactiVe Metabolites from the Mollusk Trimusculus costatus
Journal of Natural Products, 2008, Vol. 71, No. 3 423
Experimental Section
(3H, s, H3-19), 1.35 (2H, m, H-22b, H-23b), 1.15 (2H, m, H-12a,
H-23a), 1.12 (2H, m, H2-24), 1.02 (1H, m, H-22a), 0.96 (3H, d, J )
6.7 Hz, H3-21), 0.86 (6H, d, J ) 6.6 Hz, H3-26, H3-27), 0.75 (3H, s,
H3-18); 13C NMR (CDCl3, 150 MHz) δ 201.8 (C, C-9), 187.5 (C, C-6),
153.5 (C, C-8), 140.4 (C, C-5), 138.7 (CH, C-7), 135.2 (CH, C-4),
63.0 (CH, C-3), 59.0 (CH2, C-11), 50.3 (CH, C-17), 49.2 (C, C-10),
47.9 (C, C-13), 44.0 (CH, C-14), 41.3 (CH2, C-12), 39.4 (CH2, C-24),
35.5 (CH2, C-22), 34.6 (CH, C-20), 28.0 (CH, C-25), 27.2 (CH2, C-2),
27.0 (CH2, C-15), 26.8 (CH3, C-19), 26.4 (CH2, C-16), 25.0 (CH2, C-1),
24.4 (CH2, C-23), 23.0 (CH3, C-27), 22.8 (CH3, C-26), 18.8 (CH3,
C-21), 17.7 (CH3, C-18); HRFABMS m/z 430.3082 (calcd for C27H42O4
[M+], 430.3083).
General Experimental Procedures. Melting points were deter-
mined using a Reichert hot-stage microscope and are uncorrected.
Optical rotations were measured on a Perkin-Elmer 141 polarimeter
calibrated at the sodium-D line (589 nm). Infrared spectra were recorded
on a Perkin-Elmer Spectrum 2000 FT-IR spectrometer with compounds
as films (neat) on NaCl disks. NMR spectra were acquired on Bruker
400 MHz Avance and 600 MHz Avance II spectrometers using standard
pulse sequences. Chemical shifts are reported in ppm, referenced to
residual solvent resonances (CDCl3 δH 7.25, δC 77.2). HRFABMS data
were obtained on a JEOL SX102 FAB mass spectrometer. Diaion HP-
20 polystyrene divinylbenzene beads (supplied by Supelco) and
Macherey-Nagel Chroma-Bond OH Diol (0.45 µm) were used for initial
chromatographic separations. High-performance liquid chromatography
was performed on either a Whatman’s Magnum 9, Partisil 10, or a
Machery-Nagel VP 250/10 Nucleosil 100-7 OH semipreparative
column.
Animal Material. A collection of 1658 specimens of Trimusculus
costatus (Krauss, 1848) was made from the densely populated colonies
of this mollusk at Cintsa West, South Africa, in May 2007. The
specimens were immediately placed in Me2CO and exhaustively
extracted. The Me2CO extract was loaded onto a HP-20 column and
eluted with aliquots (300 mL) of increasing concentration of Me2CO
in H2O (0%, 40%, 60%, 80%, and 100% Me2CO). The 60% and 80%
aqueous Me2CO fractions were each subjected to open column diol
chromatography, eluting with aliquots (100 mL) of (i) EtOAc/95%
hexane; (ii) EtOAc/80% hexane; (iii) EtOAc/50% hexane; and (iv)
EtOAc. Normal-phase semipreparative HPLC (EtOAc/hexane, 2:3) of
a portion of fraction ii of the 60% aqueous Me2CO fraction afforded
sufficient6ꢀ,7R-diacetoxylabda-8,13E-dien-15-ol1(30mg)and2R,6ꢀ,7R-
triacetoxylabda-8,13E-dien-15-ol, 2 (41 mg), for derivatization with
camphanic chloride. Exhaustive semipreparative HPLC (EtOAc/hexane,
3:2) of fraction i of the 80% Me2CO fraction afforded both 6ꢀ,7R,15-
triacetoxylabda-8,13E-diene, 3 (3 mg, 0.002 mg/animal), and cholest-
7-en-3,5,7-triol, 5 (16 mg, 0.010 mg/animal). Fraction ii of the 80%
Me2CO fraction as well as fraction iii from both the 60% and 80%
Me2CO fractions were pooled and subjected to diol semipreparative
HPLC (EtOAc/hexane, 3:2) to afford 3R,11-dihydroxy-9,11-seco-
cholest-4,7-dien-6,9-dione, 4 (303 mg, 0.2 mg/animal).
Cholest-7-en-3,5,7-triol (5): white, amorphous solid; [R]D24 +4 (c
1
1.5, CHCl3); IR (film) νmax 3401, 2953, 1459, 1377, 1228 cm-1; H
NMR (CDCl3, 600 MHz) δ 6.49 (1H, d, J ) 8.5 Hz, H-7), 6.22 (1H,
d, J ) 8.5 Hz, H-6), 3.95 (1H, m, H-3), 2.09 (1H, ddd, J ) 13.8, 5.0,
1.9 Hz, H-4R), 1.95 (1H, m, H-1R, H-12b), 1.92 (1H, m, H-16b), 1.88
(1H, dd, J ) 13.8, 11.7 Hz, H-4ꢀ), 1.82 (1H, m, H-2R), 1.67 (1H,
ddd, J ) 13.5, 6.9, 3.5 Hz, H-1ꢀ), 1.61 (1H, m, H-11b), 1.54 (1H, dd,
J ) 12.2, 4.6 Hz, H-9), 1.51 (2H, m, H-2ꢀ, H-25), 1.48 (2H, m, H-14,
H-15b), 1.42 (1H, m, H-11a), 1.36 (2H, m, H-16a, H-20), 1.32 (1H,
m, H-22b, H-23b), 1.20 (1H, m, H-15a), 1.17 (2H, m, H-12a, H-17),
1.14 (2H, m, H-23a, H-24a), 1.01 (1H, m, H-22a), 0.88 (3H, d, J )
6.5 Hz, H3-21), 0.86 (3H, s, H3-19), 0.85 (3H, d, J ) H3-26), 0.84
(3H, d, J ) 6.6 Hz, H3-27), 0.78 (3H, s, H3-18); 13C NMR (CDCl3,
150 MHz) δ 135.4 (CH, C-6), 130.7 (CH, C-7), 82.1 (C, C-5), 79.4
(C, C-8), 66.4 (CH, C-3), 56.4 (CH, C-17), 51.6 (CH, C-9), 51.0 (CH,
C-14), 44.7 (C, C-13), 39.4 (CH2, C-12), 39.4 (CH2, C-24), 36.92 (C,
C-10), 36.89 (CH2, C-4), 35.9 (CH2, C-22), 35.2 (CH, C-20), 34.7 (CH2,
C-1), 30.1 (CH2, C-2), 28.2 (CH2, C-16), 28.0 (CH, C-25), 23.8 (CH2,
C-23), 23.4 (CH2, C-15), 22.8 (CH3, C-26), 22.5 (CH3, C-27), 20.6
(CH2, C-11), 18.6 (CH3, C-21), 18.1 (CH3, C-19), 12.6 (CH3, C-18);
HRFABMS m/z 419.3525 [M + H+] (calcd for C27H47O3, 419.3525).
Preparation of the Camphanate Ester of 2. Diterpene 2 (34 mg,
0.07 mmol), camphanic chloride (34 mg, 0.16 mmol, 2.1 equiv), Et3N
(60 µL, 7.05 mmol, 6 equiv), and DMAP (5 mg, 0.04 mmol, 0.5 equiv)
were dissolved in anhydrous CH2Cl2 (2 mL) under an Ar atmosphere
and stirred at ambient temperature (5 h). The reaction mixture was
concentrated to dryness, taken up in Et2O (5 mL), and washed with 1
M HCl (1 mL) followed by H2O (2 × 5 mL). The organic partition
was dried (MgSO4) and concentrated to an amorphous solid (69 mg).
Normal-phase HPLC (33% EtOAc, 67% hexane) afforded the cam-
phanate ester 11 (38 mg, 81%) as a yellow oil. Platelets of 11, suitable
for X-ray diffraction, were grown using the slow diffusion method
(hexane/EtOAc).
24
6ꢀ,7r-Diacetoxylabda-8,13E-dien-15-ol (1): pale yellow oil; [R]D
+93 (c 1.6, CHCl3), lit.4 +93; IR (film) νmax 3461, 2924, 2872, 1742,
1
1365 cm-1; H NMR and 13C NMR consistent with literature values;4
HRFABMS m/z 406.2718 (calcd for C24H38O5 [M+], 406.2719).
2r,6ꢀ,7r-Triacetoxylabda-8,13E-dien-15-ol (2): colorless oil;
[R]D24 +60 (c 1.7, CHCl3), lit.4 +62; IR (film) νmax 3450, 2942, 2863,
Camphanate ester (11): colorless plates (from hexane/EtOAc);
mp 154–155 °C; 1H NMR (CDCl3, 600 MHz) δ 5.40 (1H, dt, J ) 7.1,
0.9 Hz, H-14), 5.31 (1H, m, H-6), 5.12 (1H, tt, J ) 11.7, 4.1 Hz, H-2),
4.98 (1H, d, J ) 0.8 Hz, H-7), 4.74 (2H, m, H2-15), 2.42 (1H, ddd, J
) 13.5, 10.8, 4.2 Hz, H-3′b), 2.19 (1H, m, H-11b), 2.13 (1H, m, H-11a),
2.10 (2H, m, H-1b and H-12b), 2.09 (1H, m, H-12a), 2.08 (3H, s, OAc-
7), 2.03 (3H, s, OAc-6), 2.02 (3H, s, OAc-2), 2.02 (1H, m, H-3′a),
1.91 (1H, ddd, J ) 13.2, 10.8, 4.6 Hz, H-4′b), 1.80 (1H, ddd, J )
12.3, 4.1, 1.8 Hz, H-3b), 1.76 (3H, br s, H3-16), 1.68 (1H, ddd, J )
13.2, 9.2, 4.2 Hz, H-4′a), 1.62 (3H, s, H3-17), 1.51 (1H, d, J ) 1.2 Hz,
H-5), 1.36 (3H, s, H3-20), 1.27 (1H, dd, J ) 11.9, 7.0 Hz, H-3a), 1.25
(1H, dd, J ) 11.9, 7.8 Hz, H-1a), 1.10 (3H, s, H3-8′), 1.04 (3H, s,
H3-10′), 1.03 (6H, s, H3-19 and H3-20), 0.94 (3H, s, H3-9′); 13C NMR
(CDCl3, 150 MHz) δ 178.1 (C, C-6′), 170.5 (C, 2-OAc), 169.7 (C,
7-OAc), 169.5 (C, 6-OAc), 167.5 (C, C-1′), 146.5 (C, C-9), 143.2 (C,
C-13), 122.4 (C, C-8), 117.6 (CH, C-14), 91.1 (C, C-2′), 73.2 (CH,
C-7), 69.0 (CH, C-6), 68.4 (CH, C-2), 62.2 (CH2, C-15) 54.7 (C, C-7′),
54.2 (C, C-5′), 48.8 (d, C-5), 47.8 (CH2, C-3), 43.9 (CH2, C-1), 40.5
(C, C-4), 39.2 (CH2, C-12), 34.3 (C, C-10), 33.0 (CH3, C-18), 30.6
(CH2, C-3′), 29.0 (CH2, C-4′), 26.7 (CH2, C-11), 23.7 (CH3, C-19),
22.0 (CH3, C-20), 21.5 (CH3, OAc-2), 21.4 (CH3, OAc-6), 21.1 (CH3,
OAc-7), 17.0 (CH3, C-17), 16.8 (CH3, C-9′), 16.72 (CH3, C-10′), 16.67
(CH3, C-16), 9.7 (CH3, C-8′).
1
1742, 1716 cm-1; H and 13C NMR consistent with literature values;4
HRFABMS m/z 464.2760 (calcd for C26H40O7 [M+], 464.2772).
24
6ꢀ,7r,15-Triacetoxylabda-8,13E-diene (3): pale yellow oil; [R]D
+48 (c 0.2, CHCl3); IR (film) νmax 2957, 2871, 1737, 1679, 1622 cm-1
;
1H NMR (CDCl3, 600 MHz) δ 5.37 (1H, t, J ) 6.9 Hz, H-14), 5.31
(1H, br s, H-6), 4.97 (1H, br s, H-7), 4.50 (2H, d, J ) 7.1 Hz, H2-15),
2.08 (3H, s, OAc-7), 2.06 (3H, s, OAc-15), 2.03 (3H, s, OAc-6), 2.21
(2H, m, H2-11), 2.11 (2H, m, H2-12), 1.80 (1H, m, H-1b), 1.74 (3H, s,
H3-16), 1.69 (1H, m, H-2b), 1.61 (3H, s, H3-17), 1.54 (1H, m, H-2a),
1.47 (1H, d, J ) 1.1 Hz, H-15), 1.42 (1H, m, H-3b), 1.28 (3H, s, H3-
20), 1.21 (1H, m, H-1a), 1.19 (1H, m, H-3a), 0.97 (3H, s, H3-18), 0.95
(3H, s, H3-19); 13C NMR (CDCl3, 150 MHz) δ 171.1 (C, OAc-7), 169.9
(C, OAc-15), 169.7 (C, OAc-6), 147.7 (C, C-9), 142.3 (C, C-13), 121.7
(C, C-8), 118.1 (CH, C-14), 73.4 (CH, C-7), 69.6 (CH, C-6), 61.3 (CH2,
C-15), 49.2 (CH, C-5), 43.1 (t, C-3), 39.5 (C, C-10), 39.2 (CH2, C-12),
38.9 (CH2, C-1), 33.4 (C, C-4), 33.0 (CH3, C-18), 26.8 (CH2, C-11),
23.1 (CH3, C-19), 21.5 (CH3, OAc-6), 21.2 (CH3, OAc-15), 21.1 (CH3,
7-OAc), 21.0 (CH3, C-20), 19.0 (CH2, C-2), 17.0 (CH3, C-17), 16.6
(CH3, C-16); HRFABMS m/z 449.2909 (calcd for C26H41O6 [(M +
H)+], 449.2903).
3r,11-Dihydroxy-9,11-seco-cholest-4,7-dien-6,9-dione (4): bright
yellow-orange oil; [R]D +32 (c 1.5, CHCl3); IR (film) νmax 3420, 2957,
1683, 1624, 1462 cm-1; 1H NMR (CDCl3, 600 MHz) δ 6.70 (1H, dd,
J ) 4.0, 0.9 Hz, H-4), 6.69 (1H, s, H-7), 4.34 (1H, m, H-3), 3.84 (1H,
ddd, J ) 10.7, 10.0, 5.6 Hz, H-11b), 3.71 (1H, ddd, J ) 10.7, 9.4, 5.8
Hz, H-11a), 3.54 (1H, dd, J ) 11.1, 8.7 Hz, H-14), 2.17 (1H, ddd, J
) 13.5, 13.4, 3.6 Hz, H-1R), 1.89 (3H, m, H2-2, H-16b), 1.77 (1H, m,
H-1ꢀ), 1.74 (1H, m, H-17), 1.71 (1H, m, H-12b), 1.68 (2H, m, H2-15),
1.52 (1H, m, H-25), 1.50 (1H, m, H-16a), 1.43 (1H, m, H-20), 1.37
Acetylation of 1. A solution of 1 (65 mg, 0.13 mmol) in pyridine
(1 mL) and acetic hydride (1 mL) was stirred at ambient temperature
overnight (16 h). The pyridine and acetic anhydride were removed in
vacuo (0.5 mmHg) and the remaining pale yellow oil (68 mg) purified
by normal-phase HPLC (50% EtOAc, 50% hexane) to afford 3 (53
24
mg, 90%) as a pale yellow oil: [R]D +52 (c 4.8, CHCl3), 3 from T.
costatus [R]D +48; H and 13C NMR data consistent with those of
24
1

424 Journal of Natural Products, 2008, Vol. 71, No. 3
Van Wyk et al.
the natural product; HRFABMS [M + H]+ m/z 449.2909 (calcd for
C26H41O6, 449.2903).
4.21 (1H, m, H-11b), 4.16 (1H, m, H-11a), 3.43 (3H, s, H3-9′), 2.10
(1H, m, H-1b), 2.04 (2H, m, H2-2), 2.01 (3H, s, OAc-11), 1.89 (1H,
m, H-16b), 1.83 (1H, H, H-1a), 1.74 (1H, m, H-12b), 1.73 (1H, m,
H-17), 1.67 (2H, m, H2-15), 1.52 (1H, m, H-16a), 1.49 (1H, m, H-25),
1.43 (1H, m, H-20), 1.37 (1H, m, H-22b), 1.36 (1H, m, H-23a), 1.24
(1H, m, H-12a), 1.17 (1H, m, H-23a), 1.14 (2H, m, H2-24), 1.01 (1H,
m, H-22a), 0.98 (3H, d, J ) 6.7 Hz, H3-21), 0.86 (3H, d, J ) 6.6 Hz,
H3-27), 0.86 (3H, d, J ) 6.6 Hz, H3-26), 0.77 (3H, s, H3-18); 13C NMR
(CDCl3, 150 MHz) δ 199.7 (C, C-9), 186.6 (C, C-6), 170.1 (C, OAc-
11), 165.6 (C, C-1′), 153.6 (C, C-8), 143.2 (C, C-5), 138.1 (CH, C-7),
132.5 (C, C-3′), 129.7 (CH, C-7′), 129.7 (CH, C-7′), 129.6 (CH, C-4),
128.5 (CH, C-6′), 127.2 (CH, C-8′), 127.2 (CH, C-4′), 124.1 (C, C-10′),
122.2 (C, C-2′), 67.7 (CH, C-3), 60.9 (CH2, C-11), 55.4 (CH3, C-9′),
50.7 (CH, C-17), 48.9 (C, C-10), 47.8 (C, C-13), 43.8 (CH, C-14),
39.4 (CH2, C-24), 37.3 (CH2, C-12), 35.4 (CH2, C-22), 34.6 (CH, C-20),
28.0 (CH, C-25), 27.4 (CH2, C-15), 26.5 (CH3, C-19), 26.3 (CH2, C-16),
25.4 (CH2, C-1), 24.4 (CH2, C-23), 24.4 (CH2, C-2), 22.5 (CH3, C-27),
22.5 (CH3, C-26), 21.1 (CH3, 11-OAc), 19.0 (CH3, C-21), 17.6 (CH3,
C-18).
Cell Culture. Cells were routinely maintained at 37 °C and 5%
CO2. The human esophageal cancer cell line designated as WHCO1
was derived by Professor R. Veale (University of the Witwatersrand)
from South African patients with squamous cell carcinoma of the
esophagus. The WHCO1 cells were maintained in DMEM, supple-
mented with 10% fetal calf serum, 100 U/mL penicillin, and 100 µg/
mL streptomycin.
MTT Assay. IC50 determinations were carried out using the MTT
kit from Roche (cat. #1465007), according to manufacturer’s instruc-
tions. Briefly, 1500 cells were seeded per well in 96-well plates. Cells
were incubated (24 h), after which aqueous DMSO solutions of each
compound (10 µL, with a constant final concentration of DMSO )
0.1%) were plated at various concentrations. After 48 h incubation,
observations were made, and MTT (10 µL) solution added to each well.
After a further 4 h incubation, solubilization solution (100 µL) was
added to each well, and plates were incubated overnight. Plates were
read at 595 nm on an Anthos microplate reader.
Selective Acetylation of 4. The 9,11-secosterol 4 (62 mg, 0.14
mmol) and collidine (0.28 mmol, 2 equiv) were taken up in anhydrous
CH2Cl2 (1 mL) and cooled to -78 °C under an Ar atmosphere. A
solution of acetyl chloride (0.19 mmol, 2 equiv) in anhydrous CH2Cl2
(120 µL) was added in a dropwise fashion and stirred at -78 °C (3 h).
The solution was quenched with H2O (1 mL) and concentrated to
dryness in vacuo (0.5 mmHg). The resultant yellowish solid (103 mg)
was taken up in EtOAc and passed through a Teflon micropore filter
to remove the whitish precipitate that formed and the yellow filtrate
concentrated to a yellow oil (71 mg). The yellow oil was subjected to
diol HPLC (40% EtOAc, hexane 60%) to afford both 14 (6 mg, 9%)
and 3R-hydroxy-11-acetoxy-9,11-seco-cholest-4,7-dien-6,9-dione (13,
1
17 mg, 27%). Analysis of the H NMR spectrum of each of the two
fractions suggested that 14 was the diacetylated product (not further
characterized) and that 13 was the desired OAc-11 product.
3r-Hydroxy-11-acetoxy-9,11-seco-cholest-4,7-dien-6,9-dione (13):
yellow oil; [R]D34 +19 (c 0.6, CHCl3); 1H NMR (CDCl3, 600 MHz)
δ 6.69 (1H, dd, J ) 4.0, 0.9 Hz, H-4), 6.66 (1H, s, H-7), 4.33 (1H, m,
H-3), 4.19 (1H, m, H-11b), 4.15 (1H, m, H-11a), 3.44 (1H, dd, J )
11.2, 8.4 Hz, H-14), 2.18 (1H, ddd, J ) 13.5, 13.3, 3.4 Hz, H-1b),
2.01 (3H, s, OAc-11), 1.89 (1H, m, H-16b), 1.87 (2H, m, H2-2), 1.76
(1H, m, H-1a), 1.73 (2H, m, H-12b and H-17), 1.66 (2H, m, H2-15),
1.50 (2H, m, H-16a and H-25a), 1.43 (1H, m, H-20), 1.35 (2H, m,
H-22b and H-23b), 1.35 (3H, s, H3-19), 1.26 (1H, m, H-12a), 1.14
(1H, m, H-23a), 1.13 (2H, m, H2-24), 1.01 (1H, m, H-22a), 0.97 (3H,
d, J ) 6.7 Hz, H3-21), 0.85 (3H, s, H3-27), 0.85 (3H, s, H3-26), 0.77
(3H, s, H3-18); 13C NMR (CDCl3, 150 MHz) δ 200.4 (C, C-9), 187.5
(C, C-6), 171.0 (C, OAc-11), 153.4 (C, C-8), 140.5 (C, C-5), 138.2
(CH, C-7), 135.3 (CH, C-4), 63.1 (CH, C-3), 61.0 (CH2, C-11), 50.7
(CH, C-17), 49.1 (C, C-10), 47.7 (C, C-13), 43.7 (CH, C-14), 39.4
(CH2, C-24), 37.2 (CH2, C-12), 35.4 (CH2, C-22), 34.6 (CH, C-20),
27.9 (CH, C-25), 27.3 (CH2, C-15), 27.2 (CH2, C-2), 26.7 (CH3, C-19),
26.3 (CH2, C-16), 25.0 (CH2, C-1), 24.4 (CH2, C-23), 22.7 (CH3, C-26),
22.5 (CH3, C-27), 21.1 (CH3, OAc-11), 18.9 (CH3, C-21), 17.5 (CH3,
C-18); HRFABMS m/z 472.3188 (calcd for C29H44O5 [M+], 472.3189).
Preparation of the (R)- and (S)-MTPA Esters of 13. This method
is representative. (R)-R-Methoxy-R-trifluoromethylphenylacetic acid (6
mg), DCC (48 mg), and DMAP (7 mg) were added to a solution of 13
in anhydrous CH2Cl2 (2 mL). The solution was shaken periodically
over period of 1.5 h at ambient temperature, diluted with EtOAc (3
mL) and H2O (1 mL), and filtered. The resulting solution was washed
with 1 M HCl (1 mL), H2O (1 mL), saturated aqueous NaHCO3 (1
mL), and once more with H2O (1 mL) in the order presented. The
resulting colorless oil (7 mg) was purified by semipreparative HPLC
(25% EtOAc, 75% hexane) to afford 15 (6 mg). The (S)-MTPA ester
(16) was similarly prepared from 13, using (S)-R-methoxy-R-trifluo-
romethylphenylacetic acid.
Acknowledgment. Financial assistance from the South African
National Research Foundation (NRF), the Department of Environmental
Affairs and Tourism, DAAD (Ph.D. bursary A.v.W.) is gratefully
acknowledged. Professor L. Fourie (North West University) is thanked
for assistance with the acquisition of HRFABMS data. M.R.C.
acknowledges research support from the University of Cape Town and
the NRF.
Supporting Information Available: This material is available free
of charge via the Internet at http://pubs.acs.org.
References and Notes
1
(R)-MTPA ester (15): yellow oil; H NMR (CDCl3, 600 MHz) δ
(1) Walsby, J. R.; Morton, J. E.; Croxall, J. P. J. Zool. London 1973,
171, 257–283.
7.45 (2H, m, H-4′ and H-8′), 7.37 (3H, m, H-5′, H-6′ and H-7′), 6.68
(1H, s, H-7), 6.65 (1H, dd, J ) 4.4, 0.9 Hz, H-4), 5.61 (1H, m, H-3),
4.20 (1H, m, H-11b), 4.16 (1H, m, H-11a), 3.49 (3H, s, H3-9′), 3.43
(1H, dd, J ) 11.1, 8.2 Hz, H-14), 2.01 (3H, s, OAc-11), 1.99 (2H, m,
H2-2), 1.98 (1H, m, H-1b), 1.89 (1H, m, H-16b), 1.79 (1H, m, H-1a),
1.74 (1H, m, H-17), 1.72 (1H, m, H-12b), 1.67 (2H, m, H2-15), 1.53
(1H, m, H-16a), 1.51 (1H, m, H-25), 1.45 (1H, m, H-20), 1.37 (2H, m,
H-22b and H-23b), 1.37 (3H, s, H3-19), 1.26 (1H, m, H-12a), 1.14
(3H, m, H-23a and H2-24), 1.01 (1H, m, H-22a), 0.98 (3H, d, J ) 6.7
Hz, H3-21), 0.86 (3H, d, J ) 6.6 Hz, H3-27), 0.86 (3H, d, J ) 6.6 Hz,
H3-26), 0.77 (3H, s, H3-18); 13C NMR (CDCl3, 150 MHz) δ 199.7 (C,
C-9), 185.6 (C, C-6), 171.0 (C, OAc-11), 165.8 (C, C-1′), 153.7 (C,
C-8), 143.4 (C, C-5), 138.1 (CH, C-7), 132.5 (C, C-3′), 129.7 (CH,
C-4), 129.6 (CH, C-7′), 129.6 (CH, C-5′), 128.5 (CH, C-6′), 127.1 (CH,
C-8′), 127.1 (CH, C-4′), 124.1 (C, C-10′), 122.2 (C, C-2′), 67.5 (CH,
C-3), 60.9 (CH2, C-11), 55.5 (CH3, C-9′), 50.8 (CH, C-17), 48.9 (C,
C10), 47.8 (CH, C-13), 43.8 (CH, C-14), 39.4 (CH2, C-24), 37.3 (CH2,
C-12), 35.4 (CH2, C-22), 34.6 (CH, C-20), 28.0 (CH, C-25), 27.4 (CH2,
C-15), 26.4 (CH3, C-19), 26.3 (CH2, C-16), 25.2 (CH2, C-1), 24.5 (CH2,
C-23), 24.1 (CH2, C-2), 22.8 (CH3, C-26), 22.5 (CH3, C-27), 21.1 (CH3,
OAc-11), 19.0 (CH3, C-21), 17.6 (CH3, C-18).
(2) Manker, D. C.; Faulkner, D. J. Tetrahedron 1987, 43, 3677–3680.
(3) Manker, D. C.; Faulkner, D. J. J. Chem. Ecol. 1995, 22, 23–35.
(4) Gray, C. A.; Davies-Coleman, M. T.; McQuaid, C. Nat. Prod Lett.
1998, 12, 47–53.
(5) Rice, S. H. J. Exp. Mar. Biol. Ecol. 1985, 93, 83–89.
(6) Diaz-Marrero, A. R.; Dorta, E.; Cueto, M.; Rovirosa, J.; San Martin,
A.; Loyola, A.; Darias, J. Tetrahedron 2003, 59, 4805–4809.
(7) Diaz-Marrero, A. R.; Dorta, E.; Cueto, M.; Rovirosa, J.; San Martin,
A.; Loyola, A.; Darias, J. ARKIVOC 2003, 10, 107–117.
(8) San Martin, A.; Quezada, E.; Soto, P.; Palacios, Y.; Rovirosa, J. Can.
J. Chem. 1996, 74, 2471–2475.
(9) Rovirosa, J.; Quezada, E.; San Martin, A. Bol. Soc. Chil. Quim 1992,
37, 143–145.
(10) Pathak, A.; Aslaoui, J.; Morin, C. J. Org. Chem. 2005, 70, 4184–
4187.
(11) Davies-Coleman, M. T. In Progress in Molecular and Subcellular
Biology-Molluscs; Cimino, G., Gavagnin, M., Eds.; Springer Verlag:
Berlin, 2006; Vol. 43,Chapter 6, pp133–157..
(12) Crystal data for 11: C36H52O10, M ) 644.78, colorless plate, 0.12 ×
0.12 × 0.09 mm3, monoclinic, space group P21 No. 4), a ) 10.4893(2)
Å, b ) 7.7835(2) Å, c ) 21.1003(5) Å, ꢀ ) 93.518(1)°, V )
1719.46(7) Å3, Z ) 2, Dc ) 1.245 g/cm3, F000 ) 696, Nonius Kappa
CCD diffractometer, Mo KR radiation, λ ) 0.71073 Å, T ) 113(2)
1
(S)-MTPA ester (16): yellow oil; H NMR (CDCl3, 600 MHz) δ
7.46 (2H, m, H-4′ and H-8′), 7.37 (3H, m, H-5′, H-6′ and H-7′), 6.67
(1H, s, H-7), 6.56 (1H, dd, J ) 4.3, 0.9 Hz, H-4), 5.61 (1H, m, H-3),
K, 2θmax ) 51.3°, 6480 reflections collected, 6480 unique (Rint
)
0.0000). Final GooF ) 1.049, R1 ) 0.0573, wR2 ) 0.1399, R indices
based on 5199 reflections with I >2σI (refinement on F2), 430

BioactiVe Metabolites from the Mollusk Trimusculus costatus
Journal of Natural Products, 2008, Vol. 71, No. 3 425
parameters, 1 restraint. Lp corrections applied, µ ) 0.090 mm-1. H
atoms added in idealized positions in a riding model with isotropic
thermal parameters 1.2–1.5× those of the parent atoms. Crystal-
lographic data for the structure reported in this paper have been
deposited with the Cambridge Crystallographic Data Centre (CCDC
deposition number 665480). Copies of the data can be obtained, free
of charge, on application to the Director, CCDC, 12 Union Road,
Cambridge CB2 IEZ, UK (fax: + 44-(0)1223-336033 or e-mail:
deposit@ccdc.cam.ac.uk).
(19) Pickens, A.; Orringer, M. B. Ann. Thorac. Surg. 2003, 76, S1367–
S1369.
(20) Bohr, U. R. M.; Segal, I.; Primus, A.; Wex, T.; Hassan, H.; Ally, R.;
Malfertheiner, P. Helicobacter 2003, 8, 608–612.
(21) Mqoqi, N.; Kellett, P.; Madhoo J.; Sitas F. National Cancer Registry
of South Africa. Incidence and Geographical Distribution of Histologi-
cally Diagnosed Cancer in South Africa, 1996–1997; South African
Institute for Medical Research: Johannesburg, 2003.
(22) Hendricks, D. T.; Parker, M. I. Int. Union Biochem. Mol. Biol. Life
2002, 53, 263–268.
(23) Day, N. E. Cancer Res. 1975, 35, 3304–3307.
(24) Devesa, S. S.; Blot, W. J.; Fraumeni, J. F. Cancer 1998, 83, 2049–
2053.
(13) Schmidt, T. J.; Passreiter, C. M.; Wendisch, D.; Willuhn, G. Phy-
tochemistry 1995, 40, 1213–1218.
(14) Aiello, A.; Esposito, G.; Fattorusso, E.; Iuvone, T.; Luciano, P.; Menna,
M. Steroids 2003, 68, 719–723.
(15) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J. Am. Chem.
Soc. 1991, 113, 4092–4096.
(16) Gunatilaka, A. A. L.; Gopichand, Y.; Schmitz, F. J.; Djerassi, C. J.
Org. Chem. 1981, 46, 3860–3866.
(25) Rajput, J.; Moss, J. R.; Hutton, A. T.; Hendricks, D. T.; Arendse,
C. E.; Imrie, C. J. Organomet. Chem. 2004, 689, 1553–1568.
(26) Whibley, C. E.; McPhail, K. L.; Keyzers, R. A.; Maritz, M. F.; Leaner,
V. D.; Davies-Coleman, M. T.; Hendricks, D. T. Mol. Cancer Ther.
2007, 6, 2535–2543.
(17) Sera, Y.; Adachi, K.; Shizuri, Y. J. Nat. Prod. 1999, 62, 152–154.
(18) Jiménez, C.; Quinoa, E.; Castedo, L.; Riguera, R. J. Nat. Prod. 1986,
49, 905–909.
NP070612Y