Aplysiadiol from Aplysia dactylomela suggested a key intermediate for a unified biogenesis of regular and irregular marine algal bisabolene-type metabolites

Article abstract of DOI:10.1016/j.tet.2006.07.078

The rearranged compound 1, the regular 2 and 3 and the degraded 4 are novel bisabolene-type metabolites that along with the known compounds caespitol, 8-acetylcaespitol, caespitane, caespitenone, laucapyranoid A, and furocaespitane have been isolated from

Full text of DOI:10.1016/j.tet.2006.07.078

Tetrahedron 62 (2006) 9655–9660
Aplysiadiol from Aplysia dactylomela suggested a key intermediate
for a unified biogenesis of regular and irregular marine algal
bisabolene-type metabolites
*
´
Inmaculada Brito, Teresa Dias, Ana R. Dıaz-Marrero, Jose Darias and Mercedes Cueto
ꢀ
´
´
Instituto de Productos Naturales y Agrobiologıa del CSIC, Avda. Astrofısico F. Saꢀnchez, 3, Apdo. 195,
38206 La Laguna, Tenerife, Spain
Received 1 June 2006; revised 14 July 2006; accepted 25 July 2006
Abstract—The rearranged compound 1, the regular 2 and 3 and the degraded 4 are novel bisabolene-type metabolites that along with the
known compounds caespitol, 8-acetylcaespitol, caespitane, caespitenone, laucapyranoid A, and furocaespitane have been isolated from
the sea hare Aplysia dactylomela. The structures of these compounds were determined on the basis of spectroscopic evidence. The irregular
network of 1 suggested that the halogenated epoxide 6 is a key intermediate for a unified biogenetic pathway of naturally occurring marine
algal bisabolene-type metabolites.
Ó 2006 Elsevier Ltd. All rights reserved.
X
1. Introduction
X
Molluscs of the genus Aplysia are one among the best stud-
ied of all marine invertebrates, serving as a model system for
neurological studies.1 Aplysia dactylomela (Opisthobran-
chia: Anaspidea) like other sea hares thrive on algae, acquir-
ing and storing many algal metabolites in their digestive
gland.2 This species alone sequesters some 80 different sec-
ondary metabolites from its algal diet,3 but their function is
a subject of debate because the dynamics of metabolites of
sea hares is poorly understood.4 Laurencia caespitosa,
a red alga common in the diet of A. dactylomela from the
Canary Islands, was the first source5 of a marine algal bis-
abolene-type sesquiterpenoid, and since then a number of
metabolites belonging to this skeletal class has been found
in other taxa including sponges, nudibranchs, octocorals,
and, more recently, microorganisms.
Y
O
Y
X or Y: nitrogenous X and Y: halogen
functionality functionality
Figure 1. Representative structural classes of bisabolene-type metabolites
from diverse marine taxa.
are aromatic, sharing with sponges some identical but,
interestingly, antipodal metabolites related to (ꢀ)-curcu-
phenol.22–25 Recently discovered microbial bisabolene
compounds26 were also related to the curcumene skeleton
whereas bisabolene products from red algae are mostly
characterized, unlike the other taxa, by high halogen
content.5,27–29
From a survey on marine bisabolene metabolites interesting
analogies can be observed regarding structural and/or func-
tional features. For example, all bisabolene derivatives
from sponges can be grouped into two structural classes.
One class of metabolites bear an aromatic ring6–14 related
to (+)-curcuphenol and the other features a diverse nitroge-
nous substitution pattern in a non-aromatic network15–21
(Fig. 1). Moreover, all bisabolene compounds from octocoral
A. dactylomela has proved to be a rich source of structural
variety of bisabolenic metabolites, depending on where
they are located.5,30–33 We now report on the discovery of
new bisabolene derivatives 1–4 from this species, along with
the previously characterized caespitol,5 furocaespitane,27
laucapyranoid A,28 8-acetylcaespitol,32 caespitane,28,32 and
caespitenone28,32 (Fig. 2). Most of the compounds reported
here are halogenated, following the general pattern of algal
bisabolene sesquiterpenoids, 1 and 4 being a rearranged
and a degraded bisabolene skeleton derivative, respectively.
* Corresponding author. Tel.: +34 922 252 144; fax: +34 922 260 135;
e-mail: mcueto@ipna.csic.es
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.07.078

9656
I. Brito et al. / Tetrahedron 62 (2006) 9655–9660
Cl
15
14
15
7 OH
3
6
14
13
3
2
6
Br
O
Br
8 H
O
1
O
7H
1
12
8
10
H
O
10
Br
OH
12
OH 13
1
2
3
Cl
Cl
Cl
12
B
H
6
1
O
A
Br
Br
8
Br
H
O
O
10
O
11
H
OH
Br
7 OH
4: βOH
caespitol
caespitenone
4a: αOH
Figure 2. Metabolites from A. dactylomela.
2. Results and discussion
A. dactylomela was collected off the SW coast of La Palma
(Canary Islands). From the crude extracts the new com-
pounds 1–4 were isolated after gel filtration on Sephadex
LH-20 chromatography followed by successive HPLC.
Aplysiadiol 1 was a colorless oil and its NMR spectroscopic
data are summarized in Table 1. Its EIMS spectrum showed
peaks at m/z 413/415/417/419 with relative intensities
suggestive of one chlorine and two bromine atoms, which
correspond to the empirical formula C14H20O2ClBr2
[MꢀMeꢀH2O]+ (HREIMS). Absorption for a carbonyl and
a hydroxyl group at 1722 and 3448 cmꢀ1, respectively,
were observed in its IR spectrum. The 13C NMR and DEPT
spectra of 1 showed the presence of 15 carbon signals
assigned to 4ꢁCH3, 4ꢁCH2, 3ꢁCH (two bearing bromine),
and fourquaternary carbons (one ketone). The 1H NMRspec-
trum displayed signals for two protons geminal to bromine at
d 3.67 (dd, J¼2.1, 9.5 Hz) and d 4.27 (dd, J¼4.3, 12.8 Hz);
eight methylene protons between d 2.55 and 1.60; a methine
proton at 1.84 (br s); and four methyl groups [d 2.30 (s), d 1.68
(s), d 1.35 (s), d 1.32 (s)]. According to the degree of unsatu-
ration, the molecular formula of 1 expressed a monocyclic
network.
Connectivity information obtained from COSY, HSQC, and
HMBC experiments unambiguously determined a cyclo-
hexyl moiety with a halogen substitution pattern identical
with that in the corresponding ring B of caespitol (Fig. 2). The
complementary half of the molecule attached to the ring is
a highly functionalized open chain where the regiochemistry
of a carbinolic bromohydrin, involving a gem-dimethyl
group, becomes readily apparent by the mutual HMBC
correlations C-13/H3-12, C-12/H3-13 and those with H-10
and C-11, respectively. The COSY correlation of H-10 with
a remaining methylene placed a carbinolic methyl ketone
at C-8, establishing the planar structure of 1 as an irregular
sesquiterpene.
The coupling constant of H-2 (J¼4.3, 12.8 Hz) as well as the
13C-chemical shifts of C-2, C-3, and C-15 are in agreement
with the regiochemistry and configuration of a vicinal chloro-
bromo system on a cyclohexane ring with a chair confor-
mation such as that of caespitol (Table 1). An equatorial
disposition of the side chain residue on the ring was

I. Brito et al. / Tetrahedron 62 (2006) 9655–9660
9657
O
ring moiety bearing the same heteroatoms, regiochemistry
and relative configuration as those of comparable atoms of
ring A of caespitol (Table 1). The remaining two degrees of
unsaturation correspond to a trisubstituted double bond in
a residual cyclohexene ring, which suggests that the typical
chlorobromo system of ring B of caespitol was lost in com-
pound 2. On the other hand, it has been reported that the
LiAlH4 reduction of caespitol yielded a compound5 with
same planar structure as 2. The 13C NMR chemical shifts re-
ported later35 for this synthetic compound were very similar
to those of 2 except for the corresponding C-3 and C-15
carbon atoms (DdC-3¼5.6, DdC-15¼10 ppm). Since ring A
of both natural and synthetic compounds are identical in all
respects, and because such differences are improbable for a
putative diastereomer at C-6, we suspect these values were
mistaken or incorrectly assigned. In order to corroborate
this point, caespitol was efficiently dehalogenated by reac-
tion with Zn–NaI in DMF at 115 ꢂC to give a compound
identical with that depicted in 2. Thus, the correct NMR
data of 2 are assigned in Table 1.
14
15
H
7
8
R
H
10
Br
5
S
12
4
6
3
R
O
9
1
Br
Cl
S
2
11
H
S
O
H
H
H
13
H
Figure 3. Selected NOEs of compound 1.
established by the strong NOE observed between H-2 and
H-6 (Fig. 3).
The energy-minimized conformation of 1 deduced by mo-
lecular mechanics34 is shown in Figure 3. This conformer
showed a hydrogen bond between the hydroxyl groups so
that the proton of the hydroxyl at C-8 and the trans-diaxial
proton at C-9 are at a suitable interatomic distance to allow
the observed NOE between them. For the acyl group at C-8
an R configuration is given to allow an atomic cluster com-
patible with the H2-9/H-10 J-coupling and the observed
NOE. This point has been further reinforced by comparison
of the NMR data of 1 with those of a related compound 1a
generated by spontaneous rearrangement of laucopyranoid
C and whose absolute configuration was established by
X-ray crystallography.28 Whereas the 13C chemical shifts of
the carbons of the cyclohexane ring are almost identical in
both compounds, the inversion of the configuration at C-8
of 1 produced a variation in the chemical shift of this carbon
and the comparable vicinal carbons. For 13C chemical shifts
of 1a see the experimental part. This spectroscopic correla-
tion allowed us to propose for 1 the 2S, 3S, 6S, 8R, 10R
configuration depicted in Figure 3.
Deschlorobromo caespitenone 3 was isolated as an oil. NMR
spectroscopic data coupled with a molecular ion peak at m/z
234 [M]+ (HREIMS) suggested a molecular formula of
1
C15H22O2 indicating five degrees of unsaturation. The H
and 13C NMR data of 3 are similar to those of the known
compound caespitenone,28,32 also isolated in this work.
The main difference is that the chlorobromo system charac-
teristic of ring B of caespitenone has been lost in 3, leading
to the corresponding cyclohexene ring, which is in accord
with the degrees of unsaturation. Thus, 3 is the dehalogen-
ated derivative of caespitenone. This was corroborated by
oxidation–dehydration of compound 2 with Jones reagent
to give a compound identical with that depicted in 3.
Since the optical rotation of caespitol isolated in this study
is coincident with that previously reported,28 the chemical
transformation of caespitol into compounds 2 and 3
allowed us to establish their absolute streochemistries as
6S,7R,8R,10R and 6S,7R, respectively.
Cl
Cl
OH
3
2
OH
S
6
spontaneous
rearrang.
O
Br
O
S
8 HS
H
S
Br
10
S
Br
laucopyran
O
C
1a
oid
Furocaespitanelactol 4 was isolated as a white powder. Its
EIMS spectrum showed peaks at m/z 322/324/326 with
relative intensities suggestive of one chlorine and one bro-
mine atom, which correspond to the empirical formula
C12H16O3ClBr [M]+ (HREIMS). The molecular formula of
4 resembled that of furocaespitane, a degraded bisabolene-
type derivative previously isolated by us from L. caespitosa27
and also now isolated in this work from A. dactylomela.
Deschlorobromo caespitol 2 was obtained as a white powder.
The EIMS spectrum showed peaks at m/z 316/318, with rel-
ative intensities for one bromine atom. NMR data coupled
withthe[M]+ peakintheHREIMSof2suggestedamolecular
formula C15H25O2Br with three degrees of unsaturation indi-
cating that the molecule is bicyclic. The 13C NMR and DEPT
spectra of 2 (Table 1) showed the presence of 15 carbon sig-
nals assigned to 4ꢁCH3, 4ꢁCH2, 4ꢁCH (one olefinic and
two bearing heteroatoms), and three quaternary carbons
(one olefinic and two bearing heteroatoms). The following
1H NMR signals were observed: one olefinic proton at
d 5.32 (br s); two protons geminal to heteroatom [d 4.33
(dd, J¼4.1, 13.2 Hz), d 3.52 (ddd, J¼3.6, 3.6, 6.2 Hz)]; one
methine proton at d 1.70 (br s); eight methylene protons
between d 2.03 and 1.19; four methyl groups [d 1.62 (br s),
d 1.40 (s), d 1.32 (s), d 1.13 (s)], and one D2O interchangeable
proton at d 1.84 (d, J¼6.4 Hz).
Oxidation of furocaespitane with m-chloroperbenzoic acid
has been reported to give an epimeric mixture27 at C-7 of
butenolides 4 and 4a (Fig. 2). The spectroscopic data of our
single naturally occurring compound indicates that its struc-
ture is a lactol identical to one of the epimers present in the
synthetic mixture. The energy-minimized conformation of
both epimers 4 and 4a deduced by molecular mechanics is
shown in Figure 4.
The orthogonal joining of the lactone and cyclohexane rings
placed the methyl group of the respective carbinols of each
epimer at such a similar distance from H-6 that it does not
allow us to discern to which one of the epimers the NOE
Connectivity information obtained from COSY, HSQC, and
HMBC as well as NOESY experiments secured an oxane

9658
I. Brito et al. / Tetrahedron 62 (2006) 9655–9660
H
H
X
O
O
caespitol
H
O
isocaespitol
caespitenone
6-hydroxy-
caespitol
H
9
Y
H
Br
5
5
O
12
3
4
H
caespitane
H
1
6
O
O
Cl
2
Br
Cl
Br
path c
H
R
S
Br+,-OH/Cycliz.
H
H
H
H
H
O
H
Cl
O
H
rearrang.
O
4
4a
E= 25.0546 Kcal/mol
Br
[O]
X
X
E= 24.666 Kcal/mol
Figure 4. Selected NOEs of 4 and 4a.
Y
Y
H
X= Cl or Br
O
Y= Br or Cl
5
6
7
observed between CH3-11 and H-6 corresponds. However,
the interatomic distance between the hydrogen of the hy-
droxyl group and the equatorial proton H2-1 of 4 measured
by the program is 2.446 A, whereas in the epimer 4a the dis-
˚
tance to H-5eq is 3.111 A, which appears to be too large for
a clear NOE to be expected. Thus, the observed NOE in the
natural compound was assigned to the epimer with an S
stereochemistry at C-7. Because the remaining chiral centers
of the compound are identical with the corresponding cen-
ters of furocaespitane, the stereochemistry of the degraded
bisabolene metabolite was assigned as depicted in 4 (Fig. 4).
puertitol A
puertitol B
path
b
a
path
Br+/ -OH
Cl
Box B
Br+
Box A
˚
Cl
cycliz./
-H O
2
O
Br
Br
O
Br
H
H O
2
H
8
9
[O]
Cl
Br
OH
-H+
OH2
O
Cl
OH
O
Br
cycliz./
-H2O
Br
H
On standing, 4 isomerizes to 4a until a 1:1 equilibrium mix-
ture is reached (Fig. 5). This is the first time that 4 has been
isolated as a natural product. Due to the lack of 13C NMR
data and assigned 1H NMR chemical shifts, we provide com-
plete NMR data of the natural compound 4 in Table 1 and
those of 4a in the experimental part.
Br
1
H
laucopyranoid A
OH
Cl
Br
H2O
H2O
H+
OH
Cl
Br
O
H
Br
+H2O
-H2O
10
O
A bisabolyl cation is expected to be converted into the pos-
sible precursor of the naturally occurring algal bisabolene
derivatives, the true producer of this type of metabolite
found in A. dactylomela. It appears that in this process the
trienic system of the bisabolene is sequentially oxidized,
starting with chlorobromo addition to the endocyclic double
bond. This seems to be supported by finding the natural
halogenated addition products puertitols A and B36 by stabi-
lization of the intermediate 5 (Fig. 6).
Br
H
OH
Cl
B
anoid
laucopyr
Br
OH2
O
+
Br
O
H
Cl
11
O
OH
Br
OH
H
Cl
C
anoid
laucopyr
H
Br
Br
O
spontaneous
rearrang.
furocaespitane
Although 6, a putative derivative from a second oxidation
step of the central double bond of the halogenated bisabolol
5, is unknown at present, it can be postulated as a precursor
in the biogenesis of these compounds. The nascent interme-
diate 6 rearranges, by 1,2-migration of the bond connecting
both the linear and cyclic halves induced by epoxide ring
opening, to 7. The transformation of 6/7 is a key step to
account for naturally occurring marine algal irregular and
degraded bisabolene-type network metabolites.
1a
furocaespitanelactol 4
Figure 6. Unified biogenesis for marine algal bisabolene-type metabolites.
Hydration of laucopyranoid A may afford the epimeric lau-
copyranoid B and laucopyranoid C. Whereas laucopyranoid
C spontaneously rearranges to 1a, laucopyranoid B as well
as aplysiadiol 1, derived from extensive oxidation of 7,
Box B, may evolve to degraded furocaespitane and related
lactol 4 following path b indicated in Figure 6. This route
involves dehydration of 1 and/or hydration/dehydration of
laucopyranoid B to a conjugated ketone 10, formation of
a furane ring by intramolecular displacement of the bromine
atom and, finally, trapping of the resulting carbocation 11 by
loss of the terminal isopropyl carbinol as acetone.
From 7, and in virtue of the biogenetically interconvertible
849, alternative path a and/or path b may provide a non-
regular network 8 (Box A) from which laucopyranoid A
could be generated by stabilization of carbocation 8.
R
Me
R
R
O
O
O
O
O
O
O
OH
H
OH
On the other hand, the key precursor 6 may evolve by oxidiz-
ing the remaining olefin, the terminal isopropylidene, to
a fully oxidized bromohydrin derivative 12 (Box C) from
4a
4
Figure 5. Equilibrium leading to a mixture of epimers at C-7.

I. Brito et al. / Tetrahedron 62 (2006) 9655–9660
9659
which the regular sesquiterpenoid caespitol, and related
compounds such as isocaespitol,37 caespitenone,32 and 6-
hydroxycaespitol28 may originate through path c. Cyclization
of the Br+/ꢀOH addition product of 5 will produce caespi-
tane. Thus, we propose a unified pathway for the biogenesis
of regular, irregular rearranged, and degraded bisabolene-
type metabolites isolated from red algae of genus Laurencia
and the sea hares that thrive on them.
were dissected and their digestive system along with the
mantle were separated and analyzed independently.
3.3. Extraction and isolation
A. dactylomela digestive glands were extracted with acetone
at room temperature. The extract was concentrated to give
a dark green residue (31.0 g) and partitioned with H2O–
CH2Cl2. The resulting fraction of CH2Cl2 (7.5 g) was then
submitted to a gel filtration column to give fraction A
(822.9 mg), which was subjected to flash chromatography
on Si gel. The fraction eluted with hexane–EtOAc (8:2)
gave a mixture that was further separated by HPLC
(Jaigel-sil column 20ꢁ250 mm, flow 4.5 ml/min, hexane–
EtOAc (8:2)) to yield the new sesquiterpene 1 (0.7 mg)
and the known compound 8-acetylcaespitol (35.0 mg), caes-
pitane (14.0 mg), and laucapyranoid A (4.0 mg). Fraction D
was processed using flash chromatography on Si gel eluting
with hexane–EtOAc (1:1) to give a mixture that was further
separated by HPLC (Jaigel-sil column, flow 4.5 ml/min,
hexane–EtOAc (1:1)) to yield the new compounds 2
(5.5 mg) and 4 (9.0 mg), and the known compound caespite-
none (4.9 mg). Caespitol (73.6 mg) was isolated from frac-
tion F by crystallization.
Until now all reported marine algal bisabolene-type metab-
olites bear a vicinal chlorobromo system on their carbocyclic
ring. However, the corresponding dehalogenated derivatives
of caespitol and caespitenone, compounds 2 and 3, respec-
tively, have been found for the first time as naturally occur-
ring metabolites in A. dactylomela, suggesting that these
compounds have been chemically transformed by the mol-
lusc to store them as less toxic compounds. This statement
is supported by the recent finding that the naturally occurring
deschloroelatol found in A. dactylomela is less toxic than the
co-occurring chlorinated elatol.33 This resembles the
known38,39 strategy used by A. dactylomela by which some
acquired algal metabolites are occasionally acetylated to
decrease the toxicity33 of the corresponding alcohols.
Furthermore, the butenolide unit is widespread in natural
products from terrestrial to marine organisms and irrespec-
tive of an isoprenic or polyketide origin, has been purported
to possess enzyme-inhibiting functions.40 Plants appear to
use lactones to prevent being eaten and to avoid biofouling
by bacteria while some animals use lactones to regulate their
biochemical processes.40 Since the butenolide 4 does not
occur as an algal metabolite a question arises as to whether
this unit has been biosynthesized by A. dactylomela as an
additional ability (one more to add to acetylation and de-
halogenation) to enhance the fitness of the sea hare.
A. dactylomela mantles were extracted and processed
following the same scheme. Thus, the acetonic extract was
partitioned with H2O–CH2Cl2 and the resulting fraction of
CH2Cl2 (3.8 g) was submitted to a gel filtration column to
give a fraction (308.3 mg), which after flash chromato-
graphy on Si gel eluted with hexane–EtOAc (1:1) gave a mix-
ture that was further separated by HPLC (Jaigel-sil column,
flow 4.5 ml/min, hexane–EtOAc (1:1)) to yield the new com-
pounds 2 (30.0 mg), 3 (1.5 mg), and 4 (1.3 mg) together with
the known metabolites caespitol (3.1 mg), 8-acetylcaespitol
(3.3 mg), and furocaespitane (1.4 mg).
3. Experimental
3.3.1. Compound 1. Colorless oil; [a]2D5 +492 (c 0.047,
CHCl3); IR nmax (film) 3448, 1722 cmꢀ1; 1H and 13C NMR,
see Table 1; EIMS m/z 413/415/417/419 [MꢀMeꢀH2O]+
(<1, <1, <1, <1), 403/405/407/409 [MꢀMeCO]+ (<1,
<1, <1, <1), 385/387/389/391 [MꢀMeCOꢀH2O]+
(36, 87, 62, 11), 237/239/241 [C8H11OBrCl]+ (30, 39, 3),
209/211/213 [C7H11BrCl]+ (9, 12, 3), 173/175 [C7H10Br]+
(13, 13); HREIMS 412.9480 (calcd for C14H20O235Cl79Br2,
412.9518), 402.9596 (calcd for C13H22O235Cl79Br2, 402.9675),
384.9512 (calcd for C13H20O35Cl79Br2, 384.9569).
3.1. General experimental procedures
Optical rotations were measured on a Perkin–Elmer model
343 Plus polarimeter using a Na lamp at 20 ꢂC. IR spectra
were obtainedwith aPerkin–Elmer 1600/FTIR spectrometer.
EIMS and HRMS spectra were taken on a Vg-Micromass
Zab 2F spectrometer. 1H NMR and 13C NMR, HSQC,
HMBC, COSY, and NOESY spectra were measured employ-
ing a Bruker AMX 500 instrument operating at 500 MHz for
1H NMR and at 125.7 MHz for 13C NMR, using CHCl3 as an
internal standard. Two-dimensional spectra were obtained
with the standard Bruker software. HPLC separations were
performed with a Hewlett–Packard HP 1050 (Jaigel-Sil
semipreparative column, 10 m, 20ꢁ250 mm) with hexane–
EtOAc mixture. The gel filtration column (Sephadex LH-
20) used hexane–MeOH–CHCl3 (3:1:1) as solvent. Merck
Si gels 7734 and 7741 were used in column chromatography.
The spray reagent for TLC was H2SO4–H2O–AcOH
(1:4:20).
3.3.2. Compound 1a. 13C NMR (50 MHz, CDCl3) d 210.9
(C-7), 81.5 (C-8), 71.1 (C-3), 62.1 (C-2), 59.3 (C-10), 58.2
(C-11), 44.8 (C-6), 42.3 (C-4), 35.7 (C-1), 35.5 (C-9), 31.1
(C-14), 24.8 (C-13), 23.9 (C-15), 23.0 (C-5), 19.1 (C-12).
3.3.3. Compound 2. White powder; [a]2D5 ꢀ1.72 (c 1.74,
1
CHCl3); IR nmax (film) 3384 cmꢀ1; H and 13C NMR, see
Table 1; EIMS m/z 316/318 [M]+ (<1, <1), 298/300
[MꢀH2O]+ (2, 2), 236 [MꢀHBr]+ (8), 23 (28), 139 (100);
HREIMS 316.1050 (calcd for C15H25O279Br, 316.1037),
298.0900 (calcd for C15H23O79Br, 298.0932), 236.1751
(calcd for C15H24O2, 236.1776).
3.2. Animal material
Eight specimens of A. dactylomela were collected off the
SW coast of La Palma Island at ꢀ1.5 m depth. Specimens
3.3.4. Compound 3. Colorless oil; [a]2D5 ꢀ90.9 (c 0.03,
1
CHCl3); IR nmax (film) 1738 cmꢀ1; H and 13C NMR, see

9660
I. Brito et al. / Tetrahedron 62 (2006) 9655–9660
Table 1; EIMS m/z 234 [M]+ (<1), 219 [MꢀMe]+ (1), 140
(94), 125 (61), 96 (100); HREIMS 234.1665 (calcd for
C15H22O2, 234.1619), 219.1412 (calcd for C14H19O2,
219.1385).
11. El Sayed, K. A.; Yousaf, M.; Hamann, M. T.; Avery, M. A.;
Kelly, M.; Wipf, P. J. Nat. Prod. 2002, 65, 1547–1553.
12. Peng, J.; Franzblau, S. G.; Zhang, F.; Hamann, M. T.
Tetrahedron Lett. 2002, 43, 9699–9702.
13. Gaspar, H.; Feio, S. S.; Rodrigues, A. I.; Van Soest, R. Mar.
Drugs 2004, 2, 8–13.
14. Cichewicz, R. H.; Clifford, L. J.; Lassen, P. R.; Cao, X.;
Freedman, T. B.; Nafie, L. A.; Deschamps, J. D.; Kenyon,
V. A.; Flanary, J. R.; Holman, T. R.; Crews, P. Bioorg. Med
Chem. 2005, 13, 5600–5612.
3.3.5. Compound 4. White powder; [a]2D5 +55 (c 0.25,
CHCl3); IR nmax (film) 3352, 1751 cmꢀ1; 1H and 13C NMR,
see Table 1; EIMS m/z 322/324/326 [M]+ (<1, <1, <1),
307/309/311 [MꢀMe]+ (25, 33, 9), 243/245 [MꢀBr]+ (20,
6), 225/227 [MꢀBrꢀH2O]+ (77, 26), 207 [MꢀHBrꢀCl]+
(100); HREIMS 321.9934 (calcd for C12H16O335Cl79Br,
321.9971), 306.9639 (calcd for C11H13O335Cl79Br,
306.9736), 243.0723 (calcd for C12H16O335Cl, 243.1643),
225.0595 (calcd for C12H14O235Cl, 225.0682), 207.0933
(calcd for C12H15O3, 207.1021).
15. Nakamura, H.; Kobayashi, J.; Ohizumi, Y.; Hirata, Y.
Tetrahedron Lett. 1984, 25, 5401–5404.
16. Sullivan, B. W.; Faulkner, D. J.; Okamoto, K. T.; Chen,
M. H. M.; Clardy, J. J. Org. Chem. 1986, 51, 5134–5136.
17. Gulavita, N. K.; De Silva, E. D.; Hagadone, M. R.; Karuso, P.;
Scheuer, P. J.; Van Duyne, G. D.; Clardy, J. J. Org. Chem. 1986,
51, 5136–5139.
18. Kitagawa, I.; Yoshioka, N.; Kamba, C.; Yoshikawa, M.;
Hamamoto, Y. Chem. Pharm. Bull. 1987, 35, 928–931.
19. Kassulke, K. E.; Potts, B. C. M.; Faulkner, D. J. J. Org. Chem.
1991, 56, 3747–3750.
3.3.6. Compound 4a. 13C NMR (125.7 MHz, CDCl3)
d 172.6 (C-10), 170.4 (C-8), 116.3 (C-9), 106.8 (C-7), 70.2
(C-3), 60.7 (C-2), 41.9 (C-4), 39.7 (C-1), 36.8 (C-6), 28.6
(C-5), 23.9 (2C, C-11, C-12).
20. Fusetani, N.; Wolstenholme, H. J.; Shinoda, K.; Asai, N.;
Matsunaga, S.; Onuki, H.; Hirota, H. Tetrahedron Lett. 1992,
33, 6823–6826.
21. Li, C.-J.; Schmitz, F. J.; Kelly, M. J. Nat. Prod. 1999, 62, 1330–
1332.
22. McEnroe, F.; Fenical, W. Tetrahedron 1978, 34, 1661–1664.
23. Miller, S. L.; Tinto, W. F.; McLean, S.; Reynolds, W. F.; Yu, M.
J. Nat. Prod. 1995, 58, 1116–1119.
3.4. Dehalogenation of caespitol
A solution of caespitol (20.0 mg) in DMF (1.5 ml) was
treated with NaI–Zn and stirred at 115 ꢂC for 2 h. The reac-
tion was quenched with H2O and extracted with EtOAc
to give a mixture that after silica gel chromatography
(hexane–EtOAc, 95:5) gave 7.9 mg of compound 2.
24. Freyer, A. J.; Patil, A. D.; Killmer, L.; Zuber, G.; Myers, C.;
Johnson, R. K. J. Nat. Prod. 1997, 60, 309–311.
25. D’Armas, H. T.; Mootoo, B. S.; Reynolds, W. F. J. Nat. Prod.
2000, 63, 1593–1595.
26. M€ulhaupt, T.; Kaspar, H.; Otto, S.; Reichert, M.; Bringmann,
G.; Lindel, T. Eur. J. Org. Chem. 2005, 334–341.
3.5. Oxidation of compound 2
A solution of compound 2 (12.4 mg) in acetone was treated
with Jones reagent and stirred at room temperature for 1.5 h.
The reaction was quenched with H2O and extracted with
EtOAc to give 8.3 mg of compound 3.
´
27. Gonzalez, A. G.; Darias, J.; Martın, J. D. Tetrahedron Lett.
ꢀ
1973, 3625–3626.
ꢀ
28. Chang, M.; Vazquez, J. T.; Nakanishi, K.; Cataldo, F.; Estrada,
D. M.; Fernandez, J.; Gallardo, A.; Martın, J. D.; Norte, M.;
Acknowledgements
´
ꢀ
ꢀ
´
Perez,R.;Rodıguez, M.L.Phytochemistry1989, 28, 1417–1424.
29. De Carvalho, L. R.; Fujii, M. T.; Roque, N. F.; Kato, M. J.;
Lago, H. J. G. Tetrahedron Lett. 2003, 44, 2637–2640.
30. Hollenbeak, K. H.; Schmitz, F. J.; Hossain, M. B.; Van Der
Helm, D. Tetrahedron 1979, 35, 541–545.
This work was supported by the DGUI Gobierno de Canarias
(PI042005/003, PUB2005/030) and the Ministerio de Edu-
ꢀ
cacion y Ciencia (PPQ2002-02494, SAF2006-03004).
References and notes
31. Schmitz, F. J.; Michaud, D. P.; Hollenbeak, K. H. J. Org. Chem.
1980, 45, 1525–1528.
€
32. Wessels, M.; Konig, G. M.; Wright, A. D. J. Nat. Prod. 2000,
63, 920–928.
33. Dias, T.; Brito, I.; Moujir, L.; Paiz, N.; Darias, J.; Cueto, M.
J. Nat. Prod. 2005, 68, 1677–1679.
34. PCModel (v, 7.0). Serena Software: Bloomington, IN.
1. Carew, T. J.; Kandel, E. R. J. Neurophysiol. 1977, 40, 692–707.
2. Stallard, M. O.; Faulkner, D. J. Comp. Biochem. Physiol. 1974,
49B, 25–35.
3. Dias, T. Tesina de Licenciatura; CSIC-Universidade dos
Azores: La Laguna, 2004.
´
´
35. Gonzalez, A. G.; Martın, J. D.; Martın, V. S.; Norte, M.
ꢀ
Tetrahedron Lett. 1979, 2719–2722.
4. Rogers, C. N.; De Nys, R.; Charlton, T. S.; Steinberg, P. D.
J. Chem. Ecol. 2000, 26, 721–744.
36. Vazquez, J. T.; Chang, M.; Nakanishi, K.; Martin, J. D.; Martin,
V. S.; Perez, R. J. Nat. Prod. 1988, 51, 1257–1260.
ꢀ
´
5. Gonzalez, A. G.; Darias, J.; Martın, J. D.; Perez, C. Tetrahedron
ꢀ
Lett. 1974, 1249–1250.
6. Fusetani, N.; Sugano, M.; Matsunaga, S.; Hashimoto, K.
Experientia 1987, 43, 1234–1235.
7. Wright, A. E.; Pomponi, S. A.; McConnell, O. J.; Kohmoto, S.;
McCarthy, P. J. J. Nat. Prod. 1987, 50, 976–978.
8. Butler, M. S.; Capon, R. J.; Nadeson, R.; Beveridge, A. A.
J. Nat. Prod. 1991, 54, 619–623.
´
ꢀ
ꢀ
37. Gonzalez, A. G.; Darias, J.; Martın, J. D.; Perez, C.; Sims, J. J.;
Lin, G. H. Y.; Wing, R. M. Tetrahedron 1975, 31, 2429–2452.
38. Stallard, M. O.; Faulkner, D. J. Comp. Biochem. Physiol. B
1974, 49, 37–41.
39. Pennings, S. C.; Weiss, A. M.; Paul, V. J. Mar. Biol. 1996, 126,
735–743.
40. Konaklieva, M. I.; Plotkin, B. J. Mini Rev. Med. Chem. 2005, 5,
73–95.
9. Harrison, B.; Crews, P. J. Org. Chem. 1997, 62, 2646–2648.
´
10. Rodrıguez, A. D.; Vera, B. J. Org. Chem. 2001, 66, 6364–6368.