Polar Metabolites of the Tropical Green Seaweed Caulerpa taxifolia Which Is Spreading in the Mediterranean Sea: Glycoglycerolipids and Stable Enols (α-Keto Esters)

Article abstract of DOI:10.1002/(SICI)1522-2675(19980909)81:9<1681::AID-HLCA1681>3.0.CO;2-4

Examination of the polar components of the green seaweed Caulerpa taxifolia (VAHL) C. AGARDH, which is heavily spreading in the northeastern Mediterranean, led to two families of compounds. The new (2R)-3-O-β-D-galactopyranosyl-1-O-[(7Z,10Z,13Z)-hexadeca-7,10,13-trienoyl]- 2-O-[(9Z,12Z,15Z)-octadeca-1,12,15-trienoyl]-sn-glycerol (2) was isolated in low abundance, like the analogues 1 and 3 already known from freshwater cyanobacteria. The acyl positions in 1-3 were determined by enzymatic methods and the absolute configuration from the O-galactosylglycerol obtained upon alcaline methanolysis. More abundant were the (4-hydroxyphenyl)- and (3,4-dihydroxyphenyl)pyruvic acid methyl esters, occurring in the enol (Z) forms 13a and 14a accompanied by very minor (E) forms 13b and 14b. The latter became predominant on UV irradiation of 13a or 14a, allowing the determination of the C=C configuration of these isolatable, stable enols from ¹H,¹³C NMR couplings (larger H-C(3)/C(1) coupling constant in the (E) than in the (Z) isomer). Contrary to literature implications, the O-galactosylglycerolipids 1-3 lack any cholinergic or histaminergic activity; similarly, enols (= α-keto esters) 13 and 14 or terpenoids of this seaweed were also devoid of such biological activities (see Table).

Full text of DOI:10.1002/(SICI)1522-2675(19980909)81:9<1681::AID-HLCA1681>3.0.CO;2-4

Helvetica Chimica Acta ± Vol. 81 (1998)
1681
Polar Metabolites of the Tropical Green Seaweed Caulerpa taxifolia
Which Is Spreading in the Mediterranean Sea: Glycoglycerolipids and Stable
Enols (a-Keto Esters)
by Ines Mancinia), Graziano Guellaa), Andrea Defanta), M. Luz Candenasb), Cristina P. Armestob),
a
a
*
Daniela Depentori ), and Francesco Pietra )
a
Á
) Laboratorio di Chimica Bioorganica, Universita di Trento, I-38050 Povo-Trento
b
Â
) Instituto de Investigaciones Químicas Isla de la Cartuja, CSIC, Americo Vespucio s/n, Isla de La Cartuja,
E-41092 Sevilla
Examination of the polar components of the green seaweed Caulerpa taxifolia (Va h l ) C. Agardh, which is
heavily spreading in the northeastern Mediterranean, led to two families of compounds. The new (2R)-3-O-b-d-
galactopyranosyl-1-O-[(7Z,10Z,13Z)-hexadeca-7,10,13-trienoyl]-2-O-[(9Z,12Z,15Z)-octadeca-1,12,15-trienoyl]-
sn-glycerol (2) was isolated in low abundance, like the analogues 1 and 3 already known from freshwater
cyanobacteria. The acyl positions in 1 ± 3 were determined by enzymatic methods and the absolute configuration
from the O-galactosylglycerol obtained upon alcaline methanolysis. More abundant were the (4-hydroxyphen-
yl)- and (3,4-dihydroxyphenyl)pyruvic acid methyl esters, occurring in the enol (Z) forms 13a and 14a
accompanied by very minor (E) forms 13b and 14b. The latter became predominant on UV irradiation of
1
13a or 14a, allowing the determination of the CˆC configuration of these isolatable, stable enols from H,13C
NMR couplings (larger H C(3)/C(1) coupling constant in the (E) than in the (Z) isomer). Contrary to
literature implications, the O-galactosylglycerolipids 1 ± 3 lack any cholinergic or histaminergic activity;
similarly, enols ( ˆ a-keto esters) 13 and 14 or terpenoids of this seaweed were also devoid of such biological
activities (see Table).
1. Introduction. ± Spreading of alien species is being observed at increased
frequency [1] side-to-side to man-driven mass extinction [2]. Both phenomena threaten
the biological diversity at any level, either of population [2a], species, or higher taxon
[3]: mass extinction because extant species disappear faster than new species arise, and
invasions because they prevent isolation which constitutes the basis for generating
biological diversity [4].
A case in point is the current spreading of the tropical green seaweed Caulerpa
taxifolia (Va h l ) C. Agardh (Ulvophyta, Caulerpales, Caulerpaceae) in the Medi-
terranean Sea. This is an unusual invasion since this seaweed has developed toxic,
lipophilic terpenoids against herbivorous grazers to sustain its sparse and inconspicuous
populations in the pressure for food of the tropics [5]. Following the colonization of the
Mediterranean Sea, the content of toxic terpenoids of C. taxifolia has greatly increased
[6], as well as the size of the fronds, the density of the populations, and the homeostatic
capacities [5]. This incurs risks of decrease of the biological diversity and of ecotoxicity
in the Mediterranean, unless the in vitro observed photodegradation of the algal
terpenoids [5b] also occurs in nature. The prospects that this strengthened strain of C.
taxifolia may find the way back to the tropics ± in a form of natal homing [5c] ± is most
preoccupying.

1682
Helvetica Chimica Acta ± Vol. 81 (1998)
Aimed at a broader understanding of the metabolism of this seaweed, we have now
studied the polar fractions of a Mediterranean strain of C. taxifolia, finding H2O-
soluble, stable enols and polar glycoglycerolipids which were assayed in the rat-isolated
uterus and aorta.
2. Results and Discussion. ± 2.1. Glycoglycerolipids. FAB-MS of medium-polarity
products 1 ± 3 from C. taxifolia gave the same composition, C43H70O10, for 1 and 2, and
C45H74O10 for 3. The NMR data of 1 ± 3 suggested the presence of a di-O-acylglycerol
O-linked to a b-galactopyranose moiety and of six cis disubstituted CˆC bonds for all
these compounds. The nature and positions of the two polyunsaturated acyl groups
were established by a combination of enzymatic and chemical hydrolysis (Scheme 1).
Using a commercial lipase of type XI from Rhizopus arrhizus, which had already been
used both in the structural analysis of mono-O-galactosylglycerols from marine
cyanobacteria [7] and green algae [8], as well as in the synthesis of antifungal
phospholipids [9], we obtained from 1 the mono-O-acylglycerol derivative 6 with a
(7Z,10Z,13Z)-hexadeca-7,10,13-trienoyl (Acp) chain at the primary OH group, besides
linolenic acid (4; Scheme 1); it should be noticed that no reaction occurred under these
conditions in the absence of the enzyme. Compound 3 gave the O-galactosylglycerol
derivative 7 and linolenic acid (AcL ± OH; 4). That enzymatic hydrolysis occurred at
the primary-alcohol function of 1 to give the elusive intermediate 5, followed by Acp
migration, found also evidence in the slow methanolysis at the secondary-alcohol
function on storage of 2 in CD3OD to give 6 and ester 8 (Scheme 2).
Contrasting reports have appeared for processes employing the above enzymes.
Thus, selective lipase-XI-induced hydrolysis at the primary-alcohol function was first
reported not only for 1 [7a] but also for other mono-O-galactosyldi-O-acylglycerols
[7b] and di-O-acylated sulfoglycolipids and galactolipids [7c], all isolated from
freshwater cyanobacteria [7a]. Subsequently, these processes were redescribed as being
accompanied by an sn-2 !sn-1 acyl transposition of the type shown in Scheme 1, whose
extent was judged to depend on the amount of lipase, type of buffer, presence or not of
surfactants [7d,e], and reaction time [7d][10]. On the other hand, a complete
transposition of the type shown in Scheme 1 by lipase XI in similar systems required a
higher pH (8.5) [9]. Our results are not in line with these reports since we observed
complete sn-2 !sn-1 acyl transposition with the above and similar glycoglycerolipids
a) at lower pH (7.7) than indicated in the synthetic procedure [9], b) in a boric acid/

Helvetica Chimica Acta ± Vol. 81 (1998)
1683
Scheme 1
a) Lipase type XI (from Rhizopus arrhizus), Triton X-100, boric acid/borax buffer pH 7.7, 388, 40 min.
Scheme 2
a) CD3OD, 48 for 45 days (25% conversion), then 508 for 24 h (50% conversion). b) Addition of solid
NaHCO3, then HPLC. c) Sat Na2CO3/MeOH, r.t., 3 h.
borax buffer which was expected to give instead a mixture of the transposed and the
non-transposed O-galactosylmono-O-acylglycerolipid [7d], c) in a shorter time than
that indicated to give a mixture of transposed and non-transposed products [7d], and d)
using a similar amount of lipase (900 units) that brought about instead regioselective 1-

1684
Helvetica Chimica Acta ± Vol. 81 (1998)
deacylation [7c]. The possibility that the batch of lipase used in our work contained also
an acyltransferase was not investigated, being outside the scope of our work1).
Glycoglycerolipids occur widely and copiously in vascular plants [13], raphidophy-
cean microalgae [11], certain seaweeds [12], cyanobacteria [7], and marine dino-
flagellates [14]. Compounds 1 and 3 were already known [7a,b,d][8][12b][15a±c], while 2
is a new isolate. As to the glycoglycerolipids of the green seaweeds (Ulvophyta), it is
only known that the main O-galactosylglycerolipid of Monostroma nitidum (Ulo-
trichales) [12b] has a C16 :4 fatty acid residue at C(2), in sharp contrast with products
1 ± 3 from C. taxifolia.
The problem of the absolute configuration of 1 ± 3 was then addressed by subjecting
the mixture 1 ± 3 to rapid methanolysis in the presence of Na2CO3 which gave the O-
galactosylglycerol 9 and the methyl esters 11 and 12 (Scheme 2)2). Compound 9 was
determined to be (2R)-1-O-b-d-galactopyranosyl-sn-glycerol by comparing the specific
rotation and NMR spectra with published data [7a ± c][13][15a][16]. This also
established that the configuration of 2 is the same as for the O-galactosylglycerols of
terrestrial plants [11]. This additional piece of evidence as to the relatedness of green
seaweeds to vascular plants does not imply necessarily a similarity of functions of the
metabolites, however. O-Galactosylglycerolipids 1 ± 3, being present in C. taxifolia in
trace amounts, are unlikely to play a major role in stabilizing the algal membranes, in
contrast to vascular plants. Since glycerolipids are known to play various other biological
functions, the elucidation of the role they play in C. taxifolia remains a challenge.
2.2. Enols (a-Keto Esters). Aryl pyruvic esters are more abundant in C. taxifolia
than the glycerolipids. Both UV spectra (lmax 299 nm, typical of the enol form of a-keto
acids) and NMR spectra suggested that these compounds exist as mixtures of enol
tautomers [17], one stereoisomer (13a or 14a) being largely predominant in both cases.
The stereoisomers proved stable and isolatable by HPLC. UV Irradiation (350 nm) of
the predominant stereoisomer produced a mixture of stereoisomers in the ratio 45 :55
in favour of the minor natural stereoisomer (13b or 14b)3). By acetylation of the major
isomeric enol forms 13a or 14a, the diacetyl and triacetyl derivatives 15 and 16,
respectively, were obtained (Scheme 3).
1
)
)
)
The course of the 1-O-deacylation of similar O-galactosylglycerolipids of raphidophycean microalgae by a
bacterial type-XIII lipase was smoother, not accompanied by acyl transfer [11]. The enzymatic method
proved also in this case an indispensable tool in assigning the position of the acyl chains of
glycoglycerolipids, whilst chemical methods gave ambiguous results [11]. This casts doubt on the
assignment by chemical or spectroscopic methods alone of the position of the acyl chains in similar
glycoglycerolipids of brown [12a] and other seaweeds [12b].
Since the resulting O-galactosylglycerol 9 was diastereoisomerically pure (NMR), the use of the mixture
1 ± 3 was to this purpose equivalent to that of the pure O-galactosyldi-O-acylglycerols. On storage of 2 in
CD3OD in a NMR probe, a partial conversion to 6 and 8 occurred (Scheme 2, a, and Exper. Part), and on
addition of Na2CO3, 6 was converted to 9 and 10. Similarly, on storage of 3 in CD3OD, a partial conversion
to 7 and 8 occurred. It is understood that in these experiments with 2 and 3 1H-NMR observation could not
reveal D exchange at OH, whilst the isolated products had proton restored at OH following aqueous
This ratio did not change either on overnight standing of the irradiated mixture in the dark or on further
UV irradiation (350 nm) for 1 h. By irradiation under identical conditions of a 1:2 mixture 13b/14b, the
same isomer ratio (45:55) for 13a/13b and 14a/14b was obtained. In contrast, similar irradiation of
commercial (4-hydroxyphenyl)pyruvic acid resulted in only a 3% increase of the relative population of the
minor isomeric enol form. The major isomer 13b or 14b revealed 3J(H C(3),C(1)) ˆ 4.2 Hz, suggesting
(Z)-configuration (vide infra).
2
3

Helvetica Chimica Acta ± Vol. 81 (1998)
1685
Scheme 3
a) Ac2O, Py, r.t., 1 h.
3
3
The criterion of a larger J(H,Ctrans) than J(H,Ccis) heterocoupling constant in
assigning the CˆC configuration of trisubstituted alkenes [18a,b] was recently applied
to marine indole alkaloids, the aplysinopsins [18c], and was also useful in the case of the
present enol forms. Thus, following the enrichment in 14b of the natural mixture 14a/
14b by UV irradiation (350 nm), the stable stereoisomers 14a and 14b could be
separated in pure forms by HPLC and their fully coupled 13C-NMR spectra be recorded
under selective heteronuclear decoupling of the Me resonances4). It could thus be
determined that the major natural isomer 14a has a small (3.7 Hz) H C(3)/C(1) 1H,13C
heterocoupling, indicating (Z)-configuration [18]. Correspondingly, the most abun-
dant isomer from photoequilibration, 14b, showed a larger heterocoupling (9.2 Hz),
indicating (E)-configuration.
A number of different arylpropane-derived compounds have been previously
isolated from Caulerpa racemosa from the eastern coasts of India [19]. Together with
the enol forms 13a and 14a of a-keto ester, they further support the close phylogenetic
relationship of the Caulerpales to vascular plants.
2.3. Biological Assays. Neither 1 nor the mixture 1 ± 3 of natural composition had
any significant effect on preparations of rat-isolated uterus under resting tension
(Table, a). In the same myometrial strips, a maximally effective concentration of
acetylcholine (ACh) (10 3 m) [20] elicited an initial, rapid phasic contraction (83.51 Æ
6.83, expressed, here and below, in mg per mg of wet tissue) followed by rhythmic
contractions. The ACh-induced contraction was completely abolished in the presence
of atropine (10 6 m), demonstrating that the O-galactosyldi-O-acylglycerolipids
assayed here are devoid of cholinergic activity. We also analysed the relaxant effects
of C. taxifolia products in uterine strips pre-contracted by KCl or oxytocin. KCl-
Induced contraction depends primarily on Ca2‡ influx via L-type, voltage-operated
Ca2‡ channels [21], whereas the oxytocin-induced contraction involves additional
mechanisms such as activation of receptor-operated Ca2‡ channels, Ca2‡ release from
4
)
Otherwise, coupling of C(1) with the Me group led to such an extensive fragmentation of the signals that
some of them became hardly distinguishable from background noise.

1686
Helvetica Chimica Acta ± Vol. 81 (1998)
intracellular stores, and Ca2‡-independent processes [21][22]. In the uterus, isoosmotic
addition of KCl (60 mm) caused a phasic contraction (70.15 Æ 3.25) which then
declined and was followed by a prolonged sustained plateau (40.80 Æ 5.68). Oxytocin
(0.2 nm) caused rhythmic contractions of stable intensity (99.73 Æ 10.90) and duration.
Compound 1 and the mixture 1 ± 3 also failed to relax contractions evoked by either
KCl (Table, b) or oxytocin (Table, c), suggesting that they do not interact with any of
the above mentioned mechanisms. Compound 1 was also evaluated in the rat aorta, a
preparation in which histamine induced an H1-mediated contraction [23], but again
with negative results.
Table 1. Functional Assays on Rat-Isolated Myometriuma) with Products Isolated from Caulerpa taxifolia
Conc.
[m]
1
1/2/3 5 : 2 : 3
13a/14a 1 : 2b)
Caulerpenyne
Caulerpenyne/oxy-
toxin 1 2 : 1
Control
a) Effect on preparations under resting tensionc):
9
10
10
10
10
10
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.68 Æ 0.01
1.01 Æ 0.01
1.22 Æ 0.01
1.49 Æ 0.02
1.78 Æ 0.02
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
8
7
6
5
b) Effect on preparations precontracted by a depolarizing solution (60 mm KCl)d):
9
10
10
10
10
10
13.17 Æ 0.03
23.70 Æ 0.05
30.29 Æ 0.07
32.91 Æ 0.07
32.91 Æ 0.07
0.00 Æ 0.00
2.07 Æ 0.02
9.69 Æ 0.01
14.53 Æ 0.02
22.47 Æ 0.01
4.17 Æ 0.04
6.25 Æ 0.63
7.50 Æ 0.08
11.66 Æ 0.06
18.73 Æ 0.07
1.54 Æ 0.05
4.62 Æ 0.03
6.15 Æ 0.03
15.38 Æ 0.07
35.38 Æ 0.09
5.17 Æ 0.02
11.99 Æ 0.03
22.27 Æ 0.06
27.43 Æ 0.08
58.78 Æ 0.10
4.76 Æ 0.02
14.46 Æ 0.07
16.99 Æ 0.08
18.68 Æ 0.09
23.17 Æ 0.09
8
7
6
5
c) Effect on preparations precontracted by oxytocin (0.2 nm)d):
9
10
10
10
10
10
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.36 Æ 0.01
0.72 Æ 0.01
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
1.16 Æ 0.01
2.33 Æ 0.02
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
0.00 Æ 0.00
36.63 Æ 0.32
1.43 Æ 0.01
31.88 Æ 0.30
38.57 Æ 0.26
42.50 Æ 0.24
103.44 Æ 0.02
0.36 Æ 0.01
0.73 Æ 0.01
1.45 Æ 0.01
2.46 Æ 0.02
3.84 Æ 0.03
8
7
6
5
a) Values are the mean Æ s.e.m. in 5 different animals. b) Containing a small amount of 13b/14b (see text).
c) Data expressed as a percentage of the maximal contraction evoked by acetylcholine (1 nm). d) Data
expressed as a percentage of the maximal contraction evoked by either KCl or oxytoxin, respectively, prior to
the addition of the test product.
The above findings may have a bearing on work concerning an allegedly isolated
product from Mediterranean C. taxifolia which was deemed to possess an ꢀaminoacidic
skeleton in an alicyclic chain and polyenic conjugated bondsꢁ and to have cholinergic
and/or histaminergic activity [24]. Actually, the COSY plot reported [24] does not
support any such structural conjecture but, although plotted at a too high threshold
value to allow discerning all details, it is compatible with structures 1 ± 3. These cannot
be responsible for the reported [24] biological activity, however, as demonstrated
above, and therefore, we have considered possible contaminants of the isolated sample
[24]. The enol forms 13a/13b and 14a/14b of a-keto esters were such candidates, of
higher polarity than the O-galactosylglycerolipids, but they proved inactive in the rat
myometrium assay (Table). In the same vein, we further conceded that, following

Helvetica Chimica Acta ± Vol. 81 (1998)
1687
inappropriate separation procedures, the isolated sample [24] might have been
contaminated with products of far lower polarity than the enol forms of the a-keto
acids, like the algal sesquiterpenoids. Of these, the most abundant, caulerpenyne [25a]
is known to display various biological activities, such as cytotoxicity towards both
marine ciliates [25a] and human tumour cells [26][27], antibacterial activity [25a], as
well as alteration of ionic signals involved in cell dynamics by influencing the calcium
uptake [28]. Thus, both caulerpenyne and the related oxytoxin 15) were examined in
the rat-isolated-uterus assay. Oxytoxin 1 (assayed in a 1:2 mixture with caulerpenyne)
showed non-negligible activity by both abolishing rhythmic contractions induced by
oxytocin 0.2 nm and partially relaxing the KCl-induced response, whereas caulerpe-
nyne showed only marginal inhibition of the KCl- and oxytocin-induced contractions.
Since all these effects were only observed at the highest concentration assayed (10 5 m;
see Table, b and c), neither oxytoxin 1 nor caulerpenyne can be responsible for the
cholinergic and/or histaminergic activities reported [24].
One might wonder whether the summer sample of C. taxifolia from Imperia, used in
previous work [24], might have contained other metabolites than our summer sample
of the seaweed from Elba Island. Our examination of various strains of C. taxifolia ±
including tropical ones ± by very sensitive and specific MS-MS tandem ionspray
techniques [6] makes this possibility very unlikely. In any event, such a possibility is not
suggested by the reported spectral data [24].
We thank A. Sterni for recording the mass spectra and, for financial support, DGXI LIFE 1996 ± 98,
MURST, and CNR, Roma, the Spanish Ministry of Education and Science (grant PB 92-0487), and the Regione
Toscana (38 Programma di Ricerca Finalizzata Sanitaria).
Experimental Part
General. Lipase XI: from Rhizopus arrhizus (Sigma). All evaporations were carried out at r.t. under
reduced pressure. Flash chromatography (FC): Merck Si-60 (15 ± 25 mm). TLC: Merck silica gel 60 PF254 and
Merck RP-18 F254. Reversed-phase HPLC: Merck LiChrospher RP18 (7 mm), 25 Â 1 cm columns; UV
monitoring at l 220 nm and solvent flux 5 ml min 1, if not otherwise specified. Polarimetric data: JASCO-DP-
181 polarimeter; [a]D values in 10 1 deg ml g 1. UV: Perkin-Elmer-Lambda-3 spectrophotometer. NMR: Varian-
XL-300, 1H at 299.94 MHz and 13C at 75.43 MHz; d in ppm and J in Hz, in either CD3OD or CDCl3, rel. to the
solvent residual signals, d(H) 3.31 and d(C) 49.00 ppm for the first and d(H) 7.25 and d(C) 77.00 ppm for the
latter, where the solvent residual signals are rel. to SiMe4 (ˆ 0 ppm); assignments from DEPT, COSY60,
selective decoupling irradiations, and 1H,13C COSY. MS (EI and FAB): Kratos-MS80 with home-built
computerized data-acquisition system and equipped with a Vacumetrics-DIP gun for FAB spectra.
Collection and Isolations. Caulerpa taxifolia was collected (856M) in the Baia di Calenzana, Elba Island, at
the end of July 1997 and was immediately cleaned from epiphytes, immersed into 95% EtOH and stored in the
cold. After a few days, a portion of the EtOH soln., corresponding to the extract from ca. 4 kg (wet weight) of
seaweed, was homogenized and filtered, and the EtOH extract was evaporated to give 17.8 g of residue that was
subjected to FC (gradient elution with hexane/AcOEt and AcOEt/MeOH and then pure MeOH) for a total of
34 fractions of 400 ml each. Frs. 7 ± 10 contained pure caulerpenyne and Fr. 11 caulerpenyne/oxytoxin 1 2 :1,
which were used for the bioassays. Fr. 26 (300 mg (dry weight)) was subjected to reversed-phase FC (MeOH/
H2O gradient), collecting 8 fractions of 60 ml each; of these a) Fr. 6 (70 mg (dry weight)) was subjected to prep.
TLC (AcOEt/MeOH 9 :1) to give mixture 1 ± 3; the latter was subjected to reversed-phase HPLC (MeOH/H2O
97:3) yielding pure 1 (tR 9.5 min; 17.6 mg), 2 (tR 10.5 min; 6.6 mg), and 3 (tR 11.3 min; 9.2 mg); b) Fr. 1 (130 mg
5
)
Oxytoxin 1 was isolated from C. taxifolia [25a], but it might be an artifact of the hydrolysis of caulerpenyne
during workup [25b] and thus might well have been a contaminant of the mixture used in previous work
[24].

1688
Helvetica Chimica Acta ± Vol. 81 (1998)
(dry weight)) was subjected to reversed-phase HPLC (MeOH/H2O 1 :9, monitoring at l 254 nm) yielding 13b/
14b 1:2 (tR 2.9 min; 3.8 mg) and pure 14a (tR 7.5 min; 80.0 mg) and pure 13a (tR 9.8 min; 39.5 mg).
Biological Assays. Preparation of uterine strips was done as previously described [20] with slight
modifications as follows. Uterine horns from virgin female Wistar rats (200 ± 250 g) in the oestrous stage of the
hormonal cycle were removed and carefully cleaned. Longitudinal strips of uterine smooth muscle (8 ± 10 mm in
length and 1 ± 2 mm in diameter) were prepared and mounted in tissue baths containing 4 ml of Krebs soln. of
the following composition (mm): NaCl (118), KCl (5.6), CaCl2 (1.9), MgSO4 (0.95), NaHPO4 (1.0), NaHCO3
(25), and glucose (11). Preparations were suspended under an initial tension of 0.5 g, gassed with O2/CO2 95 :5
and maintained at 378. Mechanical responses were recorded isometrically by means of force-displacement
transducers (Grass FT-03) connected to a LETICA amplifier and an ABB-GOERZ-SE-130 multichannel
recorder. At the beginning of the experiment and after a 45-min equilibration period, the preparation was
induced to contract two or more times by administration of a maximally effective concentration of ACh (10 3 m)
at 30 min intervals, until constant responses were obtained. The last response served as an internal standard for
all experiments. Uterine strips were allowed to equilibrate for a further 60-min period before addition of
cumulative concentrations of a C. taxifolia product (10 9 ± 10 5 m) or its vehicle (time-matched paired control
strips). In a separate set of experiments, we examined the effect of the C. taxifolia products on uterine
‡
‡
contractions evoked by a high K depolarizing soln. (60 mm K , prepared by equimolar substitution of KCl for
NaCl in the physiological soln.) or by the receptor agonist oxytoxin (0.2 nm). When the tonic contraction
induced by KCl was sustained, or when the rhythmic contractions produced by oxytocin had attained a constant
amplitude, a test product (1 nm ± 10 mm) or its vehicle (time-matched control tissues) were added to the bath in a
cumulative manner at 15-min intervals. Some experiments were carried out in isolated rings form rat aorta,
essentially as described for the uterus, except for KCl replacing ACh as the spasmogen used as an internal
standard. In all experiments, only one product was tested on each tissue. Values are expressed as mean Æ s.e.m.
for 5 experiments in 5 different animals. Contractile responses were measured as peak increases in force and
expressed in mg per mg tissue (wet weight) or as a percentage of the peak increase in force produced by ACh
(10 3 m). Relaxation responses were expressed as percentage inhibition of the maximum tension induced by the
spasmogen (KCl or oxytocin) prior to the addition of the products.
(2R)-3-O-b-d-Galactopyranosyl-2-O-[(7Z,10Z,13Z)-hexadeca-7,10,13-trienoyl]-1-O-[(9Z,12Z,15Z)-octa-
deca-9,12,15-trienoyl]-sn-glycerol (1). Oil. [a]D25
ˆ
2.7 (c ˆ 0.2, CHCl3) [8]: [a]D
ˆ
2.8 (c ˆ 0.1, CHCl3).
‡
NMR: data matched those in [8]. FAB-MS (glycerol): 769 ([M ‡ Na] ).
(2R)-3-O-b-d-Galactopyranosyl-1-O-[(7Z,10Z,13Z)-hexadeca-7,10,13-trienoyl]-2-O-[(9Z,12Z,15Z)-octa-
deca-9,12,15-trienoyl]-sn-glycerol (2). Oil. [a]D25
ˆ
2.8 (c ˆ 0.2, CHCl3). 1H-NMR (CD3OD): 4.22 (dd, J(1a,2) ˆ
6.5, Jgem ˆ 12.1, Ha C(1)); 4.44 (dd, J(1b,2) ˆ 3.0, Jgem ˆ 12.1, Hb C(1)); 5.27 (m, H C(2)); 3.74 (dd, J(3a,2) ˆ
5.7, Jgem ˆ 11.0, Ha C(3)); 3.98 (dd, J(3b,2) ˆ 5.4, Jgem ˆ 11.0, Hb C(3)); 2.31 (t, 2 H C(2')); 1.62
(m, 2 H C(3')); 1.38 (m, 2 H C(4'), 2 H C(5')); 2.05 (m, 2 H C(6')); 5.30 (m, H C(7'),
H C(8'),
H
C(10'), H C(11'), H C(13'), H C(14')); 2.82 (m, 2 H C(9'), 2 H C(12')); 2.09 (m, 2 H C(15')); 0.97
(q, J(16',15') ˆ 7.5, Me(16')); 2.33 (t, J(2'',3'') ˆ 7.0, 2 H C(2'')); 1.62 (m, 2 H C(3'')); 1.38 (m, 2 H C(4'') to
2 H C(7'')); 2.09 (m, 2 H C(8'')); 5.36 (m, H C(9''), C(10''), C(12''), C(13''), C(15''),
H
H
H
H
H
C(16'')); 2.82 (m, 2 H C(11''), 2 H C(14'')); 2.09 (m, 2 H C(17'')); 0.97 (q, J(18'',17'') ˆ 7.5, Me(18''));
4.23 (d, J(1''',2''') ˆ 7.2, H C(1''')); 3.52 (dd, J(2''',1''') ˆ 7.2, J(2''',3''') ˆ 9.6, H C(2''')); 3.45 (dd, J(3''',2''') ˆ
9.6, J(3''',4''') ˆ 3.3, H C(3''')); 3.83 (dd, J(4''',3''') ˆ 3.3, J(4''',5''') ˆ 1.0, H C(4''')); 3.51 (ddd, J(5''',4''') ˆ 1.0,
J(5''',6'''a) ˆ 5.4, J(5''',6'''b) ˆ 6.9,
H
C(5''')); 3.71 (dd, J(6'''a,5''') ˆ 5.4, Jgem ˆ 11.3, Ha C(6''')); 3.77
(dd, J(6'''b,5''') ˆ 6.9, Jgem ˆ 11.3, Hb C(6''')). 13C-NMR (CD3OD): 63.99 (t, C(1)); 71.81 (d, C(2)); 68.72
(t, C(3)); 175.03, 174.67 (2s, C(1'), C(1'')); 132.79, 132.73, 130.92, 130.86, 130.76, 129.27 (Â 2), 129.16, 129.07
(Â 2), 128.98, 128.23 (all d, C(7'), C(8'), C(10'), C(11'), C(13'), C(14'), C(9''), C(10''), C(12''), C(13''), C(5''),
C(16'')); 35.07, 34.95, 30.85, 30.82, 30.72, 30.44, 30.34, 30.20, 30.18, 30.08, 29.81, 28.18, 28.14, 28.06, 26.55, 26.43,
26.03, 25.91 (all t, C(2') to C(6'), C(9'), C(12'), C(15'), C(2'') to C(8''), C(11''), C(14''), C(17'')); 14.69
(q, C(16'), C(18'')); 105.36 (d, C(1''')); 72.38 (d, C(2''')); 74.87 (d, C(3''')); 70.22 (d, C(4''')); 76.80 (d, C(5'''));
‡
62.46 (t, C(6''')). FAB-MS (glycerol): 769 ([M ‡ Na] ).
(2R)-3-O-b-d-Galactopyranosyl-1,2-di-O-[(9Z,12Z,15Z)-octadeca-9,12,15-trienoyl]-sn-glycerol (3). Oil.
[a]2D5
ˆ
3.8 (c ˆ 0.3, CHCl3) [15a]: [a]D
ˆ
4.0 (c ˆ 0.3, CHCl3). NMR: data matched those in [8][15a].
‡
FAB-MS (glycerol): 797 ([M ‡ Na] ).
Enzymatic Hydrolysis of Compounds 1 and 3. A soln. of 1 (4.0 mg) and Rhizopus arrhizus lipase XI
(900 units) in the presence of Triton X-100 (2.5 mg) in a boric acid/borax buffer (0.6 ml, pH 7.7) was stirred at
388 for 1 h. Then were added in sequence AcOH (0.1 ml) and EtOH (1.0 ml). The mixture was evaporated and
the residue subjected to reversed-phase HPLC (MeOH/H2O 85 :15) to give 3-O-b-d-galactopyranosyl-1-O-

Helvetica Chimica Acta ± Vol. 81 (1998)
1689
[(7Z,10Z,13Z)-hexadeca-7,10,13-trienoyl]-sn-glycerol (6, tR 5.9 min; 1.9 mg, 70%). MeOH eluates from column
washing after each HPLC cycle were combined and evaporated and the residue was subjected to reversed-phase
HPLC (MeOH/H2O 95 :5) to give (9Z,12Z,15Z)-octadeca-9,12,15-trienoic acid (ˆ linolenic acid; 4; tR 8.5 min;
1.0 mg).
On the same treatment, 3 (3.2 mg) gave 3-O-b-d-galactopyranosyl-1-O-[(9Z,12Z,15Z)-octadeca-9,12,15-
trienoyl]-sn-glycerol (7; tR 6.8 min; 1.3 mg) and 4 (0.8 mg).
Data of 4: 1H-NMR (CD3OD): 2.33 (t, J ˆ 7.0, 2 H C(2)); 1.63 (m, 2 H C(3)); 1.37 (m, 2 H C(4) to
2 H C(7)); 2.09 (m, 2 H C(8), 2 H C(17)); 5.36 (m, H C(9), H C(10), H C(12), H C(13), H C(15),
H
C(16)); 2.81 (m, 2 H C(11), 2 H C(14)); 0.98 (q, J ˆ 7.5, Me).
Data of 6: 1H-NMR (CD3OD): 4.15 (m, 2 H C(1)); 3.99 (m, H C(2)); 3.94, 3.68 (2m, 2 H C(3)); 2.31
(t, J ˆ 7.3, 2 H C(2')); 1.63 (m, 2 H C(3')); 1.38 (m, 2 H C(4'), 2 H C(5')); 2.08 (m, 2 H C(6'), 2 H C(15'));
5.35 (m, H C(7'), C(8'), C(10'), C(11'), C(13'),
H
H
H
H
H
C(14')); 2.81 (t, J ˆ 5.7, 2 H C(9'),
2 H C(12')); 0.98 (q, J ˆ 7.5, Me); 4.23 (d, J ˆ 7.4, H C(1'')); 3.51 (m, H C(2''), H C(5'')); 3.46 (dd, J ˆ 9.6,
3.3, H C(3'')); 3.81 (dd, J ˆ 3.3, 1.0, H C(4'')); 3.74 (m, 2 H C(6'')). 13C-NMR (CD3OD): 175.36 (s); 132.70
(d); 130.84 (d); 129.20 (d); 129.13 (d); 128.98 (d); 128.18 (d); 105.29 (d); 76.74 (d); 74.81 (d); 72.53 (d); 71.85
(d); 70.24 (d); 69.59 (t); 66.55 (t); 62.44 (t); 34.87 (t); 30.37 (t); 29.79 (t); 27.99 (t); 26.48 (t); 26.37 (t); 25.84 (t);
‡
21.40 (t); 14.62 (q). FAB-MS (glycerol): 509 ([M ‡ Na] ).
Data of 7: 1H-NMR (CD3OD): 4.15 (m, 2 H C(1)); 3.99 (m, H C(2)); 3.94, 3.67 (2m, 2 H C(3)); 2.33
(t, J ˆ 7.0, 2 H C(2')); 1.63 (m, 2 H C(3')); 1.37 (m, 2 H C(4') to 2 H C(7')); 2.09 (m, 2 H C(8'),
2 H C(17')); 5.36 (m, H C(9'),
H
C(10'),
H
C(12'),
H
C(13'),
H
C(15'),
H C(16')); 2.82
(m, 2 H C(11'), 2 H C(14')); 0.98 (q, J ˆ 7.5, Me); 4.22 (d, J ˆ 7.4, H C(1'')); 3.51 (m, H C(2''), H C(5''));
3.46 (dd, J ˆ 9.6, 3.3, H C(3'')); 3.82 (dd, J ˆ 3.3, 1.0, H C(4'')); 3.74 (m, 2 H C(6'')). FAB-MS (glycerol): 537
‡
([M ‡ Na] ).
Alkaline Hydrolysis of Compounds 1 ± 3. Compound 2 (3.3 mg), on storage in CD3OD (0.6 ml) for 45 days
at 48 and then 24 h at 508, was partially converted into compounds 6 (see Scheme 2, a, and Footnote 2) and
(D3)methyl (9Z,12Z,15Z)-octadeca-9,12,15-trienoate (8); NMR observation was carried out on products 6 and 8
after aqueous workup. On addition of catal., solid NaHCO3, the hydrolysis went to completion, giving 1-O-b-d-
galactopyranosyl-sn-glycerol (9) and (D3)methyl (7Z,10Z,13Z)-hexadeca-7,10,13-trienoate (10). The mixture
was subjected to reversed-phase HPLC (MeOH/H2O 95 :5): 9 (tR 2.5 min; 1.1 mg, 93%), 8 (tR 9.8 min; 1.2 mg),
and 10 (tR 6.9 min; 1.1 mg).
Under similar conditions, 8 ´ 10 3 m 3 in CD3OD was partially converted into 7 and 8 (see Footnote 2).
In another experiment, sat. methanolic Na2CO3 soln. (1 ml) was added to 1 ± 3 (3 :1:2; 10.0 mg), stirring the
mixture at r.t. until complete conversion (1 h, TLC). The residue from evaporation was subjected to reversed-
phase HPLC (MeOH/H2O 95 :5): 9 (tR 2.5 min; 3.2 mg, 90%), methyl (7Z,10Z,13Z)-hexadeca-7,10,13-trienoate
(11; tR 6.2 min; 3.0 mg), and methyl (9Z,12 Z,15Z)-octadeca-9,12,15-trienoate (12; tR 8.9 min; 3.3 mg).
Data of 8: 1H-NMR (CD3OD): 2.33 (t, J ˆ 7.0, 2 H C(2)); 1.63 (m, 2 H C(3)); 1.37 (m, 2 H C(4) to
2 H C(7)); 2.09 (m, 2 H C(8), 2 H C(17)); 5.36 (m, H C(9), H C(10), H C(12), H C(13), H C(15),
.
‡
H
C(16)); 2.82 (m, 2 H C(11), 2 H C(14)); 0.97 (q, J ˆ 7.5, Me). EI-MS: 295 (14, M ), 266(3), 252(1),
149(21), 109(119), 79(100).
Data of 9: semisolid. [aD25
ˆ
6.8 (c ˆ 0.2, H2O) ([7c]: [a]D
ˆ
7.0 (c ˆ 0.3, H2O); [15a]: [a]D
ˆ
8.0 (c ˆ
0.6, H2O)). 1H-NMR: data matched those in [15a]. 13C-NMR (CD3OD): 105.14 (d); 76.70 (d); 74.78 (d); 72.54
‡
(d); 72.09 (d); 72.01 (t); 70.36 (d); 64.01 (t); 62.51 (t). FAB-MS (glycerol): 277 ([M ‡ Na] ).
1
Data of 10: H-NMR (CD3OD): 5.37 (m, H C(7), H C(8), H C(10), H C(11), H C(13), H C(14));
2.31 (t, 2 H C(2)); 1.64 (m, 2 H C(3)); 1.39 (m, 2 H C(4), 2 H C(5)); 2.08 (m, 2 H C(6), 2 H C(15)); 2.82
(m, 2 H C(9), 2 H C(12)); 0.97 (q, J ˆ 7.5, 3 H C(16)). 13C-NMR (CD3OD): 174.60 (s); 132.73 (d); 130.86
(d); 129.22 (d); 129.16 (d); 129.04 (d); 127.31 (d); 50.41 (q); 34.78 (t); 30.68 (t); 30.36 (t); 29.79 (t); 27.99 (t);
.
‡
26.51 (t); 25.92 (t); 20.90 (t); 14.64 (q). EI-MS: 267 (14, M ), 238(2), 224(2), 149(10), 109(12), 79(100).
.
‡
Data of 11: 1H-NMR: data in agreement with [8]. EI-MS: 264 (13, M ), 235(1), 233(2), 149(8), 121(26),
109(18), 107(19), 109(18), 79(100).
Data of 12: 1H-NMR (CD3OD): 2.33 (t, J ˆ 7.0, 2 H C(2)); 1.63 (m, 2 H C(3)); 1.37 (m, 2 H C(4) to
2 H C(7)); 2.09 (m, 2 H C(8), 2 H C(17); 5.36 (m, H C(9), H C(10), H C(12), H C(13), H C(15),
H
M
C(16)); 2.82 (m, 2 H C(11), 2 H C(14)); 0.98 (q, J ˆ 7.5, 3H C(18)); 3.65 (s, MeO). EI-MS: 292 (13,
.
‡
), 263(3), 261(5), 149(15), 135(18), 121(26), 109(18), 69(20).
(Z)-2-Hydroxy-3-(4-hydroxyphenyl)prop-2-enoic Acid Methyl Ester (13a). 1H-NMR (CD3OD): 7.81
(d, J(2',3') ˆ J(6',5') ˆ 8.7,
H
C(2'),
H
C(6')); 6.76 (d, J(3',2') ˆ J(6',5') ˆ 8.7,
H C(3'), H C(5')); 7.05
(s, H C(3)); 3.82 (s, MeO). 13C-NMR (CD3OD): 137.08 (s, C(1')); 134.10 (d, C(2'), C(6')); 116.22 (d, C(3'),

1690
Helvetica Chimica Acta ± Vol. 81 (1998)
C(5')); 160.32 (s, C(4')); 128.52 (d, C(3)); 125.36 (s, C(2)); 167.37 (s, C(1)); 52.60 (q, MeO). EI-MS: 194 (26,
.
.
‡
‡
M
), 134(29), 107(100). HR-EI-MS: 194.0578 Æ 0.003 (C10H10O4 ; calc. 194.0579).
(E)-2-Hydroxy-3-(4-hydroxyphenyl)prop-2-enoic Acid Methyl Ester (13b). 1H-NMR (CD3OD): 7.34
(d, J(2',3') ˆ J(6',5') ˆ 8.7,
H
C(2'),
H
C(6')); 6.73 (d, J(3',2') ˆ J(6',5') ˆ 8.7,
H C(3'), H C(5')); 6.94
(s, H C(3)); 3.72 (s, MeO).
(Z)-3-(3,4-Dihydroxyphenyl)-2-hydroxyprop-2-enoic Acid Methyl Ester (14a). UV (MeOH): 299 (8400),
219(5200). 1H-NMR (CD3OD): 7.57 (d, J(2',6') ˆ 2.0,
H
C(2')); 6.76 (d, J(5',6') ˆ 8.4,
H C(5')); 7.19
(dd, J(6',5') ˆ 8.4, J(6',2') ˆ 2.0, H C(6')); 7.01 (s, H C(3)); 3.82 (s, MeO). 13C-NMR (CD3OD): 136.86
(s, C(1')); 118.64 (d, C(2')); 145.94 (s, C(3')); 148.70 (s, C(4')); 116.05 (d, C(5')); 125.71 (d, C(6')); 129.05
(d, C(3)); 125.80 (s, C(2)); 167.39 (s, C(1)); 52.35 (q, MeO); on selective heteronuclear decoupling of MeO
(3.80 ppm), the quint. (J ˆ 3.7) at 167.39 (C(1)) in the fully coupled 13C-NMR became a d (J ˆ 3.7). EI-MS:
.
‡
‡
210(36, M ), 150(33), 123(100). HR-EI-MS: 210.0530 Æ 0.003 (C10H10O5 ; calc. 210.0528).
(E)-3-(3,4-Dihydroxyphenyl)-2-hydroxyprop-2-enoic Acid Methyl Ester (14b). 1H-NMR (CD3OD): 7.34
(d, J(2',3') ˆ J(6',5') ˆ 8.7,
H
C(2'),
H
C(6')); 6.73 (d, J(3',2') ˆ J(6',5') ˆ 8.7,
H C(3'), H C(5')); 6.94
(s, H C(3)); 3.72 (s, MeO). 13C-NMR (CD3OD): 137.60 (s, C(1')); 117.51 (d, C(2')); 145.79 (s, C(3')); 147.40
(s, C(4')); 115.89 (d, C(5')); 123.44 (d, C(6')); 131.13 (d, C(3)); 132.35 (s, C(2)); 166.40 (s, C(1)); 52.74
(q, MeO); on selective heteronuclear decoupling of MeO (3.72 ppm), the qd (J ˆ 4.2, 9.2) at 166.40 (C(1)) in
the fully coupled 13C-NMR became a d (J ˆ 9.2).
Photoisomerization of 3-Aryl-2-hydroxyprop-2-enoic Acid Methyl Esters. Compound 14a (0.30m in
CD3OD) was UV-irradiated (350 nm) in a 5-mm NMR tube for 1 h at r.t. under H-NMR monitoring, which
1
revealed clean equilibration with 14b to give a mixture 14a/14b 45 :55; no by-products could be detected. The
composition did not change on either subjecting the mixture to a further 1-h UV irradiation (350 nm) or
allowing it to stand for 12 h at r.t. in the dark.
UV irradiation of 13b/14b 1:2 (0.02m) under similar conditions gave 13a/13b 45 :55 and 14a/14b 45 :55. The
latter mixture was separated by HPLC into the pure components.
Acetylation of 13a and 14a. Dihydroxy ester 13a (5.0 mg, 0.025 mmol) was stirred in dry pyridine (0.6 ml)
and Ac2O (0.6 ml) at r.t. for 1 h. The mixture was then quenched with H2O (2 ml) and extracted with AcOEt
(3 Â 3 ml). The org. phase was dried (Na2SO4) and evaporated: 15 (6.9 mg, 86%). Dihydroxy ester 14a gave 16
under similar conditions.
Data of 2-(Acetyloxy)-3-[4-(acetyloxy)phenyl]prop-2-enoic Acid Methyl Ester (15): 1H-NMR (CDCl3):
7.59 (d, J(2',3') ˆ J(6',5') ˆ 8.7, H C(2'), H C(6')); 7.12 (d, J(3',2') ˆ J(5',6') ˆ 8.7, H C(3'), H C(5')); 7.27
(s, H C(3)); 3.84 (s, MeO); 2.32, 2.30 (2s, AcO). 13C-NMR (CDCl3): 169.10 (2s); 162.80 (s); 151.51 (s); 136.89
.
‡
(s); 131.28 (2d); 129.66 (s); 126.32 (d); 121.98 (2d); 52.86 (q); 21.16, 20.69 (2q). EI-MS: 278 (0.6, M ), 247(2),
236(18), 194(40), 134(50), 43(100).
Data of 2-(Acetyloxy)-3-[3,4-bis(acetyloxy)phenyl]prop-2-enoic Acid Methyl Ester (16): 1H-NMR
(CDCl3): 7.46 (d, J(2',6') ˆ 2.0, H C(2')); 7.20 (d, J(5',6') ˆ 8.7, H C(5')); 7.44 (dd, J(6',5') ˆ 8.7, J(6',2') ˆ
2.0, H C(6')); 7.25 (s, H C(3)); 3.83 (s, MeO); 2.31, 2.29 (1 and 2s, resp., AcO). 13C-NMR (CDCl3): 168.38
(s); 167.92 (2s); 162.71 (s); 142.11 (s); 137.57 (s); 130.61 (s); 128.28 (d); 125.57 (d); 124.80 (d); 123.67 (d); 52.72
.
.
‡
‡
(q); 20.69, 20.65, 20.59 (3q). EI-MS: 336 (0.5, M ), 305(1.3), 294(15, [M CH2ˆCO] ), 252(44), 210(58),
‡
‡
140(58), 43(100). HR-EI-MS: 305.0661 Æ 0.003 (C15H13O7 [M MeO] ; calc. 305.0661), 294.0737 Æ 0.003
.
‡
(C14H14O7 ; calc. 294.0739).
REFERENCES
[1] J. T. Carlton, in ꢀDynamique dꢁEspeces Marines Invasives: Application a lꢁExpansion de Caulerpa taxifolia
Á
Á
Â
Â
Â
en Mediterraneeꢁ, Ed. Academie des Sciences, TEC & DOC, Paris, 1997, p. 7.
[2] a) J. B. Hughes, G. C. Daily, P. R. Ehrlich, Science (Washington, D.C.) 1997, 278, 689; b) S. Nee, R. M. May,
ibid. 1997, 278, 692.
[3] R. M. May, Phil. Trans. Royal Soc. London B 1994, 343, 105.
[4] G. L. Lövel, Nature (London) 1997, 388, 627.
Á
Á
[5] a) F. Pietra, in ꢀDynamique dꢀEspeces Marines Invasives: Application a lꢁExpansion de Caulerpa taxifolia
Â
Â
Â
en Mediterraneeꢁ, Ed. Academie des Sciences, TEC & DOC, Paris, 1997, p. 255; b) A. Guerriero,
D. Depentori, M. DꢁAmbrosio, M. Durante, F. Dini, F. Pietra, J. Chem. Soc., Chem. Commun. 1994, 2083;
c) A. Guerriero, M. DꢁAmbrosio, G. Guella, F. Dini, F. Pietra, in ꢀFirst International Workshop on
Caulerpa taxifoliaꢁ, Eds. C. F. Boudouresque, A. Meinesz, and V. Gravez, GIS Posidonie, Marseille, 1994,
p. 171.

Helvetica Chimica Acta ± Vol. 81 (1998)
1691
[6] A. Raffaelli, S. Pucci, F. Pietra, Anal. Commun. 1997, 34, 179.
[7] a) N. Murakami, T. Morimoto, H. Imamura, T. Ueda, S-i. Nagai, J. Sakakibara, N. Yamada, Chem. Pharm.
Bull. 1991, 39, 2277; b) N. Murakami, H. Imamura, J. Sakakibara, N. Yamada, ibid. 1990, 38, 3497; c)
V. Reshef, E. Mizrachi, T. Maretzki, C. Silberstein, S. Loya, A. Hizi, S. Carmeli, J. Nat. Prod. 1997, 60, 1251;
d) N. Murakami, T. Morimoto, H. Imamura, A. Nagatsu, J. Sakakibara, Tetrahedron 1994, 50, 1993; e)
H. Imamura, T. Morimoto, T. Ueda, S.-i. Nagai, J. Sakakibara, N. Yamada, Tetrahedron Lett. 1991, 32, 1331.
[8] T. Morimito, A. Nagatsu, N. Murakami, J. Sakakibara, H. Tokuda, H. Nishino, A. Iwashima, Phytochem-
istry 1995, 40, 1433.
[9] F. Van Middlesworth, M. Lopez, M. Zweerink, A. M. Edison, K. Wilson, J. Org. Chem. 1992, 57, 4753.
[10] W. Fisher, E. Heinz, M. Zeus, Hoppe-Seylerꢀs Z. Physiol. 1973, 354, 1115.
[11] M. Kobayashi, K. Hayashi, K. Kawazoe, I. Kitagawa, Chem. Pharm. Bull. 1992, 40, 1404.
[12] a) G. Falsone, F. Cateni, E. Vrech, L. Birkofer, V. Lucchini, H. Wagner, O. Seligmann, L. Coassini Lokar,
Z. Naturforsch., B 1994, 49, 1297; b) T. Arao, M. Yamada, Phytochemistry 1989, 28, 805.
[13] H. C. van Hummel, Fortschr. Chem. Org. Naturst. 1975, 32, 267.
[14] Y. Oshima, S. Yamada, K. Matsunaga, T. Moriya, Y. Ohizumi, J. Nat. Prod. 1994, 57, 534.
[15] a) A. Murikami, Y. Nakamura, K. Koshimizu, H. Ohigashi, J. Agric. Food Chem. 1995, 43, 2779; b)
P. G. Roughau, G. A. Thompson Jr., S. H. Cho, Arch. Biochem. Biophys. 1987, 259, 481; c) S. H. Cho,
G. A. Thompson, Jr., Plant Physiol. 1989, 90, 610.
[16] B. Wickberg, Acta Chem. Scand. 1958, 12, 1187; H. Kikuchi, T. Tsukitani, T. Manda, T. Fujii, H. Nakanishi,
M. Kobayashi, I. Kitagawa, Chem. Pharm. Bull. 1982, 30, 3544; Y. Oshima, S. Yamada, K. Matsunga,
T. Moriya, Y. Ohizumi, J. Nat. Prod. 1994, 57, 534.
[17] A. J. L. Cooper, J. Z. Ginos, A. Meister, Chem. Rev. 1983, 83, 321.
[18] a) C. A. Kingsbury, D. Draney, A. Sopchik, W. Rissler, D. Durham, J. Org. Chem. 1976, 41, 3863; b)
F. W. Wehrli, A. P. Marchand, S. Wehrli, in ꢀInterpretation of Carbon-13 NMR Spectraꢁ, 2nd edn., J. Wiley,
New York, 1988, p. 79; c) G. Guella, I. Mancini, H. Zibrowius, F. Pietra, Helv. Chim. Acta 1988, 71, 773.
[19] A. S. R. Anjaneyulu, C. V. S. Prakash, K. V. S. Raju, U. V. Mallvadhani, J. Nat. Prod. 1992, 55, 496.
[20] J. Magraner, F. M. Pinto, E. Anselmi, M. Hernandez, R. Perez, J. D. Martín, C. Advenier, M. L. Candenas,
Br. J. Pharmacol. 1998, 123, 259.
[21] J. Magraner, E. Morcillo, P. Ausina, F. M. Pinto, J. D. Martín, J. Moreau, E. Anselmi, M. D. Barrachina,
J. Cortijo, C. Advenier, M. L. Candenas, Eur. J. Pharmacol. 1997, 326, 211.
[22] Y. Kasai, M. Iino, O. Tsutsumi, Y. Taketani, M. Endo, Br. J. Pharmacol. 1994, 112, 1132.
[23] N. Chairungsrilerd, K. Furukawa, T. Ohta, S. Nozoe, Y. Ohizumi, Eur. J. Pharmacol. 1996, 314, 351.
Á
[24] F. Della Pieta, A. R. Bilia, R. Scatizzi, M. C. Breschi, F. Cinelli, I. Morelli, in ꢀSecond International
Â
Workshop on Caulerpa taxifoliaꢁ, Eds. M. A. Ribera, E. Ballesteros, C.-F. Boudouresque, A. Gomez, and
V. Gravez, Universitat de Barcelona, Barcelona, 1994, p. 337.
[25] a) A. Guerriero, F. Marchetti, M. DꢁAmbrosio, S. Senesi, F. Dini, F. Pietra, Helv. Chim. Acta 1993, 76, 855;
b) A. Guerriero, D. Depentori, M. DꢁAmbrosio, F. Pietra, ibid. 1995, 78, 1755.
Â
[26] R. Lemee, D. Pesando, M. Durand-Clement, D. Dubreuil, A. Meinesz, A. Guerriero, F. Pietra, J. Appl.
Phycol. 1993, 5, 485.
[27] A. Guerriero, D. Depentori, M. DꢁAmbrosio, M. Durante, F. Dini, C. Geroni, F. Pietra, in ꢀSecond
International Workshop on Caulerpa taxifoliaꢁ, Eds. M. A. Ribera, E. Ballesteros, Ch.-F. Boudouresque,
Â
A. Gomez, V. Gravez, Universitat de Barcelona, Barcelona, 1994, p. 233.
Â
[28] D. Pesando, R. Lemee, C. Ferrua, P. Amade, J.-P. Girard, Aquatic Toxicology 1996, 35, 139.
Received July 10, 1998