Algal carotenoids. Part 64. Structure and chemistry of 4-keto-19′-hexanoyloxyfucoxanthin with a novel carotenoid end group

Article abstract of DOI:10.1039/a910345g

The structural elucidation of a new carotenoid 4-keto-19′'-hexanoyloxyfucoxanthin 5 from Emiliania huxleyi is documented by chromatographic (HPLC, TLC), spectroscopic (VIS, EIMS, FABMS, FABMSMS, 2D ¹H NMR) and chemical evidence. The novel carotenoid end group exhibits particular spectroscopic and chemical properties. In particular the reactions with base and acid are investigated. Due to a very weak molecular ion upon electron impact and facile cleavage to paracentrone 20 related fragments, the new carotenoid was previously misidentified as 19′-hexanoyloxyparacentrone 3-acetate 8, also found in other prymnesiophytes (haptophytes). This novel carotenoid readily undergoes cleavage to a C₃₁-skeletal paracentrone 20 related product upon storage, preferably in methanol solution. The new end group represents a plausible precursor for C₃₁-skeletal methyl ketone apocarotenoid metabolites in animals, and differs from the previously suggested precursor.

Full text of DOI:10.1039/a910345g

View Article Online / Journal Homepage / Table of Contents for this issue
Algal carotenoids. Part 64.1 Structure and chemistry of 4-keto-
19Ј-hexanoyloxyfucoxanthin with a novel carotenoid end group
Einar Skarstad Egeland,a José Luis Garrido,b Manuel Zapata,c Miguel Angel Maestrod and
Synnøve Liaaen-Jensen*a
1
a Organic Chemistry Laboratories, Department of Chemistry, Norwegian University of Science
and Technology, Gløshaugen, NO-7491 Trondheim, Norway
b Instituto de Investigacións Mariñas, C.S.I.C., Av. Eduardo Cabello 6, ES-36208 Vigo, Spain
c Centro de Investigacións Mariñas, Consellería de Pesca-Xunta de Galicia, Apdo. 13,
ES-36620 Vilanova de Arousa, Spain
d Servicios Xerais de Apoio á Investigación, Universidade da Coruña, Campus da Zapateira,
ES-15071 A Coruña, Spain
Received (in Lund, Sweden) 20th December 1999, Accepted 22nd February 2000
Published on the Web 29th March 2000
The structural elucidation of a new carotenoid 4-keto-19Ј-hexanoyloxyfucoxanthin 5 from Emiliania huxleyi is
documented by chromatographic (HPLC, TLC), spectroscopic (VIS, EIMS, FABMS, FABMSMS, 2D 1H NMR)
and chemical evidence. The novel carotenoid end group exhibits particular spectroscopic and chemical properties.
In particular the reactions with base and acid are investigated.
Due to a very weak molecular ion upon electron impact and facile cleavage to paracentrone 20 related fragments,
the new carotenoid was previously misidentified as 19Ј-hexanoyloxyparacentrone 3-acetate 8, also found in other
prymnesiophytes (haptophytes).
This novel carotenoid readily undergoes cleavage to a C31-skeletal paracentrone 20 related product upon storage,
preferably in methanol solution.
The new end group represents a plausible precursor for C31-skeletal methyl ketone apocarotenoid metabolites in
animals, and differs from the previously suggested precursor.
published in this journal in 1969.3,6 The present paper deals
Introduction
with a novel, naturally occurring fucoxanthin related caro-
tenoid, crowded with oxygen functions, which has revealed
unexpected chemistry including exceptional instability with
cleavage in neutral solvents to C31-skeletal paracentrone 20
related products.
Whereas carotenoids with few oxygen functions readily under-
go predictable chemical reactions, xanthophylls with several
functional groups such as peridinin 1 and fucoxanthin 2 offer
interesting chemistry.2–5
Detailed accounts on the structure elucidation of fuco-
xanthin 2 and of paracentrone 20 by Weedon’s school were
Early work7,8 on the carotenoid composition of the micro-
alga Emiliania huxleyi [Coccolithus huxleyi, Prymnesiophyceae
DOI: 10.1039/a910345g
J. Chem. Soc., Perkin Trans. 1, 2000, 1223–1230
This journal is © The Royal Society of Chemistry 2000
1223

View Article Online
Fig. 1
(Haptophyceae)] has recently been reviewed,9 and also includes
subsequent HPLC studies.10
and CH-3. The interconnectivities were established by the
1H–1H COSY spectrum. Three methyl singlets were compatible
with the predicted chemical shifts for CH3-16, 17, 18. Deuter-
ium exchange of the 3-hydroxy proton was noted. An altern-
ative prasinoxanthin12 related end group 6 could be ruled out
from the lack of a hydrogen bonded hydroxy group at C-6 and
of terminal methylene.
By improved HPLC techniques Garrido and Zapata9 have
recently isolated and partly characterised a new fucoxanthin
related carotenoid from Emiliania huxleyi. The new carotenoid
had polarity similar to that of fucoxanthin 2,9 and had VIS
properties close to those of 2 and the major carotenoid 19Ј-
hexanoyloxyfucoxanthin 3,8 and had spectral fine-structure
between that of 2 and 3. A molecular weight of 786 was estab-
lished by FABMS;9 loss of 116 mass units, compatible with the
loss of hexanoic acid,8 indicated further a structural relation-
ship to 3 (M = 772).9
Observed mass spectrometric fragments by EIMS including
exact mass measurements were compatible with the cleavages
indicated in Scheme 1. Arrow heads are pointing towards
the charged fragment. The molecular ion was very weak. A
prominent peak at m/z 618 is compatible with α-cleavage of
C-6,7 with hydrogen transfer to the polyene moiety. FABMS
confirmed the molecular ion (M = 786) and FABMSMS
showed fragments compatible with combined losses of water,
acetic acid and hexanoic acid.
Additional support for structure 5 was sought by chemical
derivatisation. Acetylation provided a less polar product of
unchanged chromophore, compatible with the introduction of
a second acetate function at C-3.
A priori treatment of 5 with base was expected to result in
three sets of reactions: i) hydrolysis of the two ester func-
tions, ii) fucoxanthin 2 related chemistry with the formation
of kinetically controlled products of isofucoxanthinol 9 type,
as well as thermodynamically controlled yellow hemiketal
products 10,4 Scheme 2, and iii) astaxanthin 11 related chem-
istry with diosphenol (enolised α-diketone) formation 12,13
Scheme 2.
However, the new carotenoid behaved somewhat unexpect-
edly upon treatment with base. Even in the absence of added
base, upon prolonged storage in acetone solution at Ϫ20 ЊC, or
shorter storage in darkness at room temperature in acetone or
preferably methanol, rearrangement to a new, less polar prod-
uct 8 was observed. On the basis of structure 5 and the reported
partial racemisation of astaxanthin 11 under basic conditions,14
which must occur via enolate formation, the formation and
structure of product 8 is rationalised in Scheme 3.
The structural elucidation of this new carotenoid is reported
here.
Results and discussion
The new carotenoid represents around 19% of the total caro-
tenoids of Emiliania huxleyi, and extensive chromatographic
purification was required for separation from the major caro-
tenoid 3 for isolation of 1 mg of the new carotenoid.
1
Further spectroscopic evidence (2 D H NMR, EIMS and
FABMS) and chemical derivatisation are compatible with
structure 5 (4-keto-19Ј-hexanoyloxyfucoxanthin) for the new
carotenoid, thus possessing a novel carotenoid end group,
imposing particular 1H NMR and chemical properties.
1
2D H NMR (400 MHz, CDCl3)-Assignments are shown in
Fig. 1. Available for direct comparison was the fully assigned
1
high field H NMR spectrum of 19Ј-butanoyloxyfucoxanthin
4.11 Identification of the structural moiety from the C-8 keto
group onwards of the carotenoid skeletons of 5 and 4 was sup-
ported by comparison of chemical shifts and coupling patterns,
as well as a fingerprint similarity of the crowded 6.3–6.6 ppm
olefinic region. Attachment of an acetate function at C-3Ј and a
fatty acid ester at C-19Ј was evident. Assignment of the previ-
ously unknown end group was supported by two doublets at
δ 2.72 and 3.63 for CH2-7 and the coupling system for CH2-2
1224
J. Chem. Soc., Perkin Trans. 1, 2000, 1223–1230

View Article Online
Scheme 1
Scheme 2
The driving force of this retroaldol type reaction, following
revealing the conversion of Product 2 17 to Product 3 19. The
a weak enolisation of 5, may be the formation of the hypo-
thetical, conjugated cyclohexenedione 14.
Upon treatment of the new carotenoid 5 with added base two
of the three products, Products 1 16 and 3 (tentatively 19),
observed are also based on enolisation of the non-conjugated
α-ketol function, Scheme 4.
latter transformation (Scheme 4) serves to explain why thermo-
dynamically controlled hemiketal products, cf. conversion of 2
via 9 to 10 (Scheme 2) did not occur due to competing enolis-
ation of Product 2 17 and subsequent dehydration to provide a
conjugated system.
The evidence for the structures of Products 1 16, 2 17 and
The reaction with base was monitored by HPLC, thereby
3 19 will now be discussed. The spectroscopic properties of
J. Chem. Soc., Perkin Trans. 1, 2000, 1223–1230
1225

View Article Online
Scheme 3
vious experience astaxanthin 11 with two α-ketol end groups
readily form astacene 12,13 whereas carotenoids like flexi-
xanthin with one α-ketol end group are more resistant towards
diosphenol formation under alkaline conditions in the presence
of traces of O2.17 Presumably lack of conjugation of the keto
function in 5 and 17 reduces the tendency to direct diosphenol
formation.
It has previously been proposed18 that paracentrone 20 in
sea urchins originated from dietary fucoxanthin 2 via a hypo-
thetical 3-keto derivative of 2, and an in vitro conversion of
fucoxanthin 2 to paracentrone 20 acetate was effected under
Oppenauer oxidation conditions.18 The alternative route
presented here (Scheme 4) for the retroaldol cleavage of 4-keto-
19Ј-hexanoyloxyfucoxanthin 5 is based on enolization of a
3-hydroxy-4-keto α-ketol. The new end group represents a
plausible precursor to C31-skeletal methyl ketone apocarotenoid
metabolites in animals.
Fig. 2
Product 1 16 (19Ј-hydroxyparacentrone) were compared with
those of the synthetic methyl ketone apocarotenoid para-
centrone 20.15 VIS and MS data for Product 1 were compatible
with the structure assigned, supported by the comparative H
NMR assignments given in Fig. 2.
Fucoxanthin 2 is known to produce a dark blue colour upon
reaction with strong acids. The reaction has been rationalised
by the formation of a blue oxonium ion 21,5 Scheme 5. In
parallel experiments with 4-keto-19Ј-hexanoyloxyfucoxanthin
5 the new carotenoid developed a blue colour only slowly
when directly compared to fucoxanthin 2 upon acid treatment.
This result is predicted if acid catalysed enolisation of the
4-keto compound 5 to 22 is a competing reaction to oxonium
ion formation 23 as shown in Scheme 5. Also the blue oxonium
ion 23 formed from 5 was reacted to give a yellow product 24
upon base treatment with nucleophilic attack at C-4, cf. ref. 5
In conclusion the spectroscopic evidence and chemical
behaviour of the new algal carotenoid are consistent with struc-
ture 5. The R-chirality of the allenic end group follows from the
chemical shift of the H-8Ј proton.19 Furthermore, the relative
stereochemistry of the primed end group is compatible with
the chemical shifts, whereas the chirality proposed for the
unprimed end group is based on biosynthetic analogy, as
for fucoxanthin 2.20
1
Product 2 17, more polar than Product 1 16, had VIS absorp-
tion bathochromically shifted relative to that of 5, resembling
the VIS spectrum of isofucoxanthinol 9. Comparative 1H
NMR assignments4 are given in Fig. 3. Of diagnostic import-
ance is the H-7 proton singlet at δ 6.31 and three methyl singlets
(δ 1.11, 1.41 and 1.69) belonging to the unprimed end group.
Upon electron impact no molecular ion for 17 could be
observed, but an M Ϫ 18 ion compatible with the C40H34O7
structure was observed.
Product 3 19 (Scheme 4) had VIS absorption slightly more
bathochromically shifted than Product 2 17, and was less polar
than Product 2 17 and Product 1 16, which is compatible with
the tentative structure assigned including the hydrogen bonded
enolised α-diketone in the unprimed end group, cf. ref. 16. No
NMR data were obtained.
The particular instability of the new carotenoid 5, which is
readily cleaved upon storage in methanol or acetone solution,
is unique in carotenoid context. Obviously fast isolation and
Diosphenol formation, as for astaxanthin 11 to astacene 12,
was not observed upon base treatment of 5. According to pre-
1226
J. Chem. Soc., Perkin Trans. 1, 2000, 1223–1230

View Article Online
Scheme 4
storage in the dry state in the absence of solvents is recom-
as a new carotenoid in 1976.8 The carotenoid was isolated from
mended. Also the requirement of efficient HPLC and TLC sys-
tems for the separation of this carotenoid from the 4-deoxo
derivative should be emphasized, as well as facile retroaldol
type cleavage upon electron impact, providing no or a weak
molecular ion by EIMS.
Emiliania (Coccolithus) huxleyi clone BT-6. Later, 19-hexanoyl-
oxyparacentrone 3-acetate 8 was identified as a minor caroten-
oid in Chrysochromulina species.21 The mass spectra from these
analyses are still available. In our present study of E. huxleyi
clone CCMP 370, no 19-hexanoyloxyparacentrone 3-acetate 8
could be detected. By comparison of the mass spectra of
19-Hexanoyloxyparacentrone (as a 3-acetate, 8) was reported
J. Chem. Soc., Perkin Trans. 1, 2000, 1223–1230
1227

View Article Online
Fig. 3
Scheme 5
4-keto-19Ј-hexanoyloxyfucoxanthin 5 obtained in this work
Experimental
with the earlier obtained mass spectra of 19-hexanoyloxy-
paracentrone 3-acetate 8,8,21 it became obvious that the latter
represented a misidentification of the new carotenoid 4-keto-
19Ј-hexanoyloxyfucoxanthin 5. The prominent fragment ion at
m/z 618 had earlier been erroneously identified as the molecular
ion, instead of the proper, very weak molecular ion at m/z 786,
observed in the original mass spectrum together with weak
fragment ions at m/z 652 (M Ϫ 116 Ϫ 18) and m/z 634 (M Ϫ
116 Ϫ 18 Ϫ 18).
Biological material
For growth conditions of Emiliania huxleyi clone CCMP 370
(isolated from Oslofjorden, Norway by E. Paasche in 1959), see
ref. 9.
General methods
Solvents were of distilled or p.a./HPLC quality. All carotenoid
samples were stored in a freezer (≈Ϫ20 ЊC) under nitrogen.
Manipulations were carried out as far as possible in darkness,
in the absence of air, acids, alkali and at low temperature.
The question of further occurrence of the new fucoxanthin
derivative 5 and other 4-keto analogues will be pursued.
1228
J. Chem. Soc., Perkin Trans. 1, 2000, 1223–1230

View Article Online
Instruments
major fragments at 768.4 (M Ϫ 18), 670.5 (M Ϫ 116), 652.9
(M Ϫ 18 Ϫ 116) and 610.4 (M Ϫ 60 Ϫ 116) and minor frag-
HPLC was carried out on a Hewlett Packard instrument series
1050, detector 1040 A, and a HP 79994 HPLC ChemStation
1
ments at 706.9, 627.4, 593.4 and 577.4; H NMR (400 MHz,
CDCl3, including 1H–1H COSY), double primed numbers refer
to the fatty acid side chain: δ 0.87 (3H, t, J = 7, H-6Љ), 1.07 (3H,
s, H-17Ј), 1.10 (3H, s, H-17), 1.23 (3H, s, H-16 or H-18), 1.25
(4H, broad s, H-4ٞ, H-5Љ), 1.29 (3H, s, H-16 or H-18), 1.37 (3H,
s, H-16Ј or H-18Ј), 1.37 (1H, m, H-2Јax), 1.38 (3H, s, H-16Ј or
H-18Ј), 1.48 (1H, m, H-4Јax), 1.62 (2H, quintet, J = 7.5, H-3Љ),
1.93 (3H, s, H-19), 1.97 (3H, s, H-20Ј), 1.98 (1H, m, H-2Јeq),
1.99 (3H, s, H-20), 2.03 (3H, s, H-acetate), 2.09 (1H, dd,
J = 12.2, 14.0, H-2ax), 2.26 (1H, m, H-4Јeq), 2.28 (2H, t,
J = 7.5, H-2Љ), 2.69 (1H, dd, J = 7.4, 14.1, H-2eq), 2.72 (1H, d,
J = 18.1, H-7a), 3.39 (1H, d, J = 3.2, 3-OH, disappeared after
addition of D2O), 3.63 (1H, d, J = 18.3, H-7b), 4.44 (1H, m,
H-3, dd after addition of D2O), 4.74 (1H, distorted d, J = 11.7,
H-19Јa), 4.80 (1H, distorted d, J = 11.7, H-19Јb), 5.36 (1H, m,
H-3Ј), 6.05 (1H, s, H-8Ј), 6.30 (2H, “dd”, H-10Ј and H-14Ј),
6.41 (2H, “dd”, H-14 and H-12Ј), 6.54–6.78 (5H, m, H-11,
H-12, H-15, H-11Ј and H-15Ј), see Fig. 1.
1
program. H NMR spectra were recorded on a Bruker 400
MHz instrument with CDCl3 as solvent. EI mass spectra were
recorded on an AEI MS 902 spectrometer with a direct inlet
to the ion source. FAB mass spectra were obtained with a VG
Quattro spectrometer, using 3-nitrobenzoyl alcohol as sample
matrix. Visible light (VIS) spectra were recorded on a Perkin-
Elmer 552 spectrophotometer.
Extraction
The frozen cell concentrate was treated repeatedly with cold
(≈Ϫ10 ЊC) 30% MeOH in Me2CO on a glass sinter filter until
the filtrate was colourless.
Chromatography and spectroscopy
Systems for analytical TLC and HPLC and preparative separ-
ation by TLC were as specified elsewhere.22 In addition the
HPLC system of Rodriguez et al.23 was employed for the stabil-
ity studies of 5. Spectral fine-structure in the VIS absorption
spectra is defined as %III/II24 and Dv.25 Wavelengths given in
parentheses denote shoulders. Only diagnostically useful peaks
Acetylation gave one product with less polarity, VIS as for 5,
MS not informative.
Cleavage of 5 upon storage in solution
1
Two aliquots of 5 (each about 0.8 µg) were kept in i) acetone
or ii) methanol solution in darkness at room temperature. The
reaction mixture was analyzed by HPLC (system23) after 1, 4
and 7 days. In acetone peak ratios between 5 (tR = 21.08) and
the product (8, tR = 21.34) were 6.63, 3.36 and 0.07 respectively.
In methanol solution the corresponding peak ratios were 1.22,
0.44 and 0.39 respectively. Prolonged storage in acetone
solution in darkness at Ϫ20 ЊC resulted in complete conversion
of 5 to 8.
with m/z >100 are reported for the mass spectra. H NMR
coupling constants J are given in Hz.
Reactions
Alkali treatment was performed with 5% KOH in MeOH. After
30 to 60 min, H2O was added and the mixture extracted with
EtOAc. Acetylation was carried out in dry pyridine with Ac2O.
The mixture was extracted after 2 h with EtOAc upon dilution
with H2O.
The product 10 had VIS λmax 453 nm (round-shaped) relative
to 5 with VIS λmax 446 and 470 nm in the same HPLC solvent.
The colour test for 8-keto-5,6-epoxidic carotenoids was
performed in parallel experiments for fucoxanthin 2 and
4-keto-19Ј-hexanoyloxyfucoxanthin 5 (20 µg carotenoid in 3 ml
MeOH). The carotenoids were reacted with concentrated HCl
(1 drop). Upon shaking fucoxanthin 2 developed immediately a
blue colour, cf. ref. 5, whereas the 4-keto derivative 5 after 4 h
had attained only a green colour. VIS spectroscopy demon-
strated that a blue product (λmaxMeOH 690 nm, compared to
λmaxMeOH 700 nm for the fucoxanthin product) was present
beside carotenoid of unchanged chromophore. Upon addition
of KOH in MeOH to the reaction mixture from 5 an immediate
colour change to yellow was observed. The VIS spectrum indi-
cated the presence of new product(s) with λmaxMeOH ≈ 420 nm
in addition to product(s) of the same chromophore as 5.
Alkali treatment of 5
Conditions for the alkali treatment are specified under Reac-
tions. Products were isolated by TLC (silica Merck 5553 or
5554, 50% acetone in heptane).
19Ј-Hydroxyparacentrone (Product 1, 16). VIS λmax acetone/
nm: 440, 463, %III/II = 10, Dv = 0.90; Rf = 0.26 (TLC silica
Merck 5554, 50% acetone in heptane), tR = 7.5 (system,23 flow
1.25 ml minϪ1); EIMS 70 eV, 220Њ, m/z (rel. int.%): 478 [M]ϩ
(11), 460 [M Ϫ 18]ϩ (100), 442 [M Ϫ 18 Ϫ 18]ϩ (44), 426
1
[M Ϫ 18 Ϫ 18 Ϫ 16]ϩ (12), 424 [M Ϫ 18 Ϫ 18 Ϫ 18]ϩ (11); H
1
NMR (400 MHz, CDCl3, including H–1H COSY), see Fig. 2:
δ 1.09 (3H, s, H-17), 1.33 (3H, s, H-16), 1.39 (3H, s, H-18), 1.93
(3H, s, H-19Ј), 1.98 (≈1.5H, s, H-20), 1.99 (3H, d, J ≈ 1, H-20Ј),
2.00 (≈1.5 H, s, H-20), 2.36 (3H, s, H-7), 6.01 (≤1H, s, H-8), 6.21
(1H, d, J = 11.6, H-10), 6.29 (1H, d, J = 11.3, H-14), ≈6.40 (2H,
m, H-12 and H-14Ј), 6.50 (≈0.25H, s, H-8?), 6.60 (1H, m,
H-11Ј), ≈6.65 (3H, m, H-15, H-12Ј and H-15Ј), 6.72 (1H, m,
H-11), 7.13 (1H, d, J = 11.4, ≈1, H-10Ј).
4-Keto-19Ј-hexanoyloxyfucoxanthin 5
Available in total 1 mg; VIS λmax acetone/nm: (420), 443, 467,
%III/II = 38, Dv = 0.88, VIS λmax benzene/nm: (430), 455, 481,
%III/II = 9, Dv = 0.86, VIS λmax MeOH/nm: 443, 465, %III/
II = 28, Dv = 0.95, addition of dilute HCl caused no change;
Rf = 0.55 (TLC silica–calcium carbonate 2:1, 40% Me2CO in
heptane), Rf = 0.56 (TLC silica Merck 5553, Me2CO–EtOH–
CHCl3 12:1:87), tR = 17.2 (system,23 flow 1.25 ml minϪ1);
FABMS,9 787.47 [M ϩ H]ϩ, 786.46 [M]ϩ, 769.47 [M ϩ H Ϫ
H2O]ϩ; EIMS 70 eV, 220Њ, m/z (rel. int.%): 786 [M]ϩ (0.2), 768
[M Ϫ 18]ϩ (0.3), 750 [M Ϫ 18 Ϫ 18]ϩ (0.2), 618 [M Ϫ 168]ϩ
(17), 600 [M Ϫ 168 Ϫ 18]ϩ (7), 540 [M Ϫ 168 Ϫ 60 Ϫ 18]ϩ (3),
504 [M Ϫ 282]ϩ (5), 502 [M Ϫ 284]ϩ (4), 486 [M Ϫ 282 Ϫ 18]ϩ
(4), 484 [M Ϫ 284 Ϫ 18]ϩ (3), 424 [M Ϫ 168 Ϫ 116 Ϫ 60 Ϫ 18]ϩ
(6), 326 (8), 311 (8), 195 (26), 168 (29), 140 (71), 125 (100),
fragment ions observed also at 652 and 634 in another spec-
trum, corresponding to [M Ϫ 116 Ϫ 18]ϩ and [M Ϫ 116 Ϫ 18 Ϫ
18]ϩ.8 The FABMS showed a prominent peak at m/z 786 (Mϩ)
and peak at 810 (M ϩ Na)ϩ. FABMSMS performed by colli-
sion induced (Ar) dissociation of the molecular ion showed
4-Keto-19Ј-hydroxyisofucoxanthinol (Product 2, 17). VIS
λmax acetone/nm: 448, 463, %III/II = 13, Dv = 0.99; Rf = 0.63
(TLC silica Merck 5553, 16% EtOH in CHCl3), tR = 6.7 (sys-
tem,23 flow 1.25 ml minϪ1); EIMS 70 eV, 220Њ, m/z (rel. int.%):
626 [M Ϫ 18]ϩ (61), 608 [M Ϫ 18 Ϫ 18]ϩ (34), 602 (42), 582
1
1
(61), 551 (100); H NMR (400 MHz, CDCl3, including H–1H
COSY), see Fig. 3: δ 1.10 (3H, s, H-17Ј?), 1.11 (3H, s, H-17?),
1.34 (3H, s, H-16Ј?), ≈1.35 (1H, m, H-2Ј), 1.39 (3H, s, H-18Ј?),
1.41 (3H, s, H-16Ј?), ≈1.48 (1H, m, H-4Ј), 1.69 (3H, s, H-18?),
≈1.95 (1H, m, H-2Ј), ≈2.28 (1H, m, H-4Ј), 4.32 (1H, m, H-3Ј),
6.01 (≤1H, s, H-8Ј), 6.22 (1H, d, J = 11.6, H-10Ј), 6.30 (1H, d,
J = 11.1, H-14Ј), ≈6.40 (2H, m, H-12Ј and H-14), 6.51 (≈0.3H, s,
H-8Ј?), 6.58 (1H, dd, J = 11.1, 15.0, H-11), ≈6.65 (3H, m, H-12,
J. Chem. Soc., Perkin Trans. 1, 2000, 1223–1230
1229

View Article Online
8 N. Arpin, W. A. Svec and S. Liaaen-Jensen, Phytochemistry, 1976,
15, 529.
H-15 and H-15Ј), 6.72 (1H, m, H-11Ј), 7.13 (1H, “dd”, J = 10.3,
≈1, H-10).
9 J. L. Garrido and M. Zapata, J. Phycol., 1998, 34, 70.
10 (a) S. W. Wright and S. W. Jeffrey, Mar. Ecol. Prog. Ser., 1987,
38, 1987; (b) J. R. Nelson and S. G. Wakeham, J. Phycol., 1989, 25,
761; (c) S. W. Wright, S. Jeffrey, R. F. C. Mantoura, C. A. Llewellyn,
T. Bjørnland, D. Repeta and N. Welschmeyer, Mar. Ecol. Prog. Ser.,
1991, 77, 1991; (d) G. W. Kraay, M. Zapata and M. J. W. Veldhuis,
J. Phycol., 1992, 28, 708; (e) S. W. Jeffrey and S. W. Wright, in
The Haptophyte Algae, eds. J. C. Green and B. S. C. Leadbeater,
Clarendon, Oxford, 1994, p. 111.
11 G. Englert, T. Bjørnland and S. Liaaen-Jensen, Magn. Reson.
Chem., 1990, 28, 519.
12 (a) P. Foss, R. R. L. Guillard and S. Liaaen-Jensen, Phytochemistry,
1984, 23, 1629; (b) P. Foss, T. Skjetne and S. Liaaen-Jensen, Acta
Chem. Scand., Ser. B, 1986, 40, 172.
Product 3. VIS λmax HPLC eluent/nm: ≈ 475 (round, broad);
tR = 8.5 (system,23 flow 1.25 ml minϪ1).
In a separate experiment 5 was treated with 5% KOH in
methanol for 30 min. The reaction mixture was examined by
HPLC in the system of Rodriguez et al.23 Four peaks with
tR = 5.20 (VIS λmax in eluent 452 nm, rounded, 7% of total),
tR = 6.28 (VIS λmax 447, 465 nm, 65% of total), tR = 6.72 (VIS
λmax 460 nm, rounded, 23% of total) and tR = 8.05 (VIS λmax
470 nm, 5% of total) were observed, tentatively identified as
Product 2 17, 22 (hydrolysed 5), Product 1 16 and Product 3 19
respectively.
13 R. Kuhn, J. Stene and N. A. Sörensen, Chem. Ber., 1939, 72,
1688.
14 R. K. Müller, K. Bernhard, H. Meyer, A. Rüttimann and
M. Vecchi, Helv. Chim. Acta, 1980, 63, 1654.
Acknowledgements
15 J. A. Haugan, J. Chem. Soc., Perkin Trans. 1, 1997, 2731.
16 (a) K. Schiedt, in Carotenoids, Chemistry and Biology, ed. N. I.
Krinsky, M. M. Mathews-Roth and R. F. Taylor, Plenum,
London, 1990, p. 247; (b) G. Englert, in Carotenoids, Vol. 1B, eds.
G. Britton, S. Liaaen-Jensen and H. Pfander, Birkhäuser, Basel,
1995, p. 147.
E. S. E. was supported by a research grant from Hoffmann-
La Roche, Basel, to S. L.-J.
References
1 Part 63, E. S. Egeland, R. R. L. Guillard and S. Liaaen-Jensen,
Phytochemistry, 1997, 44, 1087.
17 A. J. Aasen and S. L. Jensen, Acta Chem. Scand., 1966, 20,
1970.
2 H. Kjøsen, S. Norga˚rd, S. Liaaen-Jensen, W. A. Svec, H. H. Strain,
P. Wegfahrt, H. Rapoport and F. T. Haxo, Acta Chem. Scand., Ser.
B, 1976, 30, 157.
3 R. Bonnett, A. K. Mallams, A. A. Spark, J. L. Tee and B. C. L.
Weedon, J. Chem. Soc. C, 1969, 429.
4 J. A. Haugan, G. Englert and S. Liaaen-Jensen, Acta Chem. Scand.,
1992, 46, 614.
5 J. A. Haugan and S. Liaaen-Jensen, Acta Chem. Scand., 1994, 48, 68.
6 G. Galasko, J. Hora, T. P. Toube, B. C. L. Weedon, D. André,
M. Barbier, E. Lederer and V. R. Villanueva, J. Chem. Soc. C, 1969,
1264.
7 (a) S. W. Jeffrey and M. B. Allen, J. Gen. Microbiol., 1964, 36, 227;
(b) S. Norga˚rd, W. A. Svec, S. Liaaen-Jensen, A. Jensen and R. R. L.
Guillard, Biochem. Syst. Ecol., 1974, 2, 7; (c) S. Hertzberg,
T. Mortensen, G. Borch, H. W. Siegelman and S. Liaaen-Jensen,
Phytochemistry, 1977, 16, 587.
18 J. Hora, T. P. Toube and B. C. L. Weedon, J. Chem. Soc. C, 1970,
241.
19 J. A. Haugan and S. Liaaen-Jensen, Tetrahedron Lett., 1994, 35,
2245.
20 Y. Yamano, C. Tode and M. Ito, J. Chem. Soc., Perkin Trans. 1,
1995, 1895.
21 B. Bjerkeng, J. A. Haugan and S. Liaaen-Jensen, Biol. Oceanogr.,
1991, 6, 271.
22 E. S. Egeland, G. Johnsen, W. Eikrem, J. Throndsen and S. Liaaen-
Jensen, J. Phycol., 1995, 31, 554.
23 F. Rodriguez, M. Zapata and J. L. Garrido, Chromatographia, 1998,
48, 677.
24 B. Ke, F. Imsgard, H. Kjøsen and S. Liaaen-Jensen, Biochim.
Biophys. Acta, 1970, 210, 139.
25 E. S. Egeland and S. Liaaen-Jensen, Phytochemistry, 1995, 40,
515.
1230
J. Chem. Soc., Perkin Trans. 1, 2000, 1223–1230