Absolute configurational assignment of 3-hydroxycarotenoids

Article abstract of DOI:10.1039/b002088p

Attempts were made to develop an exciton chirality method applicable to the unique cases represented by 3-hydroxycarotenoids. However, this approach has so far not been successful. The 3-hydroxy configurations were therefore determined by the extended Mosher ¹H-NMR method. Nine carotenoids with seven different end groups and of known 3-hydroxy configurations were derivatized with (R)- and (S)-methoxyphenylacetic acid (MPA); they all gave the expected shift differences. Three other auxiliary reagents with naphthalene and anthracene nuclei gave larger and consistent NMR shift differences, but they are not yet commercially available.

Full text of DOI:10.1039/b002088p

View Article Online / Journal Homepage / Table of Contents for this issue
Absolute configurational assignment of 3-hydroxycarotenoids
a
a
a
a
b
Thomas Andersson, Babak Borhan, † Nina Berova, Koji Nakanishi,* Jarle Andre Haugan
b
and Synnøve Liaaen-Jensen
1
a
Department of Chemistry, Columbia University, New York, NY 10027, USA
Organic Chemistry Laboratories, Department of Chemistry and Biology,
Norwegian University of Science and Technology, N-7491 Trondheim-NTNU, Norway
b
Received (in Cambridge, UK) 15th March 2000, Accepted 30th March 2000
Published on the Web 30th June 2000
Attempts were made to develop an exciton chirality method applicable to the unique cases represented by
-hydroxycarotenoids. However, this approach has so far not been successful. The 3-hydroxy configurations
3
1
were therefore determined by the extended Mosher H-NMR method. Nine carotenoids with seven different end
groups and of known 3-hydroxy configurations were derivatized with (R)- and (S)-methoxyphenylacetic acid (MPA);
they all gave the expected shift differences. Three other auxiliary reagents with naphthalene and anthracene nuclei
gave larger and consistent NMR shift differences, but they are not yet commercially available.
Introduction
X-Ray analysis has so far not been suited for stereochemical
assignments of chiral carotenoids due to difficulties involved in
1
crystallization. Stereochemical correlations have thus been
based on certain key carotenoid oxidation products with
configurations established by X-ray crystallography or by total
synthesis of optically active carotenoids.
Comparative CD spectroscopy, often in combination with
H-NMR data, has served as the most useful tool for establish-
1
ing the absolute chirality of naturally occurring carotenoids.
However, the current CD approach suffers from the disadvan-
tage that: (i) conjugation is essential between the chiral end
group and the polyene chromophore; (ii) Z-configuration in the
polyene chain inverts the Cotton effect in the case of certain
end groups; (iii) chiral allenic end groups (with two chiral
centers and one chiral axis) and chiral acetylenic end groups
give rise to only weak Cotton effects; and (iv) in carotenoids
with 5,6-epoxy or ε- end groups, the effect of the 3-hydroxy
2,3
group is practically nil.
Of the ca. 600 fully or partly characterized naturally
4
occurring carotenoids, ca. 60% are chiral. Of these about 300
are secondary alcohols, the hydroxy function most frequently
being present in substituted cyclic β- and ε- or aliphatic ψ-end
5
groups.
Results and discussion
Attempted application of the exciton chirality method
Using(3R)-3-hydroxy-β-ionine(1),‡λmax (EtOH)280nm(ε8200)
Chart 1
as a model, we attempted to apply the exciton chirality
method by acylating the 3-hydroxy with a chromophore, e.g.,
p-dimethylaminobenzoic acid (dmaBz) λmax (EtOH) 309 nm
6,7
the electric transition moments of the two chromophores con-
stitute a negative twist. However, it turns out that the unique
8
(
ε 30 4000), that would couple with the existing dienone
structure of the 3-hydroxy-β-ionine molecule does not allow
establishment of the 3-hydroxy configuration. This is because
of the flexibility of the cyclohexene ring, and the uncertainty of
the side-chain conformation around the 6-s-cis bond as evi-
chromophore (Chart 1). β-Ionone dmaBz (2) did indeed give a
negative couplet with a moderate CD amplitude (A = Ϫ8.3),
3
†
03 nm (Ϫ5.3)/273 nm (ϩ3.0) (in hexane), thus showing that
Present address: Department of Chemistry, Michigan State Uni-
denced by the multiple NOEs detected between 7-H/1-Me,
9
7
-H/5-Me, 7-H/9-Me, 8-H/1-Me, and 8-H/5-Me. However, the
overriding reason is that the two chromophores are located
para” across the cyclohexane ring and that the two chromo-
phoric electric transition moments lie in almost the same plane
versity, East Lansing, MI 48824, USA.
The IUPAC name for compound 1 is 4-(4-hydroxy-2,6,6-trimethyl-
“
‡
cyclohex-1-en-1-yl)but-3-en-2-one.
DOI: 10.1039/b002088p
J. Chem. Soc., Perkin Trans. 1, 2000, 2409–2414
This journal is © The Royal Society of Chemistry 2000
2409

View Article Online
Table 1 Shift differences δ(R) Ϫ δ(S) (ppm) of MPA esters of 3-hydroxy compounds
e
Compound
gem-diMe
2-Hax
2-Heq
4-Hax
4-Heq
5-Me
b
1
Ϫ0.048
Ϫ0.098
Ϫ0.096
Ϫ0.102
Ϫ0.101
Ϫ0.143
Ϫ0.137
Ϫ0.138
Ϫ0.16
ϩ0.177
ϩ0.173
ϩ0.17
ϩ0.173
n/a a
ϩ0.08
n/a
ϩ0.111
n/a
ϩ0.139
ϩ0.137
ϩ0.136
ϩ0.139
n/a a
ϩ0.061
ϩ0.056
ϩ0.054
ϩ0.055
ϩ0.032
ϩ0.054
Ϫ0.031
ϩ0.064
ϩ0.026
ϩ0.056
Ϫ0.061
b
8
Ϫ0.044
Ϫ0.050
Ϫ0.047
Ϫ0.034
Ϫ0.033
ϩ0.023
Ϫ0.058
Ϫ0.025
Ϫ0.045
ϩ0.119
b
a
a
9
b
1
1
1
1
1
1
1
1
0
1
2
3
4
5
6
6
Ϫ0.140
Ϫ0.135
b
b
a
a
Ϫ0.10
Ϫ0.14
ϩ0.13
n/a
c
ϩ0.085
Ϫ0.085
Ϫ0.078
Ϫ0.097
ϩ0.128
ϩ0.074
Ϫ0.101
Ϫ0.086
Ϫ0.140
ϩ0.233
c
ϩ0.099
n/a
c
c,f
c,g
ϩ0.172
n/a
ϩ0.137d
Ϫ0.142
a
1
1
b
c
d
e
f
g
Estimated from H– H COSY spectra. Solvent CDCl3. Solvent C D . Vinyl proton. Average shift difference. (β-3OH side). (ε-3OH side).
6
6
(
see 3). Depending on the rotation around the 6,7 bond and
could be used for determining the absolute configuration of
3-hydroxycarotenoids (Table 1, entry 1).
conformation of the cyclohexene ring, the chirality between the
transition moments of the side chain and the 3β-chromophore
can either be negative or positive. This situation in encountered
in the 3-hydroxy carotenoids as well. For this reason, although
the Trondheim group prepared several carotenoid 3-retinoates,
unfortunately the CD results were not conclusive.
Auxiliary chromophores other than MPA, namely chromo-
phores containing 1- and 2-naphthyl and 9-anthryl moieties
(Chart 1, 5, 6, 7) were also checked with 3-hydroxy-β-ionone
10
17
(Chart 1). Since however, only the (R)-enantiomers of these
acids were available, the induced NMR shifts were estimated
after separation of the two diastereomeric esters derived from
the racemic ionone. The largest shift differences were induced
by the 9-anthryl group, which was followed by 1-naphthyl
and then the 2-naphthyl reagents; the shifts were up to
several-fold that of MPA, which exhibited the smallest shifts.
However, the MPA esters of 3-hydroxycarotenoids yielded satis-
factory shift differences for making reliable determinations.
Furthermore, MPA is currently the only commercially available
compound.
Further CD tests revealed that insertion of a twisted auxiliary
chromophore exemplified by (R ,3R)-allenic β-ionone ester (4)
did give rise to a negative couplet around 275 nm, A = ca. Ϫ28,
associated with interaction between the biphenyl and ionone
chromophores. This approach was conceptually useful, but
difficulties in handling the reactive allenic esters, especially at
microgram levels, made it unsuited for the present carotenoid
case. However, attempts to devise a microscale CD method
applicable to cases represented by 3-hydroxy-β-ionine are still
being pursued.
a
8
All determinations of the 3-hydroxy configuration of caroten-
oids by the Mosher method gave results consistent with their
known absolute configurations. The configuration of all model
carotenoids studied here has been established by methods
including total synthesis of the optically pure carotenoids.
References for the synthesis of each compound are shown after
the compound number. The compounds used in the present
studies (Fig. 1) ranged from polyene 3-hydroxycarotenoids such
The Mosher method
In the following we report a systematic study on the application
of the extended Mosher method as an alternative for absolute
configurational assignment of 3-hydroxycarotenoids. This
derivatization/ H-NMR protocol does not suffer from the
mentioned CD related limitations and can be performed with
1
18
as zeaxanthin 8 to more complex structures containing allenes
19
2
00 microgram level samples. Nine carotenoid models of
as in paracentrone 12, triple bonds as in the apocarotenoid
20 21 22
known 3-hydroxy configuration with seven different end groups
were selected for this study (Fig. 1).
9,
triophaxanthin 10,
apohopkinsiaxanthin 11
and
24
23
22
tetradehydroastaxanthin 15, ketones in 11, astaxanthin 13,
25
Determination of the absolute configuration of chiral
tetrahydrozeaxanthin-8,8Ј-dione 14 and 15 and a Z-double
bond in 11. In some cases the compounds had opposite
chirality or two different chiral hydroxys as in astaxanthin 13
11
organic compounds by the Mosher method and exten-
sions
12,13
is performed by reacting the compound with two
26
enantiomeric auxiliary reagents and evaluating the shift differ-
ence between the two derivatives; configurations of primary
and secondary alcohols and amines and acids have been deter-
mined by this method. As depicted in Fig. 2, depending on the
chirality of the auxiliary reagent, the aromatic nuclei specific-
ally shield one side of the chiral compound. This gives rise to
shift differences of opposite signs between the two diastereo-
meric derivatives. For secondary alcohols, the frequently used
auxiliary reagent is the commercially available methoxyphen-
ylacetic acid (MPA). A conformational study of the ester
bond formed in MPA derivatives has shown that the carbonyl
oxygen is synperiplanar to both the methine hydrogen and the
methoxy group, and that the phenyl group and the C–H bond in
MPA reside in the same plane.
The applicability of the method was tested with (3R)-3-
hydroxy-β-ionone (1) synthesized as reported earlier. The
ionone 1 was converted into the corresponding (S)- and (R)-
MPA esters by reaction with excess (S)- and (R)-MPA in the
presence of 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide
hydrochloride (EDC) and 4-dimethylaminopyridine (DMAP).
A comparison of the shift differences between the two diastereo-
meric MPA esters of 1 confirmed that the Mosher method
and lutein 16. Note that the induced shifts differ only slightly
among the various carotenoids in Table 1. In cases where shift
differences are observed, it is because the presence of functional
groups such as ketones (11, 13, 15), allenes (12) or an ε-OH
group (16), might lead to conformational changes in the six
membered ring.
14
Quinoa et al. reported a method using only one MPA-
derivative and measuring the shift differences at two different
27
temperatures. At room temperature MPA is believed to con-
sist of two conformers that shield different sides of the chiral
molecule. One of the conformers is thermodynamically more
stable and more abundant at room temperature. This conformer
becomes more populated upon lowering the temperature, thus
enhancing the shielding effect on one side, while decreasing
the shielding effect from the other conformer. This results in
increased chemical shift difference between the two, hence lead-
ing to establishment of absolute configuration. To check this
method, (R)-MPA was reacted with (R,R)-zeaxanthin and the
12
15
16
1
H-NMR spectrum was taken at room temperature and at 215
12
K. However, in the present case the shift differences between
these two temperatures were all positive and hence not in
agreement with the proposed protocol.
2
410
J. Chem. Soc., Perkin Trans. 1, 2000, 2409–2414

View Article Online
Fig. 1 Structure of carotenoids 8–16.
Conclusion
previous determination of the 3ЈR-configuration which was
28,29
based on partial synthesis and CD.
The results demonstrate the usefulness of the NMR Mosher
method for assigning the absolute configurations of a variety of
3
-hydroxycarotenoids with different chain structures and end
Experimental
General experimental procedures
groups. The protocol can be performed at the microgram level
including derivatization. Z-Bonds in the polyene chain do not
affect the results, and conjugation of the chiral end group with
the polyene chain, such as zeaxaanthin 8, is not required.
Also, carotenoids with weak Cotton effects give the correct
assignment by this method. For lutein 16 the present results
represent a different and additional assignment from the
All compounds were purchased from Aldrich except the
3-hydroxycarotenoids, which were synthesized (lutein 16 was
isolated from alfalfa). Methylene chloride was distilled from
calcium hydride under nitrogen. Analytical and preparative
HPLC was performed on Waters (Rochester, MN) or Rainin
J. Chem. Soc., Perkin Trans. 1, 2000, 2409–2414
2411

View Article Online
quenched by addition of 1 ml of water and the product
extracted with diethyl ether and dried with Na CO . Yield 90%.
2
3
[
α] = ϩ72.2 (c = 0.116, 1,4-dioxane).
D
(
S)-MPA ester of (R)-3-hydroxy-ꢀ-ionone. Separated from
a reaction of racemic 3-hydroxy-β-ionone with (S)-MPA,
using a chiral column (Chiracel OD), 1:99 propan-2-ol–
1
hexane, λ = 280 nm. H-NMR (CDCl , 500 MHz) δ 1.09 (3H,
3
H
s, 1-Me), 1.13 (3H, s, 1-Me), 1.62 (1H, t, J = 12 Hz, 2-Hax),
1
1
.68 (3H, s, 5-Me), 1.81 (1H, ddd, J = 12, 3.5, 1.8 Hz, 2-Heq),
.99 (1H, dd, J = 17.5, 9 Hz, 4-Haz), 2.29 (3H, s, 9-Me), 2.36
(
1H, dd, J = 17.5, 5.6 Hz, 4-Heq), 3.42 (3H, s, O-Me), 4.74 (1H,
s, H-MPA), 5.11 (1H, dddd, J = 12, 9, 5.6, 3.5 Hz, 3-H), 6.07
(
(
1H, d, J = 16 Hz, 8-H), 7.16 (1H, d, J = 16 Hz, 7-H), 7.33–7.45
5H, phenyl). ESI-MS: 357.3 (M ϩ H).
(
S)-MPA ester of (S)-3-hydroxy-ꢀ-ionone. Separated from
a reaction of racemic 3-hydroxy-β-ionone with (S)-MPA,
using a chiral column (Chiracel OD), 1:99 propan-2-ol–hexane,
1
λ = 280 nm. H-NMR (CDCl , 500 MHz) δ 1.04 (3H, s, 1-Me),
3
H
1
.10 (3H, s, 1-Me), 1.53 (1H, t, J = 12 Hz, 2-Hax), 1.67 (1H,
ddd, J = 12, 3.6, 1.8 Hz, 2-Heq), 1.74 (3H, s, 5-Me), 2.17 (1H,
dd, J = 18, 9 Hz, 4-Hax), 2.29 (3H, s, 9-Me), 2.50 (1H, dd,
J = 18, 6 Hz, 4-Heq), 3.42 (3H, s, O-Me), 4.75 (1H, s, H-MPA),
5.10 (1H, dddd, J = 12, 9, 6, 3.5 Hz, 3-H), 6.09 (1H, d, J = 16
Hz, 8-H), 7.16 (1H, d, J = 16 Hz, 7-H), 7.33–7.46 (5H, phenyl).
ESI-MS: 357.3 (M ϩ H).
Fig. 2 Expected NMR shifts in the diastereomeric esters of methoxy-
phenylacetic acid (MPA) and the actual shifts observed in ester 10 in
CDCl3.
(
R)-1-NMA ester of (R)-3-hydroxy-ꢀ-ionone (5a). Separated
from a reaction of racemic 3-hydroxy-β-ionone with (R)-1-
NMA, using a chiral column (Chiracel OD), 1:99 propan-2-ol–
(
Woburn, MA) HPLC. The diastereomeric mixtures obtained
1
hexane, λ = 280 nm. H-NMR (CDCl , 500 MHz) δ 0.92 (3H,
by reaction of racemic 3-hydroxy-β-ionone with (S)-MPA, (R)-
1
7
propan-2-ol–hexane, λ = 280 nm. H and H– H COSY spectra
were recorded on a Bruker 500 DMX MHz spectrometer and
performed in CDCl or C D . Chemical shifts (δ) are reported
3
H
s, 1-Me), 1.05 (3H, s, 1-Me), 1.40 (1H, t, J = 12 Hz, 2-Hax), 1.57
-NMA (5a, 5b), (R)-2-NMA (6a, 6b) and (R)-9-ATMA (7a,
b) were separated using a chiral column (Chiracel OD), 1:99
(
(
1H, ddd, J = 12, 3, 1.6 Hz, 2-Heq), 1.71 (3H, s, 5-Me), 2.14
1H, dd, J = 18, 9 Hz, 4-Hax), 2.27 (3H, s, 9-Me), 2.48 (1H, dd,
1
1
1
J = 18, 6 Hz, 4-Heq), 3.46 (3H, s, O-Me), 5.09 (1H, m, 3-H),
.38 (1H, s, H-NMA), 6.05 (1H, d, J = 16 Hz, 8-H), 7.12 (1H,
d, J = 16 Hz, 7-H), 7.46–8.30 (7H, 1-naphthyl). ESI-MS: 407.3
M ϩ H).
5
3
6
6
in ppm downfield from internal TMS and coupling constants
J) in Hz. ESI mass spectra were measured on a JEOL JMS-
mate LSMS-system. [α]D values were measured on a JASCO
(
(
Ϫ1
2
Ϫ1
(
R)-1-NMA ester of (S)-3-hydroxy-ꢀ-ionone (5b). Separated
DIP-1000 polarimeter and are given in 10 deg cm g .
Preparative TLC-plates used were Uniplate Silicagel GF
2
from a reaction of racemic 3-hydroxy-β-ionone with (R)-1-
NMA, using a chiral column (Chiracel OD), 1:99 propan-2-ol–
0 × 20 cm 500 or 1000 microns, and they were bought from
1
hexane, λ = 280 nm. H-NMR (CDCl , 500 MHz) δ 1.02 (3H,
Analtech Inc.
3
H
s, 1-Me), 1.10 (3H, s, 1-Me), 1.56 (3H, s, 5-Me), 1.56 (1H, t,
General esterification procedure
J = 12 Hz, 2-Hax), 1.77 (1H, ddd, J = 12, 3, 1.6 Hz, 2-Heq), 1.83
(
1H, dd, J = 18, 9 Hz, 4-Hax), 2.25 (1H, dd, J = 18, 6 Hz,
The alcohol (normally 0.25 mg, 1 eq.), excess auxiliary acid
4
3
-Heq), 2.27 (3H, s, 9-Me), 3.46 (3H, s, O-Me), 5.10 (1H, m,
-H), 5.36 (1H, s, H-NMA), 5.99 (1H, d, J = 16 Hz, 8-H), 7.12
(
>10 eq.) and excess EDC (>10 eq.) were dissolved in ca. 1 ml
dry CH Cl and DMAP (1 eq.) was then added to the solution
2
2
(
1H, d, J = 16 Hz, 7-H), 7.46–8.29 (7H, 1-naphthyl).
and stirred at room temperature for 5–30 min. The reaction
was followed on TLC (ca. 30% ethyl acetate–hexane). When
the reaction was completed the reaction mixture was directly
applied to and eluted (20–30% ethyl acetate–hexane) on a
preparative TLC plate. Prior to NMR analysis the samples were
dried in vacuum overnight.
(
R)-2-NMA ester of (R)-3-hydroxy-ꢀ-ionone (6a). Separated
from a reaction of racemic 3-hydroxy-β-ionone with (R)-2-
NMA, using a chiral column (Chiracel OD), 1:99 propan-2-ol–
1
hexane, λ = 280 nm. H-NMR (CDCl , 500 MHz) δ 1.01 (3H,
3
H
s, 1-Me), 1.08 (3H, s, 1-Me), 1.52 (1H, t, J = 12 Hz, 2-Hax), 1.66
(
(
1H, ddd, J = 12.5, 3, 1.8 Hz, 2-Heq), 1.73 (3H, s, 5-Me), 2.17
1H, dd, J = 18, 9 Hz, 4-Hax), 2.29 (3H, s, 9-Me), 2.51 (1H, dd,
3
-Hydroxy-ꢀ-ionone (1) and its MPA esters
R)-3-Hydroxy-ꢀ-ionone. The (S)-MPA ester of (R)-3-hydroxy-
(
J = 18, 6 Hz, 4-Heq), 3.46 (3H, s, O-Me), 4.91 (1H, s, H-NMA),
5.10 (1H, m, 3-H), 6.08 (1H, d, J = 16 Hz, 8-H), 7.15 (1H, d,
J = 16 Hz, 7-H), 7.50–7.95 (7H, 2-naphthyl). ESI-MS: 407.4
(M ϩ H).
β-ionone (1) (3 mg, 8.4 µmol), was dissolved in 3 ml of meth-
anol. Excess K CO was added and the mixture was stirred at
room temperature for 30 min. The reaction was quenched
by addition of 1 ml of water and the product extracted with
diethyl ether and dried with Na CO . Yield 90%. [α] = Ϫ73.5
2
3
(R)-2-NMA ester of (S)-3-hydroxy-ꢀ-ionone (6b). Separated
from a reaction of racemic 3-hydroxy-β-ionone with (R)-2-
NMA, using a chiral column (Chiracel OD), 1:99 propan-2-ol–
2
3
D
(
c = 0.129, 1,4-dioxane).
1
(
S)-3-Hydroxy-ꢀ-ionone. The (S)-MPA ester of (S)-3-
hexane, λ = 280 nm. H-NMR (CDCl , 500 MHz) δ 1.09 (3H,
3
H
hydroxy-β-ionone (1) (2.5 mg, 7.0 µmol) was dissolved in 3 ml
s, 1-Me), 1.13 (3H, s, 1-Me), 1.62 (1H, t, J = 12 Hz, 2-Hax), 1.64
(3H, s, 5-Me), 1.81 (1H, ddd, J = 12, 3, 2 Hz, 2-Heq), 1.97 (1H,
dd, J = 18, 9 Hz, 4-Hax), 2.27 (3H, s, 9-Me), 2.35 (1H, dd,
of methanol. Excess K CO was added and the mixture was
2
3
stirred at room temperature for 30 min. The reaction was
2
412 J. Chem. Soc., Perkin Trans. 1, 2000, 2409–2414

View Article Online
J = 18, 6 Hz, 4-Heq), 3.46 (3H, s, O-Me), 4.91 (1H, s, H-NMA),
.11 (1H, m, 3-H), 6.05 (1H, d, J = 16 Hz, 8-H), 7.14 (1H, d,
J = 16 Hz, 7-H), 7.50–7.92 (7H, 2-naphthyl).
5.09 (1H, m, 3-H), 6.28 (1H), 6.38 (2H), 6.46 (1H), 6.65 (5H),
7.14 (1H), 7.33–7.45 (5H, phenyl).
5
1
(
S)-MPA ester of (all-E)-(3R)-triophaxanthin (10b). H-NMR
(
R)-9-ATMA ester of (R)-3-hydroxy-ꢀ-ionone (7a). Separated
(CDCl , 500 MHz) δ 1.19 (3H, s, 1-Me), 1.21 (3H, s, 1-Me),
3
H
from a reaction of racemic 3-hydroxy-β-ionone with (R)-9-
1.59 (1H, t, J = 12 Hz, 2-Hax), 1.84 (3H, s, 5-Me), 1.85 (1H, m,
2-Heq), 1.94 (3H, s), 1.98 (3H, s), 1.99 (1H, m, 4-Hax), 2.00
(6H, s), 2.36 (1H, dd, J = 17, 5 Hz, 4-Heq), 3.42 (3H, s, O-Me),
4.74 (1H, s, H-MPA), 5.09 (1H, m, 3-H), 6.28 (1H), 6.38 (2H),
6.46 (1H), 6.65 (5H), 7.34–7.45 (5H, phenyl).
ATMA, using a chiral column (Chiracel OD), 1:99 propan-2-
1
ol–hexane, λ = 280 nm. H-NMR (CDCl , 500 MHz) δ 0.74
3
H
(
3H, s, 1-Me), 0.98 (3H, s, 1-Me), 1.18 (1H, dd, J = 12, 11 Hz,
2-Hax), 1.41 (1H, ddd, J = 12, 3, 1.7 Hz, 2-Heq), 1.68 (3H, s,
5-Me), 2.09 (1H, dd, J = 18, 9 Hz, 4-Hax), 2.25 (3H, s, 9-Me),
2.47 (1H, dd, J = 17, 5 Hz, 4-Heq), 3.46 (3H, s, O-Me),
5.10 (1H, m, 3-H), 5.99 (1H, d, J = 16 Hz, 8-H), 6.26 (1H, s,
1
(
R)-MPA ester of (9Z)-(3S)-apohopkinsiaxanthin (11a). H-
NMR (CDCl , 500 MHz) δH 1.32 (3H, s, 1-Me), 1.37 (3H, s,
3
1
2
-Me), 1.93 (1H, m, 2-Hax), 1.95 (6H, s), 1.98 (1H, m, 2-Heq),
.00 (3H, s), 2.04 (3H, s), 2.05 (3H, s, 5-Me), 3.54 (3H, s,
H-ATMA), 7.07 (1H, d, J = 16 Hz, 7-H), 7.42–8.62 (9H,
9
-anthryl).
O-Me), 4.93 (1H, s, H-MPA), 5.57 (1H, dd, J = 14, 6 Hz, 3-H),
.32 (1H), 6.45 (4H), 6.66 (4H), 6.79 (1H), 7.12 (1H), 7.33–7.56
5H, phenyl). ESI-MS: 607.6 (M ϩ H).
6
(
(
R)-9-ATMA ester of (S)-3-hydroxy-ꢀ-ionone (7b). Separated
from a reaction of racemic 3-hydroxy-β-ionone with (R)-9-
ATMA, using a chiral column (Chiracel OD), 1:99 propan-2-
1
1
(S)-MPA ester of (9Z)-(3S)-apohopkinsiaxanthin (11b). H-
NMR (CDCl , 500 MHz) δH 1.32 (3H, s, 1-Me), 1.37 (3H, s,
1
5
ol–hexane, λ = 280 nm. H-NMR (CDCl , 500 MHz) δ 0.81
3
H
3
(
(
3H, s, 1-Me), 1.03 (3H, s, 1-Me), 1.39 (3H, s, 5-Me), 1.45
1H, dd, J = 12, 11 Hz, 2-Hax), 1.60 (1H, dd, J = 18, 8 Hz,
-Me), 1.95 (6H, s), 2.00 (3H, s), 2.04 (3H, s), 2.06 (3H, s,
-Me), 2.07 (1H, m, 2-Hax), 2.13 (1H, m, 2-Heq), 3.55 (3H,
s, O-Me), 4.91 (1H, s, H-MPA), 5.56 (1H, dd, J = 14, 6 Hz, 3-H),
.32 (1H), 6.45 (4H), 6.66 (4H), 6.79 (1H), 7.12 (1H), 7.33–7.56
5H, phenyl).
4
-Hax), 1.70 (1H, ddd, J = 12, 3, 1.7 Hz, 2-Heq), 2.10 (1H, dd,
J = 18, 5 Hz, 4-Heq), 2.24 (3H, s, 9-Me), 3.44 (3H, s, O-Me),
.09 (1H, m, 3-H), 5.83 (1H, d, J = 16 Hz, 8-H), 6.26 (1H, s,
H-ATMA), 6.98 (1H, d, J = 16 Hz, 7-H), 7.42–8.55 (9H,
-anthryl).
6
(
5
9
1
(
R)-MPA ester of (all-E)-(3S,5R,6R)-paracentrone (12a). H-
NMR (CDCl , 500 MHz) δ 1.02 (3H, s, 1-Me), 1.34 (3H, s,
-Me), 1.34 (1H, m, 2-Hax), 1.34 (3H, s, 5-Me), 1.56 (1H,
-Hax), 1.79 (3H, s), 1.87 (1H, m, 2-Heq), 1.93 (3H, s), 1.98
1
3
H
(
R)-MPA ester of (3R,3RЈ)-zeaxanthin (8a). H-NMR
1
4
(
(
CDCl , 500 MHz) δH 1.01 (6H, s, 1-Me), 1.05 (6H, s, 1-Me),
3
1.51 (2H, t, J = 12 Hz, 2-Hax), 1.65 (2H, m, 2-Heq), 1.70 (6H, s,
5-Me), 1.95 (6H, s), 1.97 (6H, s), 2.14 (2H, dd, J = 17, 8 Hz,
4-Hax), 2.45 (2H, dd, J = 17, 5 Hz, 4-Heq), 3.43 (6H, s, O-Me),
4.75 (2H, s, H-MPA), 5.11 (2H, m, 3-H), 6.08 (4H), 6.15 (2H),
6.25 (2H), 6.35 (2H), 6.63 (4H), 7.33–7.47 (10H, phenyl). ESI-
3H, s), 1.99 (3H, s), 2.27 (1H, 4-Heq), 2.36 (3H, s), 3.42 (3H, s,
O-Me), 4.74 (1H, s, H-MPA), 5.44 (1H, m, 3-H), 6.05 (1H), 6.11
1H), 6.28 (1H), 6.35 (1H), 6.39 (1H), 6.67 (5H), 7.15 (1H),
.34–7.46 (5H, phenyl).
(
7
MS: 865.4 (M ϩ H).
1
(
S)-MPA ester of (all-E)-(3S,5R,6R)-paracentrone (12b). H-
NMR (CDCl , 500 MHz) δH 1.06 (3H, s, 1-Me), 1.30 (3H, s,
1
3
(
S)-MPA ester of (3R,3RЈ)-zeaxanthin (8b). H-NMR
5-Me), 1.36 (3H, s, 1-Me), 1.44 (1H, 2-Hax), 1.47 (1H, 4-Hax),
1.79 (3H, s), 1.93 (3H, s), 1.98 (3H, s), 1.99 (3H, s), 2.01 (1H,
2-Heq), 2.14 (1H, m, 4-Heq), 2.36 (3H, s), 3.42 (3H, s, O-Me),
4.73 (1H, s, H-MPA), 5.44 (1H, m, 3-H), 6.05 (1H), 6.11 (1H),
6.28 (1H), 6.35 (1H), 6.339 (1H), 6.67 (5H), 7.15 (1H), 7.34–
7.45 (5H, phenyl).
(
1
2
CDCl , 500 MHz) δH 1.06 (6H, s, 1-Me), 1.09 (6H, s, 1-Me),
.61 (2H, t, J = 12 Hz, 2-Hax), 1.65 (6H, s, 5-Me), 1.79 (2H, m,
-Heq), 1.95 (6H, s), 1.96 (2H, 4-Hax), 1.97 (6H, s), 2.31 (2H,
dd, J = 17, 5 Hz, 4-Heq), 3.43 (6H, s, O-Me), 4.75 (2H, s, H-
3
MPA), 5.12 (2H, m, 3-H), 6.08 (4H), 6.15 (2H), 6.25 (2H), 6.35
(
2H), 6.63 (4H), 7.33–7.46 (10H, phenyl).
1
(
R)-MPA ester of (3R,3ЈR)-astaxanthin (13a). H-NMR
(
R)-MPA ester of (all-E)-(3R)-3-hydroxy-7,8-didehydro-8Ј-
(
(
1
5
(
C D , 500 MHz) δ 0.78 (6H, s, 1-Me), 0.82 (6H, s, 1-Me), 1.67
2H, m, 2-Heq), 1.74 (6H, s), 1.85 (6H, s), 1.90 (2H, m, 2-Hax),
.99 (6H, s, 5-Me), 3.57 (6H, s, O-Me), 5.05 (2H, s, H-MPA),
.68 (2H, dd, J = 14, 5 Hz, 3-H), 5.95 (2H), 6.26 (4H), 6.36
6 6 H
1
apo-ꢀ-caroten-8Ј-al (9a). H-NMR (CDCl , 500 MHz) δ 1.13
3H, s, 1-Me), 1.13 (3H, s, 1-Me), 1.49 (1H, t, J = 12 Hz,
-Hax), 1.71 (1H, m, 2-Heq), 1.90 (6H, s, 5-Me), 1.98 (3H, s),
.01 (6H, s), 2.16 (1H, dd, J = 18, 9 Hz, 4-Hax), 2.50 (1H, dd,
J = 17, 5 Hz, 4-Heq), 3.42 (3H, s, O-Me), 4.74 (1H, s, H-MPA),
.08 (1H, m, 3-H), 6.28 (1H), 6.36 (1H), 6.45 (1H), 6.56 (1H),
.71 (4H), 6.94 (1H), 7.33–7.45 (5H, phenyl), 9.46 (1H).
3
H
(
2
2
2H), 6.51 (2H), 6.68 (4H), 7.15–7.73 (10H, phenyl).
5
6
1
(
S)-MPA ester of (3R,3ЈR)-astaxanthin (13b). H-NMR
(
(
1
C D , 500 MHz) δ 0.76 (6H, s, 1-Me), 0.80 (6H, s, 1-Me), 1.59
6
6
H
2H, dd, J = 12, 5 Hz, 2-Hax), 1.74 (6H, s), 1.83 (2H, m, 2-Heq),
.85 (6H, s), 2.02 (6H, s, 5-Me), 3.51 (6H, s, O-Me), 5.05 (2H,
s, H-MPA), 5.71 (2H, dd, J = 14, 4 Hz, 3-H), 5.95 (2H), 6.26
4H), 6.36 (2H), 6.51 (2H), 6.68 (4H), 7.15–7.73 (10H, phenyl).
(
S)-MPA ester of (all-E)-(3R)-3-hydroxy-7,8-didehydro-8Ј-
1
apo-ꢀ-caroten-8Ј-al (9b). H-NMR (CDCl , 500 MHz) δ 1.16
3H, s, 1-Me), 1.19 (3H, s, 1-Me), 1.59 (1H, t, J = 12 Hz,
-Hax), 1.86 (1H, m, 2-Heq), 1.84 (3H, s, 5-Me), 1.90 (3H, s),
.98 (3H, s), 1.99 (1H, dd, J = 18, 9 Hz, 4-Hax), 2.01 (6H, s),
.36 (1H, dd, J = 17, 5 Hz, 4-Heq), 3.42 (3H, s, O-Me), 4.74
1H, s, H-MPA), 5.10 (1H, m, 3-H), 6.28 (1H), 6.36 (1H), 6.45
1H), 6.56 (1H), 6.71 (4H), 6.94 (1H), 7.34–7.45 (5H, phenyl),
.46 (1H). ESI-MS: 579.7 (M ϩ H).
3
H
(
(
2
1
2
(
(
ESI-MS: 893.3 (M ϩ H).
(R)-MPA ester of (3R,3ЈR)-(all-E)-7,8,7Ј,8Ј-tetrahydrozea-
1
xanthin-8,8Ј-dione (14a). H-NMR (CDCl , 500 MHz) δ 0.87
(3H, s, 1-Me), 0.91 (6H, s, 1-Me), 1.39 (6H, s, 5-Me), 1.64 (2H,
t, J = 12 Hz, 2-Hax), 1.73 (2H, m, 2-Heq), 1.77 (6H, s), 1.98
3
H
9
(
6H, s), 2.20 (2H, dd, J = 17, 10 Hz, 4-Hax), 2.33 (2H, dd,
1
(
R)-MPA ester of (all-E)-(3R)-triophaxanthin (10a). H-NMR
J = 17, 6 Hz, 4-Heq), 3.25 (6H, s, O-Me), 3.31 (4H), 4.75 (2H, s,
H-MPA), 5.32 (2H, m, 3-H), 6.35 (2H), 6.55 (4H), 6.69 (2H),
7.15–7.64 (10H, phenyl).
(
1
2
CDCl , 500 MHz) δH 1.13 (3H, s, 1-Me), 1.13 (3H, s, 1-Me),
.49 (1H, t, J = 12 Hz, 2-Hax), 1.71 (1H, ddd, J = 12, 3.5, 2 Hz,
-Heq), 1.90 (3H, s, 5-Me), 1.94 (3H, s), 1.98 (3H, s), 2.00 (6H,
3
s), 2.16 (1H, dd, J = 18, 9 Hz, 4-Hax), 2.36 (3H, s), 2.50 (1H, dd,
J = 17, 5 Hz, 4-Heq), 3.42 (3H, s, O-Me), 4.74 (1H, s, H-MPA),
(S)-MPA ester of (3R,3ЈR)-(all-E)-7,8,7Ј,8Ј-tetrahydrozea-
1
xanthin-8,8Ј-dione (14b). H-NMR (CDCl , 500 MHz) δ 0.94
3
H
J. Chem. Soc., Perkin Trans. 1, 2000, 2409–2414
2413

View Article Online
(
6H, s, 1-Me), 0.95 (6H, s, 1-Me), 1.33 (6H, s, 5-Me), 1.72 (2H,
t, J = 12 Hz, 2-Hax), 1.77 (6H, s), 1.83 (2H, m, 2-Heq), 1.98
6H, s), 2.09 (2H, dd, J = 16, 9 Hz, 4-Hax), 2.23 (2H, dd, J = 17,
Hz, 4-Heq), 3.21 (6H, s, O-Me), 4.75 (2H, s, H-MPA), 5.30
2H, m, 3-H), 6.35 (2H), 6.55 (4H), 6.69 (2H), 7.15–7.64 (10H,
phenyl). ESI-MS: 896.2 (M ϩ H).
Roche, Basel for samples of 8, 13 and 15, Dr G. Saucy, Roche,
Nutley, for sample 14 and Professor T. Kusumi, Tokushima
University, Tokushima for kindly providing us with 9-anthryl-
methoxyacetic acid (7).
(
5
(
References
1
F. Mo, X-Ray Crystallographic Studies, ed. G. Britton, S. Liaaen-
Jensen and H. Pfander, Birkhäuser, Basel, 1995.
S. Liaaen-Jensen, Fortschr. Chem. Org. Naturst., 1980, 39, 123.
R. Buchecker and K. Noack, “Circular dichroism”, in Carotenoids.
Vol. 1B. Spectroscopy, ed. G. Britton, S. Liaaen-Jensen and
H. Pfander, Birkhäuser, Basel, 1995.
D. Kull and H. Pfander, Carotenoids. Vol. 1A. Isolation and
Analysis, ed. G. Britton, S. Liaaen-Jensen and H. Pfander,
Birkhäuser, Basel, 1995.
(
R)-MPA ester of (3S,3ЈS)-(all-E)-7,8,7Ј,8Ј-tetradehydroasta-
1
xanthin (15a). H-NMR (C D , 500 MHz) δ 0.92 (6H, s,
-Me), 1.00 (6H, s, 1-Me), 1.62 (2H, dd, J = 13, 5 Hz, 2-Heq),
.75 (6H, s), 1.76 (2H, t, J = 13 Hz, 2-Hax), 1.81 (6H, s), 2.16
6H, s, 5-Me), 3.48 (6H, s, O-Me), 5.00 (2H, s, H-MPA), 5.71
2H, dd, J = 14, 5 Hz, 3-H), 6.27 (2H), 6.34 (2H), 6.37 (2H),
6
6
H
2
3
1
1
(
(
6
4
5
.62 (2H), 6.69 (2H), 7.15–7.70 (10H, phenyl). ESI-MS: 888.4
M ϩ H).
(
O. Straub, Key to Carotenoids, ed. H. Pfander, Birkhäuser, Basel,
1
987.
(
S)-MPA ester of (3S,3ЈS)-(all-E)-7,8,7Ј,8Ј-tetradehydroasta-
6 N. Harada and K. Nakanishi, Circular Dichroic Spectroscopy:
Exciton Coupling in Organic Stereochemistry, University Science
Books, Mill Valley, CA, 1983.
1
xanthin (15b). H-NMR (C D , 500 MHz) δ 0.94 (6H, s,
6
6
H
1
(
5
-Me), 1.03 (6H, s, 1-Me), 1.69 (2H, dd, J = 13 Hz, 2-Heq), 1.75
6H, s), 1.84 (2H, t, J = 13 Hz, 2-Hax), 1.81 (6H, s), 2.14 (6H, s,
-Me), 3.53 (6H, s, O-Me), 5.01 (2H, s, H-MPA), 5.62 (2H, dd,
J = 14, 5 Hz, 3-H), 6.27 (2H), 6.34 (2H), 6.37 (2H), 6.62 (2H),
7
K. Nakanishi and N. Berova, “The Exciton Chirality Method”, in
Circular Dichroism—Principles and Applications, ed. K. Nakanaishi,
N. Berova and R. W. Woody, VCH Publishers, Inc., New York, 1994.
8 R. Person, K. Monde, H.-U. Humpf and N. Berova, Chirality, 1995,
7, 128.
6
.69 (2H), 7.15–7.77 (10H, phenyl).
9
K. Nakanishi, in One-Dimensional and Two-Dimensional NMR
Spectra by Modern Pulse Techniques, Mill Balley, CA, 1990.
1
(
R)-MPA ester of lutein (16a). H-NMR (CDCl , 500 MHz)
3
10 S. Hertzberg and L. Liaaen-Jensen, unpublished results.
11 J. A. Dale and H. S. Mosher, J. Am. Chem. Soc., 1973, 95, 512.
12 B. M. Trost, J. L. Belletire, S. Godleski, P. G. McDougal, J. M.
Balkovec, J. J. Baldwin, M. E. Christy, G. S. Ponticello, S. L. Varga
and J. P. Springer, J. Org. Chem., 1986, 51, 2370.
δH 0.85 (3H, s, 1Ј-Me), 0.94 (3H, s, 1Ј-Me), 1.01 (3H, s, 1-Me),
.05 (3H, s, 1-Me), 1.45 (1H, dd, J = 14, 5 Hz, 2Ј-Heq), 1.51
1H, t, J = 12 Hz, 2-Heq), 1.58 (3H, s, 5Ј-Me), 1.65 (1H, m,
-Hax), 1.70 (3H, s, 5-Me), 1.84 (1H, dd, J = 14, 6 Hz, 2Ј-Hax),
.89 (3H, s), 1.96 (12H), 2.14 (1H, dd, J = 18, 9 Hz, 4-Hax), 2.35
1
(
2
1
(
3
1
3 I. Ohtani, T. Kusumi, Y. Kashman and H. Kakisawa, J. Am. Chem.
Soc., 1991, 113, 4092.
4 G. Uray, in Methods of Organic Synthesis (Houben-Weyl), vol.
E21a, Georg Thieme Verlag, Stuttgart, 1995, pp. 253–292.
1H, d, J = 10 Hz, 6Ј-H), 2.45 (1H, dd, J = 17, 5 Hz, 4-Heq),
.43 (6H, s, O-Me and OЈ-Me), 4.75 (1H, s, H-MPA), 4.77 (1H,
1
s, HЈ-MPA), 5.11 (1H, m, 3-H), 5.34 (1H, m, 4Ј-H), 5.35 (1H,
m, 3Ј-H), 6.09 (6H), 6.25 (1H), 6.35 (2H), 6.62 (4H), 7.33–7.46
15 S. K. Latypov, J. M. Seco, E. Quinoa and R. Riguera, J. Org. Chem.,
995, 60, 504.
1
1
6 H. Ok, C. Caldwell, D. R. Schroeder, A. K. Singh and
K. Nakanishi, Tetrahedron Lett., 1988, 29, 2275.
7 T. Kusumi, H. Takahashi, P. Xu, T. Fukushima, Y. Asakawa,
T. Hashimoto, Y. Kan and Y. Inouye, Tetrahedron Lett., 1994, 35,
4397.
(
10H, phenyl). ESI-MS: 863.7 (M ϩ H).
1
1
(
S)-MPA ester of lutein (16b). H-NMR (CDCl , 500 MHz)
3
δH 0.77 (3H, s, 1Ј-Me), 0.77 (3H, s, 1Ј-Me), 1.06 (3H, s, 1-Me),
.09 (3H, s, 1-Me), 1.23 (1H, dd, J = 14, 5 Hz, 2Ј-Heq), 1.61
1
1
2
8 A. Rüttimann and H. Mayer, Helv. Chim. Acta, 1989, 63, 1456.
9 J. A. Haugan, J. Chem. Soc., Perkin Trans. 1, 1997, 18, 2731.
0 J. A. Haugan and S. Liaaen-Jensen, Acta Chem. Scand., 1994, 48,
1
(
1H, t, J = 12 Hz, 2-Heq), 1.65 (6H, s, 5Ј-Me and 5-Me), 1.71
(
1H, dd, J = 14, 6 Hz, 2Ј-Hax), 1.79 (1H, m, 2-Hax), 1.89 (3H,
8
99.
s), 1.96 (12H), 1.96 (1H, m, 4-Hax), 2.31 (1H, dd, J = 17, 5 Hz,
-Heq), 2.32 (1H, d, J = 10 Hz, 6Ј-H), 3.42 and 3.43 (6H, s,
2
1 J. A. Haugan, Acta Chem. Scand., 1997, 51, 1096.
4
22 J. H. Haugan, E. Lobkovsky and S. Liaaen-Jensen, Acta Chem.
Scand., 1997, 51, 1201.
O-Me and OЈ-Me), 4.73 (1H, s, H-MPA), 4.75 (1H, s, HЈ-
MPA), 5.12 (1H, m, 3-H), 5.38 (1H, m, 3Ј-H), 5.48 (1H, m,
2
3 K. Bernhard, F. Kienzle, H. Mayer and R. K. Müller, Helv. Chim.
Acta, 1980, 63, 1473.
4
Ј-H), 6.09 (6H), 6.25 (1H), 6.35 (2H), 6.62 (4H), 7.33–7.46
2
2
4 F. Kienzle and H. Mayer, Helv. Chim. Acta, 1978, 61, 2609.
5 G. Saucy, unpublished results (formerly at Hoffman-La Roche,
Nutley, NJ, USA).
(
10H, phenyl). ESI-MS: 863.8 (M ϩ H).
2
6 H. Mayer and A. Rüttimann, Helv. Chim. Acta, 1980, 63, 1451.
Acknowledgements
We gratefully acknowledge support by the National Institute of
Health (GM 34509 to KN and NB) and the Jacob and Alice
Wallenbergs Foundation (to TA). We also thank Dr H. Mayer,
27 K. S. Latypov, J. M. Seco, E. Quinoa and R. Riguera, J. Am. Chem.
Soc., 1998, 120, 877.
2
2
8 R. Buchecker, P. Hamm and C. H. Eugster, Chimia, 1971, 25, 192.
9 A. G. Andrews, G. Borch and S. Liaaen-Jensen, Acta Chem. Scand.,
Ser. B, 1974, 28, 139.
2
414
J. Chem. Soc., Perkin Trans. 1, 2000, 2409–2414