Dicarotenoid esters of bivalent acids

Article abstract of DOI:10.1016/j.tetlet.2008.03.137

For the evaluation of the synthesis of dendritic esters from carotenoids the C₂₀ apocarotenoid retinol was chosen for model studies, being a commercially available hydroxy carotenoid. Dimers were synthesized from retinol with dicarboxyl cores and from retinol succinate with other hydroxy carotenoids.

Full text of DOI:10.1016/j.tetlet.2008.03.137

Available online at www.sciencedirect.com
Tetrahedron Letters 49 (2008) 3524–3526
Dicarotenoid esters of bivalent acids
*
´
´
´
´
´
Magdolna Hada, Veronika Nagy, Aniko Takatsy, Jozsef Deli, Attila Agocs
Department of Biochemistry and Medical Chemistry, University of Pe´cs, Medical School, Szigeti u´t 12, H-7624 Pe´cs, Hungary
Received 11 February 2008; revised 6 March 2008; accepted 27 March 2008
Available online 29 March 2008
Abstract
For the evaluation of the synthesis of dendritic esters from carotenoids the C20 apocarotenoid retinol was chosen for model studies,
being a commercially available hydroxy carotenoid. Dimers were synthesized from retinol with dicarboxyl cores and from retinol
succinate with other hydroxy carotenoids.
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords: Carotenoid; Retinol; Diester; Dimer
Carotenoids, being naturally occurring antioxidants and
having various biological effects including anti-cancer and
cardioprotective, have been the focus of food biochemists
recently.1 Dimers and dendrimers of these compounds have
not been synthesized until now, although they can have
interesting properties especially when combined with the
advantages of carotenoids. These compounds may have
enchanced antioxidant activity and could be used for
aggregation studies.2 Carotenoid diesters are not unknown
in the literature, and there are examples including enzy-
matic reactions with carotenoid acids3 and reactions
between carotenoid acids and carotenols.4 However, the
chemistry described there cannot be directly applied for
the synthesis of the esters described in this Letter.
sources or purchased except for 40-hydroxy-echinenone,
which was produced by the partial reduction of canthaxan-
thin with sodium borohydride.
Our aim was to make dicarotenoid diesters of bivalent
acids in which secondary interactions are possible between
the two polyene chains. To achieve this, retinol and various
other carotenoids were esterified with several cyclic acid
anhydrides. The results of these experiments are summa-
rized in Table 1. These anhydrides were chosen after sev-
eral unsuccessful trials with the corresponding acyl
dichlorides. In general, the reactions of carotenoids with
acyl chlorides in pyridine or in dichloromethane/pyridine
gave several by-products, whereas anhydrides gave cleaner
reactions.5,6
In our present Letter we describe the synthesis of some
retinol dimers and their precursors which can serve as a
starting point for further studies with other hydroxy
carotenoids and later form a basis for the synthesis of
carotenoid dendrimers.
Retinol was chosen as a model compound for other
carotenoids being not too expensive and easily accessible
(Fig. 1). Other carotenoids were isolated from natural
All of the hydroxyl groups of the carotenoids were ester-
ified with succinic anhydride forming the corresponding
mono- and disuccinates bearing free carboxyl groups.
Unfortunately, maleic anhydride appeared to be too reac-
tive, whereas phthalic anhydride did not give even the cor-
responding monoesters.
The retinol succinate monoester formed (1) can be cou-
pled with other carotenoid alcohols via the Steglich-
method using dicyclohexyl carbodiimide (DCC) and
4-(dimethylamino)pyridine (DMAP) in dichloromethane
(Fig. 2).7 The reaction gave homo- and heterodimers (2a–
c) in good yields, and the conversion of the alcohols was
about 70–80%. The dimers could be crystallized from
*
Corresponding author. Tel.: +36 72 536001x1864; fax: +36 72 536225.
´
E-mail address: attila.agocs@aok.pte.hu (A. Agocs).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.03.137

M. Ha´da et al. / Tetrahedron Letters 49 (2008) 3524–3526
3525
CH2OH
Retinol
15
10'
9
Q
R
9'
10
15'
O
OH
OH
HO
e
a
b
d
c
Fig. 1. Structures of the carotenoid starting materials: R = a, Q = b: b-cryptoxanthin; R = b, Q = d: lutein; R = Q = e: isozeaxanthin; R = c, Q = e: 40-
hydroxy-echinenone.
Table 1
Esterification of mono- and dihydroxy carotenoids
CH2Cl2 , r.t.
Carotenoid mono- or diester
Carotenoid-OH + 2.5 eq. anhydride
3 eq. DMAP
Retinol
Lutein
Isozeaxanthin
b-Cryptoxanthin
4-Hydroxy-echinenone
Maleic anhydride
Succinic anhydride
Phthalic anhydride
Decomp. even at ꢀ70 °C
85%
No reaction
—
74%
No reaction
Decomp. even at ꢀ70 °C
Unstable product
No reaction
Decomp. even at ꢀ70 °C
—
65%
—
81%
—
O
3 eq. DCC /
3 eq. DMAP
CH2Cl2
O
COOH
O
Ret
RO
O
Carotenoid 0.7 equiv.
1
O
2a R= Retinol, 60%
2b R= β-Cryptoxanthin, 53%
COOH
COOH
2c R= 4'-OH-echinenone, 50%
2 equiv.. Ret-OH
2a
DCC / DMAP
25%
O
O
β-Cryptoxanthin
3 eq. DCC /
3 eq. DMAP
CH2Cl2
succinate
O
+
O
β-Cryptoxanthin
42%
3
Fig. 2. Synthesis of carotenoid dimers.
methanol in reasonable yields. After the successful trials
with retinol, a homodimer of b-cryptoxanthin (3) was also
synthesized from its succinate. The yield in this case was
modest, and the conversion was only 70–80% even if 30%
excess succinate was used. The structures of dimers were
proven on the basis of NMR, MALDI-TOF, and UV data.
The purity of the compounds (>90%) was determined by
HPLC.8
This method was also tested for the direct synthesis of a
retinol homodimer (2a, Fig. 2). In this case the diacid

3526
M. Ha´da et al. / Tetrahedron Letters 49 (2008) 3524–3526
(phthalic, maleic, or succinic) was esterified with an excess
of retinol in the presence of DCC and DMAP. Maleic and
phthalic acids proved to be non-reactive again presumably
for steric reasons which suggests that diacids with a larger
distance between the two carboxylic acids can serve as
appropriate starting materials.
References and notes
1. Krinsky, N. I.; Johnson, E. J. Mol. Asp. Med. 2005, 26, 459–516.
´
2. Simonyi, M.; Bikadi, Z. S.; Zsila, F.; Deli, J. Chirality 2003, 15, 680–
698.
3. Vassilia Partali, V.; Kvittingen, L.; Sliwka, H. R.; Anthonsen, T.
Angew. Chem., Int. Ed. 1996, 35, 329–330.
Our results confirm that even a simple reaction such as
4. Levy, L. W.; Binnington, R. H.; Tabatznik, A. S., Patent WO/2002/
068385, 2002.
5. Frey, D. A.; Kataisto, E. W.; Ekmanis, J. L.; O’Malley, S.; Lockwood,
S. F. Org. Process Res. Dev. 2004, 8, 796–801.
6. Nadolski, G.; Cardonuel, A. J.; Zweier, J. L.; Lockwood, S. F. Bioorg.
Med. Chem. Lett. 2006, 16, 775–781.
esterification can have
a different outcome among
carotenoids, and carotenoids with strong structural resem-
blance can react rather differently under the same
conditions.
The new compounds are to be tested for their in vitro
and in vivo antioxidant activity.
7. Neises, B.; Steglich, W. Angew. Chem. 1978, 90, 556–557.
8. Spectroscopic data for dimer 2b: UV (hexane): 327 (421), 448 (475); MS
(MALDI-TOF) m/z = 920.6 (calcd for C64H88O4 920.67). 1H NMR
(400 MHz) d (ppm) = 0.85–2.1 (m, 62H), 2.63 (m, 4H, CH2s succ.),
4.77 (d, 2H, H-16 Ret., J16,17 = 7.2 Hz), 5.08 (m, 1H, H-30 b-Crypt.),
5.60 (m, 1H, H-17 Ret.), 6.1–6.7 (m, 18H, olefinic protons). 13C NMR
(125 MHz) d (ppm) = 12.8, 19.2, 21.4, 21.7, 25.6, 28.4, 28.9, 29.2, 29.5,
30.0, 30.3, 33.1, 33.7, 34.3, 36.7, 38.3, 39.6, 43.9, 61.5, 68.8, 124.3,
125.8, 127.0, 129.3, 129.6–130.2 (m), 130.7, 131.4, 132.3, 133.1, 135.7,
136.6, 137.5–138 (m), 138.6, 139.2, 171.9, 172.2. Anal. Calcd for
Acknowledgments
´
We thank Mrs. Andrea Bognar, Mrs. Judit Kriszt, and
Mr. Norbert Go¨tz for their assistance. This study was sup-
ported by OTKA K 60121 (Hungarian National Research
Foundation). We also thank Carotenature for providing us
with a sample of isozeaxanthin.
C
64H88O4: C, 83.43; H, 9.63; O, 6.95%. Found: C, 83.48; H, 9.60; O,
6.94%.