Synthesis of C15,C14′-ring locked all-trans-β-carotene

Article abstract of DOI:10.1016/S0040-4039(02)00663-9

Synthesis of C15,14′-ring locked all-trans-β-carotene is described. The symmetric nature of the β-carotene analog allowed for rapid construction of the carbon frame through bis-olefination of the central dialdehyde containing the ring functionality with a C15 ylide bearing the ionone moiety.

Full text of DOI:10.1016/S0040-4039(02)00663-9

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 3867–3869
Synthesis of C15,C14%-ring locked all-trans-b-carotene
Pulgam Veera Reddy and Babak Borhan*
Department of Chemistry, Michigan State University, East Lansing, MI 48824, USA
Received 30 March 2002; accepted 4 April 2002
Abstract—Synthesis of C15,14%-ring locked all-trans-b-carotene is described. The symmetric nature of the b-carotene analog
allowed for rapid construction of the carbon frame through bis-olefination of the central dialdehyde containing the ring
functionality with a C15 ylide bearing the ionone moiety. © 2002 Elsevier Science Ltd. All rights reserved.
Vitamin A and its derivatives (retinoids) are not only
essential components in vision,1,2 but also they exert
multiple effects on cell differentiation and development
with important clinical implications.3–6 b-Carotene-
15,15%-dioxygenase (BCDOX) is the enzyme responsible
for the oxidative cleavage of b-carotene to retinal in
animals (Scheme 1).7,8 Its function is of utmost impor-
tance since vertebrates cannot biosynthesize retinoids
and must rely on dietary intake or oxidation of b-
carotene to supplement their needs. Despite its pivotal
role in the production of retinoids, BCDOX has been a
largely forgotten enzyme. However, recent discovery of
genes responsible for production of BCDOX in various
species has re-ignited research in this area.9
enable us to determine the exact source of oxygen in the
products; i.e. do both oxygen atoms in a single cleavage
event come from the same molecular oxygen. The
cleavage of b-carotene yields two molecules of retinal,
and thus the loss in molecular connectivity does not
allow for analysis of products obtained from a single
oxidation event in isotopic labeling studies. Therefore,
we plan to use the b-carotene analog 1 for these studies.
Compound 1 contains a ring within the target olefin
and will preserve the molecular connectivity of the
original molecule upon oxidative cleavage. Herein, we
describe the convergent synthesis of the all-trans-
C15,C14%-ring locked b-carotene analog 1.
Our interest in BCDOX stems from its unique ability to
selectively oxidize the central double bond in b-carotene
either as a dioxygenase or a monooxygenase.9–11 In
particular, we are interested in labeling studies that will
The retrosynthetic approach depicted in Scheme 2
entails the bis-coupling the C15 ylide 2 with the C13
dialdehyde 3. The C13-dialdehyde segment is stereose-
lectively constructed by sequential Wittig and HWE
reactions starting from cyclohexenone-3-carboxalde-
hyde (5). The C15-ylide is obtained from b-ionone 4, as
described previously.12
Synthesis of 2-methyl-3-[3-(1-methyl-2-oxo-ethylidene)-
cyclohex-1-enyl]-propenal 3, depicted in Scheme 3,
began with securing cyclohexenone-3-carboxaldehyde 5,
which was prepared by previously reported proce-
Scheme 1.
dures.13 Olefination of
5
with diethyl(1-cyano-
ethyl)phosphonate afforded the dicyanide 6 in good
yield as a ꢀ2:1 mixture of EE/EZ isomers.14 The
simultaneous reduction of the two cyanide groups in 6
Keywords: b-carotene; b-carotene-15,15%-dioxygenase; Wittig; HWE
reactions.
* Corresponding author.
0040-4039/02/$ - see front matter © 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(02)00663-9

3868
P. V. Reddy, B. Borhan / Tetrahedron Letters 43 (2002) 3867–3869
nBuLi, KHMDS, and NaHMDS at −78°C or rt) did
not yield the desired diester 8. This is presumably due
to the reaction of the ylide with a less reactive conju-
gated carbonyl, and the formation of a tetrasubstituted
olefin, which is sterically a cumbersome and difficult
task for a typical HWE process.
Diethyl(1-cyanoethyl)phosphonate is a sterically less
demanding ylide (as compared to the ester analog of
the same) and could be better suited to deliver the
tetrasubstituted olefin. Gratifyingly, the HWE reaction
of the ketoester 7 with diethyl(1-cyanoethyl)phos-
phonate in the presence of NaH afforded the cyano-
ester 9 in excellent yield as a >3:1 mixture of EE/EZ
isomers.16 The geometry of the new olefinic bond was
confirmed by NOE experiments. In particular, com-
pound 9 showed a strong enhancement of the C3-
methyl group upon irradiation of the ring olefinic
proton for the EE isomer. The same enhancement was
not observed for the EZ isomer (Scheme 4).
Scheme 2.
The reduction of both the ester and cyano groups in 9
could be carried out in one step by using excess
DIBAL-H, however, the yields and recovery of the
product were not optimum. Therefore, we opted to
modify the procedure slightly by reducing both func-
tionalities sequentially without isolating the intermedi-
ate compound. The reaction was initiated by the
addition of DIBAL-H (2 equiv.) to the solution of 9 at
−78°C. A second dose of DIBAL-H (2 equiv.) was
added in 15 min, while the solution was kept at −78°C.
After 45 min, the reaction mixture was allowed to
warm up to 0°C. Wet silica gel was used to quench the
reaction by vigorously stirring the slurry (4 h), followed
by filtration of the silica. Subsequent treatment of the
filtrate with sodium-potassium tartrate afforded the
allylic alcohol 10 in 85% yield.17 Oxidation of 10 with
freshly prepared activated manganese dioxide furnished
the dialdehyde 3 in good yields.18
Scheme 3.
with DIBAL-H did not lead to a satisfactory yield of
the dialdehyde as the transformation led mostly to
decomposition of the starting material (various
conditions).
In order to overcome this problem, we turned our
attention to the synthesis of diester 8 (Scheme 4).
Treatment of cyclohexenone-3-carboxaldhyde 5 with
(carbethoxyethylidene)-triphenylphosporane provided
compound 7 in 93% yield (E/Z 99:1).15 However, the
single pot conversion of 5 to 8, utilizing excess of ylide
was not successful. For that matter, subjecting the
ketoester 7 to a second round of Wittig or HWE
olefination with ethyl 2-(diethylphosphono)propionate
or ethyl 2-(diisopropylphosphono)propionate with vari-
ous bases under different reaction conditions (NaH,
With 3 in hand the stage was set for the double
olefination with C15-triphenyl phosphonium bromide 2.
Treatment of 2 with sodium methoxide followed by
addition of dialdehyde 3 secured compound 1 in a
moderate yield (40% Scheme 4).19 Isomeric contamina-
tions were purified by HPLC. In conclusion, we have
synthesized an unnatural C15,14%-ring locked analog of
Scheme 4.

P. V. Reddy, B. Borhan / Tetrahedron Letters 43 (2002) 3867–3869
3869
b-carotene. Studies on enzymatic oxidation will be
reported in due course.
15. 2-Methyl-3-(3-oxo-cylcohex-1-enyl)-acrylic acid ethyl
1
ester (7): [E-isomer] H NMR (300 MHz, CDCl3): l 7.09
(1H, s), 6.00 (1H, s), 4.2 (2H, q, J=6.90 Hz), 2.43–2.38
(4H, m, 2×CH2), 2.05–2.01 (2H, m, CH2), 2 (3H, s, CH3),
1.28 (3H, t, J=6.90 Hz). 13C NMR (75 MHz, CDCl3): l
199.25, 167.57, 156.51, 137.10, 132.36, 129.52, 61.09,
37.14, 29.45, 22.57, 14.59, 14.04. HRMS: calcd for
C12H16O3 208.1099; found 208.1093. [Z-isomer] 1H NMR
(300 MHz, CDCl3): l 6.17 (1H, s), 5.83 (1H, s), 4.20 (2H,
q, J=6.90 Hz), 2.31–2.36 (4H, m, 2×CH2), 2.05–1.93
(2H, m, CH2), 1.99 (3H, s, CH3), 1.25 (3H, t, J=6.90
Hz).
Acknowledgements
Generous support was provided in part by Michigan
State University Startup funds to B.B. and a grant from
the Research Corporation (RI0410).
References
16. 3-[3-(Cyano-ethylidene)-cyclohex-1-enyl]-2-methyl-acrylic
acid ethyl ester (9) [EE-isomer]: 1H NMR (300 MHz,
CDCl3): l 7.13 (1H, s), 6.53 (1H, s), 4.20 (2H, q, J=7.20
Hz), 2.60 (2H, dt, J=1.50, 6.30 Hz, CH2), 2.43 (2H, t,
J=6.00 Hz, CH2), 2.05 (3H, s, CH3), 1.95 (3H, s, CH3),
1.78 (2H, quintet, J=6.30 Hz, CH2), 1.29 (3H, t, J=7.20
Hz). 13C NMR (75 MHz, CDCl3): l 168.50, 148.94,
144.05, 139.42, 129.62, 127.07, 120.27, 102.81, 60.98,
29.36 (2C), 22.26, 15.00, 14.64, 14.20. HRMS: calcd for
C15H19NO2 245.1416; found 245.1423. [EZ-isomer]: 1H
NMR (300 MHz, CDCl3): l 7.14 (1H, s), 6.81 (1H, s),
4.20 (2H, q, J=7.20 Hz), 2.39–2.35 (4H, m, 2×CH2), 2.07
(3H, s, CH3), 1.93 (3H, s, CH3), 1.77 (2H, quintet,
J=6.30 Hz, CH2), 1.29 (3H, t, J=7.20 Hz). 13C NMR
(75 MHz, CDCl3): l 168.53, 150.36, 142.87, 139.27,
130.45, 129.43, 119.58, 103.29, 60.96, 28.89, 25.82, 21.82,
16.02, 14.67, 14.24. HRMS: calcd for C15H19NO2
245.1416; found 245.1416.
17. 2-[3-(3-Hydroxy-2-methyl-propenyl)-cyclohex-2-enylidene]-
propionaldehyde (10): 1H NMR (300 MHz, CDCl3): l
10.20 (1H, s), 6.59 (1H, s), 6.10 (1H, s), 4.10 (2H, s), 2.88
(2H, dt, J=1.20, 6.30 Hz), 2.44 (2H, t, J=6.00 Hz), 1.92
(3H, s, CH3), 1.85–1.81 (2H, m), 1.84 (3H, s, CH3). 13C
NMR (75 MHz, CDCl3): l 191.10, 151.52, 147.39,
140.32, 128.94, 126.57, 126.34, 68.67, 30.33, 24.41, 22.75,
16.25, 9.91. HRMS: calcd for C13H18O2 206.1307; found
206.1308.
1. Rando, R. R. Angew. Chem., Int. Ed. Engl. 1990, 29,
461–480.
2. Rando, R. R. Chem. Biol. 1996, 3, 255–262.
3. Gudas, L. J. In The Retinoids: Biology, Chemistry, and
Medicine; Sporn, M. B.; Roberts, A. B.; Goodman, D.
S., Eds. Cellular biology and biochemistry of the
retinoids; Raven Press: New York, 1994; pp. 443–520.
4. Krinsky, N. I. Ann. NY Acad. Sci. 1998, 854, 443–447.
5. Merlet-Benichou, C.; Vilar, J.; Lelievre-Pegorier, M.;
Gilbert, T. Curr. Opin. Nephrol. Hypertens. 1999, 8,
39–43.
6. Collins, M. D.; Mao, G. E. Ann. Rev. Pharmacol. Toxi-
col. 1999, 39, 399–430.
7. Goodman, D. S.; Olson, J. A. Methods Enzymol. 1969,
15, 462–475.
8. Olson, J. A. J. Nutr. 1989, 119, 105–108.
9. von Lintig, J.; Wyss, A. Arch. Biochem. Biophys. 2001,
385, 47–52.
10. Wirtz, G. M.; Bornemann, C.; Giger, A.; Muller, R. K.;
Schneider, H.; Schlotterbeck, G.; Schiefer, G.; Woggon,
W. D. Helv. Chim. Acta 2001, 84, 2301–2315.
11. Leuenberger, M. G.; Engeloch-Jarret, C.; Woggon, W. D.
Angew. Chem., Int. Ed. 2001, 40, 2614–2617.
12. Horst, P. Angew. Chem. 1960, 72, 811–819.
13. Quesada, M. L.; Schlessinger, R. H. Synth. Commun.
1976, 8, 555–557.
18. 2-Methyl-3-[3-(1-methyl-2-oxo-ethylidene)-cyclohex-1-
14. 3-[3-(1-Cyano-ethylidene)-cyclohex-1-enyl]-2-methyl-acry-
1
enyl]-propenal (3): H NMR (300 MHz, CDCl3): l 10.24
1
lonitrile (6): [EE-isomer] H NMR (300 MHz, CDCl3): l
(1H, s), 9.46 (1H, s), 6.93 (1H, s), 6.81 (1H, s), 2.90 (2H,
dt, J=6.30, 1.20 Hz), 2.6 (2H, t, J=6.00 Hz), 2.01 (3H,
s, CH3), 1.87 (3H, s, CH3), 1.89–1.84 (2H, m). 13C NMR
(75 MHz, CDCl3): l 195.42, 191.1, 150.74, 148.96,
144.64, 138.93, 133.33, 132.22, 29.29, 24.15, 22.48, 11.31,
10.26. HRMS: calcd for C13H16O2 204.1150; found
204.1135.
6.74 (1H, s), 6.56 (1H, s), 2.70 (2H, t, J=6.00 Hz, CH2),
2.35 (2H, t, J=6.30 Hz, CH2), 2.07 (3H, s, CH3), 1.95
(3H, s, CH3), 1.78 (2H, quintet, J=6.60, 6.30 Hz). 13C
NMR (75 MHz, CDCl3): l 150.07, 145.64, 141.22,
132.47, 119.26, 119.22, 106.85, 105.19, 25.74, 25.61, 22.79,
21.3, 16.17. HRMS: calcd for C13H14N2 198.1157; found
198.1162. [EZ-isomer]: 1H NMR (300 MHz, CDCl3): l
6.59 (1H, s), 6.50 (1H, s), 2.66 (2H, t, J=6.00 Hz, CH2),
2.58 (2H, t, J=6.30 Hz, CH2), 2.07 (3H, s, CH3), 1.96
(3H, s, CH3), 1.79 (2H, quintet, J=6.46, 6.30 Hz). 13C
NMR (75 MHz, CDCl3): l 148.59, 145.28, 141.96,
128.69, 119.94, 119.03, 107.33, 104.44, 29.25, 26.64, 22.69,
21.75, 15.12. HRMS: calcd for C13H14N2 198.1157; found
198.1160.
19. 1: 1H NMR (500 MHz, CDCl3): l 7.10–5.90 (12H, m,
olefinic), 2.55–2.40 (4H, m, 2×CH2), 2.10 (3H, s, CH3),
2.20–2.00 (4H, m, 2×CH2), 1.90 (6H, s, 2×CH3), 1.78
(3H, s, CH3), 1.70 (6H, s, 2×CH3), 1.62–1.59 (4H, m,
2×CH2), 1.52 (2H, s, CH2), 1.50–1.42 (4H, 2×CH2), 1.02
(12H, s, 4×CH3). HRMS calcd for C43H60 576.4695;
found 576.4719. UV: umax (pentane): 450 nm.