A stereocontrolled synthesis of δ-trans-tocotrienoloic acid

Article abstract of DOI:10.1021/ol051849t

(Chemical Equation Presented) A consise stereoselective total synthesis of a naturally occurring polymerase β inhibitor, δ-trans-tocotrienoloic acid (2), is described. The key step in the synthesis is an acid-catalyzed cyclodehydration reaction. Additiona

Full text of DOI:10.1021/ol051849t

ORGANIC
LETTERS
2005
Vol. 7, No. 19
4297-4300
A Stereocontrolled Synthesis of
-trans-Tocotrienoloic Acid
δ
David J. Maloney and Sidney M. Hecht*
Departments of Chemistry and Biology, UniVersity of Virginia,
CharlottesVille, Virginia 22901
sidhecht@Virginia.edu
Received August 2, 2005
ABSTRACT
A consise stereoselective total synthesis of a naturally occurring polymerase
â
inhibitor,
δ
-trans-tocotrienoloic acid (2), is described. The key
step in the synthesis is an acid-catalyzed cyclodehydration reaction. Additionally, this report corrects a previously reported structural assignment,
defines the absolute stereochemistry of 2, and defines key structural requirements for polymerase inhibition.
â
DNA polymerases play a central role in the replication,
recombination, and repair of DNA.1 In particular, DNA
polymerase â (pol â) is a 39-kDa enzyme that participates
in base excision repair, by both gap-filling polymerization
and subsequent removal of the 5′-terminal deoxyribose
phosphate residues from the incised apurinic and apyrimi-
dinic sites. Importantly, it has been shown that pol â is
involved in the repair of DNA lesions formed after exposure
to DNA damaging agents such as cisplatin and bleomycin.2
Consequently, the use of otherwise noncytotoxic pol â
inhibitors in conjunction with these chemotherapeutic agents
should logically potentiate their cytotoxicity, thus allowing
for lower doses to be administered.
ketone extract of Chrysochlamys ulei.4 Compound 1 dis-
played good pol â inhibition (IC50 4.3 µM), and was found
to be unaffected by the presence of serum albumin, consistent
with the possibility that 1 may be of utility in vivo. Given
the interesting biological activity of 1, its limited availability,
and the need for structure confirmation, the total synthesis
was deemed desirable.
Upon completion of the stereoselective synthesis of 1 it
was determined that the previously reported structural
assignment was incorrect. Accordingly, this report also details
the first total synthesis of the correct structure, δ-trans-
Our laboratory has studied the isolation and biological
evaluation of naturally occurring pol â inhibitors; this has
resulted in the discovery of several novel compounds with
good inhibitory activity.3 Recently, we reported the bioassay-
guided isolation of a novel DNA polymerase â inhibitor,
chrysochlamic acid (1, Figure 1), from the methyl ethyl
(1) Kornbeg, A.; Baker. T. A. DNA Replication, 2nd ed.; W. H. Freeman
and Co.: New York, 1992.
(2) (a) Chen, J.; Zhang, Y.-Z.; Wang, L.-K.; Sucheck, S. J.; Snow, A.
M.; Hecht, S. M. Chem. Commun. 1998, 2769. (b) Sun, D.-A.; Deng, J.-Z.;
Starck, S. R.; Hecht, S. M. J. Am. Chem. Soc. 1999, 121, 6120. (c) Ma, J.;
Starck, S. R.; Hecht, S. M. J. Nat. Prod. 1999, 62, 1660.
Figure 1. Structures of chrysochlamic acid (1) and δ-trans-
tocotrienoloic acid (2).
(3) Hecht, S. M. Pharm. Biol. 2003, 41, S68 and references therein.
10.1021/ol051849t CCC: $30.25
© 2005 American Chemical Society
Published on Web 08/20/2005

tocotrienoloic acid (2),5 via an acid-catalyzed cyclodehydra-
tion reaction of a late-stage intermediate utilized for the
synthesis of 1. The synthesis permitted facile access to both
1 and 2, which permitted structure confirmation, as well as
the determination of absolute stereochemistry.
iodide 5 in 96% yield.9 Subsequent protection of the hindered
tertiary alcohol was achieved using TESOTf with 2,6-lutidine
in CH2Cl2 to afford the requisite alkyl iodide derivative 6 in
91% yield.
Several approaches to the vinyl iodide fragment 11 were
Retrosynthetically, we envisioned the separate syntheses
of both the requisite alkyl iodide 6 and vinyl iodide 11,
followed by an sp3-sp2 Suzuki coupling to assemble the
molecule in a convergent and efficient manner, as shown
below. The synthesis of alkyl iodide 6 is outlined in Scheme
1. Installation of the trisubstituted olefin was achieved in
explored; the most efficient route is outlined in Scheme 2.
Scheme 2. Synthesis of Vinyl Iodide 11a
Scheme 1. Synthesis of Alkyl Iodide 6a
Construction of the silyl-protected enyne was accomplished
in near quantitative yield via Negishi coupling of 710 and
811 by the use of methodology reported on similar systems.11
Removal of the silyl groups with tetra-n-butylammonium
fluoride (TBAF) in THF at 0 °C gave intermediate 912 in
quantitative yield. Zirconium-catalyzed carboalumination
followed by iodination provided the (E,E)-vinyl iodide in
76% yield. The primary alcohol was then oxidized using
Dess-Martin periodinane13 to yield aldehyde 10 (87%).
Wittig olefination with carbethoxyethylidene triphenylphos-
phorane in toluene at 90 °C gave the desired (E,E,E)-vinyl
iodide 11 in 85% yield with a minor amount (∼5%) of the
undesired (Z)-isomer, which could be easily separated by
column chromatography.
96% yield via a Horner-Emmons reaction of 36 with triethyl
phosphonoacetate by the use of sodium hydride as the base.7
The resulting R,â-unsaturated ethyl ester was reduced to the
requisite allylic alcohol with DIBAL-H in good yield (93%).
Asymmetric Sharpless epoxidation was accomplished by
using the (+)-DET ligand to give the (S,S)-trisubstituted
epoxide 4 in high yield (84%) and ee (95%).8 Regioselective
hydride delivery to the less substituted C of the epoxide was
achieved with Red-Al (sodium bis(2-methoxyethoxy)alumi-
num hydride) in THF at 0 °C to give the desired 1,3-diol in
97% yield. Selective mesylation of the primary alcohol
(methanesulfonyl chloride and NEt3 in CH2Cl2 at 0 °C) gave
the mesylate in 85% yield. Nucleophilic displacement of the
mesylate with NaI-saturated acetone at reflux then gave alkyl
With both the alkyl iodide 6 and vinyl iodide 11 in hand,
the Suzuki coupling was attempted using a variety of
methods. The best results were obtained using 9-MeO-9-
BBN, t-BuLi, K3PO4, Pd(dppf)Cl2, and 1.5 equiv of 11 to
give the desired coupled product in 65-81% yields (Scheme
3).14 Deprotection of the TES-protected tertiary alcohol
proceeded smoothly with 6 equiv of TBAF to afford key
intermediate 12 in 92% yield. Saponification of the ethyl
ester using K2CO3 in aqueous MeOH followed by mild
(4) Deng, J.-Z.; Sun, D.-A.; Starck, S. R.; Hecht, S. M.; Cerny, R. L.;
Engen, J. R. J. Chem. Soc., Perkin Trans. 1 1999, 1147. Structural
assignment was based on NMR (1H, 13C, 2D experiment) and HRMS [M
- H2O], and involved the assumption that the tertiary OH group had been
lost during mass spectrometric analysis. In fact the mass spectrum of
synthetic 1 did reflect some dehydration.
(5) (a) Monache, F. D.; Marta, M.; Mac-Quhae, M. M.; Nicoletti, Gazz.
Chim. Ital. 1984, 114, 135. (b) Setzer, W. N.; Green, T. J.; Lawton, R. O.;
Moriarity, D. M.; Bates, R. B.; Caldera, S.; Haber, W. A. Planta Med.
1995, 61, 275, from ToVomitopsis psychotriifolia. (c) Terashima, K.;
Shimamura, T.; Tanabayashi, M.; Aqil, M.; Akinniyl, J. A.; Niwa, M.
Heterocycles 1997, 45, 1559, from Garcinia kola.
(9) While the 1,3-diol could be converted to the alkyl iodide directly
using PPh3, imidazole, and I2, the yield was low.
(10) (a) Zheng, Y. F.; Oehlschlager, A. C.; Hartman, P. G. J. Org. Chem.
1994, 59, 5803. (b) Roush, W. R.; Barda, D. A.; Limberakis, C.; Roxanne,
K. Tetrahedron 2002, 58, 6433. (c) Groth, U.; Richter, N.; Kalogerakis, A.
Eur. J. Org. Chem. 2003, 23, 4034.
(11) (a) Foote, K. M.; Hayes, C. J.; John. M. P.; Pattenden, G. Org.
Biomol. Chem. 2003, 1, 3917. (b) Negishi, E.; Boardman, L. D.; Sawada,
H.; Bagheri, V.; Stoll, T. A. J. Am. Chem. Soc. 1998, 110, 5383.
(12) Johnson, W. S.; Yarnell, T. M.; Myers, R. F.; Morton, D. R.; Boots,
S. G. J. Org. Chem. 1980, 45, 1254.
(6) See Supporting Information for experimental procedures employed
for the synthesis of 3.
(7) The product was obtained in 6:1 (E/Z) ratio; attempts to improve
this ratio were unsuccessful.
(8) Katsuki, T.; Sharpless, K. B. J. Am. Chem. Soc. 1980, 102, 5976.
Percent ee was determined by analysis of the MTPA ester.
(13) (a) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155. (b)
Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277.
(14) (a) Johnson, C. R.; Braun, M. P. J. Am. Chem. Soc. 1993, 115,
11014. (b) Marshall, J. A.; Bourbeau, M. P. J. Org. Chem. 2002, 67, 2751.
4298
Org. Lett., Vol. 7, No. 19, 2005

Scheme 3. Suzuki Coupling and Synthesis of 1 and 2a
demethylation via cerium ammonium nitrate (CAN) oxida-
in THF-MeOH-H2O. Synthetic 2 gave spectral data in full
agreement with those of the natural product.16 Interestingly,
despite having a very similar structure, 1 displayed virtually
no polymerase â inhibition even at concentrations >120 µM,
compared to 2 (IC50 ∼4 µM) in preliminary in vitro assays.
It may be noted that other approaches to the synthesis of
tocotrienols have been reported,17 although most of these
have afforded racemic products. Considerable efforts have
also been expended to define routes to optically pure
chromans.18
While the cyclodehydration reaction has been reported
previously for the synthesis of optically active chromans,
we sought further mechanistic insight into this intriguing
reaction. Cohen and co-workers15c had previously suggested
a catalytic redox cycle similar to that shown in Scheme 4.
In this case, hydroquinone 14 is oxidized in trace quantities
to quinone 15 by O2. Subsequent nucleophilic attack by the
tertiary alcohol on the quinone forms hemiacetal 16, which
is reduced by the starting hydroquinone, generating additional
15. To obtain further evidence for the proposed mechanism,
quinone 15 was treated under the usual cyclodehydration
conditions (pTsOH, benzene, 80 °C); as expected, formation
of 13 was not observed due to the lack of the reducing
hydroquinone, as shown in Scheme 4. Interestingly, admix-
ture of quinone 15 and methylhydroquinone in the presence
of pTsOH yielded 13 along with methyl 1,4-benzoquinone
(1.0 equiv), arguing that the reaction must proceed through
the previously proposed catalytic redox cycle shown in
Scheme 4.
tion and reductive workup (Na2S2O4) gave 1 as an air-
sensitive yellow oil in 60% yield. Surprisingly, upon
comparison of their NMR spectra, it was apparent that the
spectrum of the synthetic compound differed slightly from
that of the natural product. Examination of literature reports
revealed that the NMR data from the natural product matched
that of a previously reported natural product, δ-trans-
tocotrienoloic acid (2), first isolated from the fruits of Clusia
grandiflora.5a Since this compound has not been synthesized
previously we sought to confirm the structural assignment
and determine the absolute stereochemistry. We envisioned
constructing 2 from previously synthesized intermediate 12
via an acid-catalyzed cyclodehydration reaction, thus allow-
ing for rapid access to the natural product without major
changes to the synthesis. Therefore, using the same oxidative
demethylation procedure as described above, compound 12
was demethylated (Scheme 3) to give the air-sensitive
hydroquinone derivative 14 shown in Scheme 4. Treatment
Scheme 4. Proposed Mechanism and Mechanistic Study of the
Cyclodehydration Reaction
In summary, the stereoselective synthesis, structure con-
firmation, and absolute stereochemistry of (+)-trans-toco-
(15) (a) Inoue, S.; Ikeda, H.; Sato, S.; Horie, K.; Ota, T.; Miyamoto, O.;
Sato, K. J. Org. Chem. 1987, 52, 5495. (b) Jung, M. E.; MacDougall, J. M.
Tetrahedron Lett. 1999, 40, 6339. (c) Cohen, N.; Lopresti, R. J.; Neukom,
C. J. Org. Chem. 1981, 46, 2445. (d) Mayer, H.; Vetter, W.; Metzger, J.;
Ru¨egg, R.; Isler, O. HelV. Chim. Acta 1967, 50, 1168.
(16) Synthetic 2 was identical by 1H, 13C NMR, and HRMS comparisons
with the authentic sample. The optical rotation of synthetic 2, [R]22D +19.6
(c 0.42, MeOH), had the same sign as that of the authentic sample, [R]23
D
+17.5 (c 0.18, MeOH). The somewhat lower [R]D value originally recorded
for the natural product (+6.7) undoubtedly reflected the low concentration
used for the measurement. Optical rotations measured in CHCl3 were found
to give very low values, in agreement with an early report.5c
(17) (a) Pearce, B. C.; Parker, R. A.; Deason, M. E.; Qureshi, A. A.;
Wright, J. J. J. Med. Chem. 1992, 35, 3595. (b) Terashima, K.; Takaya, Y.;
Niwa, M. Bioorg. Med. Chem. 2002, 10, 1619.
with catalytic pTsOH in benzene at reflux afforded the
desired cyclization product 13, presumably with retention
of configuration as reported previously for similar systems
(70% from 12).15 Saponification of the ethyl ester using K2-
CO3 and aqueous MeOH led to decomposition of the starting
material; however, 2 was obtained in 90% yield using LiOH
(18) Trost, B. M.; Shen, H. C.; Dong, L.; Surivet, J.-P.; Sylvain, C. J.
Am. Chem. Soc. 2004, 126, 11966 and references therein.
Org. Lett., Vol. 7, No. 19, 2005
4299

trienoloic acid (2) via an acid-catalyzed cyclodehydration
reaction has been accomplished. Additionally, this report
corrects a previously reported structural assignment (i.e., 1)
and defines structural requirements for pol â inhibition. The
route reported also provides access to analogues of 2 having
potentially improved activity as DNA polymerase â inhibi-
tors.
by NIH Research Grant 50771, awarded by the National
Cancer Institute.
Supporting Information Available: Experimental pro-
cedures and characterization data for 1 and 2, and synthetic
intermediates. This material is available free of charge via
the Internet at http://pubs.acs.org.
Acknowledgment. We thank Dr. James Ryan for the
biochemical evaluation of 1 and 2. This work was supported
OL051849T
4300
Org. Lett., Vol. 7, No. 19, 2005