General synthesis of epi-series catechins and their 3-gallates: Reverse polarity strategy

Article abstract of DOI:10.1039/c003464a

A general synthetic route to the epi-series catechins was developed based on the reverse polarity strategy. Aromatic nucleophilic substitution reaction followed by the sulfinyl-metal exchange and cyclization enabled stereo-controlled access to various members of epi-series catechins and their 3-gallates.

Full text of DOI:10.1039/c003464a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/obc | Organic & Biomolecular Chemistry
General synthesis of epi-series catechins and their 3-gallates: reverse polarity
strategy†
Ken Ohmori, Takahisa Yano and Keisuke Suzuki*
Received 23rd February 2010, Accepted 30th March 2010
First published as an Advance Article on the web 12th April 2010
DOI: 10.1039/c003464a
A general synthetic route to the epi-series catechins was
developed based on the reverse polarity strategy. Aromatic
nucleophilic substitution reaction followed by the sulfinyl–
metal exchange and cyclization enabled stereo-controlled
access to various members of epi-series catechins and their
3-gallates.
Recently, much attention has been centered on the significant
bioactivities of the epi-series catechins as represented by tea
catechins, e.g. (-)-epigallocatechin (EGC, 1), (-)-epicatechin (EC,
2), and (-)-epiafzelechin (EZ, 3) (Fig. 1).1 (-)-Epigallocatechin
gallate (EGCg, 4), one of the most abundant ingredients in green
tea, exhibits remarkable suppressive effects against various human
tumor cell lines, and also antimutagenic activity and antivirus
effects.2
Scheme 1 Stereochemical lability of catechin/epicatechin derivatives.
oxidation and reduction,4 which, however, inevitably depends on
the availability of the catechin starting materials. Other approaches
have started to appear, which, however, are still limited in scope
in terms of the versatility or the accessibility of the single
enantiomer.5
Recently, we reported a stereo-controlled synthetic approach
to the catechin derivatives via three key steps (Scheme 2);6 the
Mitsunobu-type C–O bond formation7 of iodophenol III and
epoxy alcohol syn-IV (step 1), cleavage of the oxirane ring to
give the corresponding bromohydrin VI (step 2), and the pyran-
ring closure via the selective iodine–metal exchange, constructing
the catechin skeleton VII (step 3). Ready availability of syn-IV8
via Sharpless asymmetric dihydroxylation9 makes this protocol a
general route to various catechin derivatives.
Fig. 1 Structure of epicatechin-type polyphenols.
A recent study at the molecular level revealed the specific
binding of 4 to the 67 kDa laminin receptor that mediates
its anticancer action at physiologically relevant concentrations.2c
Because of their relevance to these significant bio-functions, 4 and
other epicatechin-3-gallate derivatives have attracted a great deal
of scientific attention.
In spite of such promise, controlled synthesis of the epi-series
catechins remains a challenge, due to chemical/stereochemical
lability (Scheme 1): the C(2)-stereocenter is prone to epimerization
(IꢀAꢀII) under basic/acidic conditions providing a variable
mixture with the catechin congeners of 2,3-trans stereochemistry.
Even racemization can take place via, inter alia, the enolization
from A.3
So far, preparation of the epicatechins mainly relies on the
inversion of the C(3)-stereochemistry of the catechin derivatives via
Scheme 2 Stereo-controlled approach to catechin derivatives.
Department of Chemistry, Tokyo Institute of Technology, O-okayama,
Meguro-ku, Tokyo 152-8551, Japan. E-mail: ksuzuki@chem.titech.ac.jp
† Electronic supplementary information (ESI) available: Experimental
procedures for the preparation and spectral data of all compounds. See
DOI: 10.1039/c003464a
We envisioned that such an approach could provide access to the
epicatechins as well, given that the epimeric precursor syn-V was
accessible (Scheme 3). Among several options for the preparation
of the requisite structure syn-V,10 we focused on a change in the
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2693–2696 | 2693
©

View Article Online
Scheme 5 Conditions: (a) n-BuLi, PhSSO2Ph, Et2O, -78 ◦C, 87%.
(b) BnOH, NaH, DMF, 0 ◦C. (c) mCPBA, CH2Cl2, 0 ◦C, 2 steps 85%.
sulfoxides allowed isolation of 8 in 85% yield by silica-gel column
chromatography (hexane–EtOAc = 4 : 1). The corresponding 2,6-
regioisomer was separated in 8% yield.
With the requisite acceptor 8 in hand, the projected SNAr
reaction was attempted (Scheme 6). Upon treatment of 8 with
the sodium alkoxide of 9,6 the reaction proceeded smoothly in a
mixed solvent (toluene–DMPU = 4 : 1) at 0 ◦C. Notably, the Payne
rearrangement17 was not observed under these conditions, thereby
giving the desired SNAr-product 10 in 76% yield as a mixture of
diastereomers. Separation by silica-gel column chromatography
gave diastereomeric epoxy ethers 10a (39%) and 10b (37%), albeit
the sulfoxide stereochemistry was unspecified. Oxirane cleavage
Scheme 3 Change of the connectivity for reverse polarity strategy.
connectivity of the A/B-ring fragments: Instead of dissecting the
C ring into fragments c and d (blue lines), disconnection at the
Ar–O and Ar–C bonds (red lines) provides a new fragment pair, a
and b, which could be combined in a reversed polarity (umpolung)
fashion without affecting the C(2)-stereochemistry in b.
18
with Li2NiBr4 followed by silylation with triethylsilyl (TES)
triflate afforded the diastereomeric bromides 11a/b (97%: 11a
from 10a, 91%: 11b from 10b), respectively.
To realise this scenario, we planned the aromatic nucleophilic
substitution (SNAr)11 by employing electrophilic unit III¢ and the
alkoxide derived from syn-IV (Scheme 4).
Scheme 6 Conditions: (a) NaH, toluene, DMPU room temp., 39%
for 10a, 37% for 10b. (b) Li2NiBr4, THF, 0 ◦C. (c) TESOTf, 2,6-
lutidine, CH2Cl2, 0 ◦C, 2 steps, 97% for 11a, 91% for 11b. DMPU =
N,N¢-dimethylpropyleneurea, TES = triethylsilyl.
Scheme 4 SNAr strategy for retentive assembly of III¢ and syn-IV.
The choice of the two substituents X and Y was critical. The Y
group has to be a strongly electronegative group such as fluorine,
being capable of attracting the incoming nucleophile to form the
Meisenheimer complex B. On the other hand, the X group must
play two roles, (1) a p-electron acceptor to facilitate the conjugate
attack, and (2) an anion precursor at the stage of the pyran-ring
closure later in the synthetic scheme. We chose a sulfinyl group as
X,12 expecting that the sulfinyl–metal exchange13 would serve for
the latter purpose.
The next stage was to construct the flavan skeleton. We planned
the cyclization of 11a/b via the sulfinyl–metal exchange followed
by intramolecular nucleophilic substitution (Scheme 7). Thus,
upon treatment of diastereomeric sulfoxides 11a and 11b with
PhLi (2.0 equiv in THF, room temp., 1 h),19 smooth cyclization
of the pyran ring occurred, giving a single cyclized product 12,
respectively (81% from 11a and 62% from 11b). The C(2) and C(3)
Scheme 5 shows the preparation of fluoro-sulfoxide 8. Lithiation
of 1,3,5-trifluorobenzene (5)14 with n-BuLi15 followed by trapping
of the resulting anion with PhSSO2Ph16 gave sulfide 6 in 87% yield.
Upon careful treatment of 6 with sodium benzyloxide (2.5 mol
◦
equiv., 0 C, DMF, 1 h), two fluorine atoms were displaced with
high regioselectivity, giving the 2,4-bis-benzyl ether 7. A small
amount of the 2,6-regioisomer was also produced which was
inseparable at this stage. However, oxidation to the corresponding
Scheme 7 Construction of flavan skeleton by sulfur–metal exchange and
intramolecular nucleophilic substitution.
2694 | Org. Biomol. Chem., 2010, 8, 2693–2696
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
relative stereochemistry was unambiguously assigned by 1H NMR
analysis (J2,3 = <0.5 Hz).
It is noteworthy that the aryl lithium species generated by the
sulfinyl–metal exchange could be exploited for such an efficient
C–C bond formation, which has sizable potential for application
for other purposes.
Conversion of 12 into EGC (1) was carried out as follows
(Scheme 8). After removal of the TES group by n-Bu4NF to
give alcohol 13, all the benzyl protecting groups were removed
by hydrogenolysis over 20% Pd(OH)2/C in a mixed solvent
(THF–MeOH–H2O = 4 : 4 : 1, 12 h). Careful non-aerobic filtration
through a Celite pad (washed with MeOH) followed by purifi-
cation by Sephadex LH-20 column chromatography (MeOH)
afforded (-)-EGC (1) as a snow-white amorphous solid (64% yield
from 12), which was indistinguishable from the authentic specimen
in all respects {[a]2D1 -56.0 (c 1.02, acetone–H2O = 1 : 1), lit. [a]2D4
-57.1 (c 1, acetone–H2O = 1 : 1).3c
Scheme 9 Synthesis of (-)-epicatechin [EC (2)] and (-)-epiafzelechin
[EZ (3)] and their 3-gallates, (-)-ECg (20) and (-)-EZg (21). Conditions:
(a) NaH, toluene, DMPU, room temp. (b) Li2NiBr4, THF, 0 ◦C.
(c) TESOTf, 2,6-lutidine, CH2Cl2, 0 ◦C. (d) PhLi, THF, 0 ◦C. (e) n-Bu4NF,
THF, 0 ◦C, 99%. (f) H2, Pd(OH)2, THF, MeOH, H2O (4 : 4 : 1), room
temp.; (g) 3,4,5-tri-O-benzylgallic acid, EDCI·HCl, DMAP, Et3N, CH2Cl2
room temp., (h) H2, Pd(OH)2/C, THF, MeOH, H2O (4 : 4 : 1), room temp.
DMPU = N,N¢-dimethylpropyleneurea, TES = triethylsilyl.
In summary, the work described above entails a number of
noteworthy developments, including 1) the general and stereos-
elective access to the epi-series catechins and their 3-gallates,
2) the retentive assembly of the cyclization precursors via the
nucleophilic aromatic substitution of aryl fluoride bearing a
sulfinyl group at the ortho position, 3) the sulfinyl–metal exchange
and following nucleophilic substitution to form the corresponding
cyclized product. The efficient C–C bond formation in this
process was especially striking, which should be useful for other
purposes.
Scheme 8 Conditions: (a) n-Bu4NF, THF, 0 ◦C, 99%. (b) H2, Pd(OH)2,
THF, MeOH, H2O (4 : 4 : 1), room temp., 71%. (c) 3,4,5-tri-O-benzylgallic
acid, EDCI·HCl, DMAP, Et3N, CH2Cl2 room temp. (d) H2, Pd(OH)2/C,
THF, MeOH, H2O (4 : 4 : 1), room temp., 68% (2 steps).
Furthermore, the protection pattern of alcohol 13 is ideally
suited for straightforward access to EGCg (4): Esterification of
13 with 3,4,5-tri-O-benzylgallic acid followed by hydrogenolysis
of all benzyl groups gave 4 {[a]2D1 -218 (c 0.660, acetone–H2O =
1 : 1), lit.3c [a]D24 -216.4 (c 1, acetone–H2O = 1 : 1)} as a snow-white
amorphous solid in good yield.
Starting from epoxy alcohols 14 and 15, applicability of these
protocols was further demonstrated by the syntheses of (-)-EC (2)
and (-)-EZ (3), and its 3-gallates, respectively (Scheme 9). After
the same three-step sequence as before, the resulting bromides 16
and 1720 were subjected to the cyclization followed by desilylation
as before, giving the corresponding pyrans, 18 and 19 in good
yield, respectively. Detachment of all the benzyl groups in 18 and
19 gave (-)-EC (2)21 and (-)-EZ (3)22 as a snow-white amorphous
solid, respectively. Furthermore, esterification of 18 and 19 with
3,4,5-tri-O-benzylgallic acid followed by hydrogenolysis of all the
benzyl groups gave the 3-gallate of (-)-epicatechin [(-)-ECg (20)],23
and that of (-)-epiafzelechin [(-)-EZg (21)],24 in good yields,
respectively.
Acknowledgements
This work was partially supported by Grant-in-Aid for Scientific
Research (No. 21350050) from MEXT, Japan the Global COE
program (Chemistry) and Shorai Foundation for Science and
Technology. We are also grateful to Dr Takashi Higuchi for early
work of this project.
Notes and references
1 (a) J. Jankun, S. H. Selman and R. Swiercz, Nature, 1997, 387, 561;
(b) Y. Hara, Green Tea, CRC Press, Boca Raton, 2001.
2 (a) J.-M. Song, K.-H. Lee and B.-L. Seong, Antiviral Res., 2005, 68, 66;
(b) M. Nakayama, K. Suzuki, M. Toda, S. Okubo and T. Shimamura,
Antiviral Res., 1993, 21, 289; (c) H. Tachibana, K. Koga, Y. Fujimura
and K. Yamada, Nat. Struct. Mol. Biol., 2004, 11, 380.
3 (a) A. J. Birch, J. W. Clark–Lewis and A. V. Robertson, J. Chem. Soc.,
1957, 3586; (b) P. Kiatgrajai, J. D. Wellons, L. Gollob and J. D. White,
J. Org. Chem., 1982, 47, 2910; (c) R. Seto, H. Nakamura, F. Nanjo and
Y. H a r a , Biosci., Biotechnol., Biochem., 1997, 61, 1434; (d) P. P. Mehta
and W. E. Whalley, J. Chem. Soc., 1963, 5327.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2693–2696 | 2695
©

View Article Online
4 W. Tu¨ckmantel, A. P. Kozikowski and L. J. Romanczyk, Jr., J. Am.
Chem. Soc., 1999, 121, 12073.
1991, 4, 195; (c) T. Satoh, J. Synth. Org. Chem. Jpn., 1996, 54, 481;
(d) T. Satoh, Farumashia, 1999, 35, 1225.
14 A. Pleschke, A. Marhold, M. Schneider, A. Kolomeitsev and G.-V.
Roschenthaler, J. Fluorine Chem., 2004, 125, 1031.
15 (a) C. Heiss, E. Marzi, M. Florence and M. Schlosser, Eur. J. Org.
Chem., 2007, 669; (b) K. Shen, Y. Fu, J.-N. Li, L. Liu and Q.-X.
Guo, Tetrahedron, 2007, 63, 1568; (c) M. Stratekis, P. G. Wang and
A. Streitwieser, J. Org. Chem., 1996, 61, 3145.
5 Syntheses of epi-series catechins, see: (a) L. Li and T. H. Chan, Org.
Lett., 2001, 3, 739;(b) M. Kitade, Y. Ohno, H. Tanaka and T. Takahashi,
Synlett, 2006, 17, 2827; (c) H. Tanaka, H. Miyoshi, Y.-C. Chuang, Y.
Ando and T. Takahashi, Angew. Chem., Int. Ed., 2007, 46, 5934; (d) T.
Furuta, Y. Hirooka, A. Abe, Y. Sugata, M. Ueda, K. Murakami, T.
Suzuki, K. Tanaka and T. Kan, Bioorg. Med. Chem. Lett., 2007, 17,
3095.
16 F. Chemla, Synlett, 1998, 894.
17 (a) G. B. Payne, J. Org. Chem., 1962, 27, 3819; (b) R. M. Hanson, Org.
React., 2002, 60, 1.
6 T. Higuchi, K. Ohmori and K. Suzuki, Chem. Lett., 2006, 35, 1006.
7 O. Mitsunobu, Synthesis, 1981, 1.
8 X. Ren, X. Chen, K. Peng, X. Xie, Y. Xia and X. Pan, Tetrahedron:
Asymmetry, 2002, 13, 1799.
9 K. B. Sharpless, W. Amberg, Y. L. Bennni, G. A. Crispino, J. Hartung,
K. S. Jeong, H. L. Kwong, K. Morikawa, Z. M. Wang, D. Xu and X. L.
Zhang, J. Org. Chem., 1992, 57, 2768.
18 R. D. Dawe, T. F. Molinski and J. V. Turner, Tetrahedron Lett., 1984,
25, 2061.
19 PhLi was the reagent of choice for this reaction. In contrast, use of
n-BuLi resulted in a poorer yield and conversion (39% yield of 12, 26%
recovery of the starting material 11a).
10 Two optional approaches are conceivable: 1) employment of the C(2)-
epimer, anti-IV, as a substrate for the Mitsunobu reaction, 2) double
inversion at the C(2) position. However, these approaches were less
attractive in view of the poor accessibility of the substrate and/or the
longer steps.
11 (a) G. Bartoli and P. E. Todesco, Acc. Chem. Res., 1977, 10, 125; (b) M. J.
Strauss, Chem. Rev., 1970, 70, 667; (c) J. Miller, Aromatic Nucleophilic
Substitution, Elsevier, Amsterdam, 1968.
12 (a) A. M. Ratz and L. O. Weigel, Tetrahedron Lett., 1999, 40, 2239;
(b) K. Hiroi, Y. Suzuki, I. Abe and R. Kawagishi, Tetrahedron, 2000,
56, 4701.
13 (a) T. Durst, M. J. LeBelle, R. V. Elzen and K.-C. Tin, Can. J. Chem.,
1974, 52, 761. For reviews, see: (b) S. Oae, Rev. Heteroatom Chem.,
20 These products were obtained as diastereomer mixtures (ca. 1 : 1), which
were used for further cyclization reaction without separation.
21 K. Kamita, C. Watanabe, H. Endang, M. Umar and T. Satake, Chem.
Pharm. Bull., 2001, 49, 551.
22 K.-R. Min, B.-Y. Hwang, H.-S. Lim, B.-S. Kang, G.-J. Oh, J. Le, S.-H.
Kang, K.-S. Lee, J.-S. Ro and K. Youngsoo, Planta Med., 1999, 65,
460.
23 (a) Y. Kashiwada, G. Nonaka and I. Nishioka, Chem. Pharm. Bull.,
1984, 32, 3461; (b) R. Saijo, G. Nonaka and I. Nishioka, Phytochem-
istry, 1989, 28, 2443.
24 (a) S.-B. Wan and T.-H. Chan, Tetrahedron, 2004, 60, 8207; (b) K. K.
Khi, Z. A. Kuliev, A. D. Vdovin, M. R. Yagudaev and V. M. Mailikov,
Chem. Nat. Compd., 1991, 27, 681.
2696 | Org. Biomol. Chem., 2010, 8, 2693–2696
This journal is
The Royal Society of Chemistry 2010
©