The flavan-isoflavan rearrangement: Bioinspired synthetic access to isoflavonoids via 1,2-shift-alkylation sequence

Article abstract of DOI:10.1039/c5cc01572c

An approach to 2-substituted isoflavonoids is reported based on the 1,2-shift of the aryl group in the catechin skeleton followed by the in situ alkylation. Synthesis of (-)-equol, a natural isoflavan with estrogenic activities, was achieved.

Full text of DOI:10.1039/c5cc01572c

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
The flavan–isoflavan rearrangement: bioinspired
synthetic access to isoflavonoids via
1,2-shift–alkylation sequence†
Cite this: Chem. Commun., 2015,
51, 7012
Received 21st February 2015,
Accepted 16th March 2015
Kayo Nakamura, Ken Ohmori* and Keisuke Suzuki*
DOI: 10.1039/c5cc01572c
www.rsc.org/chemcomm
An approach to 2-substituted isoflavonoids is reported based on the
1,2-shift of the aryl group in the catechin skeleton followed by the
in situ alkylation. Synthesis of (ꢀ)-equol, a natural isoflavan with
estrogenic activities, was achieved.
Isoflavonoids constitute a class of natural products widely
found in leguminous plants. In addition to the phytoalexin
activity in the original plants,1 some compounds have been
attracting special attention in the field of human health care,
e.g., the estrogen activities identified in the soybean-derived
compounds 1–3.2 Furthermore, several elaborated compounds
with stereogenicity are found in nature, including isoflavans,
pterocarpans, and rotenonoids, such as 4–6 (Fig. 1). Due to the
diverse range of bioactivities,1 isoflavonoids have become one
of the current targets for chemical synthesis.3
The isoflavonoid biosynthesis shares its early stage with that of
the flavonoid, branching at the flavanone stage by P-450 mediated
oxygenation to induce the aryl 1,2-shift within the chroman skeleton
to form the 3-aryl derivatives (Scheme 1). Although the following
dehydration gives isoflavones,4 stereogenic compounds, e.g., 4–6 are
Scheme 1 Biosynthesis of isoflavonoids.
generated by further biosynthetic elaborations. This biogenetic
1,2-shift and the presence of the intriguing natural products 4–6
prompted us to devise a synthesis route for obtaining isoflavonoids.
In this communication, we describe a synthetic access to
isoflavonoids via a 1,2-shift and alkylation sequence within the
flavonoid scaffolds.
We took inspiration from our previous study (Scheme 2).5
Eqn (1) is a pinacol-type shift of a phenyl group by the activation
Fig. 1 Natural isoflavonoids.
Department of Chemistry, Tokyo Institute of Technology, O-okayama, Meguro,
Tokyo 152-8551, Japan. E-mail: kohmori@chem.titech.ac.jp,
ksuzuki@chem.titech.ac.jp
Scheme 2 Organoaluminum-mediated stereospecific 1,2-shift.
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5cc01572c
7012 | Chem. Commun., 2015, 51, 7012--7014
This journal is ©The Royal Society of Chemistry 2015

View Article Online
Communication
ChemComm
of a mesylate with Et3Al to effect a stereospecific 1,2-shift.5a
A
Table 1 Conversion of catechin mesylate 7 to various isoflavans
related process involves trapping of the intermediary oxonium
species by an organoaluminum ligand (eqn (2)).5b By analogy,
we asked ourselves whether such a 1,2-shift was viable within
a catechin framework, and the model study started with a
catechin-derived substrate, i.e. mesylate 7.6
Mesylate 7 was prepared from tetra-O-benzyl catechin7
(CH3SO2Cl, Et3N, CH2Cl2, 0 1C, 96% yield). Upon treatment
of 7 with Me3Al (2 equiv.) in CH2Cl2 at ꢀ78 1C followed by
gradually warming to 0 1C, a single new product 8a was
produced in 90% yield (eqn (3)). The trans stereochemistry of
the aryl and the methyl groups was assigned using the coupling
constant (J2,3 = 9.2 Hz) and the ROESY experiment.8 The
stereochemical outcome could be attributed to the inversive
1,2-shift followed by trapping by a methyl nucleophile from the
opposite side. Furthermore, the enantiomeric purity of 8a
(499% e.e.) was verified by HPLC analysis using a chiral
stationary phase.9
Run
Reagent
R
Product
Yield [%]
1
2
3
4
5
Et3Al
AlH3
i-Bu3Al
Al(CH2SiMe3)3
PhAla
Et
H
8b
8c
8d
8e
8f
86
73
74
83
95
i-Bu
CH2SiMe3
Ph
a
6
7
8
8g
8h
8i
54
76
96
b
EtAl(CRCPh)2
EtAl(CRCSiMe3)2
CRCPh
CRCSiMe3
b
a
b
Prepared from the corresponding organolithium and AlCl3. Prepared
from the corresponding alkynyllithium and EtAlCl2.
As a variation of the migrating group, an ortho-substituted
phenyl group was tested (Scheme 3). Mesylate 17 was prepared
via our method for the flavan synthesis.13 The Mitsunobu
reaction of epoxy alcohol 118 and iodophenol 12 gave ether
13 as a single product, and the subsequent cyclization gave
flavan 16. After the removal of the TES group in 16, mesylation
of the resulting alcohol gave mesylate 17. Treatment of 17 with
Me3Al (ꢀ78 - ꢀ10 1C) induced a smooth 1,2-shift, giving
isoflavan 18 in 84% yield as a single product.8
Previously, we noted that o,o0-disubstituted phenyl groups
are often sluggish to undergo the 1,2-shift, particularly when
the substrate has steric hindrance.5c To address this point, we
prepared mesylate 19 with an o,o0-dibenzyloxyphenyl group in a
similar manner. To our delight, the reaction of 198 with Me3Al
(ꢀ78 - ꢀ30 1C) gave 79% yield of the rearranged product 20
(3)
A control experiment showed the well-known importance of the
anti-relationship of the leaving group and the migrating group
(eqn (4)). Epicatechin derivative 9, upon reaction with Me3Al,
led only to a slow 1,2-shift of the hydride to give 10 in 21% yield,
and mesylate 9 was largely recovered. The ee of 10 was 0%, not
surprisingly, suggesting that the reactive species that under-
went trapping by a methyl nucleophile was the oxonium species
after the hydride shift was fully completed.
(4)
Table 1 shows the generality of the process, giving various
2-substituted isoflavans. Reaction of 7 with Et3Al proceeded
smoothly to give the ethylated product 8b in excellent yield (run 1).
AlH3, in situ prepared from LiAlH4 and AlCl3,10 induced the 1,2-shift
of 7 followed by hydride trapping, giving isoflavan 8c in 73% yield
(run 2). Reaction of 7 with i-Bu3Al gave the expected product 8d
with an i-butyl group. A small amount of 8c was obtained, arising
from the b-hydride delivery from i-Bu3Al (run 3). Furthermore,
reactions of other organoaluminum reagents gave various iso-
flavans 8e–8i in moderate to high yields and rigorous trans
selectivities. In runs 4–6, triorganoaluminum reagents were
generated in situ from the respective organolithium and AlCl3.11
In runs 7 and 8, EtAlCl2 was used for this purpose, where the
alkynyl ligands were exclusively transferred.12
Scheme 3 Keys: (a) TMAD, n-Bu3P, toluene, 0 1C, 2 h (83%, single
diastereomer), (b) Li2NiBr4, THF, 0 1C - RT, 80 h (91%), (c) TESOTf,
2,6-lutidine, CH2Cl2, 0 1C, 20 min (95%), (d) PhMgBr, PhLi, HMPA, THF,
ꢀ78 - 0 1C, 30 min (62%), (e) n-Bu4NF, THF, 0 1C, 20 min (quant.),
(f) MsCl, Et3N, CH2Cl2, 0 1C, 40 min (94%), (g) Me3Al, CH2Cl2, ꢀ78 -
ꢀ10 1C, 1.5 h (84%). TMAD = N,N,N0,N0-tetramethylazodicarboxamide.
HMPA = hexamethylphosphoric triamide.
This journal is ©The Royal Society of Chemistry 2015
Chem. Commun., 2015, 51, 7012--7014 | 7013

View Article Online
ChemComm
Communication
(eqn (5)). This is a promising result in view of the synthesis data of the synthetic sample of 4 (1H and 13C NMR, IR, [a]D)
of many natural isoflavonoids with an aryl group possessing coincided with the reported data.16
ortho-hydroxy group(s).
In conclusion, an approach to the stereoselective synthesis
of isoflavans has been established based on the 1,2-shift of aryl
groups in flavan-3-ol derivatives and in situ alkylation by
organoaluminum reagents. The method was applied in the
synthesis of (ꢀ)-equol (4).
This work was supported by a Grant-in-Aid for Specially
Promoted Research (No. 23000006) from JSPS.
(5)
Notes and references
1 D. A. Smith and S. W. Banks, Phytochemistry, 1986, 25, 979.
2 G. G. J. M. Kuiper, J. G. Lemmen, B. Carlsson, J. C. Corton, S. H. Safe,
P. T. Saag, B. Burg and J.-A. Gustafsson, Endocrinology, 1998,
139, 4252.
Finally the enantiospecific synthesis of (ꢀ)-equol (4), a soy-derived
isoflavonoid known since 1932 is described.14 Recently, sizable
phytoestrogenic activity has been found in 4,15 making it the
current target of chemical synthesis.16
´
3 (a) A. Levai, J. Heterocycl. Chem., 2004, 41, 449; (b) A. Goel, A. Kumar and
A. Raghuvanshi, Chem. Rev., 2013, 113, 1614; (c) M. A. Selepe and F. R. V.
Heerden, Molecules, 2013, 18, 4739; (d) R. L. Farmer and K. A. Scheidt,
Chem. Sci., 2013, 4, 3304; (e) Z.-G. Feng, W.-J. Bai and T. R. R. Pettus,
Angew. Chem., Int. Ed., 2015, 54, 1864, and references therein.
4 M. F. Hashim, T. Hakamatsuka, Y. Ebizuka and U. Sankawa, FEBS Lett.,
1990, 271, 219.
5 (a) K. Suzuki, E. Katayama and G. Tsuchihashi, Tetrahedron Lett.,
1983, 24, 4997; (b) Y. Honda, E. Morita and G. Tsuchihashi, Chem.
Lett., 1985, 1153; (c) T. Saito, T. Suzuki, M. Morimoto, C. Akiyama,
T. Ochiai, K. Takeuchi, T. Matsumoto and K. Suzuki, J. Am. Chem.
Soc., 1998, 120, 11633.
6 For early examples on the use of the 1,2-shift to construct 3-aryl
isoflavones, see (a) A. C. Jain, P. D. Sarpal and T. R. Seshadri, Indian
J. Chem., 1965, 3, 369; (b) W. D. Ollis, K. L. Ormand and I. O. Sutherland,
Chem. Commun., 1968, 1237; (c) O. Prakash, S. Pahuja, S. Goyal, S. N.
Sawhney and M. Moriarty, Synlett, 1990, 337; (d) O. V. Singh,
C. P. Garg and R. P. Kapoor, Tetrahedron Lett., 1990, 31, 2747;
(e) T. Kinoshita, K. Ichinose and U. Sankawa, Tetrahedron Lett.,
1990, 31, 7355; ( f ) C. D. Gabbutt, J. D. Hepworth, B. M. Heron and
J.-L. Thomas, Tetrahedron Lett., 1998, 39, 881.
Scheme 4 outlines the synthesis of 4. The key intermediate
27 was prepared from the resorcinol derivative 21. The union
of 21 with epoxide 22 (499% e.e.)7 via the Mitsunobu
reaction gave ether 23 in 78% yield as an inseparable mixture
of diastereomers (93 : 7 ratio),13b which was used for the next
step. Regioselective cleavage of oxirane 23 gave bromohydrin 24
(87% yield) and its epimer 240 (4% yield), which were separated
using flash column chromatography (hexane/toluene/EtOAc =
5/5/1). After protection of 24 as a TES ether, the cyclization
precursor 25, thus obtained, was treated with Ph3MgLi to give
flavan 26 in 88% yield.13 After two-step conversion of 26 into
mesylate 27, treatment with AlH3 (CH2Cl2, 0 1C - room temp.,
2.5 h) cleanly afforded the desired isoflavan 28 in 85% yield.
Finally, two benzyl groups were removed by hydrogenolysis [H2,
Pd(OH)2/C, THF, MeOH, H2O (2/2/1), room temp., 45 min],
giving (ꢀ)-equol (4) as a white solid (99% e.e.).17 All the physical
7 H. Kawamoto, F. Nakatsubo and K. Murakami, Synth. Commun.,
1996, 26, 531.
8 See the ESI†.
9 An authentic sample of ent-7 was prepared by the C2-epimerization
of (ꢀ)-epicatechin (EC) (pH 8 phosphate buffer, 80 1C, 2 h), giving a
mixture of (ꢀ)-catechin and (ꢀ)-EC. Peracetylation followed by the
Kawamoto protocol, separation and mesylation gave ent-7. See the
ESI†.
10 (a) A. E. Finholt, A. C. Bond and H. I. Schlesinger, J. Am. Chem. Soc.,
1947, 69, 1199; (b) S. Raghavan, S. R. Reddy, K. A. Tony, C. N. Kumar,
A. K. Varma and A. Nangia, J. Org. Chem., 2002, 67, 5838.
11 In run 6, the isoflavene (3-aryl-4H-chromene) derivative was
obtained as a byproduct (16% yield). See the ESI†.
12 (a) J. Fried, C.-H. Lin and S. H. Ford, Tetrahedron Lett., 1969, 1379;
(b) R. T. Hansen, D. B. Carr and J. Schwartz, J. Am. Chem. Soc., 1978,
100, 2244; (c) K. Taya, T. Nagasawa and K. Suzuki, Synlett, 1997, 304.
13 (a) T. Higuchi, K. Ohmori and K. Suzuki, Chem. Lett., 2006, 35, 1006;
(b) K. Ohmori, M. Takeda, T. Higuchi, T. Shono and K. Suzuki,
Chem. Lett., 2009, 38, 934.
14 G. F. Marrian and G. A. D. Haslewood, Biochem. J., 1932, 1227.
15 (a) J.-P. Yuan, J.-H. Wang and X. Liu, Mol. Nutr. Food Res., 2007, 51, 765;
(b) P. J. Magee, Proc. Nutr. Soc., 2011, 70, 10; (c) R. L. Jackson, J. S. Greiwe
and R. J. Schwen, Nutr. Rev., 2011, 69, 432; (d) D. Shor, T. Sathyapalan,
S. L. Atkin and N. J. Thaycher, Eur. J. Nutr., 2012, 51, 389.
16 (a) J. M. Heemstra, S. A. Kerrigan, D. R. Doerge, W. G. Helferich and
W. A. Boulanger, Org. Lett., 2006, 8, 5441; (b) Y. Takashima and
Y. Kobayashi, Tetrahedron Lett., 2008, 49, 5156; (c) Y. Takashima,
Y. Kaneko and Y. Kobayashi, Tetrahedron, 2010, 66, 197; (d) J.-W. Lee
and B. List, J. Am. Chem. Soc., 2012, 134, 1824; (e) S. Yang, S.-F. Zhu,
C.-M. Zhang, S. Song, Y.-B. Yu, S. Li and Q.-L. Zhou, Tetrahedron,
2012, 68, 5172.
Scheme 4 Keys: (a) TMAD, n-Bu3P, toluene, 0 1C, 1 h (78%, dr = 93/7),
(b) Li2NiBr4, THF, 0 1C, 24 h (87%), (c) TESOTf, 2,6-lutidine, CH2Cl2, 0 1C,
20 min (97%), (d) PhMgBr, PhLi, HMPA, THF, ꢀ78 - 0 1C, 45 min (88%),
(e) n-Bu4NF, THF, 0 1C - RT, 15 min (95%), (f) MsCl, Et3N, CH2Cl2, 0 1C,
10 min (99%), (g) LiAlH4, AlCl3, CH2Cl2, 0 1C - RT, 2.5 h (85%), (h) H2,
ASCA-2 [5% Pd(OH)2/C], THF, MeOH, H2O, RT, 45 min (quant.).
17 An authentic sample of (ꢁ)-4 was purchased from Tokyo Chemical
Industry Co., Ltd. (TCI) and the enantiomeric purity of synthetic 4
was assessed by HPLC analysis. See the ESI†.
7014 | Chem. Commun., 2015, 51, 7012--7014
This journal is ©The Royal Society of Chemistry 2015