Total synthesis of resveratrol-based natural products using a palladium-catalyzed decarboxylative arylation and an oxidative heck reaction

Article abstract of DOI:10.1002/anie.201310676

Controlled access to resveratrol-based natural products is offered by a novel, modular concept. A common building block readily available on a large scale serves as the starting material for the introduction of structurally important aryl groups by a Pd-c

Full text of DOI:10.1002/anie.201310676

Angewandte
Chemie
DOI: 10.1002/anie.201310676
Natural Product Synthesis
Hot Paper
Total Synthesis of Resveratrol-Based Natural Products Using
a Palladium-Catalyzed Decarboxylative Arylation and an Oxidative
Heck Reaction**
Felix Klotter and Armido Studer*
Abstract: Controlled access to resveratrol-based natural
products is offered by a novel, modular concept. A common
building block readily available on a large scale serves as the
starting material for the introduction of structurally important
aryl groups by a Pd-catalyzed decarboxylative arylation and
an oxidative Heck reaction with good yields and high
stereoselectivity. The modular approach is convincingly docu-
mented by the successful synthesis of three racemic resveratrol-
based natural products (quadrangularin A, ampelopsin D, and
pallidol).
R
esveratrol (1) and its derivatives (e.g. 2–6) belong to the
class of natural polyphenols (Figure 1). They are phytoalex-
ines which are produced by various plant species (e.g. grapes)
in different amounts in response to external stimuli (e.g.
fungal and viral infections).[1] In particular 1 has received
great attention due its high in vitro and in vivo activity against
a considerable number of diseases.[2] However, the biological
activity of resveratrol-based oligomers such as 2–6 has so far
been little explored. First investigations indicate that the
biological activities of these natural products are similar to or
even higher than that of 1 and they should therefore find
applications as valuable drugs.[3]
Only a few synthetic methods have been described for the
preparation of these polyphenolic natural products.[4] First
strategies used a biomimetic pathway resulting in a mixture of
natural and nonnatural derivatives.[5] In 2007 Snyder et al.
published an efficient approach for the synthesis of resvera-
trol oligomers, like pallidol (4; 12 steps; 7%) and ampelop-
sin F (6; 12 steps; 6%).[6] Through the use of readily
accessible building blocks other members of the family
could be accessed in succeeding work.[7] Alternative synthetic
strategies for the preparation of resveratrol-based natural
products were published by the research groups led by Sun,[8]
Hou,[9] Sarpong,[10] and Flynn.[11] Since these strategies are
mostly restricted to the preparation of a single derivative, they
are not suitable for synthesis of analogues.
Figure 1. Resveratrol (1) and some of its derivatives (2–6).
products exemplified by the syntheses of pallidol (4), quad-
rangularin A (2), and ampelopsin D (3). The retrosynthetic
analysis is illustrated in Scheme 1. Resveratrol oligomers such
as pallidol (4) and ampelopsin F (6) are structurally derived
from resveratrol dimers containing an indane core. Quad-
rangularin A (2) and ampelopsin D (3) differ only in the
substitution pattern of the trans-configured aryl groups at
positions C2 and C3. This connectivity similarity motivated us
to develop a modular approach in which the structurally
important aryl groups are introduced at a late stage of the
total synthesis to gain maximum flexibility. As one of the key
steps we chose a novel palladium-catalyzed decarboxylative
arylation,[12] which will make it possible to start with the
indene carboxylic acid 7 for the introduction of the aryl group
at position C3. The so obtained indene derivatives A will then
be transformed in the second key step to compounds of type B
with an E-configured exocyclic double bond by using our
recently developed oxidative Heck coupling process.[13]
Deprotection will provide compounds 2 and 3 directly.
After oxidation of the double bond in B, subsequent
Herein we disclose a novel highly variable approach for
the total synthesis of racemic resveratrol-based natural
[*] F. Klotter, Prof. Dr. A. Studer
Organisch-Chemisches Institut
Westfꢀlische Wilhelms-Universitꢀt Mꢁnster
Correnstrasse 40, 48149 Mꢁnster (Germany)
E-mail: studer@uni-muenster.de
[**] We thank the Fonds der Chemischen Industrie for financial support
(stipend to F.K.).
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201310676.
Angew. Chem. Int. Ed. 2014, 53, 2473 –2476
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2473

.
Angewandte
Communications
Table 1: Model studies on the decarboxylative arylation of 8/10.
Entry Indene (equiv) Equiv
of tol-I
Base
Conc. of
8 or 10 [m]
Yield [%][a]
1
2
3
4
8 (1)
8 (1)
8 (1)
8 (1)
1.1
1.1
1.1
1.1
1.1
1.1
1
1
1
1
1
Cs2CO3
Ag2CO3
LiOtBu
NaOtBu
NaOtBu
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.3
0.1
0
10
0
25
8[c]
31
57
42[c]
52[c]
54
66
5[b]
6
8 (1)
10 (1)
10 (2)
10 (1.5)
10 (1.5)
8 (1.5)[d]
8 (1.5)[d]
7
8
9
Scheme 1. Retrosynthesis of quadrangularin A (2), ampelopsin D (3),
and pallidol (4).
10
NaOtBu[e] 0.075
NaOtBu[e] 0.075
11[f]
[a] Yield of isolated product 9. [b] With P(OPh)3 (20 mol%). [c] Yield
determined by H NMR analysis with CH2Br2 as the internal standard.
[d] Prior to addition of tol-I and Pd the acid was deprotonated with
1
Friedel–Crafts alkylation and deprotection should provide
pallidol (4).
NaOtBu. [e] With 1.6 equiv base. [f] With 20 mol% Pd.
During the past years decarboxylative reactions have
emerged as highly valuable methods for selective formation
of C C and C–heteroatom bonds.[14] In particular in the field
À
of transition-metal catalysis various variants of this approach
have been described including decarboxylative Heck reac-
tions,[15] homocouplings,[16] arylations,[17] vinylations,[18] allyla-
tions,[19] and decarboxylative C-H arylations.[20]
In view of the total synthesis of resveratrol-based natural
products we first investigated the applicability of the highly
stereospecific decarboxylative coupling of 2,5-cyclohexa-
diene-1-carboxylic acids with aryl iodides, which we described
recently,[12] to benzyl-1H-indene-1-carboxylic acid (8). Opti-
mization studies were conducted with 4-iodotoluene as the
coupling partner and [Pd(dba)2] as the catalyst in toluene
(Table 1). In initial studies we varied the base. With Cs2CO3,
Ag2CO3, and LiOtBu the arylated product 9 was either not
formed or obtained in low yield (entries 1–3). NaOtBu led to
a slightly improved result (25%, entry 4). In analogy to our
previous studies on the decarboxylative coupling of 2,5-
cyclohexadiene-1-carboxylic acids we noted that the addition
of ligands leads to lower yield (8%, entry 5). Since proto-
decarboxylation was the major side reaction, we also tested
the Na salt 10 as a substrate and obtained 9 in 31% yield
(entry 6). We found that the concentration and also the excess
amount of 10 are important factors (entries 7–9). When we
used an excess of 10 and also lowered the concentration
(0.1m), 9 was obtained in 52% yield (entry 9). Upon further
dilution and in situ generation of 10, the yield was increased to
54% (entry 10). Finally, an increase of the [Pd(dba)2] loading
provided 9 in 66% yield.
Scheme 2. Reagents and conditions: a) H2SO4 (cat.), MeOH, 708C,
12 h; b) 4-MeOC6H4CH2MgCl (2.0 equiv), CuI (2.0 equiv), TMEDA
(2.2 equiv), TMSCl (5.0 equiv), Et2O/THF, À788C/À208C, 3 h;
c) NaOH (10%), MeOH, 708C, 2 h; d) MSA (10 equiv), 708C, 3 h;
e) LDA (2.2 equiv), ClCO2Me (1.3 equiv), À788C/RT, 5 h; f) NaBH4
(1.5 equiv), CeCl3 (1.1 equiv), MeOH, 08C, 0.5 h; g) MsCl (2 equiv),
NEt3 (3 equiv), CH2Cl2, 08C, 17 h; h) NaOH (30%), EtOH/H2O (5:1),
58C, 16 h. MSA: methanesulfonic acid, LDA: lithium diisopropylamide,
MsCl: methanesulfonyl chloride.
After successful optimization we focused on the total
syntheses of quadrangularin A (2), ampelopsin D (3), and
pallidol (4). Our first priority was the synthesis of the key
building block, indene carboxylic acid 7 (Scheme 2). Com-
mercially available 3,5-dimethoxycinnamic acid (11) was
readily esterified (> 99%) and the 4-methoxybenzyl group
2474 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 2473 –2476

Angewandte
Chemie
was introduced in a Cu-mediated conjugate addition accord-
ing to a procedure described by Ferreira et al.[21] Saponifica-
tion provided carboxylic acid 12 in very good yield (86% over
three steps). The indane core in 13 was built up through an
intramolecular Friedel–Crafts acylation of 12 using MSA in
good yield (89%).
natural products 2 and 3, it was essential to introduce the C2
substituent with high trans-selectivity. Moreover, the exocy-
clic double bond of 17a and 17b had to be formed
stereoselectively in E configuration. We were pleased to
find that indene 16a was readily transformed with 3,5-
dimethoxyphenylboronic acid and TEMPO as an external
oxidant into the protected quadrangularin A 17a in very good
yield and with excellent selectivity (trans- and E-selective).
The regioselectivity of the addition to 16a is controlled by
steric and electronic effects. The E-selectivity is a result of
minimizing allylic A1,3 strain during the b-H-elimination
reaction, and the trans-selectivity is determined by the C3
aryl group. Interestingly, the same reaction with O2 as the
oxidant did not lead to product formation.
For the construction of the ampelopsin D structure 4-
methoxyphenylboronic acid was used as the aryl source.
Under analogous conditions with TEMPO as oxidant product
17b was obtained as a mixture containing an inseparable
compound. However, using the bulky TEMPO derivative 18
formation of the unwanted side product was largely sup-
pressed and indane 17b was obtained from 16b in 73% yield
with excellent trans and E-selectivity.[23] Lewis acid mediated
demethylation of all phenolic methyl ethers in 17a and 17b
eventually provided the natural products quadrangularin A
(2, 45% yield starting from 7) and ampelopsin D (3, 33%
yield starting from 7), respectively.[24]
The carboxy functionality was introduced along with
a quaternary center (14) through the formation of the dianion
of 13 by treatment with LDA and subsequent addition of
methyl chloroformate. The crude product was used for the
next step without any further purification.[22] The keto
function in indanone 14 was readily reduced under Luche
conditions. Attempts to dehydrate the OH function under
acidic conditions (pTSA, silica gel, AcOH, HCl, Cu(OTf)2, or
Amberlyst 15) led to oligomerization. However, we found
that when we used an excess of MsCl and NEt3, efficient and
clean H2O elimination could be achieved to give indene ester
15 (63% over 3 steps). Due to the generally high tendency of
indene carboxylates to undergo decarboxylation, subsequent
saponification was conducted at 58C (90% yield). The
developed eight-step synthesis allowed the preparation of
the key building block 7 in a 43% overall yield in gram scale.
After having established an efficient access to indene
carboxylic acid 7, we turned to the planned key steps
(Scheme 3). Pleasingly, the palladium-catalyzed decarboxyla-
tive coupling of 7 with aryl iodides proved to be a highly
efficient method and the 1,3-disubstituted indene derivatives
16a and 16b were obtained in good yields under the
optimized conditions. In contrast to the model study the
excess of carboxylic acid in these reactions could be reduced
to 1.1 and 1.3 equiv, respectively.
To complete the synthesis of the hexacyclic natural
product pallidol (4) starting with indane 17a we chose
a known two-step reaction sequence comprising a hydrobora-
tion and a Friedel–Crafts alkylation with concomitant phenol
deprotection (Scheme 4).[8] This sequence led to the biolog-
ically active compound 4 via alcohol 19 (method A).[25]
During the studies on the hydroboration of 17a we noted
formation of trace amounts of a protected pallidol (20) during
We next investigated the applicability of the nitroxide-
mediated oxidative Heck reaction[13] for the introduction of
the C2 substituent (see Scheme 1). Due to the relative
configuration of the aryl groups at positions 2 and 3 in the
Scheme 3. Key steps in the modular approach to the natural products
starting with rac-7 (only one enantiomer drawn). Reagents and con-
ditions: a) 7 (1.1 equiv for 16a, 1.3 equiv for 16b), [Pd(dba)2]
(20 mol%), NaOtBu (1.1 and 1.3 equiv, respectively), ArI (1.0 equiv),
toluene (0.075m), 1108C, 18 h; b) Pd(OAc)2 (5 mol%), ArB(OH)2
(4 equiv), TEMPO (4 equiv) for 17a (TEMPO derivative 18 (4 equiv) for
17b), KF (4 equiv), propionic acid, RT, 18 h for 17a, 13 h for 17b: 89%
17a, 73% 17b;[23] c) BBr3 (12 equiv), CH2Cl2, 08C/RT, 6 h: 77% 2, 80%
3.[24]
Scheme 4. Total synthesis of pallidol (4). Reagents und conditions:
a) BH3·THF (10 equiv), THF, RT, 18 h, NaOH (0.25m)/H2O2 (30%)
[3:1], 1 h, basic workup; b) BBr3 (30 equiv), CH2Cl2, 08C/RT, 8 h, 52%
over 2 steps (HPLC purification); c) BH3·THF (10 equiv), THF, RT,
18 h, NaOH (0.25m)/H2O2 (30%) (3:2), 3 h, aqueous workup, 62%;
d) BBr3 (30 equiv), H2O (1 equiv), CH2Cl2, 08C/RT, 8 h, 75% (FC
purification).
Angew. Chem. Int. Ed. 2014, 53, 2473 –2476
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2475

.
Angewandte
Communications
d) Y. Takaya, K. Terashima, J. Ito, Y.-H. He, M. Tateoka, N.
Yamaguchi, M. Niwa, Tetrahedron 2005, 61, 10285; e) S. S. Velu,
I. Buniyamin, L. K. Ching, F. Feroz, I. Noorbatcha, L. C. Gee, K.
Awang, I. A. Wahab, J.-F. F. Weber, Chem. Eur. J. 2008, 14,
11376.
aqueous workup when a slight excess of H2O2 was used. After
further experimentation we were able to isolate 20 in 62%
yield when we used an excess of H2O2 (method B). We believe
that with excess H2O2 an acidic boron compound is generated
which induces a spontaneous and clean cyclization during
workup. Phenol deprotection from 20 finally afforded the
targeted natural product pallidol (4).
In conclusion we have described a novel modular
approach for the synthesis of racemic resveratrol-based
natural products. The strategy involves a novel type of
decarboxylative coupling followed by a nitroxide-mediated
oxidative Heck reaction. Using this approach structurally
relevant aryl groups could be variably introduced. The
concept was first applied to the synthesis of the resveratrol-
dimers quadrangularin A (2) and ampelopsin D (3). An
optimized reaction sequence afforded pallidol (4) in 12
steps and 13% overall yield (Ref.: 7% over 12 steps;[6] 5%
over 8 steps).[8] Importantly, our strategy can be used to
prepare other natural and nonnatural derivatives which
should be interesting for biological studies.
[6] S. A. Snyder, A. L. Zografos, Y. Lin, Angew. Chem. 2007, 119,
8334; Angew. Chem. Int. Ed. 2007, 46, 8186.
[7] a) S. A. Snyder, S. P. Breazzano, A. G. Ross, Y. Lin, A. L.
Zografos, J. Am. Chem. Soc. 2009, 131, 1753; b) S. A. Snyder,
Z. G. Brill, Org. Lett. 2011, 13, 5524; c) S. A. Snyder, A. Gollner,
M. I. Chiriac, Nature 2011, 474, 461.
[8] a) C. Zhong, X.-H. Liu, X.-D. Hao, J. Chang, X. Sun, Eur. J. Med.
Chem. 2013, 62, 187; b) C. Zhong, J. Zhu, J. Chang, X. Sun,
Tetrahedron Lett. 2011, 52, 2815.
[9] W. Li, H. Li, Y. Li, Z. Hou, Angew. Chem. 2006, 118, 7771;
Angew. Chem. Int. Ed. 2006, 45, 7609.
[10] a) J. L. Jeffrey, R. Sarpong, Org. Lett. 2009, 11, 5450; b) J. L.
Jeffrey, R. Sarpong, Tetrahedron Lett. 2009, 50, 1969.
[11] D. J. Kerr, M. Miletic, N. Manchala, J. M. White, B. L. Flynn,
Org. Lett. 2013, 15, 4118.
[12] a) C.-M. Chou, I. Chatterjee, A. Studer, Angew. Chem. 2011,
123, 8773; Angew. Chem. Int. Ed. 2011, 50, 8614; b) E. Koch, A.
Studer, Angew. Chem. 2013, 125, 5033; Angew. Chem. Int. Ed.
2013, 52, 4933.
[13] Z. He, S. Kirchberg, R. Frçhlich, A. Studer, Angew. Chem. 2012,
124, 3759; Angew. Chem. Int. Ed. 2012, 51, 3699. Review on
TEMPO-mediated processes: L. Tebben, A. Studer, Angew.
Chem. 2011, 123, 5138; Angew. Chem. Int. Ed. 2011, 50, 5034.
[14] a) L. J. Gooßen, N. Rodrꢂguez, K. Gooßen, Angew. Chem. 2008,
120, 3144; Angew. Chem. Int. Ed. 2008, 47, 3100; b) B. Song, T.
Knauber, L. J. Gooßen, Angew. Chem. 2013, 125, 3026; Angew.
Chem. Int. Ed. 2013, 52, 2954.
[15] Example: A. G. Myers, D. Tanaka, M. R. Mannion, J. Am.
Chem. Soc. 2002, 124, 11250.
[16] Example: J. Cornella, H. Lahlali, I. Larrosa, Chem. Commun.
2010, 46, 8276.
[17] Example: X. Li, D. Zou, F. Leng, C. Sun, J. Li, Y. Wu, Y. Wu,
Chem. Commun. 2013, 49, 312.
[18] Example: Z. Xu, Q. Wang, J. Zhu, Angew. Chem. 2013, 125,
3354; Angew. Chem. Int. Ed. 2013, 52, 3272.
[19] Example: C. J. Gartshore, D. W. Lupton, Angew. Chem. 2013,
125, 4207; Angew. Chem. Int. Ed. 2013, 52, 4113.
[20] Example: P. Hu, M. Zhang, X. Jie, W. Su, Angew. Chem. 2012,
124, 231; Angew. Chem. Int. Ed. 2012, 51, 227.
[21] P. S. van Heerden, B. C. B. Bezuidenhoudt, D. Ferreira, J. Chem.
Soc. Perkin Trans. 1 1997, 1141.
[22] After purification by colum chromatography, 14 was obtained
along with inseparable starting material 13 (1H NMR: 14/13 =
11:1; see the Supporting Information).
[23] An inseparable side product was formed (overall yield: 77%;
1H NMR: 17b/side product = 17:1). We believe that this side
product is an isomer containing an internal double bond (see the
Supporting Information).
Received: December 9, 2013
Published online: February 5, 2014
Keywords: decarboxylative coupling · natural products ·
.
palladium · resveratrol · total synthesis
[1] a) S. A. Adesanya, R. Nia, M.-T. Martin, N. Boukamcha, A.
Montagnac, M. Paꢀs, J. Nat. Prod. 1999, 62, 1694; b) A.
Gonzꢁlez-Sarrꢂas, S. Gromek, D. Niesen, N. P. Seeram, G. E.
Henry, J. Agric. Food Chem. 2011, 59, 8632; c) M. A. Khan, S. G.
Nabi, S. Prakash, A. Zaman, Phytochemistry 1986, 25, 1945; d) S.
Macke, G. Jerz, M. T. Empl, P. Steinberg, P. Winterhalter, J.
Agric. Food Chem. 2012, 60, 11919.
[2] a) J. A. Baur, K. J. Pearson, N. L. Price, H. A. Jamieson, C.
Lerin, A. Kalra, V. V. Prabhu, J. S. Allard, G. Lopez-Lluch, K.
Lewis, P. J. Pistell, S. Poosala, K. G. Becker, O. Boss, D. Gwinn,
M. Wang, S. Ramaswamy, K. W. Fishbein, R. G. Spencer, E. G.
Lakatta, D. Le Couteur, R. J. Shaw, P. Navas, P. Puigserver, D. K.
Ingram, R. de Cabo, D. A. Sinclair, Nature 2006, 444, 337; b) S.
Bradamante, L. Barenghi, A. Villa, Cardiovasc. Drug Rev. 2004,
22, 169; c) M. Jang, L. Cai, G. O. Udeani, K. V. Slowing, C. F.
Thomas, C. W. W. Beecher, H. H. S. Fong, N. R. Farnsworth,
A. D. Kinghorn, R. G. Mehta, R. C. Moon, J. M. Pezzuto,
Science 1997, 275, 218; d) S. Quideau, D. Deffieux, L. Pouysꢃgu,
Angew. Chem. 2012, 124, 6930; Angew. Chem. Int. Ed. 2012, 51,
6824; e) J. M. Smoliga, J. A. Baur, H. A. Hausenblas, Mol. Nutr.
Food Res. 2011, 55, 1129.
[3] a) N. Abe, T. Ito, K. Ohguchi, M. Nasu, Y. Masuda, M. Oyama,
Y. Nozawa, M. Ito, M. Iinuma, J. Nat. Prod. 2010, 73, 1499;
b) R. H. Cichewicz, S. A. Kouzi, Studies in Natural Products
Chemistry, Vol. 26, Part G (Ed.: R. Atta-ur), Elsevier, Amster-
dam, 2002, pp. 507; c) S. He, Y. Lu, B. Wu, Y. Pan, J. Chromatogr.
A 2007, 1151, 175; d) L. Li, G. E. Henry, N. P. Seeram, J. Agric.
Food Chem. 2009, 57, 7282.
[4] S. A. Snyder in Biomimetic Organic Synthesis (Eds.: E. Poupon,
B. Nay), Wiley-VCH, Weinheim, 2011, pp. 695.
[5] a) T. A. Engler, G. A. Gfesser, B. W. Draney, J. Org. Chem. 1995,
60, 3700; b) C. Li, J. Lu, X. Xu, R. Hu, Y. Pan, Green Chem. 2012,
14, 3281; c) X.-C. Li, D. Ferreira, Tetrahedron 2003, 59, 1501;
[24] Along with the product an isomer containing an internal double
bond was formed in small amounts (see the Supporting
Information). Yield: 84% (1H NMR: 2/isomer= 11:1); 90%
(1H NMR: 3/isomer= 8:1). Final purification was conducted by
preparative HPLC.
[25] An additional isomer was formed in 10% yield. It is an isomer of
quadrangularin A (2) containing an internal double bond, which
is likely formed in the reaction of 19 with BBr3 through an
elimination and rearrangement sequence (see the Supporting
Information). Surprisingly, alcohol 19 was formed as diastereo-
meric mixture (in Scheme 4 only the major product is depicted).
2476 www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 2473 –2476