Enantioselective total synthesis of (+)-brazilin, (-)-brazilein and (+)-brazilide A

Article abstract of DOI:10.1039/c3cc42385a

An enantioselective strategy for the synthesis of (+)-brazilin, (-)-brazilein and (+)-brazilide A has been developed. A Lewis acid mediated lactonization established the novel fused bis-lactone core of brazilide A and finalized the first total synthesis of (+)-brazilide A.

Full text of DOI:10.1039/c3cc42385a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Enantioselective total synthesis of (+)-brazilin,
(À)-brazilein and (+)-brazilide A†
Cite this: Chem. Commun., 2013,
49, 5405
Xuequan Wang, Hongbin Zhang,* Xiaodong Yang,* Jingfeng Zhao and
Chengxue Pan
Received 2nd April 2013,
Accepted 23rd April 2013
DOI: 10.1039/c3cc42385a
www.rsc.org/chemcomm
An enantioselective strategy for the synthesis of (+)-brazilin,
(À)-brazilein and (+)-brazilide A has been developed. A Lewis acid
mediated lactonization established the novel fused bis-lactone core
of brazilide A and finalized the first total synthesis of (+)-brazilide A.
Brazilin (1), brazilein (2), brazilide A (3) and haematoxylin (4) were
isolated from the heartwood of Caesalpinia sappan, a famous
traditional Chinese medicine used for the treatment of emmenio-
pathy, convulsion, straumatic disease and menstrual disorders.1
Recent reports have demonstrated that brazilin (1) possesses
antitumor activity and acts as a micromolar telomerase inhibitor
and produces DNA nicks.2 Haematoxylin (4) was recently proven
to be a potent inhibitor of protein tyrosine kinase.3 These natural
products belong to homoisoflavonoids and are structurally
related. Brazilein has been proven to be the oxidized form of
brazilin,4a while brazilide A might be the A-ring further oxidative
product of brazilein (Scheme 1, dashed-line arrows).4b
The structures and the diverse biological activities of brazilin
and its related compounds (Fig. 1) have attracted considerable
attention towards their synthesis. There are two accomplished
total synthesis procedures of (Æ)-brazilin in the literature; these
involve the elegant total synthesis demonstrated by Pettus et al.5a
and the synthesis conducted by Lee et al.2c An elegant enantio-
selective synthesis of O-trimethylbrazilin was also reported by
Davis and Chen in 1993.5b To date, the structurally more
challenging and interesting brazilide A (3), with a novel fused
bis-lactone ring system, has remained untouched. As part of our
ongoing program in seeking more efficient and flexible synthetic
strategies towards synthesis of bioactive natural products and
their analogues,6 we recently initiated a program towards the
syntheses of natural products isolated from Caesalpinia sappan.
Scheme 1 Retrosynthetic analysis of (1), (2) and (3).
Key Laboratory of Medicinal Chemistry for Natural Resource, Ministry of Education,
School of Chemical Science and Technology, Yunnan University, Kunming, 650091,
P. R. China. E-mail: zhanghb@ynu.edu.cn, xdyang@ynu.edu.cn;
Fax: +86-871-65035538
Fig. 1 Natural products isolated from Caesalpinia sappan.
† Electronic supplementary information (ESI) available: Details of experimental
procedure, spectral data and copies of all new compounds. CCDC 923325 and
923326. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c3cc42385a
Herein we report the first asymmetric total synthesis of
(+)-brazilin, (À)-brazilein and (+)-brazilide A based on the same
intermediate (8) as outlined in Scheme 1.
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 5405--5407 5405

View Article Online
Communication
ChemComm
The ring system of brazilide A is quite unique, with a bis-
lactone ring system fused to a cyclopentane (the red color part in
compound 3). The synthetic challenge associated with brazilide
A is the formation of the fused bis-lactone ring system. To the
best of our knowledge, no methodology has been documented
previously in the literature for the synthesis of this bis-lactone
unit bearing two fully substituted carbon centers. Therefore,
developing a synthetic strategy for the synthesis of brazilide A (3)
might also provide an access to this fused bis-lactone ring
system. We anticipated that epoxidation of 5 followed by lacto-
nization might give brazilide A (3). Selective Birch reduction of 7
followed by oxidative cleavage of the electron rich double bond
in 6 would result in intermediate 5. Compound 7 could be
obtained by a Friedel–Crafts reaction of diol 8. Diol 8, with the
key structural unit for the construction of the tetracyclic core
for brazilin and its related natural products isolated from
Caesalpinia sappan, could be prepared from compounds 9 and 10.
We started our journey by synthesizing indene 10, which was
obtained in 4 steps in 61% yield by following our previous
procedure.6c After treatment of indene 10 with 3-hydroxyphenol
benzoate (9) under Mitsunobu conditions (Scheme 2),7 the
ether (15) was obtained in 60% yield. Next, a Sharpless asym-
metric dihydroxylation with AD-mix-b8 was conducted and diol
8 was obtained. We found that addition of phenol 9 (1.0 eq.) in
this reaction could improve the enantioselectivity and acceler-
ate the reaction rate.9 Next, the Friedel–Crafts reaction of diol 8
gave the desired annulation product (16) in 80% yield. Hydro-
lysis of benzoylate 16 afforded the key intermediate 7 in 92%
yield. After recrystallization, HPLC analysis established that the
isomeric purity of compound 7 was >99%.
Scheme 3 Synthesis of (+)-brazilin (1) and (À)-brazilein (2).
Treatment of compound 7 under Birch reduction conditions10
selectively reduced the aromatic ring bearing two methoxyl
groups and afforded the desired diene compound (17). The
unstable diene (17) was then subjected to ozone oxidation,
which unfortunately gave a complex mixture. After some experi-
mentation, the diene (17) was converted to a benzoate (18)
before oxidative cleavage of the electron rich double bond.
Finally the desired intermediate 19 was obtained in 50%
isolated yield over three consecutive steps (Scheme 3).
With the key intermediate 19 in hand, we came to the key
bis-lactonization to finalize the synthesis of brazilide A. Epoxi-
dation of 19 directed by the neighboring hydroxyl group
provided 20 in high yield (92%) as a single diastereomer. To
our disappointment, Lewis acid or proton acid mediated lactoni-
zation did not yield the natural product. Instead of brazilide A,
the lactonization afforded a bis-lactone (21) with wrong
stereochemistry at the lactone ring system (Scheme 4). After
debenzoylate,11 the C-10,11 epimer of brazilide A (22) was
obtained in 90% yield.
Next we came to the final stage for the synthesis of brazilin
and brazilein. Treatment of 7 with boron tribromide in dichloro-
methane furnished the enantioselective synthesis of (+)-brazilin
(Scheme 3, 1, 9 steps, 14% overall yield). Oxidation of 1 with
(diacetoxyiodo)benzene in THF provided (À)-brazilein (2) in 76%
yield. The NMR spectra of our synthetic samples were in
complete agreement with the reported data.1c,e
In principle, the Lewis acid mediated oxirane ring opening
from both sides (C10 and C11) should result in compound 22
After having accomplished the syntheses of brazilin and
brazilein, we decided to continue our journey towards brazilde A.
Scheme 2 Synthesis of O-dimethylbrazilin 7.
Scheme 4 Synthesis of 22, C-10,11-epimer of brazilide A.
c
5406 Chem. Commun., 2013, 49, 5405--5407
This journal is The Royal Society of Chemistry 2013

View Article Online
ChemComm
Communication
In summary, an enantioselective strategy for the synthesis of
(+)-brazilin, (À)-brazilein and (+)-brazilide A has been developed.
This new route, which features a stereocontrolled Friedel–Crafts
cyclization to establish the tetracyclic ring system of brazilin and
a Lewis acid mediated formation of the novel fused bis-lactone
core of brazilide A, leads to the first total synthesis of (+)-brazilide
A in 16 steps from commercially available starting materials. The
absolute configuration of (+)-brazilide A was also confirmed by
this total synthesis. Biological studies towards brazilide A and its
analogues are currently under investigation in our laboratory.
This work was supported by grants from Natural Science
Foundation of China (20925205), National Basic Research
Program of China (973 Program 2009CB522300) and Yunnan
Provincial Science & Technology Department (2010GA014).
Scheme 5 Possible pathway for bis-lactonization.
Notes and references
1 (a) R. Robinson, Bull. Soc. Chim. Fr., 1958, 125; (b) J. C. Craig,
A. R. Naik, R. Pratt and E. Johnson, J. Org. Chem., 1965, 30, 1573;
(c) C. F. J. Yamahara, T. Shimokawa, J. Kinjo, T. Tomimatsu and
T. Nohara, Phytochemistry, 1985, 24, 2403; (d) M. Namikd, H. Nakata,
H. Yamada, M. Nagai and T. Saitoh, Chem. Pharm. Bull., 1987,
35, 2761; (e) D. S. Kim, N. Baek, S. P. Oh, K. Y. Jung, I. S. Lee and
H.-K. Lee, Phytochemistry, 1997, 46, 177; ( f ) B. O. Yang, C.-Q. Ke,
Z.-S. He, Y.-P. Yang and Y. Ye, Tetrahedron Lett., 2002, 43, 1731;
(g) H.-X. Xu and S. F. Lee, Phytother. Res., 2004, 18, 647 and
references therein.
2 (a) R. L. Chin and A. C. Tolman, Telomerase inhibitors and methods of
their use, WO. Pat., 0,193,864, 2001; (b) W. Mar, H.-T. Lee, K.-H. Je,
H.-Y. Choi and E.-K. Seo, Arch. Pharmacal. Res., 2003, 26, 147;
(c) C.-T. Yen, K. Nakagawa-Goto, T.-L. Hwang, P.-C. Wu,
S.-L. Morris-Natschke, W.-C. Lai, K. F. Bastow, F.-R. Chang,
Y.-C. Wu and K.-H. Lee, Bioorg. Med. Chem. Lett., 2010, 20, 1037.
3 L.-G. Lin, H. Xie, H.-L. Li, L.-J. Tong, C.-P. Tang, C.-Q. Ke, Q.-F. Liu,
L.-P. Lin, M.-Y. Geng, H. Jiang, W.-M. Zhao, J. Ding and Y. Ye, J. Med.
Chem., 2008, 51, 4419.
Scheme 6 Enantioselective synthesis of (+)-brazilide A (3).
4 (a) H. Hikino, T. Taguchi, H. Fujimura and Y. Hiramatsu, Planta Med.,
1977, 31, 214; (b) R. A. Hill and M. C. Parker, Nat. Prod. Rep., 2002, 19, iii.
5 (a) Y. Huang, J. Zhang and T. R. R. Pettus, Org. Lett., 2005, 7, 5841;
(b) F. A. Davis and B.-C. Chen, J. Org. Chem., 1993, 58, 1751.
6 (a) J.-B. Chen, J.-C. Chen, Y. Xie and H. Zhang, Angew. Chem., Int.
Ed., 2012, 51, 1024; (b) Y.-Y. Zhou, Y.-K. Xi, J. Zhao, X. Sheng,
S. Zhang and H. Zhang, Org. Lett., 2012, 14, 3116; (c) C. Pan,
X. Zeng, Y. Guan, X. Jiang, L. Li and H. Zhang, Synlett, 2011, 425;
(d) J. Liang, J. Chen, J. Liu, L. Li and H. Zhang, Chem. Commun.,
2010, 46, 3666; (e) Y. Wu, J. Zhao, J. Chen, C. Pan, L. Li and
H. Zhang, Org. Lett., 2009, 11, 597; ( f ) Y. Zhao, Y. Zhou, L. Liang,
X. Yang, F. Du, L. Li and H. Zhang, Org. Lett., 2009, 11, 555.
7 O. Mitsunobu, Synthesis, 1981, 1.
8 K. B. Sharpless, W. Amberg, Y. L. Bennani, 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.
9 Although the actual effect of phenol 9 is unclear, it seems to us the
additive phenol might form a complex with the substrate (15)
through hydrogen bonds thus may change the shape of the sub-
strate and also improve the solubility. The ee% value was improved
from 60% to 81%, while the reaction time was reduced from 14 days
to 3 days. For reaction conditions, see ESI†.
10 (a) L. N. Mander, Synlett, 1991, 134; (b) P. W. Rabideau and
Z. Marcinow, Org. React., 1992, 42, 1–334; (c) R. Lebeuf, F. Robert
and Y. Landais, Org. Lett., 2005, 7, 4557 and references therein.
11 N. J. DeSogza, A. N. Kothare and V. V. Nadkarny, J. Med. Chem., 1966,
9, 618.
12 Treatment of epoxide 24 with Lewis acid or proton acid again
afforded compound 22 after debenzoylate.
13 Dr Yang Ye in Shanghai Institute of Materia Medica is gratefully
acknowledged for providing the spectra of brazilide A (3).
14 Dr Xiaonian Li in Kunming Institute of Botany is gratefully acknowl-
edged for X-ray crystallography analysis of compound 21 and
brazilide A (3). CCDC 923326 (for compound 21) and CCDC
923325 (for brazilide A).
and brazilide A (3). However, after the epoxide moiety is
coordinated to boron trifluoride etherate, the methoxycarbonyl
group (Scheme 5) attacked exclusively the less hindered side of
the epoxide, namely the C10 position, thus resulting in the
observed stereochemical outcome for the bis-lactonization.
Besides the steric effects, the neighboring groups (OH or OR,
benzene ring) might also assist this process by stabilizing the
transition states. From the above analysis, we deduced that the
orientation of the epoxide unit might control the stereochemistry
for the formation of the lactone ring system.
In order to cope with this problem, the hydroxyl group in 19
was then protected by treatment with tert-butyldimethylsilyl
trifluoromethanesulfonate. The epoxidation of compound 23
occurred slowly (3 days) and afforded the epoxide 24 (from the
less hindered side) and 25 (the desired orientation), respec-
tively, in a ratio of 3 to 1.12
Treatment of epoxide 25 with ethereal boron trifluoride in
dichloromethane (Scheme 6) followed by hydrolysis of the
benzoate (one pot reaction, 82% yield) in the presence of 65%
sulfuric acid finalized the first total synthesis of (+)-brazilide A
([a]D = +3.4, c 1.36, acetone, lit.1f [a]D = +3.3, c 3.0, acetone). The
NMR spectra of our synthetic sample were in complete agree-
ment with the reported spectra.1f,13 The absolute configuration
of brazilide A (3) was established by X-ray crystallography
(CCDC 923325) using Cu-Ka radiation.14
c
This journal is The Royal Society of Chemistry 2013
Chem. Commun., 2013, 49, 5405--5407 5407