Synthesis of glaziovianin A: A potent antitumor isoflavone

Article abstract of DOI:10.1246/cl.2007.1382

Glaziovianin A (1) is a novel isoflavone derivative isolated from the leaves of the Brazilian tree Astelia glazioviana. Glaziovianin A (1) showed cytotoxic activity and was suggested to be an inhibitor of tubulin polymerization. We achieved the total synt

Full text of DOI:10.1246/cl.2007.1382

1382
Chemistry Letters Vol.36, No.11 (2007)
Synthesis of Glaziovianin A: A Potent Antitumor Isoflavone
Ichiro Hayakawa, Akiyuki Ikedo, and Hideo Kigoshiꢀ
Department of Chemistry, Graduate School of Pure and Applied Scieneces,
and Center for Tsukuba Advanced Research Alliance, University of Tsukuba, Tennodai, Tsukuba 305-8571
(Received August 28, 2007; CL-070924; E-mail: kigoshi@chem.tsukuba.ac.jp)
Table 1. Study of the borylation of arenes 4 and 5
OMe
Glaziovianin A (1) is a novel isoflavone derivative isolated
OMe
from the leaves of the Brazilian tree Astelia glazioviana. Glazio-
vianin A (1) showed cytotoxic activity and was suggested to be
an inhibitor of tubulin polymerization. We achieved the total
synthesis of glaziovianin A (1) by using the Suzuki–Miyaura
coupling as a key step.
O
O
O
O
Reagent
RO
Conditions
B
X
OR OMe
OMe
4 X = H
5 X = Br
3a R = H
3b R, R = Me2C CMe2
Entry Substrate
Reagent
Conditions
n-BuLi, THF, rt
Yield
1
4
B(OMe)3
4: 85%
Glaziovianin A (1) is a novel isoflavone derivative isolated
from the leaves of the Brazilian tree Astelia glazioviana by
Yokosuka et al. in 2007 (Figure 1).1 Glaziovianin A (1) showed
cytotoxic activity against HL-60 cells with an IC50 value of
0.29 mM. The pattern of the differential cytotoxicities in the
Japanese Foundation for Cancer Research 39 cell line of glazio-
vianin A (1), as determined using the program COMPARE, has
suggested that the activity of 1 involves the inhibition of tubulin
polymerization.2 Inhibitors of tubulin polymerization, such as
vinblastin, have become clinically important drugs against
breast cancer.3 We planned an efficient and scalable synthesis
of glaziovianin A (1), which will provide a practical supply
for further biological studies.
6
4: 62%, 10: 25%
2
4
B(O-i-Pr)3
7
n-BuLi, THF, rt
4: 79%
3
4
5
5
6
7
n-BuLi, THF, rt
t-BuLi, THF, rt
4: 72%, 10: 17%
O
B
O
Pd(OAc)2, Et3N, DPEphos
1,4-dioxane, 60 °C
5
5
H
4: 76%
8
O
O
O
O
PdCl2(dppf), KOAc
DMF, 60 °C
B
B
4: 40%, 11: 15%
6
5
9
9
PdCl2(dppf), KOAc
DMF, 100 °C
3b: 12%, 4: 39%, 11: 1%
7
8
5
5
PdCl2(dppf), KOAc
DMF, 150 °C
9
3b: 44%, 4: 40%, 11: 3%
Our synthetic route to glaziovianin A (1) involved the
Suzuki–Miyaura coupling at C-10–C-3 (Scheme 1).4 We there-
fore synthesized 3-iodochlomone derivative 2 and boronate
compound 3.
OMe
OMe
OH
O
OMe
O
O
O
OH
OMe
11
OMe
109
Preparation of the boronate compounds 3 required optimiza-
tion (Table 1). Thus, preparation of arylboronic acid 3a was
attempted by the lithiation of 45 followed by treatment with
trimethyl borate (6) or triisopropyl borate (7), but the desired
compound 3a could not be obtained (Entries 1 and 2). Bromine–
lithium exchange of aryl bromide 56 followed by reaction with
6 or 7 gave only debromo compound 4 (Entries 3 and 4). Next,
we attempted the direct borylation of the aryl bromide 5 with
pinacolborane (8) catalyzed by Pd0,7 but the arylboronate 3b
could not be obtained (Entry 5). An attempt at the cross-coupling
OMe
reaction of the bis(pinacolato)diboron (9) with 5 to 3b failed
by using PdCl2(dppf) and KOAc in DMF at 60 ꢁC (Entry 6).8
However, the reactions at higher temperature gave the desired
arylboronate 3b (Entries 7 and 8). The reaction at 150 ꢁC gave
the desired arylboronate 3b in 44% yield along with a small
amount of the homocoupling dimer 11 (Entry 8).
The 3-iodochromone derivative 2 was synthesized from
commercially available 2-hydroxy-4,5-dimethoxyacetophenone
(12) by the reported method for the preparation of 3-halogenated
chromones (Scheme 2).10 Condensation of 12 with N,N-dime-
thylformamide dimethyl acetal gave enamine 13 in quantitative
yield. The attempts to achieve iodinative cyclization of enamine
13 are summarized in Table 2. Reaction of enamine 13 with
iodine in MeOH did not give iodochromone 2 because of the
low solubility (Entry 1). The cyclization in toluene afforded
the desired iodochromone 2 in 44% yield (Entry 2).11 Treatment
of 13 with iodine in CH2Cl2 gave iodochromone 2 in 65% yield
(Entry 3). The cyclization was the most efficiently effected in
CHCl3 (Entry 4). Addition of pyridine was not effective in this
case (Entry 5).12
OMe
4'
O
O
6'
O
4
5'
2'
5
8
MeO
MeO
6
7
10
9
3
3'
1'
OMe
2
O
Figure 1. Structure of glaziovianin A (1).
OMe
O
OMe
O
O
O
O
O
MeO
MeO
3
MeO
MeO
I
O
RO
1'
B
With both segments, iodochromone 2 and arylboronate 3b,
in hand, we attempted the Suzuki–Miyaura coupling by using
PdCl2(dppf)13 to furnish glaziovianin A (1) in 80% yield
(Scheme 3).
OMe
OR OMe
O
glaziovianin A (1)
2
3
Scheme 1. Retrosynthetic analysis of glaziovianin A (1).
Copyright ꢀ 2007 The Chemical Society of Japan

Chemistry Letters Vol.36, No.11 (2007)
1383
References and Notes
O
O
O
O
MeO
MeO
MeO
MeO
MeO
MeO
I
a
b
1
A. Yokosuka, M. Haraguchi, T. Usui, S. Kazami, H. Osada,
T. Yamori, Y. Mimaki, Bioorg. Med. Chem. Lett. 2007, 17,
3091.
NMe2
OH
OH
12
13
2
2
T. Yamori, A. Matsunaga, S. Saito, K. Yamazaki, A. Komi,
K. Ishizu, I. Mita, H. Edatsugi, Y. Matsuda, K. Takezawa,
O. Nakanishi, H. Kohno, Y. Nakajima, H. Komatsu, T.
Andoh, T. Tsuruo, Canser Res. 1999, 59, 4042.
K. Komiyama, in Microbial Chemistry, 4th ed., ed. by
Reagents and conditions: a) (MeO)2CHNMe2, quant. b) I2, CHCl3, rt, 73%
(see Table 2).
Scheme 2. Synthesis of iodochromone 2.
3
Table 2. Study of iodinative cyclization
¯
Y. Ueno, S. Omura, Nankodo, Tokyo, 2003, Chap. 7A,
pp. 257–274.
I2
13
2
4
5
Y. Hoshino, N. Miyaura, A. Suzuki, Bull. Chem. Soc. Jpn.
1988, 61, 3008.
M. A. Rizzacasa, M. V. Sargent, J. Chem. Soc., Perkin
Trans. 1 1987, 2017.
R. Slamet, D. Wege, J. Chem. Res., Synop 2001, 18.
M. Murata, S. Watanabe, Y. Masuda, J. Org. Chem. 1997,
62, 6458.
T. Ishiyama, M. Murata, N. Miyaura, J. Org. Chem. 1995,
60, 7508.
The methylene acetal unit of compound 4 was hydrolyzed by
acidic workup with 1 M HCl to give catechol 10 (Entries 2
and 4), however, compound 10 was not detected in Entries
1 and 3 somehow.
Solvent, rt
Entry
Solvent
Yield
1
2
3
4
5
MeOH
Toluene
CH2Cl2
CHCl3
0% (recovery of 13: 88%)
44%
65%
73%
73%
6
7
Pyridine, CHCl3
8
9
OMe
OMe
O
O
O
O
O
O
O
O
MeO
MeO
I
MeO
MeO
B
O
a
O
OMe
OMe
10 R. B. Gammill, Synthesis 1979, 901.
2
3b
Glaziovianin A (1)
11 Y. Igarashi, H. Kumazawa, T. Ohshima, H. Satomi, S.
Terabayashi, S. Takeda, M. Aburada, K. Miyamoto, Chem.
Pharm. Bull. 2005, 53, 1088.
Reagents and conditions: a) PdCl2(dppf), 1M Na2CO3 aq., 1,4-dioxane, rt,
80%.
12 R. Hong, J. Feng, R. Hoen, G.-q. Lin, Tetrahedron 2001, 57,
8685.
13 K. Ding, S. Wang, Tetrahedron Lett. 2005, 46, 3707.
Scheme 3. Synthesis of glaziovianin A (1).
Synthetic glaziovianin A (1) gave spectral data (1H NMR,
13C NMR, HRMS, IR, and UV) in full agreement with those of
the natural glaziovianin A (1),14 completing the total synthesis.
The cytotoxic activities of synthetic glaziovianin A (1) against
HeLa cells and U87 human glioblastoma cells had the same
IC50 values as those of natural glaziovianin A (1).15
In summary, we achieved a concise and semigram scale
synthesis of glaziovianin A (1) by using the Suzuki–Miyaura
coupling as a key step. Further structure–activity relationship
studies are now in progress.
1
14 Glaziovianin A (1): H NMR (270 MHz, CDCl3) ꢀ 7.91 (s,
1H), 7.62 (s, 1H), 6.89 (s, 1H), 6.53 (s, 1H), 6.03 (s, 2H),
4.00 (s, 3H), 3.99 (s, 3H), 3.88 (s, 3H), 3.86 (s, 3H);
13C NMR (67.8 MHz, CDCl3) ꢀ 175.2, 154.1, 153.3, 152.1,
147.5, 139.0, 138.8, 136.9, 136.6, 121.6, 117.9, 117.7,
110.0, 104.8, 101.7, 99.5, 60.1, 56.9, 56.4, 56.3; HR-ESI-
MS m=z 409.0894, calcd for C20H19NaO8 [M + Na]þ
409.0899; IR (film) ꢁmax 3002, 2941, 2838, 1638, 1606,
1505, 1454, 1428, 1297, 1271, 1228, 1150, 1062,
1035 cmꢂ1; UV (MeOH) ꢂmax 317.0 (log " 4.02), 256.5 nm
(4.21), mp 185.0–187.0 ꢁC.
We thank Professor Takeo Usui and Mr. Hiroshi Kamiyama
(University of Tsukuba) for biological assay and valuable
discussion. This work was supported in part by the 21st COE
program and Grant-in-Aid for Scientific Research on Priority
Areas (No. 16073204, ‘‘Creation of Biologically Functional
Molecules’’) from the Ministry of Education, Culture, Sports,
Science and Technology, Japan.
15 Natural and synthetic glaziovianin A (1) showed cytotoxic
activities against HeLa cells with an IC50 of 5.0 and
5.3 mM, and against U87 cells with an IC50 of 7.9 and
8.3 mM, respectively. These biological assays were carried
out by Prof. T. Usui, Graduate School of Life and Environ-
mental Sciences, University of Tsukuba.
Published on the web (Advance View) October 20, 2007; doi:10.1246/cl.2007.1382