A one-pot, two-step synthesis of tetrahydro asterriquinone E

Article abstract of DOI:10.1021/ol990075b

(equation presented) Bis(indolyl)dihydroxyquinone 2, the tetrahydro analogue of naturally occurring 1a, was synthesized by a novel, expeditious route. The short synthesis was accomplished by treating p-bromanil (3) with 2 equiv of indole 4 in the presence

Full text of DOI:10.1021/ol990075b

ORGANIC
LETTERS
1999
Vol. 1, No. 3
431-433
A One-Pot, Two-Step Synthesis of
Tetrahydro Asterriquinone E
,†
G. Davis Harris, Jr.,* Ann Nguyen, Harald App, Peter Hirth,
Gerald McMahon, and Cho Tang*
SUGEN, Inc., 230 East Grand AVenue, South San Francisco, California 94080
daVe-harris@sugen.com.
Received May 5, 1999
ABSTRACT
Bis(indolyl)dihydroxyquinone 2, the tetrahydro analogue of naturally occurring 1a, was synthesized by a novel, expeditious route. The short
synthesis was accomplished by treating p-bromanil (3) with 2 equiv of indole 4 in the presence of cesium carbonate in acetonitrile at ambient
temperature to provide a 1:1 mixture of the dibromo regioisomers 7 and 8, followed by hydrolysis of the mixture to afford 2. The synthetic
compound 2 was found to inhibit the binding of the Grb2 adapter protein to tyrosine-phosphorylated EGF receptor (IC50 ) 1.2 µM).
A family of naturally occurring bis(indolyl)dihydroxyquino-
nes, known as asterriquinones, isolated from such fungal
strains as Chaetomium globosum,1 Ch. Cochlioides,1 Ch.
murorum,2 and Ch. amygdalisporum,2 has been shown to
possess a wide array of biological activity.3 Recently it was
discovered in this laboratory that the asterriquinone natural
products 1a-e (Scheme 1), isolated from Aspergillus terreus,
inhibit the binding of Grb2 adaptor protein to tyrosine-
phosphorylated EGF receptor (1a-IC50 ) 2.9 µM).4 The
Grb2 inhibitors 1a-e represent the first compounds shown
to directly inhibit the interaction between adaptor proteins
and protein tyrosine kinase molecules, an interaction impli-
cated in many classes of cancerous tumors.
Despite the therapeutic potential of the natural products
1,5 a mild, general synthetic route to these compounds and
analogues did not exist in the literature. In view of the
potential of compounds 1a-e as anticancer agents, an effort
was undertaken to formulate a general synthesis of aster-
riquinone analogues. In this communication, we describe the
novel, expeditious synthesis of the tetrahydro derivative 2
(Scheme 2) of the potent Asterriquinone E4 (1a). Greater
† Tel: 650-553-8438. Fax: 650-553-8348.
(1) Brewer, D.; Jerram, W.; Taylor, A. Can. J. Microbiol. 1968, 14, 861.
(2) Sekita, S. Chem. Pharm. Bull. 1983, 31, 2998.
(4) (a) App, H.; Fong, A.; Lipson, K.; Harris, D.; Hui, T.; Rice, A.; Wang,
H.; Chen, H.; Kim, Y.; Tang, C.; Dare, H.; Margolis, B.; Hirth, P.; Shawver,
L.; McMahon, G. Proc. Am. Assoc. Cancer Res. 1997, 38, 349. (b) Alvi,
K.; Pu, H.; Luche, M.; Dare, H.; Margolis, B.; App, H.; McMahon, G.
Submitted for publication.
(5) After submission of this paper for publication, an article appeared in
which a closely related, naturally occurring asterriquinone is described as
an insulin mimetic with antidiabetic activity: Zhang, B.; Salituro, G.;
Szalkowski, D.; Li, Z.; Zhang, Y.; Royo, I.; Vilella, D.; Diez, M. T.; Pelaez,
F.; Ruby, C.; Kendall, R. L.; Mao, X.; Griffin, P.; Calaycay, J.; Zierath, J.
R.; Heck, J. V.; Smith, R. G.; Moller, D. E. Science 1999, 284, 974.
(3) (a) Arai, K.; Shimizu, S.; Taguchi, Y.; Yamamoto, Y. Chem. Pharm.
Bull. 1981, 29, 991. (b) Shimizu, S.; Yamamoto, Y.; Inagaki, J.; Koshimura,
S. Gann 1982, 73, 642. (c) Kaji, A.; Iwata, T.; Kiriyama, N.; Wakusawa,
S.; Miyamoto, K. Chem. Pharm. Bull. 1994, 42, 1682. (d) Shimizu, S.;
Yamamoto, Y.; Koshimura, S. Chem. Pharm. Bull. 1982, 30, 1896. (e) Ono,
K.; Nakane, H.; Shimizu, S.; Koshimura, S. Biochem. Biophys. Res.
Commun. 1991, 174, 56. (f) Brewer, D.; Jerram, W. A.; Taylor, A. U.S.
Patent 3917820, 1975. (g) Mocek, U.; Schultz, L.; Buchan, T.; Baek, C.;
Fretto, L.; Nzerem, J.; Sehl, L.; Sinha, U. J. Antibiot. 1996, 49, 854.
10.1021/ol990075b CCC: $18.00 © 1999 American Chemical Society
Published on Web 06/25/1999

5 and 6 (Scheme 2). It was hoped that these reactions would
provide the bis(indolyl)dihydroxyquinone moiety in one
transformation, but in each case, employment of an extensive
set of reaction conditions failed to effect the desired indole-
quinone coupling. Analysis of the mixture from the reaction
between 4 and bromanilic acid (5) revealed a bromo-
containing bis(indolyl) product, which led us to conclude
that an unsymmetrical addition had occurred. This finding
prompted us to use the symmetrical quinone p-bromanil (3).
Scheme 1
The synthesis of 2 (Scheme 3) began with commercially
available p-bromanil (3) and indole 4. Treatment of p-
bromanil with 2 equiv of 4 in the presence of cesium
carbonate in acetonitrile at room temperature afforded a 1:1
mixture of regioisomers 7 (46%) and 8 (44%), separable by
flash chromatography.9 The 13C NMR spectrum of 7 shows
only one (CdO) signal, indicative of the para regiochemistry.
On the other hand, the 13C NMR spectrum of 8 contains
two (CdO) signals, indicative of the meta regiochemistry.
Other combinations of base and solvent were less effective
or unsuccessful (Table 1). While the reasons for the improved
yield from the Cs2CO3/CH3CN combination are unclear, there
are numerous cited examples of improving chemical pro-
ease of synthesis, the presumption of the unimportance of
the double bonds of 1a with regard to activity, and the
potential of the olefinic bonds to be metabolized led us to
synthesize the tetrahydro derivative rather than the natural
product.
Scheme 3a
Previously published complex, low-yielding syntheses of
indolylquinones were inadequate for our purposes.6,7 In
pursuit of an efficient, general methodology for synthesizing
asterriquinones, we initially attempted double C-C couplings
between 2-(3-methyl-n-butyl)indole (4)8 and benzoquinones
Scheme 2
432
Org. Lett., Vol. 1, No. 3, 1999

to be slightly more potent against Grb2 function (IC50
)
1.2 µM) than the natural product 1a.12 Thus, asterriquinone
2 was prepared in two simple steps from commercially
available compounds in 36% yield from p-bromanil. The
activity of 2 indicated that the olefinic functionality present
in the indole moiety of 1a was not necessary for Grb2
inhibition.
Alternatively, the hydrolysis of a mixture of regioisomers
7 and 8 provided pure 2 in 31% yield from p-bromanil after
HPLC purification (compound 2 was not stable to flash silica
or neutral alumina). In this procedure, aqueous KOH and
EtOH were added directly to the crude mixture of 7 and 8
in the same reaction vessel without prior workup (Scheme
4). Thus, asterriquinone 2 was synthesized in a simple, one-
Table 1. Indole-Quinone Coupling Conditions for the
Generation of 7
base
solventa
temp, °C
yield, %b,c
K2CO3
DMF
THF
THF
25
100
50
7
3
0
CH3CN
DMF
THF
CH3CN
THF
50
100
80
50
25
16
4
NaHCO3
Cs2CO3
0
0
0
DMF
CH3CN
CH3CN
25
25
25
17
28
46d
a 0.1 M concentration unless otherwise indicated. b An equal amount of
meta isomer 8 was formed in every case. c All reaction times 24 h. d 1 M
concentration.
Scheme 4
cesses by replacing sodium and potassium compounds with
their cesium counterparts.10
By monitoring the progress of the reaction via TLC, we
discovered that the addition of the first indole to p-bromanil
to afford the monoadduct 7a was relatively fast (ca. 2 h)
(Scheme 3). In contrast, the addition of the second equivalent
of 4 was slow (ca. 22 h). By virtue of this difference between
the rates of addition of the first and second equivalents of
indole, we were able to synthesize unsymmetrical aster-
riquinone analogues by adding 1 equiv of one indole to
p-bromanil followed by addition of an excess of a different
indole.11
pot transformation in yield equal to the two-step procedure.
In summary, we have described the short, efficient one-
pot synthetic route to tetrahydro Asterriquinone E (2), the
first synthetic substance to directly inhibit the interaction
between adapter proteins and protein tyrosine kinase mol-
ecules. The promising inhibitory properties of this novel
synthetic asterriquinone, as well as its ease of synthesis,
highlight the potential of this class of compounds for broad
application to the treatment of cancers.
Hydrolysis of para isomer 7 with potassium hydroxide
produced the bis(indolyl)dihydroxyquinone 2 (78%), found
(6) Noland, W. E.; Baude, F. J. J. Org. Chem. 1966, 31, 3321.
(7) Hoercher, J.; Schwenner, E.; Franck, B. Liebigs Ann. Chem. 1986,
10, 1765.
(8) Verley, A.; Beduwe, F. Bull. Soc. Chim. Fr. 1925, 37, 190.
(9) For more examples of Michael-type additions of indoles to quinones,
see: (a) Bergman, J.; Carlsson, R.; Misztal, S. Acta Chem. Scand., Ser. B
1976, 30, 853. (b) Tsaklidis, J. N.; Hofer, A.; Eugster, C. H. HelV. Chim.
Acta 1977, 60, 1033, 1053. (c) Bu’Lock, J. D.; Harley-Mason, J. J. Chem.
Soc. 1951, 703, 710. (d) Bruce, J. M. J. Chem. Soc. 1959, 2366, 2372.
(10) (a) Hofmann, H. Chem. Ind. Dig. 1993, 6, 113 and references therein.
(b) Hofmann, H. Spec. Chem. 1993, 13, 59, 62-63, 65. (c) Blum, Z. Acta
Chem. Scand. 1989, 43, 248.
Supporting Information Available: Text giving experi-
mental procedures and spectroscopic data for compounds 2,
1
7a, 7, and 8 and figures giving the H NMR spectrum and
an HPLC chromatogram of compound 2 to indicate purity.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(11) Publication in progress.
(12) Hydrolysis of pure 8 under the same conditions gave a complex
mixture containing none of the dihydroxy compound 9 (Scheme 3).
OL990075B
Org. Lett., Vol. 1, No. 3, 1999
433