Unexpected formation of the arcyriacyanin system by condensation of a 3-bromo-4-(indol-3-yl)maleimide with (2-nitrophenyl)acetates

Article abstract of DOI:10.1071/CH04043

The reaction of an N-protected 3-bromo-4-(indol-3-yl)maleimide with methyl (2-nitrophenyl)acetates in the presence of base affords condensed cycloheptatriene derivatives which can be transformed into arcyriacyanin-type alkaloids. The unusual reaction sequ

Full text of DOI:10.1071/CH04043

Rapid Communication
CSIRO PUBLISHING
www.publish.csiro.au/journals/ajc
Aust. J. Chem. 2004, 57, 625–628
Unexpected Formation of the Arcyriacyanin System by Condensation
of a 3-Bromo-4-(indol-3-yl)maleimide with (2-Nitrophenyl)acetates∗
Guido Mayer,A Claudia Hinze,A Kurt Polborn,A and Wolfgang SteglichA,B
A Department Chemie, Ludwig-Maximilians-Universität, 81377 München, Germany.
B Author to whom correspondence should be addressed (e-mail: wos@cup.uni-muenchen.de).
The reaction of an N-protected 3-bromo-4-(indol-3-yl)maleimide with methyl (2-nitrophenyl)acetates in the pres-
ence of base affords condensed cycloheptatriene derivatives which can be transformed into arcyriacyanin-type
alkaloids. The unusual reaction sequence implies a heptatrienyl–cycloheptadienyl anion rearrangement followed
by a 1,5-sigmatropic shift to yield the thermodynamically more stable cycloheptatriene derivative.
Manuscript received: 18 February 2004.
Final version: 14 April 2004.
The slime mould alkaloid arcyriacyanin A 1[1] (Scheme 1) is
a member of the bisindolylmaleimide group of alkaloids[2]
that includes protein kinase inhibitors like staurosporine and
arcyriarubinA 2.ArcyriacyaninA 1 shows unique antitumour
properties, and inhibits protein kinase C and protein tyrosine
kinase.[3] Weandothershavedevelopedsynthesesforarcyria-
cyanin A based on palladium-catalyzed reactions.[1c,3,4] In
this communication we report on a highly convergent method
of attaining the arcyriacyanin system through cyclization of
suitable nitrophenyl precursors.
H
N
H
N
O
O
O
O
NH
N
H
N
H
N
H
1
2
Scheme 1.
Our strategy is based on the unexpected discovery that
reaction of N-protected 3-bromo-4-(indol-3-yl)maleimide
3[5] with the anion of methyl (2-nitrophenyl)acetate 4[6] at
−78◦C and warming to room temperature affords the yel-
low benzo[6,7]pyrrolo[3ꢀ,4ꢀ:3,4]cyclohepta[1,2-b]indole 6 in
40% yield (Scheme 2). No trace of the expected substitution
product 5 could be detected. The structure of compound 6
was established through extensive NMR studies (see Exper-
imental) and confirmed by a single crystal X-ray analysis
(Fig. 1).[7] Upon thermolysis,[8] the Boc group can be easily
removed to yield the free indole 7, which is a close structural
analogue of arcyriacyanin 1.
In order to study the mechanism of the ring closure, we
followed the reaction between components 3 and 4 at lower
temperatures. When the reaction is carried out at −78◦C and
monitored by TLC, the rapid formation of a new product can
be detected. Upon raising the temperature to −60◦C the ini-
tial compound is transformed into a second product, which
has a characteristic green fluorescence under UV light, and
this then yields the final product 6 at temperatures above
−30◦C. The primary product could be isolated by quench-
ing the reaction with aqueous citric acid at −78◦C, and
subsequent chromatography on silica gel yielded the pure
compound in approximately 50% yield. Its spectroscopic data
were in agreement with those expected for maleimide 5.
Upon treatment with a mild base, compound 5 could be
converted into 6 in approximately 80% yield. A second inter-
mediate was obtained in pure form upon treatment of 5 with
lithium iodide in DMF followed by rapid chromatography of
the crude product on a pre-cooled silica-gel column. Immedi-
ate NMR measurements (CDCl3) at −50◦C established that
this intermediate had structure 10. In particular, at −50◦C,
signals for a 3 : 1 mixture of the (E)- and (Z)-diastereomers of
10 are observed; this indicates a hindered rotation of the Boc
residue.The doubling-up of signals due to H8, H8b, H10, and
the Boc protons disappears upon warming the NMR sample
to room temperature. Concurrently, signals for the thermo-
dynamically more stable cycloheptatriene isomer 6 gradually
emerge, and rearrangement is complete within 30 minutes.
From these results we suggest that the reaction proceeds
through the pathway depicted in Scheme 3. Thus, anion 8,
derived from the primary product 5, undergoes an electro-
cyclic heptatrienyl–cycloheptadienyl anion rearrangement[9]
∗ Dedicated to Professor Lewis Mander on the occasion of his 65th birthday.
© CSIRO 2004
10.1071/CH04043
0004-9425/04/070625

626
G. Mayer et al.
Me
N
3
Me
N
Me
N
O
O
O
O
1
O
O
13
12
CO2CH3
NO2
4
(a)
CO2Me
3b
11
CO2Me
NO2
8b
8
Br
10
5
ϩ
N
R
N
Boc
N
Boc
6
7
3
4
5
6 R ϭ Boc
7 R ϭ H
(b)
Scheme 2. Reagents and conditions: (a) 4, NaH, THF, room temp.; then −78◦C, 3, and warming to room temp., 36 h; then aqueous citric acid;
and (b) 180◦C, 10 min.
Me
N
Me
N
Me
N
O
O
O
O
O
O
Ϫ
Ϫ
CO2Me
base
CO2Me
CO2Me
NO2
5
6
8b
O2N
N
Boc
N
Boc
H
H
N
Boc
8
9
10
Scheme 3. Base-induced cyclization of the primary substituted product 5.
Me
N
Me
N
O
O
O
O
CO2CH3
NO
O
(c)
(a)
CO2Me
R
O2N
2
NH
N
R
N
Boc
14 R ϭ Boc
15 R ϭ H
12 R ϭ NO2
13 R ϭ NH2
11
(d )
(b)
Scheme 4. Reagents and conditions: (a) KH, [18]crown-6, THF, 11, 0◦C, 2 h; then 3, THF, 0◦C to room temp., NEt3, 30 h; (b) 10%
Pd/C, H2, EtOAc, 3 bar, room temp., 3 h; (c) 0.2 equiv. (HOBu2SnOSnBu2SCN)2,[12] benzene, 80◦C, 36 h; and (d) 180◦C, 5 min.
Me
N
Me
N
O
O
O
O
R
CN
(b)
(d )
O2N
NO2
CN
NH
R
N
Boc
N
R
20 R ϭ Boc
21 R ϭ H
16 R ϭ CO2Et
17 R ϭ H
18 R ϭ NO2
19 R ϭ NH2
(a)
(c)
(e)
Scheme 5. Reagents and conditions: (a) BHT, 2 N HCl (cat.), DMSO, heat until the red solution turns brown; then ice water;
(b) KH, [18]crown-6, 17, THF, −78◦C; then 3, THF, NEt3, −78◦C to room temp., 15 h; (c) 10% Pd/C, H2, EtOAc, MeOH, 4.5 bar,
80◦C, 15 h; (d) DIBAL-H, THF, 0◦C to room temp., 2 h; and (e) 180◦C, 5 min.
to yield anion 9, which after elimination of nitrite fur-
nishes the cycloheptatriene 10. The latter then undergoes a
sigmatropic [1,5]-hydrogen shift to afford the thermodynam-
ically more stable isomer 6.
For the synthesis of arcyriacyanin A 1, an additional
nitrogenatomhastobeintroducedfromthebeginning.There-
fore, we started with methyl (2,6-dinitrophenyl)acetate 11,
which can be prepared from 2,6-dinitrochlorobenzene and

Unexpected Formation of the Arcyriacyanin System
627
Boc-protected N-methyl-arcyriacyanin A 20 in 12% yield.
The low yield obtained for the indole ring closure reflects the
unfavourable geometry of the cyano group, which according
to the X-ray structural analysis of nitronitrile 18,[7] adopts a
nearly perpendicular relationship with the plane of the seven-
membered ring. Furthermore, formation of the planar pyrrole
ring causes a higher ring strain than closure of the indolinone
ring in 13. Thermal cleavage of the Boc group[8] in 20 gave
N-methyl-arcyriacyanin A 21 in 80% yield, and earlier work
within our group[1c] has shown that 21 can be converted
into 1.
Although at present the novel condensation technique can-
not compete with the Pd-catalyzed syntheses of arcyriacyanin
A 1, it offers an effective route to several seco compounds of
this series.
Experimental
Compound 5
Fig. 1. ORTEP diagram of 6 (numbering differs from that given in the
formula).
This was obtained as a light yellow, unstable solid, mp 98◦C (MeOH).
RF 0.35, EtOAc/hexanes (1 : 3), Merck silica gel 60 F254 plates.
νmax (KBr)/cm−1 2985, 2955, 1743, 1710, 1639, 1529 (NO2), 1453,
1385 (NO2), 1354, 1311, 1259, 1229, 1155, 1073, 746. λmax/nm
(log ε) (MeOH) 224 (4.44), 248 (4.26), 302 (sh, 3.81), 393 (3.44).
δH (400 MHz, CDCl3, TMS) 1.58 (9H, s, Boc), 3.14 (3H, s, NMe),
3.68 (3H, s, OMe), 6.03 (1H, s, H3b), 7.21 (1H, ddd, J 8.0, 8.0,
1.0), 7.35 (1H, ddd, J 7.3, 7.3, 1.0), 7.41–7.48 (3H, m), 7.56 (1H, s,
H8b), 7.57 (1H, ddd, J 8.0, 8.0, 1.5), 8.02 (1H, dd, J 8.0, 1.5), 8.17
(1H, d, J 8.0). δC (100 MHz, CDCl3, TMS) 24.3 (NMe), 27.8 (3C,
C(CH3)3), 43.9 (C3b), 53.0 (OMe), 84.7 (C(CH3)3), 108.4, 115.3,
120.5, 123.4, 125.3 (2C), 127.5, 128.2, 128.9, 130.3, 130.7, 133.4,
134.4, 135.5, 136.8, 148.8, 148.9, 168.9, 169.5, 170.5. m/z (FAB,
the anion of dimethyl malonate[10] followed by Krapcho de-
alkoxycarbonylation.[11] The anion derived from 11 is more
stable and less reactive than that obtained from 4. As a
result, in order to achieve cyclization 11 had to be treated
with KH at 0◦C in the presence of [18]crown-6 and tri-
ethylamine. In this way compound 12 was obtained in 40%
yield (Scheme 4), and its structure was confirmed by single-
crystal X-ray analysis.[7] Reduction of the nitro group by
catalytic hydrogenation afforded the amino ester 13, which
subsequently failed to form the oxindole ring either under
basic or acidic conditions. However, treatment of amino
ester 13 with Otera’s 1-hydroxy-3-(isothiocyanato)tetra-
butyldistannoxane catalyst[12] furnished the yellow indolin-
2-one 14 in 74% yield. Unfortunately, removal of the Boc
group[8] under thermal conditions affords an unstable prod-
uct, and after rapid chromatography, the free indole 15 could
only be isolated in 8% yield.
As oxindole 15 was too unstable to undergo further trans-
formations, we decided to prepare the indole directly from
the corresponding amino nitrile.[13] The only viable method
that we could find to prepare the necessary starting mate-
rial 17[14] involved heating cyano ester 16 in DMSO with a
hot air gun for a short time (Scheme 5). Immediately after
the red colour of the solution had changed to brown, the
reaction mixture was poured into ice water to yield a light-
brown solid, which was purified by column chromatography
on silica gel. In order to avoid polymerization of the prod-
uct, the solvents had to be removed at 0◦C. The yield of
nitrile 17 could be increased from 20 to 34% by the addi-
tion of 2,6-di-tert-butyl-4-methylphenol (BHT) as a radical
scavenger together with small amounts of 2 N HCl. The
deep blue anion of 17 was subsequently allowed to react
with bromide 3 to afford the cyclized product 18 in 48%
yield. Reduction of the nitro group with dihydrogen in the
presence of Pd/C yielded aminonitrile 19, which upon treat-
ment with diisobutyl aluminium hydride,[15] basic workup,
and gel chromatography on Sephadex LH 20 furnished the
•
m-NBA) 520 (8%, [M + 1]+ ), 519 (6, M+•), 464 (15), 432 (5), 420
(3), 388 (5), 343 (5), 314 (3).
Compound 6
This was obtained as a yellow solid, mp 178◦C (MeOH) (Found: C 68.6,
H 5.2, N 6.0. C27H24N2O6 requires C 68.6, H 5.1, N 5.9%). RF 0.46,
yellow fluorescence under UV light, EtOAc/hexanes (1 : 3); RF 0.30
EtOAc/hexanes (1 : 5). νmax (KBr)/cm−1 2984, 2952, 2939 (sh), 1739,
1707, 1640, 1520, 1458, 1435, 1386, 1313, 1289, 1227, 1152, 1120,
1064, 1006, 759. λmax/nm (log ε) (MeOH) 204 (4.63), 247 (4.41), 296
(4.34), 411 (3.33). δH (400 MHz, CDCl3, TMS) 1.36 (9H, s, Boc), 3.12
(3H, s, NMe), 3.42 (3H, s, OMe), 5.24 (1H, s, H3b), 7.37–7.53 (6H, m),
8.26 (1H, d, J 8.2, H10), 8.36 (1H, d, J 7.9, H13). δC (100 MHz, CDCl3,
TMS) 24.2 (NMe), 27.5 (3C, C(CH3)3), 45.3 (C3b), 52.6 (OMe), 84.7
(C(CH3)3), 114.0, 114.4 (C10), 123.4 (C13), 123.9, 125.2, 125.5, 126.7,
129.6, 130.1, 130.7, 131.3, 131.8, 132.5, 134.2, 137.8, 141.6, 149.7,
168.5 (C4), 169.4 (C1), 169.5 (C3). m/z (FAB, m-NBA) 473 (11%,
•
[M + 1]+ ), 472 (11, M+•), 417 (17), 372 (15), 313 (100). X-Ray crys-
tal analysis: C27H24N2O6, monoclinic, space group C2/c (No. 15), a
26.328(7), b 8.562(2), c 21.425(6) Å, β 94.43(2)◦, V 4815 (2) Å3, 3341
unique reflections, 2835 with I > 2σ(I), R1 0.0423, wR2 0.1160 for 322
refined parameters.
Compound 10
This was obtained as an oil, RF 0.51, green fluorescence under UV light,
EtOAc/hexanes (1 : 3). δH (400 MHz, CDCl3,TMS, −50◦C) (E)-isomer
(3 parts): 1.30 (9H, br s, Boc), 3.19 (3H, s, NMe), 4.13 (3H, s, OMe),
5.24 (1H, br s, H8b), 7.27 (1H, d, J 7.9, H5), 7.78 (1H, d, J 7.6, H8),
8.25 (1H, d, J 7.8, H10), 8.89 (1H, d, J 7.9, H13); (Z)-isomer (1 part):
1.68 (9H, br s, Boc), 3.19 (3H, s, NMe), 4.06 (3H, s, OMe), 5.36 (1H,
br s, H8b), 7.25 (1H, d, J 7.9, H5), 7.83 (1H, d, J 7.6, H8), 7.85 (1H, d,
J 7.8, H10), 8.92 (1H, d, J 7.9, H13) [overlapping signals: 7.20 (1H,
ddd, J 7.9, 7.6, 1.0, H12), 7.44 (1H, ddd, J 8.7, 7.6, 1.1, H7), 7.60
(1H, ddd, J 8.7, 7.9, 1.2, H6), 7.65 (1H, ddd, J 7.8, 7.6, 1.2, H11)]. δC

628
G. Mayer et al.
(100 MHz, CDCl3, TMS, −50◦C, selected) (E)-isomer (3 parts): 27.6
(C(CH3)3), 66.1 (C8b), 82.4 (C(CH3)3), 115.0 (C10), 125.7 (C8); (Z)-
isomer (1 part): 28.1 (C(CH3)3), 65.9 (C8b), 83.8 (C(CH3)3), 115.5
(C10), 126.0 (C8).
[3] M. Murase, K. Watanabe, T.Yoshida, S. Tobinaga, Chem. Pharm.
Bull. (Tokyo) 2000, 48, 81.
[4] M. Murase, K. Watanabe, T. Kurihara, S. Tobinaga, Chem.
Pharm. Bull. (Tokyo) 1998, 46, 889.
[5] M. Brenner, H. Rexhausen, B. Steffan, W. Steglich, Tetrahedron
1988, 44, 2887. doi:10.1016/S0040-4020(01)96236-2
[6] T. H. Fife, N. W. Duddy, J. Am. Chem. Soc. 1983, 105, 74.
[7] Crystallographic data (excluding structure factors) for the struc-
tures have been deposited with the Cambridge Crystallographic
Data Centre as supplementary publication numbers CCDC
190414 (6), CCDC 204892 (12), and CCDC 214318 (18).
Copies of the data can be obtained free of charge on application
to the CCDC (e-mail: deposit@ccdc.cam.ac.uk).
[8] V. H. Rawal, M. P. Cava, Tetrahedron Lett. 1985, 26, 6141.
doi:10.1016/S0040-4039(00)95036-6
Accessory Material
ORTEP diagrams for compounds 12 and 18, and experimen-
tal data for compounds 7, 12–15, and 17–20 are available
from the author or, until July 2009, the Australian Journal of
Chemistry.
Acknowledgments
We thank Dr David Stephenson for the NMR experiments
conducted at low temperatures and the Fonds der Chemischen
Industrie for financial support.
[9] R. B. Bates, W. D. Deines, D. A. McCombs, D. E. Potter, J. Am.
Chem. Soc. 1969, 91, 4608.
[10] F. J. Goetz, J. A. Hirsch, R. L. Augustine, J. Org. Chem. 1983,
48, 2468.
[11] Reviews:
References
(a) A. P. Krapcho, Synthesis 1982, 805. doi:10.1055/S-1982-
29953
(b) A. P. Krapcho, Synthesis 1982, 893. doi:10.1055/S-1982-
29991
[1] (a) M. Gill, W. Steglich, Prog. Chem. Org. Nat. Prod. 1987,
51, 216.
(b) W. Steglich, Pure Appl. Chem. 1989, 61, 281.
(c) M. Brenner, G. Mayer, A. Terpin, W. Steglich, Chem. Eur. J.
1997, 3, 70.
[12] (a) J. Otera, Chem. Rev. 1993, 93, 1449.
(b) J. Otera, N. Dan-oh, H. Nozaki, J. Org. Chem. 1991, 56, 5307.
(c) J. Otera, T. Yano, A. Kawabata, H. Nozaki, Tetrahedron Lett.
1986, 27, 2383. doi:10.1016/S0040-4039(00)84535-9
[13] (a) G. N. Walker, J. Am. Chem. Soc. 1956, 78, 3698.
(b) H. R. Snyder, E. P. Merica, C. G. Force, E. G. White,
J. Am. Chem. Soc. 1958, 80, 4622.
[2] (a) G. W. Gribble, S. J. Berthel, in Studies in Natural Products
Chemistry (Ed. Atta-ur-Rahman) 1993, Vol. 12, pp. 365–409
(Elsevier Science: Amsterdam).
(b) U. Pindur, Y.-S. Kim, F. Mehrabani, Curr. Med. Chem.
1999, 6, 29.
(c) M. Prudhomme, F. Anizon, P. Moreau, Recent Res. Dev.
Synth. Org. Chem. 1999, 2, 79.
(d) J. Bergman, T. Janosik, N. Wahlstrom, Adv. Heterocycl.
Chem. 2001, 80, 1.
(c) J. Marino, C. R. Hurt, Synth. Commun. 1994, 24, 839.
[14] S. Reich, A. Oganessian, Bull. Soc. Chim. Fr. 1917, 21, 117.
We were unable to obtain 17 by this procedure.
[15] Review: E. Winterfeldt, Synthesis 1975, 617.