Expanding the [1,2]-aryl migration to the synthesis of substituted indoles

Article abstract of DOI:10.1002/anie.200705804

(Chemical Equation Presented) Closing in on the product: A range of 2- and 2,3-substituted indoles 2 have been synthesized in moderate to excellent yields by the addition of carbon nucleophiles (organometallic reagents, RM) to readily accessible 1-(2-aminophenyl)-2-chloroethanones of type 1 under mild conditions (see scheme). A mechanism has been proposed that involves a unique [1,2]-aryl migration of intermediate 3.

Full text of DOI:10.1002/anie.200705804

Angewandte
Chemie
DOI: 10.1002/anie.200705804
Synthetic Methods
Expanding the [1,2]-Aryl Migration to the Synthesis of Substituted
Indoles**
Tao Pei,* Cheng-yi Chen,* Peter G. Dormer, and Ian W. Davies
The indole scaffold is a prevalent substructure of many
natural products and biologically active compounds.[1] The
need for efficient and practical syntheses of indoles bearing a
variety of substitution patterns provides a continual challenge
to organic chemists. Despite the many diverse and creative
approaches that have been used to assemble the indole
nucleus,[2] a general synthesis of indoles with control over the
regioselective introduction of substituents at C2 and C3, is of
tantamount importance. Herein we report a synthesis of
substituted indoles 2 from readily accessible chloroacetophe-
nones 1,[3] which contain a 1-(2-aminophenyl)-2-chloroetha-
none core structure, and commercially available organome-
tallic reagents [RM, Eq. (1)]. Of particular significance is the
alkyl Grignard reagents (RM) reacted with 3 to give C2-
substituted indoles in excellent yields (entries 2 and 3).
Table 1: Synthesis of 2-substituted indoles from ketone 3.[a]
Entry
1[f]
RM
Product
Yield [%][g]
81
5a
5b
5c
5d
5e
5 f
5g
5h
5i
2[f]
89
86
45
76
70
91
78
76
72
3[e]
4[f]
generality of this reaction and the regioselectivity achieved
under mild reaction conditions, which makes this trans-
formation viable for the preparation of many structurally
diverse indoles. This efficient synthesis of indoles takes
advantage of a pivotal [1,2]-aryl rearrangement.
Our initial encounter of the feasibility of using a [1,2]-aryl
migration to deliver 2-substituted indoles was observed in the
reaction of nPrMgCl with 1-(2-amino-3-chlorophenyl)-2-
chloroethanone (3). Surprisingly, the sole product of the
reaction was indole 5a in 81% yield (Table 1, entry 1). We
anticipated that this efficient pathway involving a unique aryl
migration would provide the basis for the direct access to 2-
and 2,3-substituted indoles upon the appropriate choice of
organometallic species as the nucleophile.
5[b,e]
6[b,f]
7[e]
nHexLi
Me3SiCH2Li
PhMgCl
8[c,e]
9[c,e]
10[e]
To explore the generality of this transformation, the
addition of different organometallic compounds to ketone 3
was studied (Table 1). Similar to nPrMgCl, other primary
5j
[*] Dr. T. Pei, Dr. C.-Y. Chen, Dr. P. G. Dormer, Dr. I. W. Davies
Department of Process Research
MerckResearch Laboratories
11[d,e]
12[d,e]
5k, R=SiMe3
5l, R=nBu
54
55
Rahway, NJ 07065 (USA)
Fax: (+1)732-594-1499
E-mail: tao_pei@merck.com
[a] Reaction conditions: All reactions were carried out without optimi-
zation: Ketone 3 (1.0 mmol) in either THF or toluene (2 mL) and
nucleophile (2.5 mmol) were stirred at ꢀ108C for 15 min, then at room
temperature for 15 min to 2 h. [b] Addition of nucleophile at ꢀ408C.
[c] The nucleophile was generated in situ from the corresponding aryl
iodide and iPrMgCl. [d] The nucleophile was generated in situ from the
corresponding alkynes and iPrMgCl. [e] Solvent=THF. [f] Solvent=tol-
uene. [g] Yields refer to isolated product based on ketone 3.
cheng_chen@merck.com
[**] We would like to thank Dr. Tom J. Novak for assistance with the
HRMS analysis.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2008, 47, 4231 –4233
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4231

Communications
Moreover, the addition of organolithium species such as n-
The versatility of this method was further extended to the
preparation of 2,3-disubstituted indoles with excellent control
over the regioselectivity; these indoles would be difficult to
prepare selectively by other means. Hence, treatment of
ketone 8 with 2.5 equivalents of nPrMgCl at ꢀ108C!228C
afforded 7-chloro-3-methyl-2-propylindole (9a) in 93% yield.
Likewise, 7-chloro-3-methyl-2-thien-2-ylindole (9b) was
readily obtained in 80% yield by employing 2.0 equivalents
of 2-thienyllithium [Eq. (2)]. These particular examples
hexyllithium and trimethylsilylmethyllithium afforded indole
5e and 5 f, in yields of 76 and 70%, respectively (entries 5 and
6).
A number of aromatic and heteroaromatic functionalities
were readily introduced at the C2-position of the indoles by
the addition of aryl and heteroaryl magnesium reagents to
ketone 3. For example, 2-phenyl-7-chloroindole (5g) was
prepared in excellent yield (91%) using phenylmagnesium
chloride (entry 7). Heteroaryl nucleophiles, such as 2- and 3-
pyridinylmagnesium chloride, and 2-thienylmagnesium bro-
mide gave the corresponding 2-heteroaryl indoles (5h–j) in
yields of 78, 76, and 72%, respectively (entries 8–10).
Furthermore, 2-alkyn-1-ylindoles were directly synthesized
from alkynylmagnesium species (entries 11 and 12).
Having successfully employed various carbon nucleo-
philes for the preparation of a range of 2-substituted indoles
from ketone 3, we decided to extend our methodology to
other substituted ketones (Scheme 1). Not surprisingly, 1-(2-
clearly highlight the remarkable efficiency of this method,
which should be broadly applicable to the regioselective
synthesis of 2,3-disubstituted indoles from other substrates.
On the basis of the regiochemistry observed in this
transformation, we propose the following reaction sequence
(Scheme 2): nucleophilic addition of an organometallic
Scheme 1. Synthesis of 2-substituted indoles from ketone 6. All reac-
tions were carried out without optimization. Yields refer to isolated
material based on ketone 6. [a] Solvent=toluene. [b] Solvent=THF.
[c] RM=Grignard reagent. [d] RM=Organolithium reagent.
aminophenyl)-2-chloroethanone (6a, X, R’ = H) worked well
in the reaction, and afforded the 2-allyl- and 2-thien-2-
ylindoles (7a and 7b) in yields of 67 and 76%, respectively.
The presence of either electron-donating or electron-with-
drawing substituents on the phenyl ring had little impact on
the transformation. 1-(2-Amino-6-methoxyphenyl)-2-chlor-
oethanone (6c, X = 6-MeO, R’ = H) which contains an
electron-rich methoxy group was converted into 4-methoxy-
2-propylindole (7c) in 63% yield, while the substrate
containing a trifluoro-substituted phenyl ring reacted with
nPrMgCl to form 4,5,6-trifluoro-2-propylindole (7d) in 68%
yield. 5,7-Dimethyl-2-phenylethynylindole (7e) and 7-fluoro-
5-methyl-2-phenylindole (7 f) were readily prepared by the
addition of either phenylethynyl- or phenylmagnesium chlo-
ride, respectively. N Substitution was also tolerated in this
reaction, as demonstrated in the reaction of 1-(N-methyl-2-
aminophenyl)-2-chloroethanone (6g, X = H , R’ = Me) with
nPrMgCl to form N-methyl-2-propylindole (7g) in very high
yield (91%).
Scheme 2. Proposed mechanism illustrated with 6a as the starting
material.
reagent to the chloroketone 6a to form tertiary alkoxide 10,
followed by a facile [1,2]-aryl rearrangement[4] to form ketone
12, then ring-closure and dehydration to form 2-substituted
indole 7. We believe the aniline moiety serves as a key driver
for the success of the reaction, wherein the aniline nitrogen
atom promotes the net [1,2]-aryl rearrangement, either
through a conventional aryl migration (path a) or the
formation of a phenonium ion intermediate, such as 11,
which favors its migration (path b).[5,6] The aniline nitrogen
atom could also serve as an effective trap for the ketone in
intermediate 12, thus preventing reaction of the ketone with
an additional equivalent of the organometallic reagent.
This net [1,2]-aryl migration mechanism is supported by
two experiments. When the dianion 13[7] derived from N-2-
bromophenylpivaloylamide reacts with ketone 6a, the only
4232
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 4231 –4233

Angewandte
Chemie
product observed, 15, is derived from migration of the
unprotected aniline group. Under the assumption that the
intermediate 14 forms, there are two possible aniline groups
which could participate in this [1,2]-aryl migration. Clearly,
the electron-withdrawing effect of the N-pivaloyl group
disfavors its migration [Eq. (3)].[7] Another experiment
chloro ketones. This simple and mild procedure renders the
method a valuable addition to the arsenal of indole syntheses.
Received: December 18, 2007
Revised: February 25, 2008
Published online: April 29, 2008
Keywords: anilines · indoles · rearrangement · regioselectivity ·
.
synthetic methods
[1] For recent reviews on indole alkaloids and references to the
biological activity of compounds containing the indole substruc-
ture, see a) T. Kawasaki, K. Higuchi, Nat. Prod. Rep. 2007, 24, 843,
and references therein; b) J. E. Saxton in The Alkaloids, Vol. 51
(Ed.: G. A. Cordell), Academic Press, San Diego, 1998; c) J. E.
Saxton, Nat. Prod. Rep. 1997, 14, 559; d) A. Brancale, R. Silvestri,
Med. Res. Rev. 2007, 27, 209; e) S. Harper, S. Avolio, B. Pacini, M.
Di Filippo, S. Altamura, L. Tomei, G. Paonessa, S. Di Marco, A.
Carfi, C. Giuliano, J. Padron, F. Bonelli, G. Migliaccio, R.
De Francesco, R. Laufer, M. Rowley, F. Narjes, J. Med. Chem.
2005, 48, 4547.
which supports the proposed rearrangement mechanism
involves the reaction of deuterium-labeled alcohol 16 with
iPrMgCl in THF, which produces [2-D]-7-chloroindole (17) in
69% yield [Eq. (4)]. An alternative mechanism involving the
[2] For recent reviews on indole synthesis, see a) G. R. Humphrey,
J. T. Kuethe, Chem. Rev. 2006, 106, 2875; b) S. Cacchi, G. Fabrizi,
Chem. Rev. 2005, 105, 2873; c) G. Zeni, R. C. Larock, Chem. Rev.
2004, 104, 2285; d) J. A. Joule in Science of Synthesis, Vol. 10 (Ed.:
E. J. Thomas), Thieme, Stuttgart, 2000, p. 361; e) T. L. Gilchrist, J.
Chem. Soc. Perkin Trans. 1 2001, 2491; f) G. W. Gribble, J. Chem.
Soc. Perkin Trans. 1 2000, 1045; g) R. J. Sundberg, Indoles,
Academic Press, San Diego, 1996.
[3] T. Sugasawa, T. Toyoda, M. Adachi, K. Sasakura, J. Am. Chem.
Soc. 1978, 100, 4842.
[4] R. L. Huang, J. Org. Chem. 1954, 19, 1363.
[5] a) D. J. Cram, J. Am. Chem. Soc. 1949, 71, 3863; b) S. Winstein, R.
Baird, J. Am. Chem. Soc. 1957, 79, 756.
formation of an epoxide and subsequent rearrangement to
form an aldehyde is ruled out since it would lead to a 3-
substituted indole after a [1,2]-hydride shift.[8]
In summary, we have discovered a new and efficient
method for the regioselective synthesis of substituted indoles.
The reaction proceeds from readily available 1-(2-amino-
phenyl)-2-chloroethanones by a [1,2]-aryl rearrangement
followed by intramolecular condensation to form indoles.
The method introduces substituents at the C2-position of
indoles and tolerates different substitution patterns on a-
[6] The chlorohydrin was isolated in high yield in the reaction of 2-
chloroacetophenone with a Grignard reagent even after pro-
longed aging; see J. Barluenga, J. Florez, M. Yus, J. Chem. Soc.
Perkin Trans. 1 1983, 3019.
[7] The dilithium reagent generated from N-2-bromophenylpivaloy-
lamide reacted with a-chloroketone to form the 3-substituted
indole; see P. A. Wender, A. W. White, Tetrahedron 1983, 39,
3767.
[8] a) T. A. Geissman, R. I. Akawie, J. Am. Chem. Soc. 1951, 73,
1993; b) L. Crombie, R. Hardy, D. W. Knight, J. Chem. Soc. Perkin
Trans. 1 1985, 1373.
Angew. Chem. Int. Ed. 2008, 47, 4231 –4233
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4233