A practical synthesis of indoles via a Pd-catalyzed C-N ring formation

Article abstract of DOI:10.1021/ol5018118

A method for the synthesis of N-functionalized C2-/C3-substituted indoles via Pd-catalyzed C-N bond coupling of halo-aryl enamines is described. The general strategy utilizes a variety of amines and β-keto esters which are elaborated into halo-aryl enamin

Full text of DOI:10.1021/ol5018118

Letter
pubs.acs.org/OrgLett
A Practical Synthesis of Indoles via a Pd-Catalyzed C−N Ring
Formation
Rishi G. Vaswani,* Brian K. Albrecht, James E. Audia,‡ Alexandre Cote, Les A. Dakin,§ Martin Duplessis,
̂
́
Victor S. Gehling, Jean-Christophe Harmange,‡ Michael C. Hewitt, Yves Leblanc,
Christopher G. Nasveschuk, and Alexander M. Taylor
†Department of Chemistry, Constellation Pharmaceuticals, Inc., 215 First Street, Suite 200, Cambridge, Massachusetts 02142, United
States
‡Constellation Pharmaceuticals, Inc. 215 First Street, Suite 200, Cambridge, Massachusetts 02142, United States
S
* Supporting Information
ABSTRACT: A method for the synthesis of N-functionalized
C2-/C3-substituted indoles via Pd-catalyzed C−N bond
coupling of halo-aryl enamines is described. The general
strategy utilizes a variety of amines and β-keto esters which are
elaborated into halo-aryl enamines as latent precursors to
indoles. The preferred conditions comprising the RuPhos
precatalyst and RuPhos in the presence of NaOMe in 1,4-
dioxane tolerate a variety of substituents and are scalable for the construction of indoles in multigram quantities.
Scheme 1. General Strategy
he indole nucleus is among the most ubiquitous structural
Tmotifs found in nature. Additionally, the indole
architecture has consistently proven to be a valuable and
privileged scaffold for drug discovery and development efforts
in the pharmaceutical industry.1 Consequently, methods
devoted to the chemical synthesis of indoles have been an
area of active research for over a century.2
As part of our ongoing drug discovery and development
programs, we were interested in the generation of C2-/C3-
substituted indole scaffolds with systematic variations of the N-
substituent.3 While numerous classical methods for the
construction of indoles have been well-established, the use of
transition-metal-catalyzed reactions, in particular palladium
catalysis, for the production of the indole nucleus has gained
significant attention in recent years.2a In the context of
palladium-catalyzed indole formation, the intramolecular C−
N bond arylation of enamines onto pendant aryl halides was of
particular interest.4,5 We reasoned that appropriately function-
alized halo-aryl enamines could serve as precursors for a Pd-
mediated C−N bond arylation reaction in the production of
substituted indole scaffolds (Scheme 1). Interestingly, similar
C−N bond cyclizations utilizing either iron6 or copper7
catalysts have previously been disclosed.8 We anticipated that
the use of Pd catalysis, generally tolerant of a wide range of
functionalities and applicable to complex molecules, coupled
with a modular incorporation of substituted amines onto acyclic
carbon frameworks, should allow for the production of C2-/
C3-substituted indoles with varying N-substituents. Herein, we
report an efficient, highly reactive, and scalable Pd-catalyzed
intramolecular C−N bond formation protocol for the
construction of indoles, under operationally simple conditions
with fast reaction times (NaOMe, 1,4-dioxane, 100 °C).
We initially prepared N-benzyl enamine 7 as a model
substrate in order to screen a variety of palladium sources and
commercially available ligands for the intramolecular C−N
bond formation (Scheme 2). To that end, the lithium enolate
of o-bromophenylacetic ester was subjected to a Claisen
condensation reaction with acetylimidazole to provide β-keto
Scheme 2. Generation of Halo-aryl Enamines
Received: June 23, 2014
Published: July 28, 2014
© 2014 American Chemical Society
4114
dx.doi.org/10.1021/ol5018118 | Org. Lett. 2014, 16, 4114−4117

Organic Letters
Letter
ester 6.9 The condensation of benzylamine with β-keto ester 6
in the presence of AcOH and t-BuOH (or EtOH) cleanly
afforded enamine 7 as a single isomer.10 Similarly, the aryl
chloride and aryl iodide derived enamines, 10 and 13
respectively, were generated using the reaction sequences
described within Scheme 2. Enamines 7, 10, and 13 were
assigned as Z-isomers based on the observation of the
downfield chemical shift of the N−H proton (8.50−11.50
ppm) within the respective 1H NMR spectra. The large
downfield chemical shift of the N−H proton can be attributed
to the presence of a hydrogen bond between the N−H proton
and the ester carbonyl.11,12
LCMS) and complete conversion to 14 was only achieved after
aging the reaction mixture overnight at 100 °C. In contrast to
the complex generated in situ from Pd2(dba)3 and t-Bu3P·HBF4
(see entry 7), the cyclization of enamine 7 with the
commercially available [t-Bu3P−Pd−Br]2 proved markedly
inferior (entry 2). Buchwald’s XPhos precatalyst (Gen III)
system (entry 1), Pd2(dba)3 with the DavePhos ligand (entry
3), and Pd(dppf)Cl2·CH2Cl2 (entry 5) were also far less
effective catalysts for this transformation, affording only 60%−
70% conversion to 14 with increasing quantities of proto-
dehalogenation byproduct 15. Other commonly utilized ligands
for C−N bond couplings, such as XantPhos and ( )-BINAP,
provided the lowest levels of conversion within the 1 h time
frame (entries 4 and 6, respectively).15
With enamine 7 in hand, we initiated our catalyst screen for
the intramolecular C−N bond formation (Table 1). Buchwald’s
Although the RuPhos precatalyst (Gen III) and the
Pd2(dba)3 and t-Bu3P·HBF4 combination (1:1 Pd/Ln mol
ratio) were equally effective catalyst systems for the cyclization,
we chose to focus on the RuPhos precatalyst (Gen III) system
and examined the incorporation of various amines to yield N-
substituted indoles under our reaction conditions (Scheme
3).16 The cyclization of enamine 7 onto the pendant aryl
Table 1. Catalyst Screen for Intramolecular C−N Bond
a
Formation
Scheme 3. Examination of Amines on the Indole Ring
Formation
(%)
conversion
ratio of
a
a
b
entry Pd (2 mol %) Ln (2 mol %)
14:15:7
1
2
3
4
5
6
7
8
9
XPhos precat.
XPhos
−
DavePhos
XantPhos
−
( )-BINAP
t-Bu3P·HBF4
−
72
27
5:1:1
5:0:14
100:6:1
10:1:8
5:1:5.5
33:1:3
44:1:0
100:1:0
70:1:0
11:1:0
c
[t-Bu3PPdBr]2
d
Pd2(dba)3
74
d
Pd2(dba)3
16
Pd(dppf)Cl2
62
d
Pd2(dba)3
2
e
e
e
d
Pd2(dba)3
100
100
100
RuPhos precat.
RuPhos precat. RuPhos
d
f
10
Pd2(dba)3
RuPhos
68
a
Percent conversion was monitored by LCMS at 215 nm and reported
as conversions (uncorrected) relative to starting enamine 7 after 1 h.
b
1
Relative ratios of 14:15:7 were measured by H NMR integration of
c
benzylic CH2 protons after reaction was allowed to age over 24 h. 1
d
mol % of [t-Bu3PPdBr]2 was used. 1 mol % of Pd2(dba)3 was used.
e
f
Consumption of enamine 7 occurred at <2 h. Percent conversion
reported by LCMS at 215 nm as conversions (uncorrected) relative to
g
starting enamine 7 after 2 h. The stereochemistry of the resultant des-
halo enamine 15 was not determined.
a
Unless otherwise noted, cyclizations were performed on aryl
bromides (i.e., X = Br). Yields were recorded for isolated products
b
after an average of two experiments. The cyclization of methyl 3-
(benzylamino)-2-(2-chlorophenyl)but-2-enoate was heated to 100 °C
c
for 2 h.
See Supporting Information for the synthesis of
corresponding enamine from trifluoroethylammonium hydrochloride.
RuPhos precatalyst (Gen III), either in the absence or presence
of exogenous sources of the RuPhos ligand, provided clean and
complete conversions of enamine 7 to the desired indole 14
within 1 h (entries 8 and 9, respectively).13 Similarly, a catalyst
system derived from Pd2(dba)3 and t-Bu3P·HBF4 (1:1 Pd/Ln
mol ratio) also afforded complementary conversion to 14
within 1 h (entry 7).14 Under either set of conditions only trace
quantities of proto-dehalogenation product 15 was observed.
The cyclization reaction utilizing Pd2(dba)3 and the RuPhos
ligand, while productive, afforded slower conversions to 14
(entry 10). After 2 h, 68% conversion to 14 was observed (by
bromide afforded the desired indole 14 in 95% yield.
Additionally, aryl iodides and aryl chlorides proved to be
competent substrates for the C−N bond arylation, affording the
indole 14 in 90% and 64% yields, respectively. In the C−N
bond formation with aryl chloride typically 85% conversion to
the desired indole was observed with ∼15% of starting material
remaining unreacted (as judged by LCMS analysis). Increases
in the catalyst loading or quantities of base failed to advance the
conversion beyond 85% (as judged by LCMS analysis).
Nonetheless, aryl chloride proved to be a viable substrate
under the Pd-catalyzed intramolecular C−N bond formation, in
4115
dx.doi.org/10.1021/ol5018118 | Org. Lett. 2014, 16, 4114−4117

Organic Letters
Letter
contrast to the previously reported Cu-catalyzed indole
formation.
synthetically useful yields (see indoles 27 and 28). Interest-
ingly, the inclusion of the C3-nitrile functionality was also
successful. The cyclization of 3-(benzylamino)-2-(2-bromo-
phenyl)but-2-enenitrile afforded 1-benzyl-2-methyl-1H-indole-
3-carbonitrile (29) in 86% yield.
From the outset of this methodology, our goal was to
develop a practical and scalable protocol for the construction of
N-substituted indoles. Using the optimal procedure(s) outlined
above, we set out to examine the feasibility of the current
method for the large scale production of N-substituted 2-
methyl-1H-indole-3-carboxylate; of particular interest to us was
the construction of methyl (R)-1-(1-(1-(tert-butoxycarbonyl)-
piperidin-4-yl)ethyl)-2-methyl-1H-indole-3-carboxylate
(Scheme 5). To implement this vision, β-keto ester 6 was
Generally, a variety of sterically encumbered or electronically
deactivated amines were tolerated and incorporated into
indoles (Scheme 3). Of particular interest was the use of chiral
branched amines for the generation of chiral N-substituted
indoles. Gratifyingly, (S)-(−)-α-methylbenzylamine was cleanly
assimilated to afford indole 16 in 85% yield with no net loss of
stereochemical integrity under these conditions (99% ee). N-
Alkyl amines with attenuated reactivity were converted to N-
alkyl indoles in generally acceptable synthetic yields (substrates
20−22).17 Additionally, aromatic amines such as anilines were
also examined for the C−N bond formation. Electron-neutral
or -deficient anilines afforded indoles, 17 and 19 respectively, in
>90% yields. The incorporation of electron-rich aniline, such as
p-anisidine, provided indole 18 in lower but yet synthetically
viable yields (76% yield).
Scheme 5. Large Scale Pd-Catalyzed C−N Bond Cyclization
The effect of varying the electronic or the steric environ-
ments for the Pd-catalyzed intramolecular C−N bond
formation was subsequently investigated (Scheme 4). The
Scheme 4. Indole Formation with Various Aromatic
a
Systems
condensed with chiral amine18 30 to yield enamine 31.
Enamine 31 was subsequently subjected to the Pd-catalyzed
intramolecular C−N bond formation to cleanly afford indole 32
on greater than 100 g scales. The N-Boc piperidine 33 was
deprotected with HCl to furnish the chiral indole piperidine
hydrochloride salt 33 in 80% yield over two steps. The
operationally simple approach allowed for the production of
requisite indole 33 on a large scale.
In summary, we have developed a practical, robust, and
scalable Pd-catalyzed intramolecular C−N bond forming
strategy for the construction of N-functionalized C2-/C3-
substituted indole scaffolds. We demonstrated that enamines
derived from a broad variety of amines (alkyl, branched/chiral,
or aromatic) can be assimilated into indole rings with ease.
Additionally, indoles with diverse substitutions (around the
aromatic ring and C2-positions) and electronic properties may
also be synthesized. Notably, we have also established that the
existing protocol can be used effectively in the large scale
production of indoles.
a
Yields were recorded for isolated products after an average of two
b
experiments. Isolated as an average of 9:1 mixture of 6-chloroindole
23: des-chloroindole 14. Reaction was only performed once on 500
mg scale, and yield was recorded after one reaction.
c
C−N bond cyclization of enamine substrates with orthogonally
reactive aryl halides to yield halo-indoles was also examined.
For example, the cyclization of an enamine substrate that
contained a 2-bromo-4-chlorophenyl motif afforded the 6-
chloroindole 23 in 75% yield, as an average of a 9:1 mixture of
23 to des-chloroindole 14. The intramolecular C−N bond
formation onto both electron-deficient and -rich aryl bromides
provided the corresponding substituted indoles, 24 and 25 in
78% and 77% yields, respectively. These results suggest that
electronic factors inherent within the aromatic ring contribute
minimally during the cyclization event.
Notably, the current synthetic scheme allows for the
controlled incorporation of varying substituents at the C2-
position of indoles. Using the general protocol, the des-methyl
1H-indole 3-carboxylate ester 26 could be produced in 84%
yield, comparable to previously established Cu-catalyzed
protocols (see ref 7b). Further diversification of the C2-
position with other alkyl groups was also achieved in
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental procedures, compound characterization, and
spectral data for all compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
4116
dx.doi.org/10.1021/ol5018118 | Org. Lett. 2014, 16, 4114−4117

Organic Letters
Letter
(9) Other activated acids (such as anhydrides or esters) have also
been utilized for the Claisen Condensation reaction. See Supporting
Information for details.
AUTHOR INFORMATION
Corresponding Author
■
(10) The use of other lower boiling solvents, such as MeOH,
generally led to lower yields and conversion of the enamine. t-BuOH
and EtOH afforded the cleanest and best yield for the enamine
formation.
(11) The observation of downfield chemical shifts of hydrogen
bonded N−H protons on similar substrates have previously been
reported; see: (a) Du, Y.; Liu, R.; Linn, G.; Zhao, K. Org. Lett. 2006,
*E-mail: rishi.vaswani@constellationpharma.com.
Present Address
§Pfizer, Inc., 610 Main St., Cambridge, MA 02412.
Notes
The authors declare no competing financial interest.
26, 5919. (b) Casarrubios, L.; Per
J. L. J. Org. Chem. 1996, 61, 8358.
́
ez, J. A.; Brookhart, M.; Templeton,
ACKNOWLEDGMENTS
■
(12) While enamines 7, 10, and 13 were isolated as single isomers,
several enamines generated within this manuscript were isolated as
mixtures of Z- versus E-isomers. The mixtures Z- versus E-enamines
proved to be competent substrates in the Pd-mediated cyclization. See
Supporting Information for examples.
(13) For references on the use of diakylbiaryl phosphine ligands in
Pd-catalyzed C−N bond formations, see (a) Surry, D. S.; Buchwald, S.
L. Chem. Sci. 2011, 2, 27. (b) Maiti, D.; Fors, B. P.; Henderson, J. L.;
Nakamura, Y.; Buchwald, S. L. Chem. Sci. 2011, 2, 57. (c) Fors, B. P.;
Buchwald, S. L. J. Am. Chem. Soc. 2010, 132, 15914. (d) Harris, M. C.;
Huang, X.; Buchwald, S. L. Org. Lett. 2002, 4, 2885.
(14) For references on the use of bulky electron-rich phosphine
ligands in Pd-catalyzed C−N bond formations, see: (a) Hartwig, J. F.;
Kawatsura, M.; Hauck, S. I.; Shaughnessy, K. H.; Alcazar-Roman, L. M.
J. Org. Chem. 1999, 64, 5575. (b) Nishiyama, M.; Yamamoto, T.; Koie,
Y. Tetrahedron Lett. 1998, 39, 617. (c) Stambuli, J. P.; Kuwano, R.;
Hartwig, J. F. Angew. Chem., Int. Ed. 2002, 41, 4746.
(15) (a) Klingensmith, L. M.; Streiter, E. R.; Barder, T. E.; Buchwald,
S. L. Organometallics 2006, 25, 82. (b) Michele, H. C.; Geis, O.;
Buchwald, S. L. J. Org. Chem. 1999, 64, 6019. (c) Yang, B. H.;
Buchwald, S. L. Org. Lett. 1999, 1, 35.
This manuscript is dedicated to Professor A. R. Chamberlin
(University of California, Irvine) on the occasion of his 65th
birthday. We would like to thank Dr. Jin Hong and Jake Zhao
from Custom NMR Services for assistance in acquisition of all
NMR spectra. We also acknowledge WuXi AppTech for
preparation of intermediates 6 and 30 on 100 g scales and for
chiral analysis of sample(s) via SFC.
REFERENCES
■
(1) Rodrigues de Sa
́
Alves, F.; Barreiro, E. J.; Fraga, C. A. M. Mini-
Rev. Med. Chem. 2009, 9, 782.
(2) For selected reviews on the syntheses of indoles, see: (a) Cacchi,
S.; Fabrizi, G. Chem. Rev. 2011, 111, PR215. (b) Cacchi, S.; Fabrizi, G.
Chem. Rev. 2005, 105, 2873. (c) Humphrey, G. R.; Kuethe, J. T. Chem.
Rev. 2006, 106, 2875. (d) Sundberg, R. J. Indoles; Academic Press:
London, 1996; p 175. (e) Taber, D. F.; Tirunahari, R. K. Tetrahedron
2011, 67, 7195. (f) Gribble, G. W. J. Chem. Soc., Perkin Trans. 1 2000,
1045. (g) Vicente, R. Org. Biomol. Chem. 2011, 19, 6469. (h) Inman,
M.; Moody, C. J. Chem. Sci. 2013, 4, 29.
(3) Our initial attempts to functionalize the relatively non-
nucleophilic indole N−H group via direct alkylation, thereby varying
the N-substituent, proved to be problematic (not shown). As such, we
were prompted to seek alternative synthetic strategies to achieve this
goal.
(4) For examples of related Pd-catalyzed C−N bond syntheses of
indoles, see: (a) Yamazaki, K.; Nakamura, Y.; Kondo, Y. J. Org. Chem.
2003, 68, 6011. (b) Yamazaki, K.; Nakamura, Y.; K, Y. J. Chem. Soc.,
Perkin Trans. 1 2002, 2137. (c) Kim, J. H.; Lee, S.-g. Synthesis 2012,
44, 1464. (d) Brown, J. Tetrahedron Lett. 2000, 41, 1623. (e) Kim, J.
H.; Lee, S.-g. Org. Lett. 2011, 13, 1350. (f) Willis, M. C.; Brace, G. N.;
Holmes, I. P. Angew. Chem., Int. Ed. 2005, 39, 2501. (g) Fletcher, A. J.;
Bax, M. N.; Willis, M. C. Chem. Commun. 2007, 4764. (h) Willis, M.
C.; Brace, G. N.; Findlay, T. J. K.; Holmes, I. P. Adv. Synth. Catal.
2006, 348, 851. (i) Siebeneicher, H.; Bytschkov, I.; Doye, S. Angew.
Chem., Int. Ed. 2003, 42, 3042. (j) Barluenga, F.; Jimen
Valdes, C.; Aznar, F. Angew. Chem., Int. Ed. 2007, 46, 1529.
(k) Barluenga, F.; Jimenez-Aquino, A.; Valdes, C.; Aznar, F. J. Am.
Chem. Soc. 2009, 131, 4031. (l) Barluenga, F.; Jimenez-Aquino, A.;
Valdes
(16) In our hands, addition of exogenous amounts of RuPhos ligand
afforded more reproducible results for the indole formation.
(17) Our attempts to condense tert-butylamine or 1-adamantlyamine
with β-keto ester 6 failed to deliver the corresponding tert-butyl or
adamantyl enamine for Pd-catalyzed cyclization. This failure in
enamine formation was attributed to considerably attenuated
nucleophilicity of these amines and an increase in the steric hindrance
in the substrates.
(18) (a)
Chiral amine 30 was synthesized according to the
procedure outlined by: Albrecht, B. K.; Audia, J. E.; Cook, A. S.;
Dakin, L. A.; Duplessis, M.; Gehling, V. S.; Harmange, J.-C.;
Nasveschuk, C. G.; Vaswani, R. G. PCT Int. Appl. (2013), WO
2013120104 A2 Aug 15, 2013. (b) Chiral amine 30 was resolved
according to the procedure outlined by: Mortimore, M.; Young, S. C.;
Everitt, S. R. L.; Knegtel, R.; Pinder, J. L.; Rutherford, A. P.; Durrant,
S.; Brenchley, G.; Charrier, J. D.; O’Donnell, M. PCT Int. Appl.
(2008), WO 2008079346 A1 Jul 03, 2008.
́
ez-Aquino, A.;
́
́
́
́
́
, C.; Aznar, F. Chem.Eur. J. 2010, 16, 11707.
(5) For reviews on the C−N bond arylation, see: (a) Bariwalab, J.;
Van der Eycken, E. Chem. Soc. Rev. 2013, 42, 9283. (b) Fischer, C.;
Koenig, B. Beilstein J. Org. Chem. 2011, 7, 59. (c) Wolfe, J. P.; Wagaw,
S.; Marcoux, J.-F.; Buchwald, S. L. Acc. Chem. Res. 1998, 31, 805.
(6) For references on the construction of indoles via Fe-catalyzed
intramolecular amination of aryl bromides, see: Melkonyan, F. S.;
Topolyan, A. P.; Karchava, A. V.; Yurovskaya, M. A. Chem. Heterocycl.
Compd. 2010, 46, 1158.
(7) For references on the construction of indoles via Cu-catalyzed
intramolecular amination of aryl bromides, see: (a) Melkonyan, F. S.;
Kuznetsov, D. E.; Yurovskaya, M. A.; Karchava, A. V. RSC Adv. 2013,
3, 8388. (b) Melkonyan, F. S.; Karchava, A. V.; Yurovskaya, M. A. J.
Org. Chem. 2008, 73, 4275. (c) Besandre, R.; Jaimes, M.; May, J. A.
Org. Lett. 2013, 15, 1666.
(8) While the Cu catalyzed reactions offer the advantage of
inexpensive reaction conditions, the transformation is limited by the
use of anhydrous bases (see ref 7c) and aryl bromides (see ref 7a−c).
4117
dx.doi.org/10.1021/ol5018118 | Org. Lett. 2014, 16, 4114−4117