Dirhodium(II) carboxylate catalyzed formation of 1,2,3-trisubstituted indoles from styryl azides

Article abstract of DOI:10.1002/anie.201308611

Dirhodium(II)-carboxylate complexes were discovered to promote the selective migration of acyl groups in trisubstituted styryl azides to form 1,2,3-trisubstituted indoles. The styryl azides are readily available in three steps from cyclobutanone and 2-iod

Full text of DOI:10.1002/anie.201308611

Angewandte
Chemie
DOI: 10.1002/anie.201308611
Heterocycles
Dirhodium(II) Carboxylate Catalyzed Formation of 1,2,3-
Trisubstituted Indoles from Styryl Azides**
Crystalann Jones, Quyen Nguyen, and Tom G. Driver*
Abstract: Dirhodium(II)-carboxylate complexes were discov-
ered to promote the selective migration of acyl groups in
trisubstituted styryl azides to form 1,2,3-trisubstituted indoles.
The styryl azides are readily available in three steps from
cyclobutanone and 2-iodoaniline.
T
he development of new processes to construct polysubsti-
tuted indoles continues to motivate the work of synthetic
groups because of the fundamental importance of this
structural motif across diverse disciplines.[1] Because of the
poor selectivity in Fischer-Indole-type processes, the synthesis
of 6- or 4-(poly)substituted indoles as single isomers remains
a synthetic challenge.[2] In particular, reactions that create the
À
indole through formation of two new C C bonds to the
nitrogen atom are exceedingly sparse.[3] Our laboratory has
focused on developing transition-metal-catalyzed methods
which transform vinyl or aryl azides into heterocycles by
Scheme 1. Development of a new metal-catalyzed electrocyclization/
migration tandem reactions of substituted styryl azides. MG=migrat-
[4]
À
exploiting the mechanism of C N bond formation. Our
investigation of b,b-disubstituted styryl azides (1) revealed
that 2,3-disubstituted indoles could be synthesized as single
isomers through a selective 1,2-shift from 3, a shift which
occurred after electrocyclization of the rhodium nitrene
(Scheme 1).[5] During aromatization to give the indole 5, we
recognized that the a-hydrogen migrates to the nitrogen
atom. At the conclusion of these studies we were curious to
determine what the effect would be on the reaction when this
a-hydrogen atom was replaced with an alkyl or aryl group. We
anticipated that trisubstituted styryl azides (6) would be ideal
substrates for this query because our migratorial aptitude
studies predict that the carbonyl group should preferentially
shift.[5b] Once reaction conditions are found to form a metal
aryl nitrene from 6, a cascade reaction might then be triggered
to produce the N-heterocycles 7–9. If the reactivity of 6
mirrors that of the b,b-disubstituted styryl azide 1 then either
7 or 8 would be produced. In contrast to our expectations,
however, we report herein that rhodium(II) carboxylate
complexes catalyze the formation of the 1,2,3-trisubstituted
indoles 9, as single isomers from trisubstituted styryl azides
through exclusive carbonyl migration. The substitution pat-
ing group.
tern embedded in 9 is present in a variety of natural products
including the Strychnos and Kopsia alkaloid families.[6,7]
To discover the optimal reaction conditions for metal aryl
nitrene formation, the reactivity of the trisubstituted styryl
azide 11 towards a range of transition-metal complexes was
examined (Table 1). This azide is readily constructed in two
steps through a palladium-catalyzed Heck cross-coupling
ring-expansion reaction between 2-iodoaniline and the cyclo-
butanol 10, with subsequent azidation of the resulting ani-
line.[8,9] The trisubstituted styryl azide 11 proved to be less
reactive than the b,b-disubstituted styryl azides we previously
investigated: effervescence was not observed at 808C (entry 1
and 2). Increasing the temperature to 1008C, however,
produced the 1,2,3-trisubstituted indole 12 as the only product
when 11 was exposed to rhodium carboxylates (entries 3–
6).[10] Among the rhodium complexes surveyed, we found that
[Rh2(OAc)4] and [Rh2(esp)2][11] were the most efficient
catalysts for indole formation (entries 2 and 6). While we
attribute the success of [Rh2(esp)2] to the thermal robustness
inherent in its tetradentate ligands,[12] the activity of [Rh2-
(OAc)4] was surprising because its insolubility rendered it
impotent in our previous methods.[4,5] Its increased solubility
at elevated temperatures, however, appears to transform it
into a competent catalyst. On larger reaction scales, however,
we found that [Rh2(esp)2] outperformed [Rh2(OAc)4] to
provide the indole product with higher yields and greater
reproducibility. Lowering the catalyst loading or reducing the
reaction temperature led to attenuated yields of the 1,2,3-
trisubstituted indole product (entries 7 and 8). Examination
[*] C. Jones, Q. Nguyen, Prof. Dr. T. G. Driver
Department of Chemistry, University of Illinois at Chicago
845 W. Taylor St., Chicago (USA)
E-mail: tgd@uic.edu
Homepage: http://www.chem.uic.edu/driver
[**] We are grateful to the National Institutes of Health NIGMS
(R01GM984945) and the University of Illinois at Chicago for their
generous support.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201308611.
Angew. Chem. Int. Ed. 2014, 53, 785 –788
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
785

.
Angewandte
Communications
Table 1: Determination of optimal reaction conditions for indole
Table 2: Scope of rhodium(II) carboxylate catalyzed carbonyl shift.
formation.
Entry
11
R1
R2
R3
12
Yield [%][a]
Entry
[MXn]
mol%
T [8C]
Yield [%][a]
1
2
3
4
5
6
7
8
11a
11b
11c
11d
11e
11 f
11g
11h
11i
H
H
H
H
H
H
H
H
H
Me
Cl
H
H
H
H
H
H
H
H
H
87
93
96
93
95
95
81
90
97
87
70
1
2
3
4
5
6
7
8
none
n.a.
5
5
5
5
5
5
5
1
10
10
10
5
130
100
100
100
100
100
80
130
100
100
100
100
100
100
n.r.
60
76
90
89
70
n.r.
90
75
n.r.
42
35
38
54
OMe
Me
F
Cl
Br
CF3
NO2
H
H
H
[Rh2(O2CCF3)4]
[Rh2(O2CC3F7)4]
[Rh2(O2CCH3)4]
[Rh2(O2CC7H15)4]
[Rh2(esp)2]
[Rh2(esp)2]
[Rh2(esp)2]
[Rh2(esp)2]
CuI
9
10
11
H
H
OMe
9
11j
11k
10
11
12
13
14
FeBr2
RuBr3·nH2O
CoTPP
[a] Yield of isolated 12 after filtration through silica gel. It was the only
product obtained.
[{Ir(cod)(OMe)}2]
10
[a] As determined by 1H NMR spectroscopy using CH2Br2 as an internal
standard; only N-heterocycle obtained. esp=a,a,a’,a’-tetramethyl-1,3-
benzene-dipropionate. cod=cyclo-1,5-octadiene, esp=a,a,a’,a’-tetra-
methyl-1,3-benzenedipropionic acid, TPP=tetraphenylporphyrin.
product (entry 7). Using our optimal reaction conditions an a-
pyridyl substituent proved to be too Lewis basic: the pink
color of its reaction mixture indicated that it sequestered the
rhodium catalyst (entry 7). By raising the catalyst loading and
the reaction temperature, however, the yield of the indole 14g
was increased from 20% to 59%. The styryl azides 13h and
13i illustrated that the a-aryl group could be replaced with an
alkyl group without deleterious effects, thus illustrating the
generality of our method (entries 8 and 9). For the substrates
surveyed here, migration of the b-carbonyl group occurred
exclusively to the indole nitrogen atom to afford the 1,2,3-
trisubstituted indole 14 to illustrate the generality and
selectivity of our process.
A catalytic cycle for our rhodium(II) carboxylate cata-
lyzed carbonyl migration reaction can be proposed to account
for the exclusive formation of the 1,2,3-trisubstituted indole
product in our tandem electrocyclization/carbonyl migration
reaction (Scheme 2). Coordination of the styryl azide to the
rhodium carboxylate complex produces 15.[19] Extrusion of N2
forms the rhodium nitrene 16,[20] which undergoes a 4p-
electron-5-atom electrocyclization to form the spirocyclic
cation 17. While a 1,2 carbonyl shift from 17 could occur to
form the 2H-indole 18, this product was reported by Ban and
co-workers to be unstable and prone to fragmentation upon
exposure to Lewis or Brønsted acids.[21] Consequently, it
appears likely that the Lewis-acidic rhodium(II) carboxylate
triggers fragmentation of 17 or 18 to produce the acylium ion
19. This acylium ion could be attacked by the indole at the C3-
position to reform the kinetic,[22] but unstable 18,[23] or at the
anionic nitrogen atom to afford the thermodynamic indole
product 12 and regenerate the rhodium(II) carboxylate
catalyst.
of other established nitrogen-atom-transfer catalysts, includ-
ing copper,[13] iron,[14] ruthenium,[15] cobalt,[16] or iridium[17]
complexes did not result in improved reaction outcomes
(entries 9–13). Changing the reaction media to ethereal,
chlorinated, or other aromatic solvents also did not increase
the yield of the migration reaction.[18]
Using [Rh2(esp)2] as the catalyst, a series of styryl azides
(11) were examined to determine the scope and limitations of
our carbonyl 1,2-shift reaction (Table 2). Our optimal reac-
tion conditions enabled the indole product to be purified by
simply filtering the reaction mixture through a pipette column
of silica gel. We found that a range of electron-releasing and
electron-withdrawing R1 substituents are tolerated in our
reaction to give the 1,2,3,5-substituted indoles 12a–h
(entries 1–8). Even substrates bearing strong electron-with-
drawing substituents, such as the nitro group, were efficiently
converted into the indole product. Pleasingly, our reaction
also tolerates R2 substituents to enable access to the 6-
substituted indoles 12i and 12j, which cannot be made
regioselectively using the Fischer–Indole reaction (entries 9
and 10).[2] Finally, the influence of increasing the steric
environment around the azide was examined using 11k
(entry 11). This reaction proceeded smoothly, albeit slower, to
afford 12k in 70% yield.
Next, the identity of the a-substituent on the styryl azide
13 was varied to determine its influence on the reaction
(Table 3). These substrates were readily synthesized by simply
varying the identity of the magnesium acetylide substituent.[8]
We found that changing the a-aryl group, including increasing
its steric nature, did not change the outcome of reaction
(entries 1–6). In contrast to our earlier studies, substrates
bearing Lewis-basic substituents could be converted into
To investigate the mechanism of the 1,2-acyl shift,
a double crossover experiment was performed using the
styryl azides 20 and 21 [Eq. (1)]. We anticipated that if N-
acylation was slow that the acylium ion might diffuse and four
indole products would be observed. If N-acylation occurred
786
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 785 –788

Angewandte
Chemie
Table 3: Effect of changing the a-substituent on the reaction yield.
Entry 13
14
Yield [%][a]
1[c]
13a
13b
13c
13d
13e
14a 77
2
3
4
5
14b 89
14c 75
14d 85
14e 72
Scheme 2. Potential catalytic cycle.
In conclusion, we discovered that rhodium(II) carboxylate
complexes catalyze the formation of 1,2,3-trisubstituted
indoles from a,b,b-trisubstituted styryl azides through the
exclusive migration of the b-carbonyl group. The requisite
styryl azides are readily available in two steps from 2-
iodoaniline, and our reaction tolerates a range of different
substituents at the a-position as well as on the aryl azide.
Together these attributes combine for an efficient synthesis of
polysubstituted indoles, which cannot be accessed selectively
using Fischer-Indole-type processes. Our future work will
target exploiting this new reactivity in the design of methods
which access complex, functionalized N-heterocycles from
readily available trisubstituted styryl azides.
6
7
13 f
14 f 96
Received: October 2, 2013
Published online: November 26, 2013
13g
14g 20 (59)[b]
Keywords: azides · electrocyclic reactions · indole · rhodium ·
.
synthetic methods
8
9
13h
13i
14h 88
[1] For reviews, see: a) G. W. Gribble, J. Chem. Soc. Perkin Trans.
1 2000, 1045; b) S. Cacchi, G. Fabrizi, Chem. Rev. 2005, 105, 2873;
c) C. R. Edwankar, R. V. Edwankar, O. A. Namjoshi, S. K.
Rallapappi, S. J. Yang, J. M. Cook, Curr. Opin. Drug Discovery
Dev. 2009, 12, 752; d) D. F. Taber, P. K. Tirunahari, Tetrahedron
2011, 67, 7195.
[2] For a discussion on the limits of regioselectivity in the Fischer
indole reaction, see: a) R. R. Phillips, Org. React. 1959, 10, 1143;
b) B. Robinson, Chem. Rev. 1969, 69, 227, and references therein.
[3] a) S. D. Edmondson, A. Mastracchio, E. R. Parmee, Org. Lett.
2000, 2, 1109; b) T. Jensen, H. Pedersen, B. Bang-Andersen, R.
Madsen, M. Jørgensen, Angew. Chem. 2008, 120, 902; Angew.
Chem. Int. Ed. 2008, 47, 888; c) L. Ackermann, A. Althammer, P.
Mayer, Synthesis 2009, 3493.
14i
55
[a] Yield of isolated 14 after chromatography on silica gel. It was the only
product obtained. [b] Obtained using 10 mol% of [Rh2(esp)2] at 1408C.
faster than diffusion, or was concerted then only two indole
products would be obtained. Exposure of 20 and 21 to the
reaction conditions produced only the indoles 22 and 23, thus
confirming that the 1,2-shift is either concerted or that N-
acylation occurs faster than diffusion.
[4] a) B. J. Stokes, H. Dong, B. E. Leslie, A. L. Pumphrey, T. G.
Driver, J. Am. Chem. Soc. 2007, 129, 7500; b) M. Shen, B. E.
Leslie, T. G. Driver, Angew. Chem. 2008, 120, 5134; Angew.
Chem. Int. Ed. 2008, 47, 5056; c) B. J. Stokes, K. J. Richert, T. G.
Angew. Chem. Int. Ed. 2014, 53, 785 –788
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
787

.
Angewandte
Communications
Driver, J. Org. Chem. 2009, 74, 6442; d) B. J. Stokes, T. G. Driver,
Eur. J. Org. Chem. 2011, 4071; e) A. L. Pumphrey, H. Dong, T. G.
Driver, Angew. Chem. 2012, 124, 6022; Angew. Chem. Int. Ed.
2012, 51, 5920.
[14] a) E. R. King, E. T. Hennessy, T. A. Betley, J. Am. Chem. Soc.
2011, 133, 4917; b) Q. Nguyen, T. Nguyen, T. G. Driver, J. Am.
Chem. Soc. 2013, 135, 620; c) E. T. Hennessy, T. A. Betley,
Science 2013, 340, 591.
[5] a) K. Sun, S. Liu, P. M. Bec, T. G. Driver, Angew. Chem. 2011,
123, 1740; Angew. Chem. Int. Ed. 2011, 50, 1702; b) B. J. Stokes,
S. Liu, T. G. Driver, J. Am. Chem. Soc. 2011, 133, 4702; c) C.
Kong, N. Jana, T. G. Driver, Org. Lett. 2013, 15, 824.
[15] a) W. G. Shou, J. Li, T. Guo, Z. Lin, G. Jia, Organometallics 2009,
28, 6847; b) H. Dong, R. T. Latka, T. G. Driver, Org. Lett. 2011,
13, 2726.
[16] a) H. Lu, H. Jiang, L. Wojtas, X. P. Zhang, Angew. Chem. 2010,
122, 10390; Angew. Chem. Int. Ed. 2010, 49, 10192; b) V.
Lyaskovskyy, A. I. O. Suarez, H. Lu, H. Jiang, X. P. Zhang, B.
de Bruin, J. Am. Chem. Soc. 2011, 133, 12264.
[17] a) K. Sun, R. Sachwani, K. J. Richert, T. G. Driver, Org. Lett.
2009, 11, 3598; b) Y. Nishioka, T. Uchida, T. Katsuki, Angew.
Chem. 2013, 125, 1783; Angew. Chem. Int. Ed. 2013, 52, 1739.
[18] Refer to the Supporting Information for more details.
[19] For leading crystal structures of metal azide complexes, see:
a) M. G. Fickes, W. M. Davis, C. C. Cummins, J. Am. Chem. Soc.
1995, 117, 6384; b) R. Waterman, G. L. Hillhouse, J. Am. Chem.
Soc. 2008, 130, 12628.
[20] For computational studies on the mechanism of related copper
and cobalt nitrenoid formation from azides, see Ref. [15b] and
Ref. [8a].
[21] a) Y. Ban, K. Yoshida, J. Goto, T. Oishi, J. Am. Chem. Soc. 1981,
103, 6990; b) Y. Ban, K. Yoshida, J. Goto, T. Oishi, E. Take-
da (nꢁe Ishigamori), Tetrahedron 1983, 39, 3657.
[22] The C3-position is established as the kinetic site of reaction in
indoles, see: a) F. Terrier, M.-J. Pouet, J.-C. Halle, S. Hunt, J. R.
Jones, E. Buncel, J. Chem. Soc. Perkin Trans. 2 1993, 1665; b) S.
Lakhdar, M. Westermaier, F. Terrier, R. Goumont, T. Boubaker,
A. R. Ofial, H. Mayr, J. Org. Chem. 2006, 71, 9088.
[23] Our efforts to trap either of the catalytic intermediates 17, 18, or
19 through the addition of a nucleophile (e.g. allyl stannane) or
by performing the reaction in 4-methoxyanisole were not
successful.
[6] “The Strychnos Alkaloids”: J. B. Hendrickson in The Alkaloids:
Chemistry and Physiology, Vol. 6 (Ed.: R. H. F. Manske),
Academic Press, New York, 1960, p. 179. For a recent review
of syntheses, see: M. Mori, Heterocycles 2010, 81, 259.
[7] Kopsia alkaloid Mersicarpine: T.-S. Kam, G. Subramaniam, K.-
H. Lim, Y.-M. Choo, Tetrahedron Lett. 2004, 45, 5995; Syntheses,
see: a) J. Magolan, C. Carson, M. A. Kerr, Org. Lett. 2008, 10,
1437; b) R. Nakajima, T. Ogino, S. Yokoshima, T. Fukuyama, J.
Am. Chem. Soc. 2010, 132, 1236; c) Y. Iwama, K. Okano, K.
Sugimoto, H. Tokuyama, Org. Lett. 2012, 14, 2320.
[8] a) R. C. Larock, C. K. Reddy, Org. Lett. 2000, 2, 3325; b) R. C.
Larock, C. K. Reddy, J. Org. Chem. 2002, 67, 2027.
[9] K. Barral, A. D. Moorhouse, J. E. Moses, Org. Lett. 2007, 9, 1809.
[10] For previous reports of the synthesis of this indole scaffold, see:
a) A. Fꢀrstner, D. N. Jumbam, Tetrahedron 1992, 48, 5991; b) J.
Liu, M. Shen, Y. Zhang, G. Li, A. Khodabocus, S. Rodriguez, B.
Qu, V. Farina, C. H. Senanayake, B. Z. Lu, Org. Lett. 2006, 8,
3573; c) J. Liu, Y. Zhang, G. Li, F. Roschangar, V. Farina, C. H.
Senanayake, B. Z. Lu, Adv. Synth. Catal. 2010, 352, 2667.
[11] C. G. Espino, F. W. Fiori, M. Kim, J. Du Bois, J. Am. Chem. Soc.
2004, 126, 15378.
[12] D. Zalatan, J. Du Bois, J. Am. Chem. Soc. 2009, 131, 7558.
[13] a) Y. Badiei, A. Dinescu, X. Dai, R. M. Palomino, F. W.
Heinemann, T. R. Cundari, T. H. Warren, Angew. Chem. 2008,
120, 10109; Angew. Chem. Int. Ed. 2008, 47, 9961; b) R. T.
Gephart III, D. L. Huang, M. J. B. Aguila, G. Schmidt, A. Shahu,
T. H. Warren, Angew. Chem. 2012, 124, 6594; Angew. Chem. Int.
Ed. 2012, 51, 6488.
788
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 785 –788