Interrupted fischer-indole intermediates via oxyarylation of alkenyl boronic acids

Article abstract of DOI:10.1021/ol401416r

The oxyarylation of alkenyl boronic acids with N-arylbenzhydroxamic acids has been achieved under both copper-mediated and copper-catalyzed conditions to provide access to interrupted Fischer-indole intermediates. This transformation is believed to proceed through a copper-promoted C-O bond forming event followed by a [3,3] rearrangement. The scope of the method is described and mechanistic experiments are discussed.

Full text of DOI:10.1021/ol401416r

ORGANIC
LETTERS
2013
Vol. 15, No. 13
3362–3365
Interrupted Fischer-Indole Intermediates
via Oxyarylation of Alkenyl Boronic Acids
Heng-Yen Wang and Laura L. Anderson*
Department of Chemistry, University of Illinois, Chicago, Illinois 60607, United States
lauralin@uic.edu
Received May 19, 2013
ABSTRACT
The oxyarylation of alkenyl boronic acids with N-arylbenzhydroxamic acids has been achieved under both copper-mediated and copper-catalyzed
conditions to provide access to interrupted Fischer-indole intermediates. This transformation is believed to proceed through a copper-promoted
CÀO bond forming event followed by a [3,3] rearrangement. The scope of the method is described and mechanistic experiments are discussed.
The Fischer-indole reaction is a powerful transforma-
tion that has been widely used to form new CÀC bonds
through the [3,3] rearrangement of an aryl hydrazone and
new CÀN bonds via the intramolecular 1,2-addition of an
aniline to an imine.1,2 Unfortunately, under the conditions
of the Fischer-indole reaction, it is impossible to separate
these two events. A method designed to halt the process
after CÀC bond formation would provide a simple route
to R-arylated ketones with o-amide substituents. These
compounds are challenging to synthesize due to the lack
of a general preparative method and because they repre-
sent a limitation of transition-metal-catalyzed R-arylation
procedures.3À7
Recently, we investigated the use of N-enoxyphthali-
mides as precursors to R-hydroxy ketones.8 To further
exploit O-alkenyl hydroxamates as rearrangement precur-
sors for the construction of challenging bonds, we decided
to target an alkenyl ester of N-phenylbenzhydroxamic acid
1a by copper-mediated CÀO bond coupling (Scheme 1).9
We were intrigued by this intermediate because we rea-
soned that 3aa could potentially undergo two distinct [3,3]
(1) For Fischer-indole reaction reviews, see: (a) Downing, R. S.;
Kunkeler, P. J. The Fischer Indole Synthesis. In Fine Chemicals through
Heterogeneous Catalysis; Sheldon, R. A., Van Bekkum, H., Eds.; Wiley-
VCH: Weinheim, 2001; pp 178À183. (b) Hughes, D. L. Org. Prep. Proc.
Int. 1993, 25, 607. (c) Ambekar, S. Y. Curr. Sci. 1983, 52, 578. (d)
Robinson, B. The Fischer Indole Synthesis: John Wiley & Sons: Hoboken,
1983.
(2) For a mechanistic study of the Fischer-indole reaction, see:
Hughes, D. L.; Zhao, D. J. Org. Chem. 1993, 58, 228.
(3) For reviews of transition-metal-catalyzed R-arylation reactions,
see: (a) Johansson, C. C. C.; Colacot, T. J. Angew. Chem., Int. Ed. 2009,
49, 676. (b) Mazet, C. Synlett. 2012, 23, 1999. (c) Culkin, D. A.; Hartwig,
J. F. Acc. Chem. Res. 2003, 36, 234.
(4) For selected examples of transition-metal-catalyzed R-arylation
of ketones, see: (a) Ge, S.; Hartwig, J. F. J. Am. Chem. Soc. 2011, 133,
16330. (b) Lou, S.; Fu, G. C. J. Am. Chem. Soc. 2010, 132, 1264. (c)
Hamada, T.; Chieffi, A.; Ahman, J.; Buchwald, S. L. J. Am. Chem. Soc.
2002, 124, 1261. (d) Hesp, K. D.; Lundgren, R. J.; Stradiotto, M. J. Am.
Chem. Soc. 2011, 133, 5194. (e) Huang, K.; Li, G.; Huang, W.-P.; Yu,
D.-G.; Shi, Z.-J. Chem. Commun. 2011, 47, 7224. (f) Kale, A. P.; Pawar,
G. G.; Kapur, M. Org. Lett. 2012, 14, 1808. (g) Danoun, G.; Tlili, A.;
Monnier, F.; Taillefer, M. Angew. Chem., Int. Ed. 2012, 51, 12815.
(5) For intramolecular examples of R-arylation of N-phenylamides to
form oxindoles, see: (a) Ju, X.; Liang, Y.; Jia, P.; Li, W.; Yu, W. Org.
Biomol. Chem. 2012, 10, 498. (b) Wu, L.; Falivene, L.; Drinkel, E.;
Grant, S.; Linden, A.; Cavallo, L.; Dorta, R. Angew. Chem., Int. Ed.
2012, 51, 2870. (c) Beyer, A.; Buendia, J.; Bolm, C. Org. Lett. 2012, 14,
3948.
(6) For examples of Cu-catalyzed R-arylation of malonates, β-keto-
esters, and 1,3-diones with o-haloanilides, see: (a) Chen, Y.; Wang, Y.;
Sun, Z.; Ma, D. Org. Lett. 2008, 10, 625. (b) Chen, Y.; Xie, X.; Ma, D.
J. Org. Chem. 2007, 24, 9329. (c) Xie, X.; Chen, Y.; Ma, D. J. Am. Chem.
Soc. 2006, 128, 16050. (d) Huang, Z.; Hartwig, J. F. Angew. Chem., Int.
Ed. 2012, 51, 1028.
(7) For a mechanistic discussion describing the inability to use Pd-
catalyzed R-arylation to convert an o-haloaryl amine to an R-(o-
aminoaryl)ketone and the inability to use a Buchwald-Hartwig reaction
to convert an R-(o-haloaryl)ketone to an R-(o-aminoaryl)ketone, see:
Knapp, J. M.; Zhu, J. S.; Tantillo, D. J.; Kurth, M. J. Angew. Chem., Int.
Ed. 2012, 51, 10588.
(8) Patil, A. S.; Mo, D.-L.; Wang, H.-Y.; Mueller, D. S.; Anderson,
L. L. Angew. Chem., Int. Ed. 2012, 51, 7799.
(9) For reviews and examples of O-alkenylation with boronic acids,
see: (a) Lam, P. Y. S.; Vincent, G.; Clark, C. G.; Deudon, S.; Jadhav,
P. K. Tetrahedron Lett. 2001, 42, 3415. (b) Lam, P. Y. S.; Vincent, G.;
Bonne, D.; Clark, C. G. Tetrahedron Lett. 2003, 44, 4927. (c) Qiao, J. X.;
Lam, P. Y. S. Synthesis 2011, 829. (d) Quach, T. D.; Batey, R. A. Org.
Lett. 2003, 5, 1381. (e) Shade, R. E.; Hyde, A. M.; Olsen, J.-C.; Merlic,
C. A. J. Am. Chem. Soc. 2010, 132, 1202. (f) Winternheimer, D. J.;
Merlic, C. A. Org. Lett. 2010, 12, 2508. (g) Chan, D. G.; Winternheimer,
D. J.; Merlic, C. A. Org. Lett. 2011, 13, 2778. (h) Winternheimer, D. J.;
Shade, R. E.; Merlic, C. A. Synthesis 2010, 2497.
r
10.1021/ol401416r
Published on Web 06/25/2013
2013 American Chemical Society

rearrangements to give either a new CÀC bond 4aa or a
new CÀO bond 5aa.10 To our delight, we discovered that
treatment of 1a with cyclohexenyl boronic acid 2a in the
presence of CuBr exclusively gave R-arylated ketone 4aa.
This transformation implies that copper-mediated CÀO
bondformation triggersa spontaneous and chemoselective
[3,3] rearrangement to provide direct access to isolable
interrupted Fischer-Indole intermediates.11 Herein we
present the optimization and scope of this operationally
convenient method for the oxyarylation of alkenyl boronic
acids and insight into the mechanism of this process for the
preparation of R-(o-anilido)ketones.
Table 1. Optimization of Coupling and Rearrangementa,b
additive
entry copper salt
[Cu]
(1 equiv) yield (%) 4aa/6aa
1
Cu(OAc)2
CuBr
1 equiv
1 equiv
none
none
none
none
none
none
76
84
65
nr
69
47
nr
dec
60
nr
50
33
41
75
only 4aa
only 4aa
only 4aa
2
3
CuBr SMe2 1 equiv
3
4
CuI
1 equiv
1 equiv
1 equiv
5
CuOTf Tol
only 4aa
only 4aa
3
Scheme 1. Chemoselective [3,3] Rearrangements
6
CuTC
CuBr
CuBr
CuBr
CuBr
Cu(OAc)2
CuCl
7
20 mol % BQ
8
20 mol % PhI(OAc)2
20 mol % Zn dust
20 mol % Mn dust
20 mol % Zn dust
20 mol % Zn dust
20 mol % Zn dust
20 mol % Zn dust
9
2/1
10
11
12
13
14
5/1
5/2
9/1
7/1
CuI
c
CuSO4
a Reaction mixtures were prepared as 1:2:5:1 mixture of 1/2/pyridine/
4 A MS in DCE (0.1 M) and were run in air. b % yield determined by 1H
˚
NMR spectroscopy using CH2Br2 as a reference. c CuSO4 5H2O.
3
The coupling of N-phenylbenzhydroxamic acid 1a with
cyclohexenyl boronic acid 2a was tested with a variety of
copper(I) and copper(II) salts to determine optimal
conditions for the preparation of R-(o-anilido)ketones
(Table 1). Several copper reagents were shown to effi-
ciently convert mixtures of 1a and 2a to 4aa in the presence
of pyridine, a halogenated solvent, and air. The most
efficient reagent was identified as CuBr (entries 1À6).
While unassisted, substoichiometric amounts of CuBr
did not exhibit any turnover for the desired transforma-
tion. Considering that inefficient oxidation state changes
may have been responsible for the lack of catalyst turn-
over, common oxidants and reductants were screened as
additives (entries 7À10). While the use of benzoquinone
and PhI(OAc)2 failed to promote the conversion of 1a to
4aa in the presence of 20 mol % of CuBr, we were delighted
to discover that the addition of 1 equiv of Zn dust provided
the desired product in good yield. Surprisingly, under these
copper-catalyzed conditions, the preparation of 4aa was
accompanied by R-aminated ketone 6aa (entry 9).12 This
competing side product could be generated via a formal
[1,3] rearrangement of the proposed intermediate 3aa.13
Compound 6aa was removed from the crude oxyarylation
product mixtures by column chromatography and was not
converted to 4aa under the reaction conditions. Other
common reductants such as Mn(0), Al(0), and SnCl2 were
tested for comparison to Zn(0) but none promoted the
desired catalysis.14 Further screening of the choice of
copper catalyst led to increased conversion and varying
ratios of 4aa and 6aa (entries 11À14). The best combina-
tion of yield and selectivity for CÀC bond formation was
provided by 20 mol % of CuSO4 5H2O when used with a
3
single of equivalent of Zn(0). The optimal conditions
described in entries 2 and 14 of Table 1, were chosen to
further explore the scope of the oxyarylation of alkenyl
boronic acids with N-arylhydroxamic acids.
Several alkenyl boronic acids were screened under
both copper-catalyzed and copper-mediated conditions
(Table 1, entries 2 and 14, respectively) to determine the
tolerance of the oxyarylation for the alkenyl coupling
partner.15 As shown in Scheme 2, a variety of cyclic and
(10) For examples of similar [3,3] rearrangements of N-arylhydrox-
amate-substituted ketene acetals, see: (a) Endo, Y.; Uchida, T.; Shudo,
K. Tetrahedron Lett. 1997, 38, 2113. (b) Endo, Y.; Uchida, T.; Hizatate,
S.; Shudo, K. Synthesis 1994, 1096. (c) Almeida, P. S.; Prabhakar, S.;
Lobo, A. M.; Marcelo-Curto, M. J. Tetrahedron Lett. 1991, 32, 2671. (d)
Miller, S. J.; Bayne, C. D. J. Org. Chem. 1997, 62, 5680. (e) Santos, P. F.;
Srinivasan, N.; Almeida, P. S.; Lobo, A. M.; Prabhakar, S. Tetrahedron
2005, 61, 9147. (f) Mao, Z.; Baldwin, S. W. Org. Lett. 2004, 6, 2425. (g)
Duguet, N.; Slawin, A. M. Z.; Smith, A. D. Org. Lett. 2009, 11, 3858.
(11) For examples of the use of other types of interrupted Fischer-
indole intermediates in synthesis, see: (a) Boal, B. W.; Schammel, A. W.;
Garg, N. K. Org. Lett. 2009, 11, 3458. (b) Schammel, A. W.; Boal, B. W.;
Zu, L.; Mesganaw, T.; Garg, N. K. Tetrahedron 2010, 66, 4687. (c) Zu,
L.; Boal, B. W.; Garg, N. K. J. Am. Chem. Soc. 2011, 133, 8877. (d)
Bunders, C.; Cavanagh, J.; Melander, C. Org. Biomol. Chem. 2011, 9,
5476. (e) Schammel, A. W.; Chiou, G.; Garg, N. K. J. Org. Chem. 2012,
77, 725. (f) Schammel, A. W.; Chiou, G.; Garg, N. K. Org. Lett. 2012, 14,
4556.
(12) Compound 6aa was identified by comparison of spectroscopy
data to R-(N-phenylamino)cyclohexanone: Liu, S.; Xie, J.-H.; Li, W.;
Kong, W.-L.; Wang, L.-X; Zhou, Q.-L. Org. Lett. 2009, 11, 4994.
(13) For examples of formal [1,3] rearrangements of O-vinyl oximes
that we have observed previously, see: (a) Wang, H.-Y.; Mueller, D. S.;
Sachwani, R. M.; Kapadia, R.; Londino, H. N.; Anderson, L. L. J. Org.
Chem. 2011, 76, 3203. (b) Wang, H.-Y.; Mueller, D. S.; Sachwani, R. M.;
Londino, H. N.; Anderson, L. L. Org. Lett. 2010, 12, 2290.
(14) Zn(0) and Zn(II) salts did not promote the oxyarylation in the
absence of CuBr or CuSO4 5H2O.
3
(15) Literature procedures were used to prepare alkenyl boronic
acids 2. See the Supporting Information for details.
Org. Lett., Vol. 15, No. 13, 2013
3363

Scheme 2. Scope of Alkenylboronic Acid Reagents for
Oxyarylation
Scheme 3. Scope of N-Arylhydroxamic Acid Reagents for
Oxyarylation
a Percent isolated yield. b 20 mol % of CuSO4 5H2O and 1 equiv of
3
Zn. c 1 equiv of CuBr. d 20 mol % of Cu(OAc)2 and 1 equiv of Zn.
a Percent isolated yield. b 20 mol % of CuSO4 5H2O and 1 equiv of
3
Zn. c 1 equiv of CuBr.
linear, Z-disubstituted alkenyl boronic acids were success-
fully transformed into the corresponding R-(o-anilido)-
ketones. Cyclic alkenyl boronic acids provided good to ex-
cellent yields of 4aa, 4ab, and 4ad under copper-catalyzed
conditions and larger ring sizes were well-tolerated under
copper-mediated conditions. Z-2-Butenyl boronic acid
2e was identified as the most efficient coupling partner
with 1a and longer alkyl chain analogues and branched
substituents were similarly well-tolerated (see 4afÀah).
Strylboronic acids 2iÀl were efficiently converted to
R-arylated ketones 4aiÀal under copper-mediated condi-
tions, while copper-catalyzed conditions were also shown
to be effective for 4ak. To the best of our knowledge, none
of the products illustrated in Scheme 2 have previously
been reported in the literature. For comparison, we at-
tempted to synthesize 4ai using literature procedures for
palladium-catalyzed R-arylation but no reaction was ob-
served and starting materials were recovered.16 These
control experiments emphasize the practical importance
of the new method in addressing current limitations of
R-arylation reactions.
1c was similarly effective, it gave a 3:2 regioisomeric mix-
ture of 4ce.18 N-p-Chlorophenyl hydroxamic acid 1d was
converted to analogous R-arylated ketones 4da and 4de, in
somewhat attenuated yields compared to the tolyl ana-
logues. The oxyarylation method was also shown to tolerate
both electron-rich and electron-poor N-benzoyl protecting
groups under both copper-catalyzed and copper-mediated
conditions (4ea and 4fa). N-Perfluorobenzoyl hydroxamic
acids, in particular, proved to be excellent substrates when
treated with boronic acid 2e. Alkyl-, halogen-, and OMe-
substituted N-aryl groups gave R-arylated ketones 4heÀke
in excellent yields while the efficiency of the conversion of
N-(p-CF3)phenyl hydroxamic acid 1l to 4le was moderate.
N-Aryl acetohydroxamic acids were also shown to be
amenable tothe oxyarylation processbut were lessefficient
than N-benzoyl substrates (4me). Carbamate derivatives
were tested but these compounds provided mixtures of the
desired products and the corresponding indole. These
results demonstrate that the mild conditions of the oxyar-
ylation reaction enable broad functional group tolerance
to access a range of R-(o-anilido)ketones.
The scope of the oxyarylation reaction was further
evaluated by testing a variety of N-arylhydroxamic acids
for the coupling and rearrangement process (Scheme 3).17
Benzhydroxamic acids with N-p-tolyl substituents pro-
vided excellent yields of R-arylated cyclohexanone 4ba
and acetone 4be. While N-(m-tolyl)benzhydroxamic acid
To better understand the copper-catalyzed oxyarylation
process, several experiments were designed to compare the
individual steps of the proposed mechanism illustrated
in Scheme 1.19 An intramolecular competition experi-
ment was used to assess the CÀH functionalization event.
(16) No reaction was observed when a mixture of N-(2-bromophenyl)-
benzamide and butyrophenone was subjected to the R-arylation condi-
tions provided in the following references: (a) Kawatsura, M.; Hartwig,
J. F. J. Am. Chem. Soc. 1999, 121, 1473. (b) Fox, J. M.; Huang, X.;
Chieffi, A.; Buchwald, S. L. J. Am. Chem. Soc. 2000, 122, 1360.
(17) Literature procedures were used to prepare N-arylhydroxamic
acids 1. See the Supporting Information for details.
(18) For a discussion on the limits of regioselectivity in the Fischer-
indole reaction, see: Phillips, R. R. Org. React. 1959, 10, 1143 (ref 1d)
and references cited therein.
(19) In changing from copper-mediated to copper-catalyzed condi-
tions, the only step of the proposed mechanism that is thought to be
affected in our current working model is catalyst regeneration.
3364
Org. Lett., Vol. 15, No. 13, 2013

Scheme 4. Intramolecular Competition Experiment
As shown in Scheme 4, when hydroxamic acid 1i-d1 was
treated with 2e under the reaction conditions, a primary
kinetic isotope effect was not observed. This suggests
that CÀH functionalization is not involved in the rate-
determining step of the oxyarylation. To distinguish between
the CÀO and CÀC bond forming events, the dependence
of the rate of product formation on the electronic nature of
the N-aryl group was evaluated. Para-substituted N-aryl
groups were used to examine both the CÀO and the CÀC
bond-forming steps of the transformation by comparison
of relative rates to σp and σm values, respectively. Competi-
tion experiments between para-substituted N-hydroxyani-
lides 1hÀl and N-hydroxyanilide 1gwere usedtodetermine
relative initial rates of oxyarylation. As shown in Figure 1,
the relative rate dependence on the type of N-aryl sub-
stituent had a linear correlation with σp values with a
F-value of À1.01 and a nonlinear correlation with σm
values.20 These data suggest that CÀO bond formation is
the slow-step of the oxyarylation process and are consis-
tent withthe factthat intermediate3 wasnotobserved. The
negative F-value may be indicative of electron-donating
groups facilitating the transmetalation of the boronic acid
which has been suggested to be the slow step of a Chan-
Lam-Evans coupling.21 Previous studies have shown that
mechanistic data supports cyclization of the imino anilide
as the rate-determining step of the Fischer-indole reaction
if the substrate has an N-acetyl substituent.22 This implies
thatthe keytoourability toisolate the interruptedFischer-
Indole intermediates described above is that our coupling
conditions are mild enough to suppress cyclization to the
corresponding indole.
Figure 1. Correlation of relative initial rates to Hammett σp values.
In summary, we have shown that interrupted Fischer-
indole intermediates can be prepared by copper-mediated
or copper-catalyzed oxyarylation of alkenyl boronic acids
with N-arylbenzhydroxamic acids. These transformations
provide access to R-(o-anilido)ketones which are challeng-
ing to prepare because of their propensity to cyclize to the
corresponding indole and the inability of R-arylation
procedures to access these types of structures. Mechanistic
data suggests that CÀO bond formation is the slow-step of
the oxyarylation process and that the mild conditions used
to affect this transformation allow for the isolation of an
interrupted Fischer-indole intermediate after the [3,3] re-
arrangement and before a cyclization to the corresponding
indole can occur.
Acknowledgment. We acknowledge generous funding
from ACS-PRF (50491-DNI), NSF (NSF-CHE 1212895),
and the University of Illinois at Chicago. We also thank
Prof. T. Driver (UIC) and Prof. R. J. Thomson (NU) for
insightful discussions and Mr. Furong Sun (UIUC) for
mass spectrometry data.
Supporting Information Available. Experimental pro-
cedures and compound characterization data. This in-
formation is available free of charge via the Internet at
http://pubs.acs.org.
(20) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165.
(21) (a) Qiao, J. X.; Lam, P. Y. S. Synthesis 2011, 829. (b) King, A. E.;
Brunold, T. C.; Stahl, S. S. J. Am. Chem. Soc. 2009, 131, 5044. (c) King,
A. E.; Huffman, L. M.; Casitas, A.; Costas, M.; Ribas, X.; Stahl, S. S.
J. Am. Chem. Soc. 2010, 132, 12068.
(22) (a) Douglas, A. W. J. Am. Chem. Soc. 1978, 100, 6463. (b)
Douglas, A. W. J. Am. Chem. Soc. 1979, 101, 5676.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 13, 2013
3365