Extending Pummerer reaction chemistry. Application to the oxidative cyclization of indole derivatives

Article abstract of DOI:10.1021/ol0493406

Matrix presented. Treatment of 2-(phenylsulfinyl)indoles bearing a pendant nucleophile at C(3) with Tf₂O/lutidine triggers a Pummerer-like cyclization to furnish 3,3-spirocyclic-2-(phenylthio)indolenine products, which can, in turn, be hydrolyzed to 3,3-spirocyclic oxindoles.

Full text of DOI:10.1021/ol0493406

ORGANIC
LETTERS
2004
Vol. 6, No. 11
1869-1871
Extending Pummerer Reaction
Chemistry. Application to the Oxidative
Cyclization of Indole Derivatives
Ken S. Feldman* and Daniela Boneva Vidulova
Department of Chemistry, The PennsylVania State UniVersity,
UniVersity Park, PennsylVania 16802
ksf@chem.psu.edu
Received April 11, 2004
ABSTRACT
Treatment of 2-(phenylsulfinyl)indoles bearing a pendant nucleophile at C(3) with Tf2O/lutidine triggers a Pummerer-like cyclization to furnish
3,3-spirocyclic-2-(phenylthio)indolenine products, which can, in turn, be hydrolyzed to 3,3-spirocyclic oxindoles.
The development of a new strategy for the synthesis of 3,3-
spirosubstituted oxindoles 6 via regioselective cyclization of
3-substituted 2-(phenylsulfinyl)indoles 1 is described, Scheme
1. This indirect oxidative cyclization sequence evolved from
oxidation.1 A new twist on the venerable Pummerer reaction2
serves as the pivotal transform that both enforces complete
regioselectivity upon cyclization and ensures that the oxida-
tion-sensitive product survives unscathed. Oxidation is
confined to the sulfur atom in the sulfide precursor to 1,
and since oxidation and cyclization are now two separate
events, there is no possibility for undesired product (over)-
oxidation. A priori, two limiting mechanisms, termed viny-
logous and additive,3 can be posited for this sequence. These
pathways differ principally by whether the leaving group OR
in 2 is discharged prior to (vinylogous), or simultaneously
Scheme 1. Extended Pummerer-Mediated Functionalization of
Indole C(3): A New Approach to 3,3-Disubstituted Oxindoles
(1) (a) Braun, N. A.; Ousmer, M.; Bray, J. D.; Bouchu, D.; Peters, K.;
Peters, E.-M.; Ciufolini, M. A. J. Org. Chem. 2000, 65, 4397-4408. (b)
Somei, M.; Noguchi, K.; Yamagami, R.; Kawada, Y.; Yamada, K.; Yamada,
F. Heterocycles 2000, 53, 7-10. (c) Wang, H.; Ganesan, A. J. Org. Chem.
2000, 65, 4685-4693. (d) Irikawa, H.; Mutoh, S.; Uehara, M.; Okumura,
Y. Bull. Chem. Soc. Jpn. 1989, 62, 3031-3033. (e) Gu¨ller, Borschberg,
H.-J. HelV. Chim. Acta 1993, 76, 1847-1862.
(2) (a) Moiseen, A. M.; Dragan, V. A.; Veselovskii, V. V. Russ. Chem.
ReV. 1991, 60, 643-657. (b) de Lucchi, O.; Miotti, U.; Modena, G. In
Organic Reactions; Paquette, L. A., Ed.; John Wiley and Sons: New York,
1991; Vol. 40, pp 157-405. (c) Padwa, A.; Gunn, D. E., Jr.; Osterhout, M.
H. Synthesis 1997, 1353-1377.
(3) Leading references to additive and vinylogous Pummerer reactions
can be found in: (a) Padwa, A.; Kuethe, J. T. J. Org. Chem. 1998, 63,
4256-4268. (b) Shibata, N.; Fujimori, C.; Fujita, S.; Kita, Y. Chem. Pharm.
Bull. 1996, 44, 892-894. (c) Marino, J. P.; Bogdan, S.; Kimura, K. J. Am.
Chem. Soc. 1992, 114, 5566-5572.
the recognition that direct oxidative cyclization onto C(3)
of the indole nucleus is often frustrated by low yields,
insufficiently controlled regiochemistry, and product over-
10.1021/ol0493406 CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/23/2004

Table 1. Examples of 3,3-Disubstituted Oxindole Synthesis via
Table 2. Variation in Yield of 8 and 10 upon Attempted
Extended Pummerer Reaction of 2-(Phenylsulfinyl)indoles
Pummerer-Initiated Cyclization of Indole Allylsilane Sulfoxide 7
entry activator
base
temp (°C) solvent 8a (%) 10a (%)
1
2
3
4
5
6
7
TFAA
TFAA
Tf2O
Tf2O
Tf2O
Tf2O
Tf2O
i-Pr2NEt
DTBPb
i-Pr2NEt
i-Pr2NEt
i-Pr2NEt
DTBP
-40
-75
-75
-75
-75
-75
-75
CH2Cl2
CH2Cl2
toluene
Et2O
CH2Cl2
CH2Cl2
CH2Cl2
64
58
<5
42
30
56
82
19
2,6-lutidine
a Yields reported are for chromatographically pure, characterized material.
b DTBP ) 2,6-di-tert-butylpyridine.
temperature led to varying amounts of the desired spiro-
methylenecyclohexane 2-(phenylthio)indolenine 8, Table 2.
These optimization studies revealed that use of 2 equiv of
Tf2O and 3 equiv of 2,6-lutidine with 0.005 M substrate in
CH2Cl2 at -75 °C afforded the highest yield of the desired
spirocyclic product. Interestingly, use of TFAA as an
activator provided only the trifluoroacetate-trapped product,
isolated as the unstable alcohol 10 after SiO2-induced
trifluoroacetate hydrolysis or solvolysis upon attempted
chromatographic purification (Table 2, entries 1-2). It
appears that the electrophilicity at C(3) of the indole core
(cf. 3) is expressed under these conditions, but then the weak
nucleophile trifluoroacetate surprisingly out-competes the
pendant alkene nucleophile (allylsilane or propenyl; the
timing of TMS loss is not known) for this reactive intermedi-
ate. The hydrolysis of the thioimidate function of 8 proceeded
in higher yield using an oxidative protocol (CAN, H2O)4
rather than the standard Hg2+-promoted procedure.5 The
thioimidate within 10 lacks the sensitive exocyclic alkene,
and mercury-assisted hydrolysis cleanly provides the ex-
pected 3-hydroxyoxindole 11 without complication.
A second substrate of interest, 12, featured a silyl enol
ether nucleophilic trap (Table 1, entry b). Treatment of this
species under the conditions optimized for the allylsilane
analogue led to formation of the expected spirocyclohexa-
none product 13 in good yield. Unlike the related alkene-
bearing species 8, ketone 13 could be readily hydrolyzed
with Hg2+ assistance to provide the oxindole product 14 in
excellent yield.
A departure from C-C bond-forming nucleophiles fea-
tured ketone 15 (Table 1, entry c) as the Pummerer
cyclization substrate. In this instance, cyclization at oxygen
is readily accomplished, presumably via the derived enol.
The O-C over C-C bond-forming selectivity observed
within this putative enol-bearing Pummerer intermediate
might be attributed to the facility of 6-enol-exo cyclizations
over the alternative C-C bond-forming 6-enol-endo pathway.
a Conditions: 7, see Table 2; 12, 15, 17, Tf2O, 2,6-lutidine CH2Cl2, -75
°C; 20, Tf2O, 2,6-lutidine, indicated solvent/temperature. b Conditions: 9,
CAN, H2O; 10, 13, 18, HgCl2, H2O.
with (additive), nucleophilic attack. In either circumstance,
the transposition of “oxidation” results in electrophilic
activation at C(3) (cf. 1 f 2 or 3). This electrophilicity can
be quenched through internal trapping by a pendant nucleo-
philic atom to provide a thioimidate product 5 en route to
the ultimate spirocyclic oxindole product 6. Nucleophilic
addition at other accessible and conceivably electrophilic sites
in 2 or 3 such as C(2) or C(4) is likely to be dissuaded by
the loss of aromaticity upon bond formation. In this way,
the regiochemistry of nucleophile addition is assured.
A search to uncover carbon-based nucleophiles that would
tolerate both sulfoxide formation and subsequent Pummerer
reaction began with the allylsilane sulfoxide 7, Table 1.
Exposure of this sulfoxide to either of the two common
Pummerer activators, triflic anhydride (Tf2O) or trifluoro-
acetic anhydride (TFAA), in the presence of a variety of
bases in moderately polar to nonpolar solvents at low
The occurrence of several 3,3-disubstituted oxindole-
derived natural products formally originating from tryptamine6
provided the motivation to explore this Pummerer cyclization
chemistry of the tryptamine derivative 17 (Table 1, entry
(4) Ho, T.-L.; Ho, H.; Wong, C. M. J. Chem. Soc., Chem. Commun.
1972, 791.
(5) Satchell, D. P. N. Chem. Soc. ReV. 1977, 6, 345-371.
1870
Org. Lett., Vol. 6, No. 11, 2004

Scheme 2. Postulated Origin of Diastereoselectivity with 17
Scheme 3. 2-(Phenylsulfinyl)indole Cyclization Substrate
Syntheses
d). Triflic anhydride treatment of 17 furnished the desired
cyclized material 18 in excellent yield as a mixture of
diastereomers strongly favoring the equatorial-phenyl isomer
shown. Resubmission of partially purified samples of either
isomer to reaction conditions did not alter the diastereomer
ratio. This kinetic stereochemical outcome can be rationalized
by invoking differential steric interactions in the chairlike
transition state models shown in Scheme 2.7
The successful Pummerer-initiated cyclization of the
N-methyl indole substrate 20 (Table 1, entry e) focuses
attention on the aforementioned mechanistic dichotomy
between the additive and the vinylogous Pummerer pathways
(cf. Scheme 1). Since the activated sulfoxide corresponding
to 2 no longer has a proton to lose (NH is now NCH3),
passage through a now dicationic intermediate (cf. 3, with
N replaced by +NCH3) in the vinylogous Pummerer route
might exact a substantial energetic penalty, whereas the
alternative additive pathway does not engage the indole
nitrogen’s electrons until after C-C bond formation. There
is no experimental basis for distinguishing between these
proposals at present. Irrespective of the mechanistic details,
the formation of the N-methyl oxindole product 21 directly
from a C(2)-sulfur-containing starting material broadens the
utility of the chemistry because a second discrete hydrolysis
step is avoided.
The syntheses of the 2-(phenylsulfinyl)indole substrates
used in this study are detailed in Scheme 3. Notable features
include (1) the formation of both the silyl enol ether unit of
12 and the silyl ketene iminal unit of 17 in the presence of
the potentially interfering 2-(phenylsulfinyl)indole moiety’s
acidic N-H and (2) the use of the naphthylcarbamate
protecting group8 in 26 to overcome the problem of partial
PhS- loss that accompanied attempted hydrogenolysis of a
-CO2Bn analogue. The low yield reported for formation of
silyl ketene iminal 17 was a consequence of product loss
(hydrolysis) that occurred during chromatographic purifica-
tion. Typically, the crude substrate 17, which was formed
in ∼60% yield (1H NMR assay) was used in the Pummerer
sequence without further purification to furnish the thioimi-
date product in excellent overall yield.
Acknowledgment. We thank the NIH (GM35727) for
financial support.
Supporting Information Available: Experimental pro-
cedures, spectral data (1H NMR, 13C NMR, IR, MS, selected
combustion analyses), and copies of 1H and 13C NMR spectra
for 7-9, 11-16, 18-21, 24a, 24b, and 26. This material is
available free of charge via the Internet at http://pubs.acs.org.
(6) (a) Joshi, B. S.; Gundu Rao, P.; Rogers, D.; Singri, B. P.; Williams,
D. J. Ind. J. Chem. 1984, 23B, 101-102. (b) Numata, A.; Takahashi, C.;
Ito, Y.; Takada, T.; Kawai, K.; Usami, Y.; Matsumura, E.; Imachi, M.; Ito,
T.; Hasegawa, T. Tetrahedron Lett. 1993, 34, 2355-2358. (c) Verbitski,
S. M.; Mayne, C. L.; Davis, R. A.; Concepcion, G. P.; Ireland, C. M. J.
Org. Chem. 2002, 67, 7124-7126.
(7) Similar selectivity for C(3) cyclization through an indole aryl-
equatorial construct (cf. 22b) can be cited to rationalize observations in
related systems. (a) Atarabashi, S.; Choi, J.-K.; Ha, D.-C.; Hart, D. J.;
Kuzmich, D.; Lee, C.-S.; Ramesh, S.; Wu, S. C. J. Am. Chem. Soc. 1997,
119, 6226-6241. (b) Earley, W. G.; Oh, T.; Overman, L. E. Tetrahedron
Lett. 1988, 29, 3785-3788.
OL0493406
(8) Papageorgiou, E. A.; Gaunt, M. J.; Yu, J.-q.; Spencer, J. B. Org.
Lett. 2000, 2, 1049-1051.
Org. Lett., Vol. 6, No. 11, 2004
1871