Use of Stang's reagent, PhI(CN)OTf, to promote Pummerer-like oxidative cyclization of 2-(phenylthio)indoles

Article abstract of DOI:10.1016/j.tetlet.2004.04.182

Treatment of various 2-(phenylthio)indoles, bearing nucleophilic groups tethered to C(3), with PhI(CN)OTf initiated an oxidative cyclization sequence that afforded 3,3-spirocyclic 2-(phenylthio)indolenines or oxindoles in moderate yield.

Full text of DOI:10.1016/j.tetlet.2004.04.182

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 5035–5037
Use of Stang’s reagent, PhI(CN)OTf, to promote Pummerer-like
oxidative cyclization of 2-(phenylthio)indoles
Ken S. Feldman* and Daniela Boneva Vidulova
Chemistry Department, 152 Davey Lab, The Pennsylvania State University, University Park, PA 16802, USA
Received 21 April 2004; revised 30 April 2004; accepted 30 April 2004
Abstract—Treatment of various 2-(phenylthio)indoles, bearing nucleophilic groups tethered to C(3), with PhI(CN)OTf initiated an
oxidative cyclization sequence that afforded 3,3-spirocyclic 2-(phenylthio)indolenines or oxindoles in moderate yield.
Ó 2004 Elsevier Ltd. All rights reserved.
The oxidative cyclization of 2-(phenylsulfinyl)indoles
bearing a nucleophile tethered to C(3) provides a range
of 3,3-spirocyclic 2-(phenylthio)indolenine products as
exemplified by the conversion of 1a into 2, Eq. 1.1 These
transformations likely proceed via a classical Pummerer-
type reaction initiated by exposure of the sulfoxide to
Tf2O. Pummerer-like oxidative cyclizations with aryl
sulfide substrates have been reported as well,2 and a
range of oxidants, including Cl2,2a ArIF2,2b and
PhI(OTFA)2,2c;d appear to find productive use in this
regard. This shortcut to achieving a net Pummerer
rearrangement has seen little use in synthesis, perhaps as
a consequence of the relatively harsh oxidants em-
ployed, and a lack of demonstrated substrate generality.
Attempts to apply either of the common hypervalent
iodine oxidants PhI(OAc)2 or PhI(OTFA)2 to the 2-
(phenylthio)indole system 1b failed to furnish any spi-
rocyclic product 2, but rather returned low yields of the
C(3) oxidized oxindole 3.
However, switching to PhI(CN)OTf, a reagent intro-
duced by Stang and Zhdankin,3 led to a significant
improvement in reaction outcome with the sulfide 1b
(Table 1). A survey of reaction parameters led to the
optimized conditions described in entry ‘e’ of the table.
Control experiments demonstrated that the thioimidate
product 2 was stable to the reaction conditions, even
when the hypervalent iodine oxidant was used in sig-
nificant excess. The imidate resonance in 2 may diminish
the reactivity of the sulfur’s lone pairs in this instance.
The oxidative cyclization proceeded equally well in
CH2Cl2 and CH3CN (0.01 M), but did not work at all in
toluene. Lutidine premixed with the iodonium reagent
€
was more advantageous than Hunig’s base, and warmer
temperatures led to slightly higher yields of product.
Excess Stang reagent was required to consume all of the
starting material (TLC), and the highest yields accom-
panied portionwise addition over 90 min, suggesting that
unidentified processes were consuming the hypervalent
iodine reagent in competition with the desired oxidative
cyclization sequence. The yield of 2 was only mildly
Table 1. Optimization studies for the PhI(CN)OTf-mediated conver-
sion of 1b into 2
Entry Solvent
Temp Base
(°C)
Ratio PhI(CN)
OTf to 1b
Yield
(%)
a
b
c
d
e
f
CH3CN
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
)45
)80
0
Lutidine 2.2
Lutidine 2.5
Lutidine 2.2
51
48
61
41
ð1Þ
0
iPr2NEt
Lutidine
4.1
Keywords: Pummerer; Hypervalent iodine; Oxindole.
* Corresponding author. Tel.: +1-814-8634654; fax: +1-814-8638403;
e-mail: ksf@chem.psu.edu
0
20
4.1 (portionwise) 67
Lutidine 2.2 61
0040-4039/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2004.04.182

5036
K. S. Feldman, D. Boneva Vidulova / Tetrahedron Letters 45 (2004) 5035–5037
Table 2. Survey of 2-phenylthioindole substrates subjected to PhI(CN)OTf-mediated oxidative cyclization
Entry
a
Substrate
Cond.a
Productb
Yield (%)
46
A
b
B
42
c
C
B
55
63
d
a A: 4 equiv PhI(CN)OTf, 3 equiv lutidine, CH3CN, )40 °C; B: 1 equiv PhI(CN)OTf, 3 equiv lutidine, CH3CN, 0 °C; C: 4 equiv PhI(CN)OTf, 3 equiv
lutidine, CH2Cl2, )78 °C.
b All new compounds exhibited satisfactory spectral data (see Ref. 1, supporting information).
sensitive to concentration variance over the range
0.001 M (56%) to 0.1 M (44%).
intermediate 14 whereas the additive Pummerer alter-
native avoids such a potentially high-energy species. At
present, no basis for distinguishing between these two
hypotheses for 4, or any other substrate examined in this
study, is in hand.
Extension of this 2-(phenylthio)indole oxidative cycli-
zation procedure to other substrates led to mixed but
generally favorable results (Table 2). Reaction of the N-
methyl analogue of 1b, 2-(phenylthio)indole 4, under the
optimized conditions provided the oxindole 5 directly,
entry ‘a’. The yield is slightly lower than with the N–H
species 1b, perhaps reflecting the more complicated
multi-step reaction course in this case. This N-methyl
species brings into focus an interesting mechanistic
question as detailed in Scheme 1. A priori, both additive
and vinylogous Pummerer-like pathways can be invoked
to rationalize the formation of 5.4 However, the vinyl-
ogous Pummerer sequence passes through a dicationic
The use of the much more nucleophilic silyl enol ether 6
(Mayr nucleophilicities:allylsilane ¼ 1.8, silyl enol
ether ¼ 5.4)5 raises the possibility that competitive enol
ether/iodonium reaction may divert the transform. The
product spirocycle 7 is formed in relatively more modest
yield, but no evidence for a-keto iodonium ylide for-
mation was detected.
An even more significant challenge can be found in
substrate 8, where the presence of the potent silyl ketene
iminal nucleophile (Mayr nucleophilicities for silyl
ketene acetals ꢀ9–10, and for enamines ꢀ10–11)5 places
even more stringent requirements on the selectivity of
electrophile addition. Stang’s reagent evidently is capa-
ble of meeting this challenge, as its apparent thiophi-
licity leads to appropriate Pummerer initiation and
eventual formation of the spirocyclic imide product 9 in
good yield.
The final substrate examined in this study has no formal
alkene nucleophile at all: the simple ketone 10. Inter-
pretation of the mechanistic course of this reaction is
clouded by uncertainty about the identity of the actual
participating nucleophile. Either the carbonyl oxygen
itself or the hydroxyl of a derived enol may constitute
the reactive nucleophile that traps the electrophilic C(3)
site of the oxidized substrate.
Scheme 1. Mechanistic speculation for the conversion of 4 into 5.

K. S. Feldman, D. Boneva Vidulova / Tetrahedron Letters 45 (2004) 5035–5037
5037
In summary, these examples of Pummerer-like oxidative
References and notes
cyclization demonstrate the feasibility of using the
unconventional initiator PhI(CN)OTf with a sulfide
substrate. The use of this reactive yet highly selective
oxidant offers two advantages over existing indole oxi-
dative cyclization triggers: (1) oxidative activation is
confined to C(3), and (2) no potentially interfering
nucleophilic counterions are introduced.
1. Feldman, K. S.; Vidulova, D. B. Org. Lett. 2004, in press.
2. (a) Decker, D.; Mayer, R. Zeit. Chem. 1990, 30, 367–368;
(b) Motherwell, W. B.; Greaney, M. F.; Edmunds, J. J.;
Steed, J. W. J. Chem. Soc., Perkin Trans. 1 2002, 2816–
2826; (c) Tamura, Y.; Yakura, T.; Shirouchi, Y.; Haruta, J.
i. Chem. Pharm. Bull. 1986, 34, 1061–1066; (d) Wang, H.-
M.; Lin, M.-C.; Chen, L.-C. Heterocycles 1994, 38, 1519–
1526.
3. Stang, P. J.; Williamson, B. L.; Zhdankin, V. V. J. Am.
Chem. Soc. 1991, 113, 5870–5871.
4. For a discussion of the additive versus vinylogous Pum-
merer reaction pathways, with leading references, see
Padwa, A.; Kuethe, J. T. J. Org. Chem. 1998, 63, 4256–
4268.
5. Mayr, H.; Kempf, B.; Ofial, A. R. Acc. Chem. Res. 2003,
36, 66–77.
Acknowledgements
We thank the Institute of General Medical Sciences
within the National Institutes of Health (GM35727) for
financial support of this work.