Development of a novel, highly efficient halide-catalyzed sulfenylation of indoles

Article abstract of DOI:10.1021/ol052615c

The reaction of a variety of indoles with N-thioalkyl- and N-thioarylphthalimides to produce 3-thioindoles is reported. Catalytic quantities of halide-containing salts are crucial to the success of this reaction. This highly efficient reaction provides sulfenylated indoles from bench-stable, readily available starting materials in good to excellent yields.

Full text of DOI:10.1021/ol052615c

ORGANIC
LETTERS
2006
Vol. 8, No. 4
565-568
Development of a Novel, Highly Efficient
Halide-Catalyzed Sulfenylation of Indoles
Matthew Tudge,* Minoru Tamiya, Cecile Savarin, and Guy R. Humphrey
Department of Process Research, Merck Research Laboratories, P.O. Box 2000,
Rahway, New Jersey 07065
matthew_tudge@merck.com
Received October 28, 2005
ABSTRACT
The reaction of a variety of indoles with N-thioalkyl- and N-thioarylphthalimides to produce 3-thioindoles is reported. Catalytic quantities of
halide-containing salts are crucial to the success of this reaction. This highly efficient reaction provides sulfenylated indoles from bench-
stable, readily available starting materials in good to excellent yields.
Sulfenylated indole motifs are ubiquitous in many biologi-
cally important compounds.1a-d In particular, the therapeutic
value of numerous 3-thioindoles has been assessed in several
disease areas, including obesity,1a cancer,1b heart disease,1c
and bacterial infection.1d Preparation of such compounds is
generally achieved by electrophilic aromatic sulfenylation
chemistry. Sulfenylating agents such as disulfides,2a sulfenyl
halides,3b and quinone mono-O,S-acetals4c have all been
employed in this reaction, but are often impractical on both
small and large scale due to the accessibility, substrate
compatibility, and stability of these reagents. To overcome
these problems, sulfenylations utilizing thiols activated in
situ by N-chlorosuccinimide5 or transition-metal catalysts6
have been developed.
In the context of an ongoing drug discovery program, we
required an efficient sulfenylation method that could be
carried out on a range of electronically distinct indoles, using
bench-stable sulfenylating agents at relatively high concen-
tration (Scheme 1).7
Scheme 1
(1) (a) Acton, J. L.; Meinke, P. T.; Wood, H.; Black, R. M. PCT Int.
Appl. WO 2004/019869 A2, 2004. (b) (i) Avis, I.; Mart´ınez, A.; Tauler, J.;
Zudaire, E.; Mayburd, A.; Abu-Ghazaleh, R.; Ondrey, F.; Mulshine, J. L.
Cancer Res. 2005, 65, 4181. (ii) De Martino, G.; La Regina, G.; Coluccia,
A.; Edler, M. C.; Barbera, M. C.; Brancale, A.; Wilcox, E.; Hamel, E.;
Artico, M.; Silvestri, R. J. Med. Chem. 2004, 47, 6120. (c) Funk, C. D.
Nat. ReV. Drug DiscoVery 2005, 4, 664. (d) Khandekar, S. S.; Gentry, D.
R.; Van Aller, G. S.; Doyle, M. L.; Chambers, P. A.; Konstantinidis, A.
K.; Brandt, M.; Daines, R. A.; Lonsdale, J. T. J. Biol. Chem. 2001, 276,
30024.
(2) (a) Atkinson, J. G.; Hamel, P.; Girard, Y. Synthesis 1988, 480. (b)
(i) Scoffone, E.; Fontana, A.; Rocchi, R. Biochemistry 1968, 7, 971. (ii)
Anzai, K. J. Heterocycl. Chem. 1979, 16, 567. (c) Matsugi, M.; Murata,
K.; Gotanda, K.; Nambu, H.; Anilkumar, G.; Matsumoto, K.; Kita, Y. J.
Org. Chem. 2001, 66, 2434.
To date, however, no general sulfenylation protocol has
been reported that is economical with respect to the sulfur
component,8 avoids the use of harsh chlorination conditions,6
(5) For sulfenylations utilising and/or liberating >1 equiv of sulfide, see
refs 3 and 4.
(3) Schlosser, K. M.; Krasutsky, A. P.; Hamilton, H. W.; Reed, J. E.;
Sexton, K. Org. Lett. 2004, 6, 819.
(4) Maeda, Y.; Koyabu, M.; Nishimura, T.; Uemura, S. J. Org. Chem.
2004, 69, 7688.
(6) In our hands, sulfenylations involving the use of SO2Cl2 (ref 2b, ii)
or NCS (ref 3) gave significant quantities of 3-chloroindole byproducts.
(7) Trost, B. M. Chem. ReV. 1978, 78, 363.
(8) Kuehen, M. E. J. Org. Chem. 1963, 28, 2124.
10.1021/ol052615c CCC: $33.50
© 2006 American Chemical Society
Published on Web 01/19/2006

and is compatible with sensitive functionality. Having
investigated several of the existing procedures,2-4 we sur-
veyed the literature for appropriate sulfenylating agents to
meet our needs. Of particular interest were several reports
detailing the sulfenylation of enolates7 and enamines8 using
readily prepared,9 stable thiophthalimide reagents. Further-
more, the reaction of a thiosuccinimide reagent with an indole
moiety had been observed; however, stoichiometric quantities
of BF3‚OEt2 in refluxing methylene chloride were required
to achieve any appreciable conversion.10 Prompted by these
reports, we sought to develop a more benign and general
sulfenylation protocol that circumvented the need for large
quantities of strong Lewis acids or chlorinating reagents.
Herein, we report the successful outcome of our investiga-
tions into this reaction.
were also ineffective catalysts, possibly due to their insolu-
bility in DMAc. Interestingly, however, trace amounts of hard
Lewis acids such as MgBr2 and LiCl effectively promoted
the transformation. Indeed, as detailed in Figure 1, when
Our initial investigations focused on establishing reactivity
between N-methylindole and phenylthiophthalimide. We
quickly determined that the reaction was sluggish at tem-
peratures below 70 °C, perhaps due to the insolubility of
the thiophthalimide reagent. Initial solvent screens11 showed
that polar solvents were essential for reactivity, with di-
methylacetamide (DMAc) proving optimal (Scheme 2).
Figure 1. Effect of hard Lewis acids on the rate of reaction.
compared with the control experiment at 90 °C reaction times
were cut from 24 h to 30 min, clearly indicating that catalysis
is occurring.
Scheme 2. Initial Result
Table 2. Effect of Catalytic Non-metal Halide Salts on the
Reaction
Although after 24 h the reaction in DMAc would reach
completion, with more sensitive substrates significant quanti-
ties of reagent decomposition or byproduct formation were
observed.12
In an attempt to accelerate the reaction, we probed the
effect of Lewis and Brønsted acid additives on the reaction
(Table 1).
additive
LiBF4
Bu4NCl + LiBF4
assay yield (%)
additive assay yield (%)
12
43
Bu4NCl
Bu4NBr
43
40
Table 1. Survey of Catalytic Brønsted and Lewis Acid Sources
At this point, we postulated that a simple Lewis acid
catalyzed mechanism was in operation, where the hard metal
center was coordinating to the phthalimide carbonyl, thus
weakening the N-S bond and increasing the electrophilicity
of the sulfur.
Based on this hypothesis, we believed that metal salts
bearing noncoordinating counterions would prove to be more
effective catalysts. Surprisingly, however, lithium tetrafluo-
roborate failed to promote the reaction. However, when
additive
none
TsOH‚OH2
AcOH
BF3‚OEt2
NaCl
KCl
assay yield (%)
additive
assay yield (%)
13
10
12
8
13
11
TiCl4
AlCl3
MgCl2
MgBr2
LiBr
29
36
38
54
48
49
(9) For a convenient synthesis of thiophthalimide reagents, see: Klose,
J.; Reese, C. B.; Song, Q. Tetrahedron 1997, 53, 14411.
(10) Silvestri, R.; De Martino, G.; La Regina, G.; Artico, M.; Massa,
S.; Vargiu, L.; Mura, M.; Loi, A. G.; Marceddu, T.; La Colla, P. J. Med.
Chem. 2003, 46, 2482.
LiCl
(11) For details regarding solvents screened, see the Supporting Informa-
tion.
(12) Over extended reaction times diphenyl disulfide was observed.
Unfortunately, protic acids had no pronounced effect on
the reaction, and potassium chloride and sodium chloride
566
Org. Lett., Vol. 8, No. 4, 2006

Table 3. Indole Scope and Limitations
Table 4. Thiophthalimide Scope
a Reaction performed at 100 °C.
be seen in Table 3, a wide variety of indoles can be
sulfenylated, including the less reactive indole-2-carboxylates
and 2-formyl-indoles (entries 9 and 10, respectively). Fur-
thermore, indoles containing potentially base and/or Lewis
acid sensitive functionalities such as nitriles, carboxylic
esters, aromatic chlorides, and aldehydes were effectively
sulfenylated in good yields (entries 4, 5, 7, and 10).
Having established the scope with respect to the indole
moiety, we next turned our attention to the thiophthalimide
component (Table 4). Gratifyingly, a range of alkyl and aryl
thiophthalimide reagents were effective sulfenylating agents.
Again, potentially base or Lewis acid sensitive functionalities
(entry 2) were well tolerated under the reaction conditions.
Thiophthalimide reagents containing hindered alkyl groups
such as isopropyl and cyclohexyl reacted smoothly to provide
the corresponding sulfenylated indole product (entries 3 and
4).
a Reaction was performed at 120 °C.
tetrabutylammonium chloride was added during the course
of the reaction, a significant increase in conversion was
observed (Table 2).
We therefore assumed the reaction was being promoted
by the halide counterion rather than the metal. Consequently,
we attempted the reaction in the presence of catalytic
tetrabutylammonium halide salts. In all cases, reactivity
almost comparable to the metal salts was observed, indicating
that the halide counterion was indeed the catalytically active
species.
After demonstrating the reaction with a range of thioph-
thalimide moieties, we began to assess more challenging
substrate combinations.
For example, the sulfenylation of indole-2-carboxylates
with a sulfur-transfer reagent bearing a hindered alkyl group
has proved to be a particularly difficult transformation.3,4
Under our standard reaction conditions, 2-carboxyethylin-
dole reacted sluggishly (∼5% conversion over 24 h),
resulting in the increased decomposition of the reagent and
formation of unwanted byproducts. To increase the reactivity
of the indole, we screened a range of common bases.
With our optimized set of conditions in hand, we began
to assess the scope and limitations of the reaction. As can
Org. Lett., Vol. 8, No. 4, 2006
567

We were pleased to find that NEt3 in combination with
MgBr2 successfully promoted the reaction (Scheme 3). Under
Scheme 4. Proposed Mechanism for Halide-Catalyzed
Sulfenylation
Scheme 3. Sulfenylation of Deactivated Indoles with
Hindered Thiophthalimide Reagents
to produce the transient sulfenyl halide. This then undergoes
reaction with the indole moiety to produce the sulfenylated
product.
In summary, we have developed an efficient, general
indole sulfenylation protocol that is catalytic in halide. Most
notably, this procedure can be used to sulfenylate deactivated
indoles with sterically demanding sulfur groups in an efficient
manner.
our modified conditions, indole-2-carboxylates react smoothly
with N-(isopropyl)thiophthalimide in the presence of 0.5
equivalents of MgBr2 and 0.7 equivalents of NEt3 to afford
previously inaccessible sulfenylated indoles in good yield.
Acknowledgment. We thank Tom Novak (Merck & Co.
Rahway, N.J.) for obtaining mass spectral data.
Studies are now underway to extend this methodology to
other aromatic systems and determine the role of the halide
in the reaction. The proposed mechanism, shown in Scheme
4, details our current hypothesis where catalytic quantities
of halide salt can react with the N-thiophthalimide reagent
Supporting Information Available: Experimental in-
formation and data for all new compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL052615C
568
Org. Lett., Vol. 8, No. 4, 2006