Switchable Catalysis: Modular Synthesis of Functionalized Pyrimidinones via Selective Sulfide and Halide Cross-Coupling Chemistry

Article abstract of DOI:10.1021/ol035649y

(Equation presented) A general demonstration of orthogonal selectivity of the Liebeskind-Srogl cross-coupling protocol compared to the Suzuki-Miyaura and Stille variants is reported.

Full text of DOI:10.1021/ol035649y

ORGANIC
LETTERS
2003
Vol. 5, No. 23
4349-4352
Switchable Catalysis: Modular
Synthesis of Functionalized
Pyrimidinones via Selective Sulfide and
Halide Cross-Coupling Chemistry
,‡
Carrie Kusturin,† Lanny S. Liebeskind,* Hayat Rahman,† Kirby Sample,†
,†
Barbara Schweitzer,† Jiri Srogl,‡ and William L. Neumann*
Pfizer Global Research and DeVelopment, St. Louis Laboratories,
800 North Lindbergh, St. Louis, Missouri, and Emory UniVersity,
Department of Chemistry, 1515 Pierce DriVe, Atlanta, Georgia 30322
wlneumann@mindspring.com
Received August 28, 2003
ABSTRACT
A general demonstration of orthogonal selectivity of the Liebeskind−Srogl cross-coupling protocol compared to the Suzuki−Miyaura and Stille
variants is reported.
Heterocyclic structures have long been important to the
pharmaceutical industry. In recent years, heterocycles have
been successfully utilized as scaffold systems for lead
discovery and prospecting for biological activities via the
combinatorial and parallel medicinal chemistry formats.1 As
heterocycles can often be considered “condensed, confor-
mationally constrained” derivatives of amino acids, they are
ideal central platforms for the generation of peptide sur-
rogates.2 These systems can be designed to spatially display
remnants of peptide secondary structure and side-chain
functionality to enable the preferred ligand-receptor con-
formational ensemble in concordance with the strategic
movement from peptides to peptidomimetics possessing the
required pharmacokinetic characteristics.3
for traditional analogue synthesis and may have application
to parallel methodologies. This strategy can be envisioned
as a combination of “graded functionality” and “orthogonal
reactivity”.4 For selectivity, it relies upon the mechanistic
differences between the Suzuki-Miyaura5 and Stille6 cross-
coupling of halides compared to the Liebeskind-Srogl
protocol for thioorganics.7
Pyrimidinone scaffold 4 was readily prepared from com-
mercially available thiouracil 1 in three steps. Conversion
(1) Horton, D. A.; Bourne, G. T.; Smythe, M. L. Chem ReV. 2003, 103,
893. Villalgordo, J. M.; Heras, M.; Font, D. J. Comb. Chem. 2003, 5, 311.
Krchnak, V.; Holladay, M. W. Chem ReV. 2002, 102, 61. Franzen, R. G. J.
Comb. Chem. 2000, 2, 195.
(2) Lewis, J. G.; Bartlett, P. A. J. Comb. Chem. 2003, 5, 278. Ripka, A.
S.; Rich, D. H. Curr. Opin. Chem. Biol. 1998, 2, 441.
(3) Bursavich, M. G.; Rich, D. H. J. Med. Chem. 2002, 45, 541.
(4) Abell, C.; Oliver, S. F. Curr. Opin. Chem. Biol. 1999, 3, 299 and
referenced cited therein.
Herein, we report a modular synthesis of highly function-
alized pyrimidinone heterocycles that has proven efficient
(5) Suzuki, A. Pure Appl. Chem. 1994, 66, 213. Suzuki, A.; Miyaura,
N. Chem. ReV. 1995, 95, 2457.
(6) Stille, J. K. Pure Appl. Chem. 1985, 57, 1771.
† Emory University.
‡ Pfizer Global Research and Development.
10.1021/ol035649y CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/18/2003

Scheme 1
Table 1. Suzuki and Stille Couplings of Scaffold 4
to thioether 2 by reaction with sodium hydroxide and methyl
iodide followed by bromination afforded the known com-
pound 5-bromo-2-methylthiouracil (3).8 Synthesis of 4 was
accomplished by regioselective N-alkylation with tert-butyl
bromoacetate as described in Scheme 1. It is important to
emphasize the simplicity of this rapid large scale assembly
of a heterocyclic scaffold with three orthogonal points of
reactivity.
Table 1 depicts the isolated and purified yields9 of standard
Suzuki-Miyaura and Stille couplings at the bromide position
of 4. In all cases, the principal product observed by analysis
of the crude LCMS was formed via the desired bromide
cross-coupling reaction. No lone cross-coupling at the
thiomethyl center was detected, and only traces of double
coupling products were observed. In some cases (entries 1
and 2), anhydrous conditions with Cs2CO3 afforded accept-
able results. In reactions for which conversions were under
75% using Cs2CO3 as base (entries 3-7), aqueous carbonate
conditions proved very effective for acceleration of the
transmetalation step. The Stille variants (entries 8-10) were
somewhat sluggish compared to the Suzuki cases and
required higher temperatures and additional catalyst to
proceed to completion. Still, under these more stringent
conditions, good conversions and excellent selectivities were
observed.
Table 2 depicts the isolated and purified yields9 of selective
cross-couplings at the thiomethyl position of 4 in the presence
of the Suzuki- and Stille-active bromide. This unique
selectivity is effected by the use of a Cu(I) carboxylate as a
metal cofactor of higher thiophilicity than the Pd catalyst.7
In addition to facilitating transmetalation from boron or tin
a 10% catalyst. b 3%catalyst. c 105 °C.
to the -Pd-SMe bond,7 the results reported herein suggest
that interaction of the soft sulfur atom with the soft Cu(I)
metal facilitates selective oxidative addition at the thiomethyl
center through direct polarization of the C-S bond and/or
through coordination of the adjacent pyrimidine nitrogen.10
An elegant study reported by Jacobi, which supports this
hypothesis, details the activation of a methylthioimidate C-S
bond for oxidative addition to Pd(0) by Lewis acid coordina-
tion of the imidate nitrogen. The study also found that direct
(7) (a) Heteroarylthioethers: Liebeskind, L. S.; Srogl, J. S. Org. Lett.
2002, 4, 979. Egi, M. Liebeskind, L. S. Org. Lett. 2003, 5, 801. Alphonse,
F.-A.; Suzenet, F.; Keromnes, A.; Lebret, B.; Guillaumet, G. Org. Lett.
2003, 5, 803. For other thioorganic cross-couplings, see: Savarin, C.; Srogl,
J.; Liebeskind, L. S. Org. Lett. 2001, 3 (1), 91. Savarin, C.; Liebeskind, L.
S. Org. Lett. 2001, 3 (14), 2149. Srogl, J.; Liebeskind, L. S. Org. Lett.
2002, 4 (6), 979. Kusturin, C. L.; Liebeskind, L. S.; Neumann, W. L. Org.
Lett. 2002, 4 (6), 983. Liebeskind, L. S.; Srogl, J.; Savarin, C.; Polanco, C.
Pure Appl. Chem. 2002, 74 (1), 115. Egi, M.; Wittenberg, R.; Srogl, J.;
Liebeskind, L. S. Org. Lett. 2003, in press.
(10) An alternative explanation of the unique selectivity requires rapid
and reversible oxidative addition to both the C-Br and C-SMe bonds
followed by highly selective Cu(I) carboxylate activation of transmetalation
at the C-Pd-SMe center under these nonbasic conditions.
(8) Barrett, H. W.; Goodman, I. Dittmer, K. J. Am. Chem. Soc. 1948,
70, 1753.
(9) Unoptimized.
4350
Org. Lett., Vol. 5, No. 23, 2003

By proceeding under nonbasic conditions, the Liebeskind-
Srogl cross-coupling is distinguished from the Suzuki reac-
tion, which requires a base or fluoride additive to activate
transmetalation.5 Furthermore, it is suggested by the ac-
cumulated results that the Cu(I) carboxylate not only activates
the carbon-sulfur bond for oxidative addition, but also
activates transmetalation by coordinative delivery of the
carboxylate to the trivalent boron atom.7a For the boronic
acid cases (entries 1-7), all reactions afforded very high
conversions to the desired thiomethyl cross-coupling products
(often >85% purities were observed by direct analysis of
the crude LCMS traces). In all cases, less than 10% Suzuki
coupling was detected, which was presumed to occur after
the thiomethyl coupling reaction was complete.13
Table 2. Liebeskind-Srogl Couplings of Scaffold 4
Remarkably, for the organostannane cases (entries 8-10),
which generally do not require an additive for transmetala-
tion, very high selectivity for thiomethyl cross-coupling was
still observed. These results are a further testament to the
ability of the Cu(I) carboxylate to switch on catalysis at the
carbon-sulfur center. As was the case for the Suzuki-
Miyaura and Stille reactions demonstrated in Table 1, a wide
range of functionality is compatible with the Liebeskind-
Srogl protocol as evident by the variety of substituted boronic
acids and stannanes exemplified in Table 2. A particular
noteworthy example is that of entry 7. Not only does cross-
coupling occur selectively at the thiomethyl center but also
generates a molecule with two differentially reactive bro-
mides.
To fully illustrate the switchable selectivity and iterative
nature of this carbon-carbon bond forming chemistry, the
rapid assembly of compound 8 with three new points of
diversity is presented in Scheme 2. In this sequence, product
6g was selected for by employing the previously described
Cu(I) cofactor to activate the carbon-sulfur bond. By
substituting a base (Na2CO3) in place of the Cu(I) carboxy-
late, selective Suzuki-Miyaura coupling chemistry can now
be switched on at the more reactive 5-bromopyrimidinone
position. Thus, cross-coupling of 6g with 4-acetylboronic
acid afforded compound 7 in 78% yield. The sequence can
also be reversed by conducting the Suzuki-Miyaura coupling
first. As described in Table 1 (entry 4), 4-acetylphenylboronic
acid was coupled selectively at the bromide position using
basic conditions to afford compound 5d. Intermediate 5d was
then converted to compound 7 by cross-coupling with
3-bromophenylboronic acid in the presence of CuMeSal.14
Final Suzuki-Miyaura coupling with an olefinic boronic acid
at the remaining 3-bromophenyl position afforded function-
alized pyrimidinone example 8 in 91% yield.
a 2% catalyst. b THF solvent/54 °C. c Dioxane solvent/54 °C. d Dioxane
solvent/80 °C. e 10% catalyst.
In summary, the design and synthesis of an orthogonally
functionalized pyrimidinone system 4 has been reported.
Using this system, we have demonstrated the selective nature
coordination of the sulfur with Zn is also effective at
promoting Pd(0) oxidative addition.11 Thus, in the pres-
ence of CuTC or CuMeSal,12 substrate 4 undergoes selec-
tive Pd-catalyzed thiomethyl cross-coupling with boronic
acids and organostannanes in the presence of the bromide
which participated in standard coupling reactions shown in
Table 1.
(12) CuTC ) copper(I) thiophene-2-carboxylate. CuMeSal ) copper(I)
3-methylsalicylate.
(13) There was no evidence of bromide cross-coupling products in these
reactions in the absence of thiomethyl coupling. Only small amounts of
double couplings could be detected as side products even though in most
cases 1.5-2.0 equiv of boronic acid was used.
(14) This not only demonstrates the full selectivity of the coupling
methods but provides chemical correlation of product 7 with regard to the
differential reactivity of each bromide in 6g.
(11) Ghosh, I.; Jacobi, P. A. J. Org. Chem. 2002, 67, 9304.
Org. Lett., Vol. 5, No. 23, 2003
4351

Scheme 2
of the Liebeskind-Srogl thiomethyl cross-coupling protocol
Acknowledgment. We thank Dr. Samuel Tremont for
many helpful discussions and Drs. James Doom and David
Masters-Moore of PfizersSt. Louis for HRMS data.
versus the Suzuki-Miyaura and Stille reactions. Due to the
inherent mechanistic differences, coupling chemistry can be
switched on and off between halide and thiomethyl positions
by employing a base or a Cu(I) carboxylate cofactor as
appropriate. As this chemistry can be used in an iterative
fashion to rapidly build functionalized carbon frameworks
around a heterocyclic core, it has great promise in the parallel
synthesis format. Solid-phase synthesis applications of these
systems will be reported in the future.
Supporting Information Available: Complete descrip-
tion of experimental details and product characterization. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL035649Y
4352
Org. Lett., Vol. 5, No. 23, 2003