Scheme 2
Table 1. Optimization of Buchwald Amination of 16
entry
catalyst/liganda
base
solvent (T, °C) % yieldb
1
2
3
4
5
6
7
8
Pd(OAc)2/BINAP
Pd(dba)2/BINAP
Pd(OAc)2/BINAP
Pd(OAc)2/BINAP
Pd(dba)2/P(t-Bu)3
Pd(OAc)2/BP(t-Bu)2 tBuOK
Pd(OAc)2/BP(t-Bu)2 tBuOK
Pd(dba)2/ BP(t-Bu)2 tBuOK
Cs2CO3
Cs2CO3
tBuOK
Cs2CO3
Cs2CO3
toluene (100)
toluene (100)
toluene (100)
aminec (100)
toluene (80)
toluene (80)
amine (80)
55
57
49
53
70
75
92
90
amine (80)
a Pd catalyst (0.1 equiv) was used for all entries. Ratio of Pd to ligand
was 1:1.5-2 except for entry 5, where the ratio was 1:1. Reactions were
run overnight at the indicated temperature. BP(t-Bu)2 is (o-biphenyl)P(t-
Bu)2. b Yields are isolated yields after silica column purification. c Reactions
were run in neat amine at the indicated temperature. For all entries in Table
1, the amine is cyclopentylamine. A variety of other primary and secondary
amines also couple under the conditions outlined.
While Scheme 2 gives 2 in 30% overall yield from 12,
we felt that we could improve the efficiency of the synthesis
by avoiding the low-yielding formation of the vinylogous
amide 15.
As such, we examined Stille methodology to couple the
pyrimidine directly to the 3-iodoimidazopyridine 17, thereby
avoiding construction of the pyrimidine.15 Compound 17 was
formed in nearly quantitative yield by treating 13 with
N-iodosuccinimide. The required stannane 18 was synthe-
sized in two steps as described in the literature.16 Coupling
of 17 and 18 afforded only a modest 50% yield of 19.17
Oxidation of 19 with m-CPBA, followed by treatment with
cyclopentylamine at elevated temperatures, gave 16. The poor
yield for the Stille coupling, the tedious synthesis of the
coupling partner 18, and the undesirable aspects of tin use
made this route less appealing.
We then decided to investigate an alternative route for
building the pyrimidine, avoiding the vinylogous amide.
Looking at the mechanism by which vinylogous amides
condense with guanidines to form pyrimidines, we reasoned
that an alkynyl ketone could serve as a surrogate for the
vinylogous amide.18 Thus, formylation of 13 under Vils-
meier-Haack conditions gave an excellent yield of the
remarkably stable aldehyde 20. This aldehyde was treated
with the commercially available ethynyl Grignard reagent
at low temperatures to give the propargyl alcohol in excellent
yield. This alcohol was easily oxidized to the ketone 21 using
of 2-amino-3-chloropyridine with 2-bromoacetophenone 4
gave 13, which was subjected to acetylation conditions as
described above to give 14 in excellent yield. Treatment of
14 with DMF-DMA gave the desired vinylogous amide 15,
but in only a moderate 57% yield (similar to the yield
previously observed for the conversion of 8 to 9). Condensa-
tion of 15 with 10 gave the 8-chloroimidazopyridine 16.
Initial attempts at thermally displacing the 8-chloro substitu-
ent with cyclopentylamine failed, but the 8-position could
be coupled with cyclopentylamine using standard Buchwald
amination conditions (Pd(OAc)2, rac-BINAP, Cs2CO3).13
These conditions gave a modest 55% yield of the desired
substituted imidazopyridine 2 along with a significant
quantity of undesired dechlorinated imidazopyridine byprod-
uct. The amination conditions were optimized as outlined
in Table 1.
Attempts to increase the yield of 2 by varying the
palladium catalyst (entry 2), base (entry 3) or by running
the reaction in neat cyclopentylamine (entry 4) failed. By
replacing the BINAP ligand with P(t-Bu)3 or more conve-
niently the air-stable (o-biphenyl)P(t-Bu)2 ligand,14 we were,
however, able to improve the yields to >70% (entry 5 and
6). Under these conditions we still observed formation of
small quantities of the dechlorinated imidazopyridine. Finally,
by using neat cyclopentylamine instead of toluene, we were
able to obtain >90% of the desired product and did not
observe formation of the dechlorinated byproduct.
(14) (a) Hartwig, J. F.; Kawatsura, M.; Hauck, S. I.; Shaughenessy, K.
H.; Alcazar-Roman, L. M. J. Org. Chem. 1999, 64, 5575. (b) Wolfe, J. P.;
Tomori, H.; Sadighi, J. P.; Yin, J.; Buchwald, S. L. J. Org. Chem. 2000,
65, 1158.
(15) For Suzuki-type cross-coupling of 3-iodoimidazopyridines, see:
Enguehard, C.; Renou, J.-L.; Collot, V.; Hervet, M.; Rault, S.; Gueiffier,
A. J. Org. Chem. 2000, 65, 6572.
(16) (a) Sandosham, J.; Undheim, K. Tetrahedron 1994, 50, 275. (b)
Majeed, A. J.; Antonsen, O.; Benneche, T.; Undheim, K. Tetrahedron 1989,
45, 993.
(17) Suzuki couplings of 17 with pyridylboronic acids gave 50-60%
yield of the desired coupled product.
(18) A similar cyclization protocol appeared in the literature during the
course of our work: (a) Adlington, R. M.; Baldwin, J. E.; Catterick, D.;
Pritchard, G. J. J. Chem. Soc., Perkin Trans. 1 1999, 855. (b) Adlington,
R. M.; Baldwin, J. E.; Catterick, D.; Pritchard, G. J. Chem. Commun. 1997,
1757.
(12) We prepared 2-amino-3-chloropyridine by treating 2,3-dichloropy-
ridine with aqueous ammonia at 190 °C for 48 h in a steel bomb. Also see:
Trapani, G.; Franco, M.; Latrofa, A.; Ricciardi, L.; Carotti, A.; Serra, M.;
Sanna, E.; Biggio, G.; Liso, G. J. Med. Chem. 1999, 42, 3934.
(13) (a) Wolfe, J. P.; Wagaw, S.; Buchwald, S. L. J. Am. Chem. Soc.
1996, 118, 7215. (b) Harris, M. C.; Geis, O.; Buchwald, S. L. J. Org. Chem.
1999, 64, 6019. (c) Wolfe, J. P.; Buchwald, S. L. J. Org. Chem. 2000, 65,
1144.
Org. Lett., Vol. 5, No. 8, 2003
1371