Scheme 4
Scheme 5
ridines are also susceptible to nucleophilic aromatic substi-
tution with alkoxides, and as such the chlorine provides a
handle for the introduction of the methoxy group at the
4-position of the pyridine. Compound 21 was prepared in
three steps from 2-picolinic acid by oxidation to 4-chloro-
picolinyl chloride using Sundberg’s procedure,9 condensation
of the acid chloride with diisopropylamine to provide the
diisopropyl amide, and ortho-lithiation of the amide with
LDA followed by trapping with iodine. Unfortunately,
treatment of 21 with either LDA or lithium 2,2,6,6-tetra-
methylpiperidide (LTMP) at -78 °C followed by an aqueous
workup provided variable yields of products. While some
of the desired product (22) was observed, the deiodinated
product 20 was always the major product, and at times it
was the sole product. We were unable to discover conditions
that reliably provided 22, and we speculate that deiodination
occurs due to the enhanced stability of the 3-lithio species
23, rendering 21 unusually susceptible to nucleophilic de-
iodination. We therefore decided to examine a different non-
exchangeable group at the 4-position, one that is less effective
as a carbanion stabilizing group so that de-iodination would
not be a problem, and chose a methoxy group (Scheme 5).10
The 4-methoxypyridine derivative 24 was prepared in 89%
yield from 20 by nucleophilic aromatic substitution using
sodium methoxide. Metalation of 24 with n-BuLi and
trapping with iodine provided 25 in 75% yield. Treatment
of 25 with LDA smoothly induced a 1,3-migration of the
iodide from the 3- to the 5-position and provided the more
stable 3-lithiopyridine, which upon aqueous workup provided
compound 26 in 88% yield and none of the deiodinated
compound (24). Nucleophilic aromatic substitution of the
iodide at the 5-position proved to be more difficult than at
the 4-position.11 We therefore utilized a modified Ullmann
coupling procedure that consisted of treating 26 with sodium
methoxide in dimethylformamide containing CuI at 80 °C
to provide 27 in 92% yield.12,13 In the absence of Cu(I) salts,
or in less polar solvents, significant quantities of the reduced
byproduct 24 were observed. With 27 in hand, we required
the installation of our coupling substituent at the 6-position
of the pyridine. This was accomplished as planned by
introduction of a bromine at the 3-position of 27 (n-BuLi,
Br2, 84%), followed by a 1,4-dance of the bromine from the
3-position to the 6-position (LDA -78 °C, CH3OH, 80%).
Thus, the 1,4-dance proceeded cleanly and in high yield to
provide 29 ready for coupling with a pyridine derivative.
Completion of the synthesis of caerulomycin C required
a cross-coupling with a suitably functionalized pyridine,
followed by functional group manipulations to convert the
amide to an oxime. We found that the best procedure for
this cross-coupling was a Negishi coupling,14 using Pd2(dba)3/
Ph3P as the catalyst, and obtained an 80% yield of bipyridyl
(11) Nucleophilic aromatic substitution at the 3- or 5-position of pyridines
is known to be more difficult than at the 2- or 4-position. See: Schofield,
K. Hetero-Aromatic Nitrogen Compounds Pyrroles and Pyridines; Ple-
num: New York, 1967; p 244.
(12) Keegstra, M. A.; Peters, T. H. A.; Brandsma, L. Tetrahedron 1992,
48, 3633.
(13) This product is contaminated with about 5% of the reduced
compound 27, which is difficult to separate at this stage, but is readily
removed in the next step.
(14) For a review, see: Negishi, E.-i. In Metal-catalyzed Cross-coupling
Reactions; Diederich, F., Stang, P. J., Eds.; Wiley-VCH: New York, 1998;
Chapter 1. The corresponding Stille reaction using 2-tributlystannyl pyridine
was less efficient and provided the desired product in 50% yield under
optimized conditions (Pd2(dba)3, P(2-furyl)3, CuI, DMF, 80 °C). See: Farina,
V.; Krishnan, B. J. Am. Chem. Soc. 1991, 113, 9585. Liebeskind, L. S.;
Fengl, R. W. J. Org. Chem. 1990, 55, 5359.
(8) Gschwend, H. w.; Rodriguez, H. R. Heteroatom Facilitated Lithia-
tions. Org. React. (N.Y.) 1979, 26, 1-30. Mongin, F.; Queguiner, G.
Tetrahedron 2001, 57, 4059.
(9) Sundberg, R. J.; Jiang, S. Org. Prep. Proced. Int. 1997, 29, 117.
(10) A methoxy group can be a better kinetic ortho-directing group than
a chlorine atom; however, chlorine can be a better thermodynamic carbanion
stabilizing group. For a discussion, see: Iwao, M. J. Org. Chem. 1990, 55,
3622. For a related competition experiment, see: Slocum, D. W.; Dietzel,
P. Tetrahedron Lett. 1999, 40, 1823. Iwao ascribed the kinetic preference
for deprotonation ortho- to a methoxy group to complexation to the
alkyllithium base; however, Collum has recently provided evidence that
complexation-induced proximity effects are not important in the lithiation
of anisole. See: Chadwick, S. T.; Rennels, R. A.; Rutherford, J. L.; Collum,
D. B. J. Am. Chem. Soc. 2000, 122, 8640.
Org. Lett., Vol. 4, No. 14, 2002
2387