Organic Letters
Letter
of 1,5-diketones and an additional oxidative aromatization step.
Other reported [5+1] cycloadditions require commercially
unavailable starting materials or/and transition metal catalysts
in combination with organohalides.9c,10 Developing versatile
and efficient methods for the construction of functionalized
pyridines would be highly desirable, especially to fuse a
pyridine moiety to a target molecule during a late stage from
readily accessible starting materials. In this paper, we report a
metal-free [5+1] procedure for the synthesis of highly
functionalized pyridines.
Scheme 3. Studies of the [5+1] Cycloadditions
In our previous synthesis of Vinca alkaloids, we have
developed a practical method for accessing 1,3-dioxinone-
derived tert-butanesulfinamides.11 We envisioned that the 1,3-
dioxinone-containing sulfonamides could serve as 4C−1N
building blocks for the synthesis of functionalized pyridines
(Scheme 2). We postulated that the tert-butanesulfinyl group
Scheme 2. Proposed [5+1] Cycloaddition for the Synthesis
of Polysubstituted Pyridines
reactions with amide 10b, obtained from Boc protection of
10a, but failed again to give any pyridine derivatives. These
results suggested that pyridine formation might follow a
different pathway. It was deduced that the tert-butanesulfina-
mide might first experience a Pummerer-type rearrangement13
to yield sulfenimine (14) in the presence of excess Vilsmeier
reagent, and the newly generated sulfenimine would attack the
Vilsmeier reagent followed by aza-Prins cyclization of the 1,3-
dioxinone moiety (Scheme 3). With this in mind, we next
conducted a reaction with 2.0 equiv of Vilsmeier reagent in
DMF. To our delight, sulfenimine 14 was isolated. Under
standard Pummerer-type rearrangement conditions,13 sulfeni-
mine 14 was obtained in 90% yield. Treatment of sulfenimine
with Vilsmeier reagent provided pyridine in 75% yield. It was
clear that sulfenimine 14 was involved as an intermediate in
this cascade process.
To gain further insight into the reaction mechanism, several
control experiments were conducted with sulfenimine 14
(Scheme 4). To exclude the possible oxidation by oxygen in
this process, the reaction system was degassed, and the
reaction was carried out under the protection of argon. There
was no difference observed between reactions conducted in
argon and air. Next, sulfenimine 14 was treated with m-CPBA
(1.1 equiv) in dichloromethane to afford sulfinamide 15 and
sulfonamide 16 [existing as a mixture of imine and enamine
sulfonamide 16 with Vilsmeier reagent did not give the
desired pyridine. With substrate 15, pyridine 12a was obtained
in only 5% yield, with chloro-pyridine product 12b being
obtained in 36% isolated yield. In this case, a Pummerer-type
rearrangement occurred and resulted in a Michael acceptor;
the chloride anion thus attacked the conjugated system
would be removed under acidic conditions on the way to
generate Vilsmeier reagent, and the resulting primary amine
(10) would act as a nucleophile to attack the Vilsmeier
reagent, the C1 unit in the desired pyridine ring, to form the
aza-Prins cyclization precursor. After Prins cycloaddition of the
1,3-dioxinone moiety followed by late stage oxidative
dehydrogenation of dihydropyridines (11) as in Hantzsch’s
pyridine synthesis, the desired multifunctional pyridines would
be produced [general structure 12 (Scheme 2)].
We initially investigated the proposed reaction in Scheme 2
with tert-butanesulfinylimine 7a derived from condensation of
commercially available benzaldehyde and tert-butanesulfina-
mide.12 Mannich addition of the dioxinone-derived lithium
dienolate to sulfinyl imine 7a provided sulfinamide 9a in 91%
yield according to our previous conditions.11 Next, we
prepared Vilsmeier reagent by addition of phosphoryl
trichloride (6.0 equiv) to DMF. Introduction of sulfinamide
9a to Vilsmeier reagent in DMF resulted in a new compound
(12a) in 70% yield. To our surprise, the NMR spectra of this
new compound clearly indicated that pyridine was formed
rather than dihydropyridine. This result prompted us to ask
whether an air-induced concomitant autoxidative aromatiza-
tion occurred. To answer the question, we decided to carry out
the reactions step by step (Scheme 3). The sulfinyl group in
compound 9a was removed under acidic conditions to give
10a in 82% yield. Treatment of 10a with Vilsmeier reagent in
DMF unfortunately failed to yield the desired dihydropyridine,
with olefin 13 being formed instead. We also conducted the
B
Org. Lett. XXXX, XXX, XXX−XXX