imidazolones such as 2 using electron-deficient aryl fluorides
(Scheme 1). Imidazolones 2 can themselves be prepared via
cyclization of amino amides 3 with a suitable ortho ester 4.4
Scheme 3. Synthesis of 5-Aryloxyimidazole 14
Scheme 1. Retrosynthesis
Our first attempt using this strategy on a model system
gave an unexpected result (Scheme 2). Treatment of amino
Scheme 2. Undesired C-Arylation of 6
the next step. However, unlike 6, 11 was actually stable when
stored crude at room temperature for long periods (at least
2 months). Arylation of 11 using 3,5-dicyanofluorobenzene
7 proceeded regioselectively as we predicted to provide 12,
and none of the C-arylated material was detected. To aid
deprotection and to provide the desired N1-Et isomer
regioselectively, we performed the N-alkylation first. Indeed,
we have shown that if imidazole 12 is deprotected first there
is a preference for alkylation of the 5-aryloxy-NH-imidazole
to proceed via the “undesired” nitrogen regioselectively (∼3:
1). Thus, treatment of 12 with EtI provided imidazolonium
salt 13 and although the deprotection proved troublesome,
amide 55 with triethyl orthoacetate provided the known
imidazolone 6.6 However, subsequent treatment with com-
mercially available 3,5-dicyanofluorobenzene 7 surprisingly
gave a poor yield of C-arylated material 8 (proved by HMBC
and IR; see the Supporting Information) and none of the
desired adduct, despite trying various solvents and bases.
Poor yields for this reaction may be a reflection of the
instability of imidazolone 6 which slowly dimerizes when
stored at room temperature.6 We reasoned that introduction
of a protecting group to N3 would not only force the arylation
to the relatively less encumbered oxygen of the imidazolone,
but it may also improve the stability of this reactive
intermediate.
Following extensive examination of benzyl and 4-meth-
oxybenzyl protecting groups,7 we discovered that the 2,4-
dimethoxybenzyl group was ideal for our purposes as it was
resilient to the somewhat harsh conditions used for imid-
azolone formation, but which could be removed later in the
synthesis. The synthesis of 14 (Scheme 3), a representative
of this series, commenced with the known bromide 9,8 which
was reacted with 2,4-dimethoxybenzylamine to provide
amino amide 10. Cyclization with triethyl orthoformate
provided imidazolone 11, which was used immediately in
9
it could be effected using BBr3 to yield compound 14 in a
respectable overall yield.
With 14 in hand, we next turned our attention to func-
tionalization of the C1-position, which would allow us to
prepare various derivatives from a late-stage intermediate.
Reaction of 14 with paraformaldehyde proceeded smoothly
to give the hydroxymethyl derivative 15 (Scheme 4). The
hydroxy group now gives us a handle to prepare various
analogues, which is ongoing work within our group and will
Scheme 4. Preparation of Carboxamide 16
(4) For example: Maurer, F.; Hammann, I. DE 3136328.
(5) Conley, J. D.; Kohn, H. J. Med. Chem. 1987, 30, 567.
(6) Maquestiau, A.; Van Haverbeke, Y.; Flammang-Barbieux, M.;
Beaufays-Bar, F. Bull. Soc. Chim. Belg. 1976, 85, 573.
(7) These protecting groups could not be removed later in the synthesis
despite using various deprotection conditions, including BBr3, Pd-C/H2,
DDQ, CAN, and AlCl3. We believed the more electron-rich 2,4-dimethoxy-
benzyl group could be deprotected more readily later in the synthesis.
(8) Yamazaki, H.; Harada, H.; Matsuzaki, K.; Yoshioka, K.; Takase, M.;
Ohki, E Chem. Pharm. Bull. 1987, 35, 2243.
1726
Org. Lett., Vol. 8, No. 8, 2006