synthesis of urea-tethered neoglycopeptides, the strategy for
which is outlined in Scheme 1. Starting with urea-linked
Scheme 3. Preparation of N-Acetyl-D-glucosamine Isonitrile 2
Scheme 1. Synthetic Strategy of Neoglycopeptides
With the synthesis of isonitrile 2 established, we initially
examined the synthesis of urea glycosides by the reaction
of 1 with six different amines as summarized in Table 1. In
a typical case (entry A), oxidation of 2 was carried out with
pyridine N-oxide (3 equiv) and a catalytic amount of iodine
(7 mol %) in acetonitrile in the presence of water scavenger
(MS 3 Å).
The resulting solution, containing highly reactive isocy-
anate 1, was immediately treated with phenethylamine (2.0
equiv).14 To our delight, the urea glycoside 7a was isolated
in 92% yield. Primary amines having alkyl branches at the
R-carbon (entry B and C) as well as secondary amine (entry
D) smoothly reacted with 1 to afford the corresponding urea
glycosides 7b-d in good yields (>90%). Even in the case
of a sterically hindered secondary amine, such as diisopro-
pylamine (entry E), the corresponding urea glycoside 7e was
obtained in 91% yield. An application of 1 for the synthesis
of urea-tethered disaccharide is represented in entry F, where
aminosugar reacted with 1 to yield the urea-tethered pseudo-
disaccharide 7f in 92% yield.
N-acetyl-D-glucosamine amino acid conjugate A, solid-phase
synthesis would give the N-acetyl-D-glucosaminyl peptide
B. Transglycosylation of B using endo-â-GLcNAc-ase would
provide the urea-tethered neoglycopeptide C.9 In this paper,
we report on the synthesis of the urea-tethered glycosyl
amino acid (A, P1 ) Fmoc) and its use as a building block
for solid-phase synthesis based on the Fmoc-strategy.
Our retrosynthetic analysis of A is shown in Scheme 2. A
key feature of our plan was to construct the urea-glycosyl
bond through a coupling reaction between isocyanate 1 and
various R,â-diamino acid derivatives. Since such a highly
functionalized isocyanate 1 was only postulated to be a
transient intermediate by Pinter,10 the crucial step in our
approach was the synthesis of the reactive isocyanate 1. To
solve this problem, we planned to employ the oxidation of
isonitrile 2 for the generation of 1 under mild reaction
conditions.11
(8) Ichikawa, Y.; Nishiyama, T.; Isobe, M. Synlett 2000, 1253.
(9) (a) Seitz, O.; Wong, C.-H. J. Am. Chem. Soc. 1997, 119, 8766. (b)
Wang, L.-X.; Tang, M.; Suzuki, T.; Kitajima. K.; Inoue, Y.; Inoue, S.; Fan,
J.-Q.; Lee, Y. J. Am. Chem. Soc. 1997, 119, 11137. (c) Mizuno, M.; Haneda,
K.; Iguchi, R.; Muramoto, I.; Kawakami, T.; Aimoto, S.; Yamamoto, K.;
Inazu, T. J. Am. Chem. Soc. 1999, 121, 284.
Scheme 2. Retrosynthetic Analysis of A
(10) Pinter reported reaction of azide 5 with triphenylphosphine and CO2
and isolated the symmetrical carbodiimide ii. Although the intermediacy
of isocyanate 1, formed by reaction of iminophosphorane i with CO2, was
proposed, isolation or even detection of 1 has never been successful due to
the rapid reaction of i with reactive isocyanate 1. See: Kovacs, J.; Pinter,
I.; Messmer, A. Carbohydr. Res. 1987, 166, 101.
Starting with commercially available N-acetyl-D-glu-
cosamine 3 (Scheme 3), Horton’s protocol afforded R-chloro-
N-acetyl-D-glucosamine acetate 4,12 which was further
transformed into glycosyl azide 5 by the displacement
reaction with sodium azide under phase-transfer conditions.13
Catalytic hydrogenation of azide 5 followed by treatment of
the resulting glycopyranosylamine with acetic formic anhy-
dride furnished the formamide 6 in 73% yield over two steps.
Dehydration of 6 with triphosgene/triethylamine gave the
N-acetyl-D-glucosamine isonitrile 2 in good yield (81%).
(11) Ichikawa, Y.; Nishiyama, T.; Isobe, M. J. Org. Chem. 2001, 66,
4200.
(12) Horton, D., Johnson, A. L.; McKusick, B. C. Organic Syntheses;
Wiley: New York, 1973; Collect. Vol. V, p 1.
(13) Tropper, F. D.; Anderson, F. O.; Braun, S.; Roy, R. Synthesis 1992,
619.
(14) Although silica gel TLC analysis of the reaction mixture showed
the consumption of 2, isocyanate 1 could not be observed by TLC.
Accordingly, we employed 2 equiv of amines to optimize the yields.
5010
Org. Lett., Vol. 8, No. 22, 2006