Glycosyl sulfonylcarbamates: new glycosyl donors with tunable reactivity.

Abstract

The advent of the Koenigs-Knorr method of glycosylation enabled the chemical synthesis of oligosaccharides.1 Since that advance, the differences in regiochemistry and stereochemistry of glycosyl linkages and variations in the stereoelectronic properties of glycosyl donors and acceptors have impelled the search for milder and more selective procedures.2 Successes in this pursuit have expanded the variety of complex, saccharidecontaining natural products accessible through chemical synthesis. The method of choice for a given donor-acceptor pair, however, depends on a number of variables. One such parameter is the nature of the anomeric leaving group. A divergent synthetic strategy to generate various glycosyl donors from a single intermediate would facilitate the construction of different glycosidic linkages. In search of such a donor, we explored the glycosylation reactions of glycosyl sulfonylcarbamates. We hypothesized that glycosyl sulfonylcarbamates could serve as glycosyl donors with a unique feature: the reactivity of these donors could be tuned by postsynthetic modification. Allyl3 and phenyl4 glycosylcarbamates have previously been shown to act as glycosyl donors. Likewise, we envisioned that treatment of glycosyl sulfonylcarbamates with an electrophilic promoter would result in a loss of CO2 and sulfonamide with production of a reactive glycosyl donor (Figure 1). A unique feature of the sulfonylcarbamate group is that it can be selectively altered by N-alkylation. We postulated that donors of differing reactivity could be generated through alteration of the characteristics of the resulting N-alkyl group.5 This approach offers significant advantages over the current methods of reaction tuning, which involve the independent syntheses of differently functionalized glycosyl donors of varying reactivities.6 Glycosyl sulfonylcarbamates are readily synthesized from the reaction of a sulfonyl isocyanate with the anomeric hydroxyl group of a protected saccharide (Figure 2). The putative glycosyl donors are formed in quantitative yield as a mixture of R and â isomers. The resulting compounds can be purified by silica gel chromatography, and are extremely stable; no decomposition is observed for samples stored at room temperature for three months. Although the two anomers can be separated, methods for generating the Ror â-isomer preferentially were developed. The â-anomer can be prepared by treatment of the starting material with excess 1,4-diazabicyclo[2.2.2]octane (DABCO) in toluene followed by addition of p-toluenesulfonyl isocyanate (TsNCO). The resulting glycosyl p-toluenesulfonylcarbamate is obtained in high selectivity (1:13 R: â ratio).4 When DBU (1,8-diazabicyclo[5.4.0]undec-7-ene) is used as a base, the R-isomer is favored (5:1 R: â). Since the â-isomer can be produced most efficiently, it was used to explore the feasibility and scope of the proposed glycosylation reaction. Compound 1 was tested as a glycosyl donor (Figure 3). We reasoned that 1 could be activated with Lewis acids to promote glycosylation. The most effective promoter was found to be trimethylsilyl triflate (TMSOTf). Treatment of 1 with TMSOTf results in the production of a silylated intermediate; presumably the trifluoromethanesulfonic acid that is generated activates this intermediate for glycosylation. Using these conditions, a variety of primary and secondary alcohols (Figure 4) were glycosylated in high yields, including hindered hydroxyl groups (Table 1, entries 8 and 9). As with typical glycosyl donors, however, the yields of some reactions were lower. For example, phenols were especially poor acceptors for glycosylation (entry 10). To optimize the donor reactivity for less nucleophilic acceptors, donors with alternative anomeric leaving groups were generated in a single step. The anomeric substituents that were employed vary in their ability to serve as a leaving group. Because of the low pKa of the sulfonylcarbamate group, we anticipated that N-alkylation of 1 could produce a variety of different donors.7 Backes et al. demonstrated in reactions of related acylsulfonamides that the electronic properties of an N-alkyl substituent could perturb the ability of the sulfonamide to serve as a leaving group.5 Given † Department of Chemistry. ‡ Department of Biochemistry (1) Koenigs, W.; Knorr, E. Chem. Ber. 1901, 34, 957. (2) For recent reviews, see: (a) Toshima, K.; Tatsuta, K. Chem. ReV. 1993, 93, 1503-1531. (b) Boons, G. J. Tetrahedron 1996, 52, 1095-1121. (c) Whitfield, D. M.; Douglas, S. P. Glycoconjugate J. 1996, 13, 5-17. (d) Garegg, P. J. AdV. Carbohydr. Chem. Biochem. 1997, 52, 179-205. (e) Davis, B. G. J. Chem. Soc., Perkin Trans. 1 2000, 2137-2160. (f) Seeberger, P. H.; Haase, W. Chem. ReV. 2000, 100, 4349-4394. (3) Kunz, H.; Zimmer, J. Tetrahedron Lett. 1993, 34, 2907-2910. (4) Prata, C.; Mora, N.; Lacombe, J. M.; Maurizis, J. C.; Pucci, B. Tetrahedron Lett. 1997, 38, 8859-8862. (5) Backes, B. J.; Virgilio, A. A.; Ellman, J. A. J. Am. Chem. Soc. 1996, 118, 3055-3056. Figure 1. Proposed formation of oxonium ion upon addition of an electrophile.