The tumor-associated carbohydrate antigens Globo-H, SSEA-3, and Gb3 were synthesized in a linear fashion using glycosyl phosphate monosaccharide building blocks. All of the building blocks were prepared from readily available common precursors. The difficult alpha-(1-->4-cis)-galactosidic linkage was installed using a galactosyl phosphate donor with high selectivity. Introduction of the beta-galactosamine unit required the screening a variety of amine protecting groups to ensure good donor reactivity and protecting group compatibility. An N-trichloroacetyl-protected galactosamine donor performed best for the installation of the beta-glycosidic linkage. Conversion of the trichloroacetyl group to the N-acetyl group was achieved under mild conditions, fully compatible with the presence of multiple glycosidic bonds. This synthetic strategy is expected to be amenable to the synthesis of the globo-series of tumor antigens on solid-support.

Dynemicin A is a member of the family of enediyne natural products. It is unique in that it combines a ten‐membered enediyne with an anthraquinone substructure. These features stimulated the development of synthetic approaches to the natural product itself and of analogs thereof. This review summarizes the total syntheses of dynemicin A. In addition, an overview of the known analogs is presented. The analogs can be classified according to the designed trigger mechanism. Most of the analogs contain a removable carbamate on the nitrogen atom. Others are quite similar to the natural lead in that they contain a quinone substructure, which upon reduction causes opening of the oxirane ring. In addition, there are analogs that contain an aromatic sector, the enediyne, and the oxirane ring but lack the nitrogen heterocycle. In these compounds the aryl ring assumes a different conformation from that in dynemicin A. Many of the simplified analogs proved to be quite active in vitro as well in vivo against murine tumor models. A highlight is compound 30 which is much more active than dynemicin A itself. However, looking at all analogs there is no clear‐cut correlation between the DNA‐cleaving ability at neutral pH and the in vitro results. From this one might conclude that there are possibly two mechanisms for antitumor activity. One involves diradical formation whereas the other might be due to a ligand‐receptor interaction.

No abstract

No abstract

No abstract