ABSTRACT
The emergence of information on the role that carbohydrates play in a variety of molecular recognition processes in biological systems [l) and the awareness of the convenient exploitation of these compounds as precursors to chiral building blocks for organic synthesis [2), have stimulated increased demand for practical synthetic methods in carbohydrate chemistry. In response, new chemical and biological synthetic strategies have been developed in recent years for the stereocontrolled synthesis of natural oligosaccharides and glyconjugates and their unnatural analogues [3). Among monosaccharides, considerable attention has been given to the synthesis of modified sugars and to compounds containing seven or more carbon atoms, the so-called higher-carbon sugars [4). Syntheses have been carried out by various approaches: (1) from nonsugar precursors [5); (2) direct coupling of two monosaccharide subunits ( C-glycosidation), or construction of a second unit on a preexisting one [6); (3) chain-elongation of either sugar or nonsugar starting materials by installation of a carbon chain bearing an apparent or masked functional group [7). Each of these synthetic strategies has its own validity, provided it involves chemically and stereochemically efficient steps. However, prominent approaches are those based on the homologation of readily available natural sugars because, in principle, they should permit accessing various types of compounds and higher-carbon sugars of any required length. Quite interesting, although multicarbon homologation appears attractive and spectacular because of the rapid growth of the chain, sometimes iterative one-carbon homologation is more convenient, for it permits a full range of stereochemical variations and the introduction of differentially protected functional groups. While moving from one-carbon homologation based on the cyanohydrin and nitromethane syntheses (i.e., the roots of a great deal of carbohydrate chemistry), we will first shortly review recent methods reported from other laboratories. For a critical and comparative evaluation, only those methods will be described wherein the whole sequence, starting from an aldose or a ketose, and ending with the corresponding one-carbon higher homologue has been completed. However, formal homologations (i.e., methods dealing with intermediates that have been previously transformed in aldoses or ketoses) will be also considered. We will then describe in some more detail our own method that is based on the application of the thiazole-aldehyde synthesis [8).
