A [2,3]-Wittig rearrangement approach towards the stereoselective synthesis of the C(10)–C(20) backbone of the fumonisins

Abstract

Fusarium verticillioides (= Fusarium moniliforme) a common fungal contaminant of maize throughout the world has been associated with diseases in both man and animals. The structure of the fumonisins, a family of structurally related mycotoxins isolated from cultures associated with the high incidence of human oesophageal cancer in the Transkei region in South Africa and with equine leucoencephalomalacia, a neurological disorder in horses and donkeys, has been established. The main mycotoxin, fumonisin B1 consists of the diester formed by the C(14) and C(15) hydroxyl groups of (2S,3S,5R,10R,12S,14S,15R,16R)-2- amino-12,16-dimethyleicosane-3,10,14,15-pentaol with the Si carboxy group of propane- 1,2,3-tricarboxylic acid. A comparison of the structures of the 28 known fumonisins reveals that they share a common structural motif for the C(11)-C(20) unit, and probably also the same stereochemistry for the 4 stereogenic centres present in this unit. Disconnection of the C(9)–C(10) bond in a retrosynthetic analysis of the fumonisins C20 backbone (C19 in the fumonisin C series)identifies (3S,5S,6R,7R)-3,7-dimethylundecane-1,5,6-triol as a common building block for the synthesis of any of the fumonisins. In the dissertation the retrosynthetic analysis of this 3,7-dimethylundecane-1,5,6-triol building block identifies (3S,4R,5R)-5-methylnonane-1,3,4-triol as a viable target which in turn could be derived from a simple starting material trans-4-hexen-3-one. Key reactions identified to realise the required transformations leading to the identified target included kinetic enzymatic resolution of the racemic alcohol obtained from trans-4-hexen-3-one, and a pivotal role for both the [2,3]-Wittig rearrangement and the use of Sharpless asymmetric epoxidation methodology as these reactions generated the requisite stereogenic centres present in (3S,4R,5R)-5-methylnonane-3,4-diol. In this manner a synthetic route from trans-4-hexen-3- one to (2R,3R,4R,5R,6E)-4-(benzyloxy)-2,3-epoxy-5-methylnon-6-en-1-ol using appropriate functional group transformations and protective group strategies, with complete stereochemical control, were developed in this work. Alternative strategies to overcome problems encountered during the synthesis are presented for future work. The conversion of the 4-(benzyloxy)-2,3-epoxy-5-methylnon-6-en-1-ol intermediate to the protected 5- methylnonane-1,3,4-triol target could not be carried out due to time constraints and material shortages.