Sucrose analogues modified at position 3: chemoenzymatic synthesis and inhibition studies of dextransucrases.

Conditions for the large-scale (molar) oxidation of sucrose by Agrobacterium tumefaciens were improved, thus leading to homogeneous solutions of 3-ketosucrose in 40% yield. Treatment of this solution with hydroxylamine or methoxylamine afforded the corresponding oximes 3a and 3b (isolated as acetates) in excellent yield. Dissolving-metal reduction of these oximes gave mixtures of amino disaccharides in which the gluco epimer (3-amino-3-deoxysucrose) was predominant. A more efficient approach to this amino sucrose was provided by the highly stereoselective hydrogenation of 3-ketosucrose peracetate (7), which gave exclusively the allo isomer 8 (2,4,6-tri-O-acetyl-alpha-D-allopyranosyl 1,3,4,6-tetra-O-acetyl-beta-D-fructofuranoside). Upon reaction with lithium azide, the triflate derived from 8, compound 9, afforded 3-azido-3-deoxysucrose peracetate (10) which was converted into 3-amino-3-deoxysucrose (12). The reaction of triflate 9 with potassium ethylxanthate led to a mixture of products (the expected 3-S-ethoxythiocarbonyl-3-thiosucrose derivative and the peracetates of 3-thiosucrose and of 3-thiosucrose disulfide), which could be all converted into 3-thiosucrose (17). Sucrose analogues 12 and 17 were not substrates of dextransucrases from various strains of L. mesenteroides, nor did they participate in glycosyl transfer reactions to an acceptor (maltose). Compounds 3a and 12 were found to be strong competitive inhibitors of the dextran synthesis process (dextransucrase from strain B-1397). These results indicate that 3a and 12 compete effectively with sucrose for the sucrose binding site but are unable to participate as glycosyl donors in the polymerization or glycosyl-transfer processes.