Enzymatic synthesis and isolation of thymidine diphosphate-6-deoxy-D-xylo-4-hexulose and thymidine diphosphate-L-rhamnose

A two-step enzymatic synthesis of dTDP-l-rhamnose is developed using enzymes from sonicated extracts of cultures of Escherichia coli K12 strains harboring plasmids containing different parts of the rfb gene cluster of Salmonella enterica LT2. The intermediate dTDP-6-deoxy-d-xylo-4-hexulose was isolated after a 1-h reaction, using only dTDP-d-glucose and dTDP-d-glucose 4,6-dehydratase, followed by protein precipitation and desalting by gel chromatography (yield 89%). In a two-step reaction using dTDP-d-glucose and dTDP-d-glucose 4,6-dehydratase in the first step, and with NADPH, dTDP-6-deoxy-d-xylo-4-hexulose 3,5-epimerase and NADPH:dTDP-6-deoxy-l-lyxo-4-hexulose-4-reductase in the second hour of incubation, the dTDP-d-glucose was fully converted to dTDP-l-rhamnose. The hexoses of both products were identified by mass spectroscopy. The molar yield of dTDP-l-rhamnose, after protein precipitation, anion-exchange chromatography and desalting by gel chromatography, was 62%, corresponding to more than 150 mg, starting from 250 mg of dTDP-d-glucose. When stored lyophilysed under nitrogen, these products were found to be stable for several months. Both dTDP-6-deoxy-d-xylo-4-hexulose and dTDP-l-rhamnose have light absorption maxima at 267 nm, with molar absorption coefficients close to that of dTMP. However, the absorption coefficient of dTDP-6-deoxy-d-xylo-4-hexulose at the absorption maximum of 320 nm (specific for sugars containing keto groups) was found to be approximately 20% higher than values presented earlier. Furthermore, an HPLC technique is presented for determining the net activity of dTDP-6-deoxy-d-xylo-4-hexulose 3,5-epimerase and NADPH:dTDP-6-deoxy-l-lyxo-4-hexulose-4-reductase, based on separation of dTDP-6-deoxy-d-xylo-4-hexulose and dTDP-l-rhamnose. The HPLC technique is also suitable for determination of all the nucleotide components involved in the synthesis.