P41: Directed evolution of non-natural nucleoside analog drug biosynthesis from a dideoxyribose precursor

Sunday, August 12, 2012
Columbia Hall, Terrace Level (Washington Hilton)
William R. Birmingham1, Timothy D. Panosian2, David P. Nannemann3, Tina M. Iverson2 and Brian O. Bachmann3, (1)Biochemistry, Vanderbilt University, Nashville, TN, (2)Pharmacology, Vanderbilt University, Nashville, TN, (3)Chemistry, Vanderbilt University, Nashville, TN
Nucleoside analog drugs are mainstays in the treatment of viral infections but they are also expensive to manufacture, with up to 99% of treatment costs stemming from direct costs. Herein we describe a strategy to produce dideoxynucleosides using E. coli via a bioretrosynthetically engineered three step pathway. Having completed evolution of the product forming enzyme, purine nucleoside phosphorylase, we are currently engineering phosphopentomutase from Bacillus cereus, the penultimate enzyme, to continue the bioretrosynthesis paradigm. Phosphopentomutase catalyzes the interconversion of α-D-ribose 5-phosphate and α-D-ribose 1-phosphate. This simple transformation that activates ribose for nucleoside biosynthesis via a nucleoside phosphorylase can also be utilized to activate the non-natural sugar 2,3-dideoxyribose 5-phosphate for dideoxynucleoside biosynthesis. Analysis of phosphopentomutase substrate co-crystal structures suggested multiple residues as potential modulators of substrate binding.  Targeted saturation mutagenesis of these residues generated variants with up to >500-fold change in substrate specificity compared to the wild-type enzyme. Subsequent evolution by random mutagenesis has increased phosphopentomutase activity in cell free extract. The engineered phosphopentomutase, combined with the previously engineered purine nucleoside phosphorylase in vitro, shows capable production of the nucleoside analog reverse transcriptase inhibitor 2’,3’-dideoxyinosine (ddI, Videx®), a nucleoside analog currently prescribed as treatment for HIV. Furthermore, we report the in vitro production of dideoxyinosine using the complete three step biosynthetic pathway beginning from the precursor 2,3-dideoxyribose. Once fully optimized, this biosynthetic pathway may offer a complementary or possibly an entirely alternative route for production of dideoxynucleosides, with the potential to be additionally tailored to generate other clinically relevant nucleoside analogs.