Nucleoside phosphorylases from thermophiles - Recombinant expression and biocatalytic use for modified nucleosides

Abstract

Modified nucleosides are valuable pharmaceutical agents used in the treatment of cancer and viral infections. Moreover, they serve as building blocks in the synthesis of therapeutic oligonucleotides with advanced properties. While the chemical modification of pyrimidine nucleosides is generally well established, the synthesis of modified purine nucleosides is often rather challenging, resulting in multistage processes with low yield. Alternative synthetic routes include the chemo-enzymatic synthesis of purine nucleosides from a pyrimidine nucleoside serving as pentofuranosyl donor and a purine base functioning as pentofuranosyl acceptor. As biocatalysts, nucleoside phosphorylases (NPs) are used to catalyze the regioand stereoselective transfer reaction, whereby natural or chemically prepared artificial precursors can be applied as substrate. Unfortunately, a number of highly interesting nucleoside analogues are hardly recognized as substrate by NPs that are currently in use. Moreover, high temperatures are desirable to increase the concentration of poorly soluble purine bases, but many enzymes are rapidly deactivated by heat. Both factors limit the scope and the efficiency of NP mediated syntheses of modified nucleosides and prompted us to study novel, thermostable nucleoside phosphorylase variants as potential biocatalysts. Therefore a set of 5 NPs from 4 different thermophilic microorganisms (Deinococcus geothermalis, Geobacillus thermoglucosidasius, Thermus thermophilus, Aeropyrum pernix) has been overexpressed in E. coli. The recombinant proteins were characterized in order to assess their potential application as biocatalysts. Thermal properties (temperature optima, stability) varied significantly and were dependent on the source microorganism and the type of enzyme. Investigations of the substrate specificities revealed striking differences in the ability to tolerate modified nucleosides as substrate. The data allowed us to select and test the most promising combinations of enzymes for enzymatic transglycosylation reactions. In focus of the present work was thereby the synthesis of 2′-fluorinated purine nucleosides as well as 2,6-dihalogenated purine nucleosides. 2′-Fluorinated nucleosides were found to have valuable pharmaceutical properties and impart favourable characteristics to synthetic oligonucleotides. On the other hand, 2,6-dihalogenated purine nucleosides are versatile precursors for a variety of purine modified nucleosides. In comparison to E. coli enzymes that are described in literature as biocatalysts for the synthesis of 2′-fluorinated purine nucleosides, the application of the novel, thermostable enzymes permits the operation at higher temperature, and appears to be more efficient in the synthesis 2′-fluorinated purine nucleosides. Furthermore, 2,6-dihalogenated purines were readily accepted as substrates and the respective (deoxy-)ribosides were rapidly produced by the novel enzyme preparations. The results corroborate the general potential of thermostable NPs in the synthesis of modified nucleosides and specifically pave the way towards improved, environmentally friendly synthetic procedures affording valuable 2′-fluorinated and 2,6-dihalogenated purine nucleoside analogues. The present work was performed from April 2009 – April 2012 in the research group of Prof. Dr. Peter Neubauer (Laboratory of Bioprocess Engineering) at the Department of Biotechnology, Technische Universitat Berlin.