Molecular modelling tools to design artificial enzymes

Xenobiotic nucleic acids (XNAs) are chemical analogues of natural nucleic acids, modified in at least one of their three main chemical moieties: nucleobase, sugar and phosphate backbone. As analogues of natural DNA and RNA with increased biological activity, they have diverse applications that are traditionally categorized according to their mechanism of action whether they interact directly with the cellular nucleic acid, as antisense or RNAi ONs, or have their own specific structure and function, as ribozymes and aptamers. The latter are single-stranded nucleic acids, capable of folding into specific three-dimensional structures that can be harnessed for the development of highly-specific high-affinity ligands. For the selection of XNA aptamers, a minimum set of enzymatic tools to manipulate XNA in vitro is required. However, biostable XNA are typically not recognized by enzymes found in nature. The further XNA structure diverges from its natural congeners, the more difficult it becomes for directed evolution to evolve native enzymes to process XNA. This project aims the development of a software tool for high-throughput ‘in silico’ structure-based domain swapping to create proteins with predefined functionalities. As a proof of concept, the developed software will be applied to design new enzymes to manipulate XNA with fully modified backbone in vitro and in vivo. Biochemical characteristics of obtained enzymes will be compared to those of enzymes that are engineered with available software tools such as molecular docking and full mutation analysis.