< Terug naar vorige pagina

Publicatie

Rational Design of XNA Processing Enzymes

Boek - Dissertatie

Xenobiotic Nucleic Acids (XNA) are synthetic analogues of natural nucleic acids (DNA and RNA). They are developed for applications such as therapeutics, diagnostics and alternative carriers of genetic information. The main goal of the PhD research is to develop a 'toolbox' with enzymes for XNA in molecular biology that is based on rational insights. In chapter 2, currently available computer-driven algorithms were explored for their potential use to alter the affinity of DNA-binding enzymes for DNA towards XNA polymers. The Chlorella virus (ChVLig) DNA ligase was used as a test model for this. Due to the low uniformity between the results, these were categorized and compared within each cluster of algorithms to make a rational initial selection of potential mutants. After determining the optimal conditions to express wild-type ligase and its variants, selected mutants were screened for their activity to join (ligate) broken XNA fragments with different hybrid setups. Although no activity was observed for XNA-modified fragments, the generated results emphasized the general importance of the structure of the enzyme rather than its amino acid sequence. Based on these results, it was concluded that not only the amino acid composition of the enzyme but also the general structure and dynamics of the enzyme-substrate complex are critical to be included in rational protein redesign experiments. To study this, a reliable model for the desired XNA complex is necessary. In chapter 3, a new computer-aided method was developed to generate reliable models for nucleic acids that are centrally bound in toroidal proteins. This approach was applied again to ChVLig, the test model also used in Chapter 2. The in silico results indicated that a stable enzyme complex with XNA could only be obtained if the central cavity in the torus was enlarged with an extra amino acid. The in vitro results confirmed that the 189insG mutant of ChVLig clearly shows the intended enzyme activity. This 189insG mutant from ChVLig is an important new tool for synthetic genetics that enables the synthesis of longer XNA gene fragments, complementing the 'toolbox' with existing XNA polymerases. In chapter 4, the method developed in chapter 3 was generalized into a publicly accessible software platform where a model can be generated for XNA-bound nucleic acid binding proteins starting from a PDB structure available in complex with DNA or RNA. In addition, the concept was further expanded to perform mutation analyzes in silico to optimize XNA binding and obtain the orthogonality required for use in xenobiology (PREDICT). Along with a new nucleic acid fragment screening test, the initial results obtained for ligases, endonuclease and polymerases provide a solid groundwork for the rational development of XNA-processing enzymes.
Jaar van publicatie:2020
Toegankelijkheid:Open