부산대학교 도서관

The most common conformation of double-stranded DNA (dsDNA) in biological systems is right-handed B-DNA with Watson and Crick base pairing. However, the considerable flexibility of the phosphodiester bonds at DNA structure allows conformational variations of DNA structure (Non-B-DNA) such as G-quadruplex, Cruciform and Z-form in specific conditions. Particularly, left-handed Z-DNA is one of the double helical structures which adopt left-handed conformation; all bases are in alternate anti- and syn- conformation compared with anti-conformation in B-DNA, which results in the zig-zag phosphate backbone. Since its discovery in 1979, the left-handed Z-DNA has drawn great interest in biological field. With the progress of the studies, the left-handed Z-DNA has been demonstrated that Z-DNA indeed dose exist in vivo. Moreover, it was also suggested that Z-forming sequences are widely found near transcription start sites in which play a functional role in biological processes such as gene expression and chromatin remodeling. In addition, the Z-DNA binding protein domain that extracted by a Z-DNA affinity column with a 5-bromodeoxycytosine-modified probe in the chicken blood nuclei was identified as double-stranded RNA adenosine deaminase 1 (ADAR1). From the further characterization of the ADAR1 protein, it was shown to possess a Z-DNA binding specific binding domain, named as Zα, at the N-terminus. Furthermore, the Zα domain exists in several other proteins such as E3L of poxviruses, DAI, and PKZ proteins. These important findings made it possible to study biological function of Z-DNA in cells. In these studies, Z-DNA binding proteins were further characterized to elucidate their correlation between structure and function by structural and biochemical experiments.In Chapter 2, changes in the kinetic behavior of the Zα-induced B-to-Z conversion by introducing single-stranded regions: In this study, the effect of unpaired region of DNA such as overhangs and hairpins on the B-to-Z conversion by Zα was examined. We used ds-(CG)6 core with 5’ or 3’ overhang, which are from 1 to 6 nucleotide-long with either A, C, G or T. The result showed that overhang DNAs were more rapidly converted to Z-DNA by comparing with that of non-overhang dsDNA. Especially, the B-to-Z conversion of ds-(CG)6-T series DNAs depended on the length of the overhang region and temperature. In view of these findings, we extended the investigation on the kinetic behaviors of B-to-Z conversion of hairpin-DNA. The CD analysis reveals the kinetics of B-to-Z conversion was also the length of loop-dependent. Taken together, Z-DNA formation by Zα via introducing single-stranded region may play a critical role in kinetics of the B-to-Z conversion process.In Chapter 3, effect of the positively charged and hydrophobic residues in the α3 helix of the Z-DNA binding domains from human ADAR1 and vaccinia virus E3L on the B-to-Z conversion: So far, it has been known that the Z-DNA binding activity of vvZαE3L was shown to play an important role in viral pathogenesis in mice and anti-apoptotic activity in HeLa cells. However, there is still an intractable-issue that why does the vvZαE3L cannot flip from B-DNA to Z-DNA in vitro? Until now, previous studies were only restricted to the key Z-DNA contacting residues which have been identified by the co-crystal structure. In this study, various chimeric and point mutants were created and characterized for B-to-Z conversion activity. The results showed that both positively charged residues in the C-terminal and hydrophobic residues of N-terminal of the α3 helix from hZαADAR1 and vvZαE3L play important roles in the B-Z transition. Moreover, these positively charged residues are possibly critical to the B-DNA binding affinity and further influence on the ability of B-to-Z conversion. These findings open new lines of inquiry that the positively charged and hydrophobic residues in the a3 helix region largely affect on the B-to-Z conversion process. Furthermore, it provides a new template for B-DNA binding-guided mechanism of B-to-Z conversion by Zα.In Chapter 4, reconstitution and characterization of Zα domains from Zα half-parts by anti-parallel heterodimeric leucine zipper were attempted. To investigate a role of each part of Zα in Z-DNA binding and B-Z conversion activity, we split hZADAR1 into two halves, the N-terminal half (hZADAR1N) and the C-terminal half (hZADAR1C) subdomains. These subdoamins were then fused to one of anti-parallel dimeric leucine zippers (for example, NZIP- hZADAR1C and hZADAR1N-CZIP) for stable expression and reassembly, respectively. Assisted by the interaction of anti-parallel heterodimeric leucine zippers, noncovalently-assembled chimeric proteins were created and characterized. The mix of NZIP- hZADAR1C and hZADAR1N-CZIP shows to have B-to-Z conversion activity as like the wild-type hZαADAR1. However, it is inconclusive that the C-terminal subdomains of Z domains alone could have B-to-Z conversion activity.