Date Published: January 01, 2018
Publisher: International Union of Crystallography
Author(s): Bohdan Schneider, Paulína Božíková, Iva Nečasová, Petr Čech, Daniel Svozil, Jiří Černý.
Large deformations of the DNA double helix induced by interactions with proteins and small molecules are necessary to support the biological function of DNA. Here, the software tools available at https://dnatco.org that classify the dinucleotide building blocks into 44 distinct structural classes and 11 letters of a first DNA structural alphabet are presented and are used to analyze several prototypical DNA structures.
The prevailing DNA architecture, a right-handed double helix composed of sequentially complementary antiparallel strands, is structurally very plastic. Local deformations of DNA induced by interactions with binding partners, proteins and small-molecule drugs are necessary to convey its biological function, conservation and transfer of genetic information. The DNA strand has the ability to accommodate large conformational changes by forming kinks or bends or to undergo radical rearrangements into loops or folded forms such as quadruplexes. These conformational changes cannot be understood without going beyond the ‘A–B–Z’ classification traditionally used to describe DNA structural diversity. The necessity of understanding the DNA conformational space in its full complexity increases with the increasing number of biologically important DNA structures other than the double helix: tetraplexes with complicated and variable topologies, single-stranded hairpins and cruciforms, and DNA junctions involved in recombinant processes such as Holliday junctions.
Our analysis of DNA structures led to the identification of 44 DNA step conformer classes called NtC (Supplementary Table S1). Structurally similar NtC classes were grouped into 11 letters of the DNA structural alphabet CANA (Table 1 ▸, Fig. 3 ▸), which makes the analysis of DNA structure more comprehensible yet does not compromise the impartiality of the structural description and provides a tool to characterize the DNA structure beyond a rough classification into BI-, BII-, A- and Z-DNA types. Annotation of the conformational properties of a few archetypal types of DNA structures revealed some unexpected features. The Dickerson–Drew dodecamer, which is often considered to be a typical B-DNA duplex, is conformationally rich, with a high proportion of features mixing B and A forms. Our analysis of duplex models based on the fibre-diffraction data discloses the need for their critical evaluation before they are used for computer modelling (Fig. 7 ▸, Supplementary Fig. S4). Conformational analysis of guanine quadruplexes demonstrates the universality of the most frequent B conformer, BB00, which builds the tetrad cores of these folded DNA in combination with more exotic conformers such as BBS1 or BB1S.
The following references are cited in the Supporting Information for this article: Adhireksan et al. (2014 ▸), Ahlrichs et al. (1989 ▸), Aymami et al. (1999 ▸), Boer et al. (2006 ▸), Davis et al. (2007 ▸), Frouws et al. (2016 ▸), Glas et al. (2009 ▸), Gleghorn et al. (2008 ▸), Goedecke et al. (2001 ▸), Johnson & Beese (2004 ▸), Jurečka et al. (2007 ▸), Kondo et al. (2014 ▸), Larkin et al. (2007 ▸), Mondragón & Harrison (1991 ▸), Nicoludis et al. (2012 ▸), Patikoglou et al. (1999 ▸), Schäfer et al. (1994 ▸), Schellenberg et al. (2012 ▸), Sriram et al. (1992 ▸), Swan et al. (2009 ▸), Tao et al. (2003 ▸), Temperini et al. (2003 ▸) and Williams et al. (2008 ▸).