The self-assembly of DNA Holliday junctions studied with a minimal model

arXiv:0807.3280v2 [cond-mat.soft] 22 Dec 2008 The self-assembly of DNA Holliday junctions studied with a minimal model Thomas E. Ouldridge, Iain G. Johnston, and Ard A. Louis Rudolf Peierls Centre for Theoretical Physics, 1 Keble Road, Oxford, UK OX1 3NP Jonathan P. K. Doye Physical and Theoretical Chemistry Laboratory, Department of Chemistry, University of Oxford, South Parks Road, Oxford OX1 3QZ, United Kingdom (Dated: October 25, 2018) In this paper, we explore the feasibility of using coarse-grained models to simulate the self-assembly of DNA nanostructures. We introduce a simple model of DNA where each nucleotide is represented by two interaction sites corresponding to the sugar-phosphate backbone and the base. Using this model, we are able to simulate the self-assembly of both DNA duplexes and Holliday junctions from single-stranded DNA. We find that assembly is most successful in the temperature window below the melting temperatures of the target structure and above the melting temperature of misbonded aggregates. Furthermore, in the case of the Holliday junction, we show how a hierarchical assembly mechanism reduces the possibility of becoming trapped in misbonded configurations. The model is also able to reproduce the relative melting temperatures of different structures accurately, and allows strand displacement to occur. PACS numbers: 87.14.gk,81.16.Dn,87.15.ak I. INTRODUCTION The ability to design nanostructures which accurately self-assemble from simple units is central to the goal of engineering objects and machines on the nanoscale. Without self-assembly, structures must be laboriously constructed in a step by step fashion. Double-stranded DNA (dsDNA) has the ideal properties for a nanoscale building block,1,2 with structural length scales deter- mined by the separation of base pairs, the helical pitch and its persistence length (approximately 0.33 nm, 3.4 nm (Ref. 3) and 50 nm,4 respectively). Over these distances, dsDNA acts as an almost rigid rod and so it is capable of forming well-defined three dimensional struc- tures. It is the selectivity of base pairing between single strands, however, that makes DNA ideal for controlled self-assembly. By designing sections of different strands to be complementary, a certain configuration of a system of oligonucleotides can be specified as the global mini- mum of the energy landscape. In this way the target structure (usually consisting of branched double helices) can be ‘programmed’ into the sequences. This approach was initially demonstrated for a four-armed junction by the Seeman group in 1983.5 Such junctions and more rigid double crossover motifs6 can then be used to cre- ate two-dimensional lattices.7,8 Yan et al.9 have also con- structed ribbons and two dimensional lattices from larger four-armed structures, each arm consisting of a junction of four strands. Furthermore, using Rothemund’s DNA “origami” approach an almost arbitrary variety of two- dimensional shapes can be created.10 Progress in forming three-dimensional DNA nanos- tructures was initially much slower. The Seeman goup managed to synthesize a DNA cube11 and a truncated octahedron,12 but only after a long series of steps and with a low final yield. More recently, approaches have been developed that allow polyhedral cages, such as tetrahedra,13 trigonal bipyramids,14 octahedra,15,16 do- decahedra and truncated icosahedra,17 to been obtained in high yields simply by cooling solutions of appropri- ately designed oligonucleotides from high temperature. Additional structures have also been produced using pre- assembled modular building blocks incorporating other organic molecules.18,19 In designing strand sequences, it is important to min- imize the stability of competing structures with respect to the stability of the target configuration. In addition, if systems can be designed to follow certain routes through configuration space—for example, by the hierarchical as- sembly of simple motifs20—the target can potentially be reached more efficiently. A standard approach to hier- archical assembly, such as that described by He et al.,17 involves choosing sequences so that bonds between dif- ferent pairs of oligonucleotides become stable at different temperatures. This allows certain motifs to form in iso- lation at high temperatures before bonding to each other as the solution is cooled. An alternative, elegant system for programming assembly pathways has been proposed by Yin et al.21, which relies on the metastability of single stranded loop structures and the possibility of catalyzing their interactions using other oligonucleotides. Given these recent experimental advances in creating DNA nanostructures, it would be useful to have theo- retical models that allow further insights into the self- assembly process. In particular, a successful model would be able to provide information on the formation path- ways and free energy landscape associated with the self- assembly, and as such would be of use

…(Full text truncated)…

This content is AI-processed based on ArXiv data.