A Computational Study of DNA Four-way Junctions and Their Significance to DNA Nanotechnology

The field of DNA nanotechnology has rapidly evolved over the past three decades, reaching a point where researchers can conceive of and implement both bioinspired and biomimetic devices using the programmed self-assembly of DNA molecules. The sophisticated natural systems that these devices seek to interrogate and to imitate have Angstrom-level organizational precision, however, and the nanotechnology community faces the challenge of fine-tuning their design principles to match. A necessity for achieving this level of spatial control is an understanding of the atomic-level physico-chemical interactions and temporal dynamics inherent to fundamental structural motifs used for nanodevice design.

The stacked configurational isomers of four-way junctions, the motif on which DNA nanotechnology was founded, are the focus of this work; initially in isolation and then as part of larger DNA nano-assemblies. The first study presented here investigates the impact of sequence on the structure, stability, and flexibility of these junction isomers, along with their canonical B-form duplex, nicked-duplex and single cross-over topological variants. Using explicit solvent and counterion molecular dynamics simulations, the base-pair level interactions that influence experimentally-observed conformational state preferences are interrogated and free-energy calculations provide a detailed theoretical picture of isomerization thermodynamics.

Next, the synergy of single molecule imaging, computational modelling, and a novel enzymatic assay is exploited to characterize the three-dimensional structure and catalytic function of a DNA tweezer-actuated nanoreactor. The analyses presented here show that rational redesign of the four-way junctions in the device enables the tweezers to be more completely and uniformly closed, while the sequence-level design strategies explored in this study provide guidelines for improving the performance of DNA-based structures.

Finally, MD simulations are used to inform finite-element method coarse-grained models for the ground-state structure determination and equilibrium Brownian Dynamics of large-scale DNA origamis. Together, this thesis presents a set of guidelines for the rational design of nanodevices comprising arrays of constrained four-way junctions. (Copies available exclusively from MIT Libraries, libraries.mit.edu/docs - docs@mit.edu)