Construction, analysis, ligation, and self-assembly of DNA triple crossover complexes

This paper extends the study and prototyping of unusual DNA motifs, unknown in nature, but founded on principles derived from biological structures. Artificially designed DNA complexes show promise as building blocks for the construction of useful nanoscale structures, devices, and computers. The DNA triple crossover (TX) complex described here extends the set of experimentally characterized building blocks. It consists of four oligonucleotides hybridized to form three double-stranded DNA helices lying in a plane and linked by strand exchange at four immobile crossover points. The topology selected for this TX molecule allows for the presence of reporter strands along the molecular diagonal that can be used to relate the inputs and outputs of DNA-based computation. Nucleotide sequence design for the synthetic strands was assisted by the application of algorithms that minimize possible alternative base-pairing structures. Synthetic oligonucleotides were purified, stoichiometric mixtures were annealed by slow cooling, and the resulting DNA structures were analyzed by nondenaturing gel electrophoresis and heat-induced unfolding. Ferguson analysis and hydroxyl radical autofootprinting provide strong evidence for the assembly of the strands to the target TX structure. Ligation of reporter strands has been demonstrated with this motif, as well as the self-assembly of hydrogen-bonded two-dimensional crystals in two different arrangements. Future applications of TX units include the construction of larger structures from multiple TX units, and DNA-based computation. In addition to the presence of reporter strands, potential advantages of TX units over other DNA structures include space for gaps in molecular arrays larger spatial displacements in nanodevices, and the incorporation of well-structured out-of-plane components in two-dimensional arrays.

Duke Scholars

Author John H. Reif Computer Science