Programmable RNA tile self-assembly

Jinwen Yu, Purdue University


RNA self-assembly has emerged as a powerful building material in nanoconstruction, mainly because of their programmability and great potentials in various applications, especially cell-host applications, such as nanomedicines. However, currently RNA self-assembly strategies are limited by available building blocks or motifs. It’s desirable to expand the list of RNA motifs by designing novel and artificial RNA motifs. In this thesis, we firstly designed an artificial RNA motif, which can self-assemble themselves into a pre-defined nanoprism. This novel motif breaks current limitation in RNA tile based self-assembly, which requires multiple different copies of RNA tiles in a self-assembly process. We use PAGE , AFM, Cryo-EM and single particle reconstruction to analyze our data. The self-assembly process produced uniform and well-defined nanoprism as we designed. We also observed a synergistic effect in the self-assembly process which helps achieve the most energetically favorable products. Based on the success of the first project, we further design artificial RNA motifs folded from a single stranded RNA molecule. We demonstrate that by incorporating T-junction interaction with Kissing Loop interaction, we can design three types of xiv RNA motifs that can self-assemble themselves into 1D ladder, 2D arrays and 3D prism. During our exploration of siRNA delivery, we made our discovery in hybrid siRNA. Conventional siRNA duplex is formed by hybridizing two RNA molecules. We found out that after about 75% RNA nucleotides were substituted by DNA, the hybrid siRNA was still capable of suppressing GFP expression in HeLa-GFP cells. We further designed experiments to investigate this observation and proposed possible explanations. Finally, I reported the work on the design and self-assembly of novel double multi-arm junction (DMaJ) tiles. The DMaJ tiles are inspired by classic Double-crossover (DX) molecules. By adding sticky ends to the tiles, they could self-assemble into 1D chain and 2D arrays, matching the initial designs well.




Mao, Purdue University.

Subject Area


Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server