;RNAJunction Database: The design of RNA based nanostructures is at least partially reliant on the components that comprise the individual building blocks. A good way of determining these components is by utilizing naturally occurring RNA motifs. Using this philosophy we developed the RNAJunction relational database, which consists of the PDB (Protein Data Base) representations of over 13,000 RNA kissing loops and n-way junctions. The database is available from our website www.ccrnp.ncifcrf.gov/bshapiro/. These junctions were found by scanning the entire PDB database of RNA structures using our JunctionScanner algorithm. This algorithm relies on results that are obtained by running the RNAview software that parses the PDB structure into standard Watson/Crick base pairs as well as several other non-canonical base interactions. The results from this parse are then used to determine the connectivity of the junctions and kissing loops. The database itself is divided into six sub-databases each of which can be searched in a multitude of ways. Three of these sub-databases categorize the motifs based upon the degree of well-formedness of the projecting helical stubs. The other categories reduce redundancy by clustering motifs based upon the criteria that each cluster must contain motifs with the same sequences and each of the motifs must be conformationally very similar to each other. Several search modes are available to the user. One mode of search that has proven to be very useful involves searching for motifs that have specific angle ranges between their helical stubs. Using this information, it is possible to add A-form helical connectors to attach these found motifs to others, ultimately forming a desired shape. The existence of this database has proven to be very valuable for the manual and combinatoric generation of RNA based nanostructures. RNA Hexagonal Nanoring and Nanotube: One of our goals is to design functional RNA nanoparticles that can be used, for example, for therapeutic purposes, as substrates for crystallography or for nanosensors. One of our first designs was of an RNA hexagonal ring and RNA nanotube (patent pending). Besides elucidating the properties of the RNA that forms such structures, the ring and tube may be engineered to include functional entities such as siRNAs, molecular beacons or aptamers. The existence of the RNAJunction database made it possible to computationally design an RNA hexameric nanoring and ultimately an RNA nanotube. A significant issue was to find a motif that could form approximately a 120 degree angle at each of six corners. By scanning the RNAJunction database the Col E1 kissing loop motif was discovered to have a 122 degree angle. This kissing loop structure was determined by NMR (one half designated as RNAIi and the other half as RNAIIi). Two forms of the building blocks were computationally designed. One form consists of two building block components (Nanoring A+B). The first component contains an RNAIi loop on both ends and the other contains an RNAIIi loop at both ends. The other form contains an RNAIi loop on one end and an RNAIIi loop on the other (self-dimer). The concept of the hexagonal nanoring was extended by adding appropriately engineered dangling ends oriented perpendicular to the ring plane in alternating patterns. This permitted the placement of multiple rings on top of each other by utilizing complementary dangling ends. The placement of these stacked rings result in the RNA based nanotube. NanoTiler - Software for RNA Nanostructure Design: To better facilitate the design of RNA based nanoparticles an extensive software system, NanoTiler, has been developed that permits RNA nanodesign at several different conceptual levels. The user can interface with the system via a graphical user interface or with a scripting language. A key feature of NanoTiler is its ability to accomplish combinatorial search of 3D RNA structure spaces by utilizing motifs derived from the RNAJunction database. A specified set of motifs can be placed in space and joined with A-form helix connectors. The connector lengths can be varied. This leads to a large combinatorial space of structures. Some of these structures form closed rings; others can form dendrimer-like conformations. Ring-formation can be detected automatically. Constraint satisfaction methods are also applied to improve ring closure and proper fit of connected helices. In addition, a graph that indicates the desired topology can be input into the design process of a structure. A graph matching algorithm is used to determine when a designed structure matches the desired topology. Once a desired topology is realized, NanoTiler can then be focused on producing a set of sequences that can be experimentally tested for the formation of the designed structure. A sequence-fusing algorithm connects fragments that were used in the generation of the conformational topologies. Next the sequence optimization algorithm can be applied in order to limit the amount of cross talk between the designed sequences. Sequences are repeatedly mutated, except for the portions that have to be maintained to preserve important motifs such as those obtained from the RNAJunction database, scoring each set of mutated sequences. NanoTiler in conjunction with other programs measures the degree of hybridization that occurs between the sequences and the degree of folding into the target secondary structure. Limitations on GC content and repetitious runs of sequence are also taken into account in the score. Once an optimized set of sequences is generated, mutations are substituted back into the 3D structure. This is accomplished by an algorithm in NanoTiler that searches known structures for the same base pairs that have a conformation similar to that needed in the generated structure. Once all fragments are designed, they are subjected to molecular mechanics minimization to fix bond lengths and angles. If desired, the entire structure or portions of the structure are subjected to molecular dynamics to characterize the dynamical qualities of the designed nanostructure. RNA2D3D - Software for RNA Nanostructure Exploration and RNA 3D Modeling: We developed another software package, RNA2D3D, that is being used for exploratory design of RNA based nanoparticles. An example of its use was illustrated in the design of a tecto-square and teco-mesh, which have been shown experimentally by others to be capable of self-assembly. Part of the modeling process allows the definition of a kissing loop interaction by the appropriate sculpting of the hairpins involved in the kissing loop. A pair of complementary bases can be specified allowing the loops to dock in a coaxial fashion. A list of these loop-loop interactions can be specified in a file that indicates the connectivity. In this way, one can establish the docked elements of, in this case, the four L-shaped components that make up the tecto-square. A similar approach can be taken to specify the base pairing interactions that make up the single stranded tails that bring multiple squares together to form a mesh. Modeling these components in this fashion immediately indicates that treating the building blocks as rigid bodies does not produce a closed ring [summary truncated at 7800 characters]

Agency
National Institute of Health (NIH)
Institute
National Cancer Institute (NCI)
Type
Intramural Research (Z01)
Project #
1Z01BC011061-01
Application #
7733458
Study Section
Project Start
Project End
Budget Start
Budget End
Support Year
1
Fiscal Year
2008
Total Cost
$512,890
Indirect Cost
Name
National Cancer Institute Division of Basic Sciences
Department
Type
DUNS #
City
State
Country
United States
Zip Code
Dennison, Jennifer B; Shahmoradgoli, Maria; Liu, Wenbin et al. (2016) High Intratumoral Stromal Content Defines Reactive Breast Cancer as a Low-risk Breast Cancer Subtype. Clin Cancer Res 22:5068-5078