Accurate Determination of the RNA Three-Way Junctions Via Single-Molecule High-Precision Fret Measurements

Abstract

In nature RNA helices are often linked via various junctions and bulges that determine overall 3D structure which are used as building blocks and functional components in nanotechnology applications. Hence visualization of RNA conformations is crucial for understanding their biological functions. Förster-Resonance-Energy-Transfer (FRET) restrained high-precision structural modeling can be used to determine the structure of these junctions. Multi-parameter fluorescence detection (MFD) of single molecules and ensemble Time-Correlated Single Photon Counting (eTCSPC) measurements were applied to perform FRET study on systematic series of 6 RNA three- way-junctions (3WJs) derived from the hairpin ribozyme. Bulge and sequence variations were considered as dominant factors influencing junction, two of which have bulge (two and five unpaired nucleotides) in the junction region. We have generated a database of overall 252 different RNA 3WJs labeled with Alexa488 and Cy5 fluorescent dyes. The analysis toolkit included probability distribution analysis (PDA) for FRET distance determination and FRET position and screening (FPS) toolkit for structural model generation. We detected only one predominant conformer for RNAs 3WJ. Furthermore we report that bulges in the junction determine orientation and rotation of helices, inducing coaxial stacking. We provide a geometrical description of the junction geometry and our results show that small changes in the sequence make dramatic changes in RNA 3WJ tertiary structures which are expected to have significant impact on the functionality. As the accuracy of FRET restrained modeling depends on the correct description of the dye behavior, we characterized the position dependent fluorescent properties of the coupled dyes and statistically analyzed the systematic deviations of the calculated FRET distances from the model generated distances. This allowed us to identify the problematic labeling positions. Thus we could improve the accuracy of the FRET method by careful selection of labeling sites.