TIRF-Based FRET with One Base-Pair Resolution

Abstract

Single-molecule FRET (smFRET) is commonly used as a “nanoscale ruler” for the measurement of biomolecular distances and distance changes. However, the limits of FRET resolution for measurements on surface-immobilized molecules have not been rigorously explored. Using total-internal reflection fluorescence (TIRF) microscopy on a set of DNA standards and advanced image analysis software, we have quantified and extended the limits of FRET resolution associated with the use of electron-multiplying CCD (EMCCD) cameras. For such measurements, we derived a novel theoretical description of the major sources of noise ( photon shot noise, background, CCD noise and pixelation effects); we find excellent agreement between our experimental results and predictions from theory and Monte Carlo simulations. For FRET measurements on a truly single-molecule basis (as opposed to measurements on an ensemble of single molecules), analysis of the experimental noise allows us to predict a resolution of 4% FRET within the linear FRET range (20-80%), sufficient to directly observe a distance difference equivalent to one DNA base-pair separation (3.4 A). For FRET distributions obtained from an ensemble of single molecules (which exhibits broadening due to presence of static heterogeneity), we demonstrate the ability to distinguish between distances differing by as little as 2 base pairs (∼7A). Current work focuses on real-time observation of single-base-pair translocation steps of Escherichia coli RNA polymerase within single early transcription-elongation complexes; such observations are crucial for understanding the mechanisms of DNA and RNA polymerases.Our work paves the way for ultra-high resolution studies of processes involving conformational changes and protein translocation on nucleic acids.