Recovering Absolute Fret Efficiency from Single Molecules: Comparing Methods of Gamma Correction

Abstract

Fluorescence resonance energy transfer is widely thought of as a “spectroscopic ruler.” Because biological processes and cellular assemblies occur on the nanometer scale, FRET is a popular tool for structural biology. In contrast to ensemble solution FRET measurements which record the entire emission spectrum, microscopy- based FRET experiments separate donor and acceptor intensity by passing the emission through a series of optical elements. Observed FRET efficiency, determined from the uncorrected donor and acceptor intensities, has been called a relative proximity ratio, which is internally consistent only if the photophysical properties and instrument remain unchanged. However, it is desirable to measure absolute distances using FRET, which requires that FRET efficiency be corrected for both instrument response and fluorophore properties. Thus, “gamma” correction adjusts for differences between the donor and acceptor dyes in their probability of photon emission upon excitation and the probability that emitted photons will be detected. Methods of gamma correction vary depending on the single molecule methodology. To test different methods for correcting FRET efficiency, we recorded smFRET distributions for protein and DNA on different instruments and with different filter sets which altered the observed FRET efficiency. Knowledge of filter set transmission allows for comparison of results between groups using different instruments. Applying empirically-derived corrections for instrument response and quantum yield was only slightly better than corrections based solely on filter set transmission data. We found that gamma correction based on single molecule photobleaching was the most effective particularly when gamma was determined for each sample or even each molecule. Variations in focus of the two colors and sub-pixel errors in image mapping affect both FRET and gamma. As such, per molecule correction affects distribution width because FRET outliers may also have anomalous gamma values.