Robustness of FCS (Fluorescence Correlation Spectroscopy) with Quenchers Present

Abstract

Inspired by recent publications doubtful of the FCS technique, we scrutinize how irreversible ("static") and reversible ("dynamic") quenching can influence the interpretation of such data. Textbook presentations often emphasize only how to analyze data in extremes, the absence of quenching or the presence of substantial quenching. Here, we consider intermediate cases where the assessment of photophysics (static quenching, blinking-like triplet-state relaxation) influence on autocorrelation curves can be delicate if dye-labeled objects diffuse on comparably rapid time scales. We used the amino acid, tryptophan, as the quencher. As our example of small-molecule dye that diffuses rapidly, we mix the quencher with the fluorescence dye, Alexa 488. The translational diffusion coefficient, inferred from fit to the standard one-component Fickian diffusion model, speeds up without the loss of quality of fit, but quenching is reflected in the fact that the data become exceptionally noisy. This reflects the bidisperse population of quenched and unquenched dyes when the time scales overlap between the processes of translational diffusion, quenching, and blinking. As our example of the large-molecule dye-labeled object that diffuses relatively slowly, we mixed the quencher with dye-labeled BSA, bovine serum albumin. Diffusion, static quenching, and blinking time scales are now separated. In spite of quenching contribution to the autocorrelation function when the delay time is relatively short, the inferred translational diffusion coefficient now depends weakly on the presence of a quencher. We conclude that when the diffusing molecule is substantially slower to diffuse than the time scale of photophysical processes of the fluorescent dye to which it is attached, the influence of quenching is self-evident and the FCS autocorrelation curves give an appropriate diffusion coefficient if correct fitting functions are chosen in the analysis.