Superior robustness of ExEm-spFRET to IIem-spFRET method in live-cell FRET measurement.

We recently developed two quantitative fluorescence resonance energy transfer (FRET) measurement methods based on spectral unmixing of emission spectra (IIem-spFRET) and excitation-emission spectra (ExEm-spFRET), respectively. We here evaluated robustness of the two methods by implementing them on a self-assembled quantitative FRET measurement system using the cells expressing different constructs. For the cells with larger signal-to-noise (S/N) ratio (>9), the two methods obtained consistent FRET efficiency (E) values and total concentration ratio (RC) values of acceptor to donor for all constructs; for the cells with 3 < S/N < 9, IIem-spFRET obtained bigger RC values than the expected value for VCV construct; for the cells with S/N < 3, although IIem-spFRET method obtained inaccurate E and RC values for VCV construct, the two methods also obtained consistent E and RC values for all other constructs. Collectively, both ExEm-spFRET and IIem-spFRET methods are very applicable for live-cell FRET measurement, and ExEm-spFRET has superior robustness especially for the cells with low S/N ratio. LAY DESCRIPTION: Fluorescent proteins (FPs)-based fluorescence resonance energy transfer (FRET) has been widely used as a powerful technique to study protein-protein interaction and stoichiometry of macromolecular complexes in living cells. There are two key issues for quantitative FRET measurement especially in living cells: donor emission crosstalk (donor fluorescence is collected in acceptor detection channel) and acceptor excitation crosstalk (acceptor is excited directly under donor excitation) due to the spectral overlap of FPs. Two-wavelengths excitation-based spectral linear unmixing of emission spectra can resolve donor emission crosstalk due to the obvious difference in emission spectra between donor and acceptor, but additional reference is necessary for the correction of acceptor excitation crosstalk. Spectral unmixing of excitation-emission spectra has inherent ability to resolve donor emission crosstalk and acceptor excitation crosstalk simultaneously without additional reference. We recently developed two quantitative FRET measurement methods based on spectral unmixing of emission spectra (IIem-spFRET) and excitation-emission spectra (ExEm-spFRET), respectively. We here evaluate robustness of the two methods by implementing them on a self-assembled quantitative FRET measurement system using the same cells expressing different constructs under different signal-to-noise (S/N) ratios. For the cells with S/N > 9, the two methods obtained consistent FRET efficiency (E) values and total concentration ratio (RC) values of acceptor to donor for all constructs; for the cells with 3 < S/N < 9, IIem-spFRET obtained bigger RC values than the expected value for VCV construct; for the cells with S/N < 3, although IIem-spFRET method obtained inaccurate E and RC values for VCV construct, the two methods also obtained consistent E and RC values for all other constructs. Collectively, our experimental results demonstrate that both ExEm-spFRET and IIem-spFRET methods are very applicable for live-cell FRET measurement, and ExEm-spFRET has superior robustness especially for the cells high E and low S/N ratio.