Translational diffusion, molecular brightness, and energy transfer analysis of mEGFP-linker-mScarlet–I crowding biosensor using fluorescence correlation spectroscopy

Abstract

Recently, we have investigated the sensitivity of an mEGFP-linker-mScarlet-I construct (GE2.3) in response to macromolecular crowding using ensemble time-resolved two-photon (2P) fluorescence measurements [Mersch et al., Phys. Chem. Chem. Phys. 2024, 26(5), 3927–3940] as a point of reference for developing a single-molecule approach for Förster resonance energy transfer (FRET). Here, we investigate the fluorescence fluctuations, FRET, molecular brightness, and translational diffusion of GE2.3 as a model system using fluorescence correlation spectroscopy (FCS), at the single molecule level, as a function of the excitation and detection wavelengths of the donor (mEGFP) and the acceptor (mScarlet-I). We hypothesize that the molecular brightness (number of fluorescence photons per molecule) of the donor of GE2.3, in the presence and absence of the acceptor, would be distinct due to FRET at the single-molecule level. To test this hypothesis, we used wavelength-dependent FCS to quantify the molecular brightness of intact and enzymatically cleaved GE2.3 as a function of Ficoll-70 (a crowding agent, 0–300 g L⁻¹) at room temperature. Our results indicate that the molecular brightness of intact GE2.3 in a buffer is smaller than that of the cleaved counterpart under 488-nm excitation of the donor, which is attributed to FRET. In contrast, the molecular brightness of both cleaved and intact GE2.3 seems to be the same under the 561-nm excitation of the acceptor due to the absence of FRET. Our results also show that the FRET efficiency of GE2.3 increases as the concentration of Ficoll increases up to 200 g L⁻¹, which agrees with our previous time-resolved 2P-fluorescence measurements. Fluctuation autocorrelation analysis shows that the translational diffusion of intact and cleaved GE2.3 sensors deviates from the Stokes–Einstein model in Ficoll crowded solutions. Additionally, we highlight the multiscale translational and rotational diffusion coefficients of GE2.3 in terms of the average distance between neighboring Ficoll molecules, over the same concentration range, to elucidate the spatio-temporal scaling aspect of FRET and protein–protein interactions. These single-molecule studies would be beneficial for future studies in living cells, where very low GE2.3 expression levels will be required as compared with ensemble, time-resolved 2P-fluorescence measurements.