Restricted State Selection in Fluorescent Protein Förster Resonance Energy Transfer

Abstract

The measurement of donor lifetime modification by Förster resonance energy transfer (FRET) is a widely used tool for detecting protein-protein interactions and protein conformation change. Such measurements can be compromised by the presence of a significant noninteracting fraction of molecules. Combining time-resolved intensity and anisotropy measurements gives access to both molecular distance and orientation. Fluorescent proteins frequently used to detect energy transfer in biological systems often exhibit decay characteristics indicative of more than one excited state. However, little attention has thus far been given to the specific modes of energy transfer, in particular, which states are predominantly coupled. Here, we use a previously characterized dimerization system to study energy transfer between EGFP and mCherry. Optically excited EGFP and mCherry both exhibit biexponential decays, and FRET should therefore involve dipole-dipole transfer between these four states. Analysis of the sensitized fluorescence anisotropy and intensity decays indicates that FRET transfer is predominantly from the shorter lived EGFP emitting state (2.43 ns) to the longer lived (ca. 2.77 ns) minority component (ca. 16%) of the optically excited mCherry emission. This high degree of state selection between these two widely used FRET pairs highlights the fundamental differences that can arise between direct optical excitation of an isotropic molecular population and dipole-dipole coupling in a far from isotropic interaction geometry and has consequences regarding the accurate interpretation of fluorescent protein FRET data.

Quantifying intracellular equilibrium dissociation constants using single‐channel time‐resolved FRET

“…Although orientational heterogeneity as a result of the flexibility of the linkers between the fluorophores and the proteins they are fused to average out any orientational mismatch while adding to the complexity of the decay [13], its effect is expected to be small due to physical restrictions arising from the combination of the short linker length (less than 17 aminoacids, SM) and the size of the fluorophore barrels. This is consistent with previous in-vitro observations of little local motion of eGFP and mCherry when fused to PDK1 [12]. Finally, we cannot rule out that a small fraction of green-emitting non-matured acceptor with a short lifetime is responsible for the slight decrease of the non-FRET lifetime of eGFP in the presence of mCherry, although this can also be accounted for by the effect of the approximation in equation 18 or by a change of the molecular environment of FRET-inactive eGFP.…”

“…The in-vitro K d measured by FRET was estimated to be between 1 to 5 mM (Supporting Material) [12]. This range is consistent with the intracellular K d measured for the stimulated cells irrespective of the cell line, indicating that stimulation drives a large population of PDK1 in the cell to homodimerise in the same favourable FRET geometry than in-vitro.…”

“…Optically excited eGFP exhibits a biexponential fluorescent decay in the absence of transfer [12,[15][16][17], indicative of two electronic state transitions:…”

“…The decrease of the lifetimes of 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 eGFP in the presence of an acceptor and the increase of the weight of the short component were the signature of FRET happening. As transfer from eGFP to mCherry is restricted to one state in eGFP [12], this indicates that the long-lived state of unperturbed eGFP, energy-state 1, must be involved in FRET, since otherwise its relative weight would not have changed. The short lifetime is thus the effective FRET lifetime t To test the validity of equation 18, we re-analysed the donor-only cells fixing the short lifetime of eGFP-PDK1, t Figure 2D).…”

“…mCherry is the fastest-maturing mFruit protein [24], although the presence of a relevant fraction of nonmatured mCherry has been established in yeast [25]. However, this protein has also been shown to exist in two non-dark excited states [2,12] with no interconversion between them [2], and it has been reported that misfolded mCherry or mCherry dark states could not explain mCherry brightness in eukaryotic cells [2]. Our own observations on the mCherry-6Gly-eGFP tandem construct indicate that, relative to eGFP, the probability of the construct carrying an intact mCherry copy is roughly 0.8 (SM and Supporting Figure S5), too large to account for a donor FRET-inactive population close to Figure S3).…”