Modelling Förster resonance energy transfer (FRET) using anisotropy resolved multi-dimensional emission spectroscopy (ARMES)

Abstract

BackgroundFörster Resonance Energy Transfer (FRET) is widely used to study the structure and dynamics of biomolecular systems and also causes the non-linear fluorescence response observed in multi-fluorophore proteins. Accurate FRET analysis, in terms of measuring changes in donor and acceptor spectra and energy transfer efficiency is therefore critical. MethodsWe demonstrate a novel quantitative FRET analysis using anisotropy resolved multidimensional emission spectroscopy (ARMES) in a Human Serum Albumin (HSA) and 1,8-anilinonaphathalene sulfonate (ANS) model. ARMES combines 4D measurement of polarized excitation emission matrices (pEEM) with multivariate data analysis to spectrally resolve contributing fluorophores. Multivariate analysis (Parallel Factor, PARAFAC and restricted Tucker3) was used to resolve fluorophore contributions and for modelling the quenching of HSA emission and the HSA-ANS interactions. ResultspEEM spectra were modelled using Tucker3 which accommodates non-linearities introduced by FRET and a priori chemical knowledge was used to optimise the solution, thus resolving three components: HSA emission, ANS emission from indirect FRET excitation, and ANS emission from direct excitation. Perpendicular emission measurements were more sensitive to indirectly excited acceptor emission. PARAFAC modelling of HSA, donor emission, separated ANS FRET interacting (Tryptophan) and non-interacting (Tyrosine) components. This enabled a new way of calculating quenching constants using the multi-dimensional emission of individual donor fluorophores. ConclusionsFRET efficiency could be calculated using the multi-dimensional, resolved emission of the interacting donor fluorophores only which yielded higher ET efficiencies compared to conventional methods. General significanceShows the potential of multidimensional fluorescence measurements and data analysis for more accurate FRET modelling in proteins.