Confocal single molecule fluorescence detection - methodical developments and applications to biological specimens
Aachen (2017, 2018) [Dissertation / PhD Thesis]
Page(s): 1 Online-Ressource (xix, 130 Seiten) : Illustrationen, Diagramme
Over the last few decades, single-molecule FRET has become a valuable tool enlightening the fields of molecular conformational dynamics, folding and structure determination of biological macro-molecules. However, quantitative statements about any parameter obtained using FRET are based on aprecise calibration of the acquired data, among others taking into account the properties of the FRET-pair used. In this work, two methods are developed and established that facilitate this calibration by means of confocal microscopy, automatically allowing a sample characterization under application relevant conditions, i.e. close to the single-molecule level. To assess the orientation factor entering the Förster radius calculation, time-resolved anisotropy measurements are performed. In this regard, the depolarization effects induced by the use of a high numerical aperture objective have to be taken into account by two correction factors. These are precisely determined by combining an extended experimental calibration procedure adjusted to the temporal resolution of the setup at hand with theoretical predictions considering the exact measurement conditions. To determine the fluorescence quantum yields of donor and acceptor used in the FRET efficiency calculation, the linear relation between the molecular brightness, made accessible by Fluorescence Correlation Spectroscopy, and the fluorescence quantum yield in the limit of low excitation intensities is exploited. Based on this, the presented quantum yield determination method lowers the needed amount of sample by a factor of around thirty as compared to a commonly applied optical method, but still provides at least the same precision. As compared to fluorescence lifetime based quantum yield determination methods, the novel approach is more comprehensive as it is sensitive not only to dynamic, but also to static fluorescence quenching. Hence, apart from its relevance for smFRET, a reliable characterization of biological samples with limited expression yields is made possible. With these two methodical developments available, smFRET data of structurally rigid, double-stranded DNA oligonucleotides in aqueous buffer and in buffers with specific amounts of glycerol, guanidine hydrochloride and sodium chloride added are analyzed. It is demonstrated that the calculation of inter-dye distances, without taking into account solvent-induced spectral and photo-physical changes of the labels, leads to deviations of up to 4 Å from the real inter-dye distances and furthermore to a misinterpretation of the underlying structural changes. Additionally, it is experimentally shown that electrostatic dye dye repulsions are negligible for the inter-dye distance regime considered here (> 50 Å ). Expanding the given framework of accessible volume calculations by taking into account the electrostatic interaction potential of donor and acceptor in the respective solvent environment, these findings are supported by theoretical predictions. Finally, all methodical approaches and experimental/theoretical findings are combined to validate the further compaction of the already unfolded state of the protein Phosphoglycerate Kinase (PGK) with decreasing concentrations of denaturant, a mechanism known as coil-globule transition.
Fitter, Jörg Ludwig