Abdullin, Dinar: Localization of Paramagnetic Metal Ions in Biomolecules ny means of Electron Paramagnetic Resonance. - Bonn, 2017. - Dissertation, Rheinische Friedrich-Wilhelms-Universität Bonn.
Online-Ausgabe in bonndoc: https://nbn-resolving.org/urn:nbn:de:hbz:5n-46416
urn: https://nbn-resolving.org/urn:nbn:de:hbz:5n-46416,
author = {{Dinar Abdullin}},
title = {Localization of Paramagnetic Metal Ions in Biomolecules ny means of Electron Paramagnetic Resonance},
school = {Rheinische Friedrich-Wilhelms-Universität Bonn},
year = 2017,
month = mar,

note = {Localization of metal ion binding sites in biomolecules is an important objective in structural biology. Experimental methods that are commonly used to determine the location of metal ions, such as X-ray crystallography and NMR, are not always applicable due to their limitations related to sample preparation and data analysis. It is, therefore, of high importance to have complementary methods that can be applied to difficult cases. This thesis reports on the first application of EPR spectroscopy for the localization of metal ions in biomolecules via trilateration. Since the trilateration principle implies the use of distance constraints to determine the location of a metal ion, a significant part of the work is related to distance measurements between metal ions and spin labels by means of pulsed EPR techniques. The most popular technique for such measurements, pulsed electron-electron double resonance (PELDOR), is often affected by orientation selectivity effects which significantly complicate the derivation of distances from the experimental time-domain data. To enable the extraction of distances from orientation-selective PELDOR data without any preliminary knowledge about the relative orientation of spin centers, a method based on a simplified geometric model of a spin system was developed. The method was implemented in a computer program called PeldorFit, and its validity was confirmed by multiple tests on PELDOR data sets from the literature and from the present work. In particular, the new method allowed extracting Cu(II)-MTSSL distance distributions from PELDOR data sets of six MTSSL-labeled azurin mutants. These distance distributions were then used for benchmarking the EPR-based trilateration on the Cu(II) ion of azurin. The trilateration problem was solved with the newly developed program mtsslTrilaterate. The obtained Cu(II) coordinates were compared to the available crystal structure of azurin, revealing that the calculated Cu(II) location is very close to the corresponding crystallographic site. The precision of the trilateration was estimated at 0.26 nm. This value was shown to be affected by the spin density delocalization of the Cu(II) center, the precision of the average MTSSL coordinates, the number and precision of the distance constrains, and the accuracy of the structural model of the metal-free protein. The main source of the trilateration error was, however, attributed to the average MTSSL coordinates. These coordinates were determined using the existing in silico spin labeling methods which were shown to have an average error of 0.15 nm. More accurate MTSSL coordinates were obtained using the PELDOR constraints and measuring the crystal structures of the azurin mutants. Furthermore, the possibility to extend the EPR-based trilateration to other metal ions, such as Fe(III), was investigated in the context of the distance measurements for these ions. Since PELDOR distance measurements on the Fe(III)/nitroxide spin pair are complicated due to the large spectral width of Fe(III), the use of an alternative pulsed EPR technique RIDME was considered. Both techniques, RIDME and PELDOR, were compared using a MTSSL-labeled mutant of the heme-containing protein cytochrome P450cam. The results of the comparison are that RIDME allows avoiding the orientation selectivity effects and yields a seven times higher signal-to-noise ratio than PELDOR. This demonstrates the great potential of the RIDME technique for future trilateration studies. Besides distance constraints, the availability of spin labels with well-studied structure and dynamics is required for the trilateration. Since one of the potential applications of the EPR-based trilateration is the localization of metal centers in ribozymes, the structure and dynamics of a nitroxide-labeled uracil nucleobase were investigated in this work. Using the information obtained by EPR spectroscopy and DFT calculations, a structural model of this spin-labeled nucleobase was proposed and implemented in the program mtsslWizard. This model was successfully applied for the prediction of the mean distance between two such nucleobases in an RNA duplex. The comparison of the predicted distance with the corresponding PELDOR-derived distance revealed only a minor difference of 0.1 nm.},
url = {http://hdl.handle.net/20.500.11811/7130}

The following license files are associated with this item: