Starting Solid-State NMR where X-Ray Crystallography Ends: Opportunities for Studying Complex Biomolecules
Embargoed until 2025-05-23
Author
Date
2023Type
- Doctoral Thesis
ETH Bibliography
yes
Altmetrics
Abstract
In the last ten to 15 years, several developments have made solid-state nuclear magnetic resonance (NMR) spectroscopy a valuable technique in the field of structural biology. Stronger magnets (up to 28.2 T) have improved both sensitivity and resolution of the experiments, allowing the investigation of increasingly large and complex biomolecules. Fast magic-angle spinning (MAS) at frequencies exceeding 100 kHz is now capable of averaging the strong dipolar coupling between protons and other nuclei to an extent that enables the recording of well-resolved proton-detected spectra. The combination of sample rotation and simultaneous radio-frequency (rf) irradiation in solid-state NMR opens up the possibility of generating different effective Hamiltonians at different parts of an experiment, thus creating a huge and constantly growing variety of customised pulse sequences that are able to selectively correlate the nuclei of interest. Finally, the establishment of sedimentation as a sample preparation technique has made many soluble and non-crystallisable proteins accessible to solid-state NMR. In chapter 1, the theoretical foundations for solid-state NMR experiments are summarised. A particular focus is put on recent developments as well as aspects that are relevant for the ensuing parts of this work. The following chapters provide a biological background and summarise the available literature on the various biomolecular systems that were investigated in the projects presented in this work. These include the bacterial RNA helicase and acetyltransferase TmcA (chapter 2), the non-structural protein 5A of the hepatitis C virus (chapter 3), membrane proteins of the SARS coronavirus 2 (chapter 4) and reversible lysozyme fibrils (chapter 5). Chapter 6 encompasses the characterisation of the apo-state of TmcA. It starts with the work that had been done to establish this protein system for solid-state NMR studies, including cloning of the construct, the purification of the protein and sedimentation. An important aspect in this regards was also the resonance assignment, which was however limited by the size of the protein and accordingly spectral overlap. Still, it was possible to obtain heavy atom resonance assignments for 84 out of 671 residues, corresponding to about 13%. While the resonance assigment was done on ADP-bound TmcA, which yielded well-resolved carbon-detected spectra, sedimentation also allowed to analyse the non-crystallisable apo-state. Besides chemical-shift perturbations, which were located in the helicase portion of the protein, the resonances in spectra of apo-TmcA were considerably broader. Bulk relaxation measurements enabled to pinpoint this difference to a larger inhomogeneous linewidth, suggesting more structural heterogeneity in the absence of the nucleotide. This can also serve as an explanation why crystallisation attempts failed. Chapter 7 builds upon the work on sedimented TmcA presented in the preceding chapter to study the effect of the binding of various types of ligands to TmcA. First, the diamagnetic Mg(II) in the ATPase active site was substituted with paramagnetic Mn(II) and Co(II) to localise it within the protein, since the metal ion was absent in the model from X-ray crystallography. Paramagnetic relaxation enhancements (PRE) were used as distance restraints to triangulate the position of the metal ion, which was then confirmed by pseudo-contact shifts. The metal ion was found to be coordinated by conserved Walker motif residues and the beta-phosphate of the nucleotide, which is in agreement to structures of similar proteins. The PREs also allowed an estimation of the Mn(II) electron relaxation time. Secondly, TmcA was co-sedimented with various ATP analogues. This enabled to follow the ATP hydrolysis cycle from the perspective of the protein by carbon-detected experiments and from the perspective of the nucleotide by phosphorus-detected experiments. Finally, the binding of RNA and acetyl-coenzyme A was analysed, which were previously known to stimulate ATP hydrolysis. Interestingly, chemical-shift perturbations (CSPs) were not only observed close to the expected binding site, but also in the ATPase active site. While the solid-state NMR measurements on the bacterially expressed TmcA were carried out in 3.2 mm rotors, the next two chapters deal with proteins that were produced by cell-free synthesis using wheat-germ extract. The comparably low yields of this expression technique are better compatible with proton-detection in 0.7 mm rotors at fast MAS of 100 kHz. In chapter 8, continuing work on NS5A of the hepatitis C virus is presented. Our group has previously established a membrane-anchored NS5A construct reconstituted in liposomes for solid-state NMR. This was the basis for investigation of Daclatasvir binding, which is an approved drug against hepatitis C that targets NS5A. Spectra recorded at the highest available field of 28.2 T revealed that already in the absence of Daclatasvir, NS5A exists in two forms, with a major and minor species, potentially a dynamic equilibrium between monomer and dimer. Upon Daclatasvir binding, the previously minor form is stabilised to make up almost 100%. The distribution of chemical-shift changes on the crystal structure of a monomer suggests that the dimer is asymmetric in nature. Chapter 9 continues on the discussion of viral membrane proteins, this time of the SARS-CoV-2. These proteins proved to be challenging targets, with short T2' relaxation times and broad, unresolved spectra. Further analyses showed that the broadening predominantly occurs in the proton dimension, while carbon and nitrogen dimensions are better resolved. Broad spectra were also observed in the case of the reversible lysozyme fibrils, which are presented in chapter 10. Bulk relaxation measurements again shed light on the different contributions to the linewidth. It turned out that both the homogeneous and the inhomogeneous linewidths are exceptionally large. This suggests that the fibrils are characterised by extensive polymorphism, and the individual polymorphs are additionally highly dynamic. The different projects presented in this work constitute examples for the application of solid-state NMR spectroscopy to the study of complex biomolecules. In the conclusions, the opportunities and limitations of solid-state NMR as a tool in structural biology are discussed and it is contrasted with other techniques, in particular X-ray crystallography, but also cryo-electron microscopy and in-silico structure prediction. Show more
Permanent link
https://doi.org/10.3929/ethz-b-000613001Publication status
publishedExternal links
Search print copy at ETH Library
Publisher
ETH ZurichSubject
Solid-state NMR; NMR; NMR assignment; Protein NMR; Magic-angle spinning (MAS); Proton detection; Hepatitis C virus; SARS-CoV-2; Escherichia coli (E. coli); Recombinant protein expression; Protein purification; Sedimentation; Isotope labellingOrganisational unit
03496 - Meier, Beat H. (emeritus) / Meier, Beat H. (emeritus)
More
Show all metadata
ETH Bibliography
yes
Altmetrics