Solid-State NMR Studies of Large Protein Assemblies

Amongst other applications, solid-state nuclear magnetic resonance (NMR) spectroscopy can provide atomic-resolution data for the determination of protein structures and dynamics. This is true in particular for protein assemblies that cannot be crystallised or are too large for solution-state NMR. While understanding the dynamics of a protein is of great importance for a full appreciation of its molecular machinery, this thesis has a focus on the structural aspects of a number of large protein assemblies. The first study, presented in Chapter 2, builds on previously determined chemical shifts of a mutant form of amyloid-β, which is compared to a brain-seeded form of the wild-type. This comparison suggests that the determined mutant fold can also be adopted by the wild-type, with small conformational adaptations which accommodate the E22 deletion in the Osaka mutant. In addition, this chapter illustrates how other mutants could conform to this model. The stabilisation of the N-terminal part of the protein via an intermolecular salt bridge to K28 may represent a common structural motif for the mutants that are related to early-onset Alzheimer’s disease. This feature may connect to the observed increased toxicity of the mutant forms compared to wild-type Amyloid-β1-40, where the salt bridge involving K28 is intramolecular instead of intermolecular. A continuation of amyloid studies follows in Chapter 3, but for a different kind of amyloid: in recent years the idea that a number of peptide hormones and neuropeptides are transiently stored in aggregated form has accumulated support. These reversible, functional amyloids are believed to be packed into dense-core vesicles, which function as temporary depots of messenger peptides in secretory cells. Somatostatin (SST) is such a peptide hormone that occurs physiologically both aggregated and as a soluble monomer. The structure of human SST-14 in the context of a fibril was determined to atomic resolution using magic-angle spinning (MAS) solid-state NMR spectroscopy. In addition to scanning transmission electron microscopy data, the complete backbone resonance assignment is presented in this chapter. Subsequently, dipolar-based experiments that provide spectrally unambiguous long-range distance restraints are combined with a prediction of secondary-structure elements by secondary chemical-shift calculations and dihedral-angle restraints. The collective data culminate in the molecular structure presented in this chapter. In Chapter 4, both 13C- and 1H-detected experiments are presented. Both approaches are compared in general, and more specifically in the context of several proteins related to the hepatitis B virus (HBV). HBV is a small enveloped DNA virus whose genomic information encodes only a few genes: the envelope proteins S, M and L (collectively known as hepatitis B surface antigen/HBsAg), the core protein (Cp), the polymerase (P), and the X protein (HBx). This chapter presents structural studies of the envelope protein S and the core protein Cp in its full-length (including C-terminal domain (CTD)) and reduced (without CTD) forms. Proton detection is applied to probe interactions between protein and nucleic acids (ATP analogues and the deoxyribonucleotides of DNA) in combination with phosphorus-detected experiments in Chapter 5. Protein-nucleic acid interactions play important roles not only in energy-providing reactions such as ATP hydrolysis, but also in reading, extending, packaging or repairing genomes. While they can often be analysed in detail with X-ray crystallography, complementary methods are necessary to visualise these interactions in complexes which are not crystalline. This chapter describes how solid-state NMR can detect and classify protein-nucleic acid interactions via site-specific 1H- and 31P-detected spectroscopy. The sensitivity of 1H chemical-shift values for non-covalent interactions involved in these molecular recognition processes is exploited to directly probe the chemical bonding state, a characteristic that cannot be directly obtained from an X-ray structure. Despite its rather challenging size, the method is applied to study interactions in the 669 kDa dodecameric DnaB helicase in complex with ADP:AlF4-:DNA. Finally, Chapter 6 investigates proton-detection in solid-state NMR rather from a more methodological point of view in the context of MAS and resolution. Spectral resolution is key to unleash the structural and dynamic information contained in NMR spectra. The advent of ever faster MAS, today exceeding 100 kHz, is what enabled proton detection in solid-state NMR. In this respect, it is valuable to evaluate the benefit of a continued investment in faster spinning. To address this question, MAS up to 150 kHz is used to investigate a protein complex of archaeal RNA polymerase subunits 4 and 7. Using a rotor with an outer diameter of 0.5 mm and a sample content of approximately 170 µg, the total linewidth of Rpo4/7 improves by a factor of 1.23 ± 0.05 by going from 100 to 150 kHz, and signal intensity increases by a factor 1.48 ± 0.13 in the same MAS range. With some further considerations demonstrated in this chapter, the conclusion is that continued investment in faster MAS is indeed meaningful