Rapid Determination of High-Resolution Protein Structures by Solution and Solid-state NMR Spectroscopy

Abstract

NMR spectroscopy provides high-resolution structural information of biomolecules in near-physiological conditions. Structural studies of proteins and nucleic acids are critical for understanding biological processes at the molecular level. Although significant improvements were achieved in NMR spectroscopy in the last 20 years, the increase in genome sequencing data has created a need for rapid and efficient methods of NMR-based structure determination. NMR data acquisition can be accelerated significantly, when sensitive spectrometers are combined with new methods for sampling chemical shifts in multidimensional NMR experiments. Therefore, data analysis and in particular the requirement to assign side chain chemical shifts to specific atoms is the major bottleneck of rapid NMR-based structure determination. In chapter 2, a method termed FastNMR (FAst STructure determination by NMR), is described in detail, which enables automatic, high-resolution NMR structure determination of domain-sized proteins starting from unassigned NMR data. Using FastNMR the de novo structure of the 65-residue cone snail neurotoxin conkunitzin-S2 was determined automatically. Large classes of proteins, such as membrane proteins and insoluble aggregates of peptides and more complex systems, cannot be investigated with the above method, because the proteins cannot be made soluble for liquid-state NMR. Therefore, there is a considerable interest in the development of methods for protein structure determination that do not have these limitations. In chapter 3 and 4 of this thesis, it is demonstrated that, combining the knowledge obtained in solution-state NMR, a rapid determination of high-resolution protein structure of globular proteins, such as, potassium channel blocker, Kaliotoxin existing in free form and also in complex with KcsA-Kv1.3, from solid-state NMR data could be obtained. Also in chapter 4, an improved model of KTX-KcsA-Kv1.3 complex is proposed based on functional and solid-state NMR data. Finally, chapter 5 sheds light on understanding the mechanism of alignment of proteins and efforts in improving the accuracy of prediction of charge-induced molecular alignment from the protein's known 3D structure, by employing more atomistically detailed electrostatic models. Preliminary results suggest that the accuracy in predicting RDCs and magnitude of alignment using detailed electrostatics might improve in comparison with the simplified model implemented in PALES.