Methods for Membranes and Membrane Proteins in Solution State Nuclear Magnetic Resonance Spectroscopy

The work described here is a collection of research into novel techniques to be used in the biophysical study of protein and lipid biomolecules by solution state NMR. These methods are developed on the extensive background work in the field and look to extend the capabilities of solution NMR. The first chapter provides an introduction to the basics of NMR, the fundamentals to lipid structure and aggregation, and an in depth look at the history and mathematics of diffusion experiments that are central to many of the subsequent chapters. The second chapter describes the development and application of the bicelle induced curvature and sorting (BICS) method, which presents an alternative quantitative measure of biomolecular curvature in biologically relevant lipid phases and temperatures. Prior to this study, curvature was measured either qualitatively by differential scanning calorimetry or quantitatively by x-ray crystallography. X-ray curvature methods require measurement in the presence of hexagonally inverted lipids to be able to calculate the effect of a perturbant on the lipid structure. Maintaining the inverted phase has at times required researchers to push the limits of relevance by measuring in extreme pHs. BICS is founded on the concept that the insertion of a curvature perturbant into a bilayer, and the pivotal plane being located on the cis leaflet of the perturbation, correlates to a change in size that can be used to quantify curvature induced in a more physiologically relevent bicelle system. The application of this method begins with a proof of concept with lipid curvature, and finishes with the use of BICS to confirm the method of curvature induction for a viral fusion protein, (HAfp23). The third chapter introduces a new tool of membrane mimicry that retains many of the valuable lipid-protein interactions that are present with membrane bound and associating proteins. Prior to this work, in vitro biophysical studies have been performed in non-ideal membrane-like environments. Detergents produce positive curvature strain and can denature protein domains. Bicelles reduce positive curvature strain but provide a homogenous lipid environment, a similar problem found in lipodiscs and nanodiscs. Nanodiscs and lipodiscs introduce either a polymer or membrane scaffolding protein that additionally can have undesired interactions with analytes. The concept of the native bicelle study is that by extracting the whole lipid profile from a eukaryotic tissue and resolubilizing with a detergent component can produce a lipid environment that better describes the true behavior in an in vivo environment. This study includes the extensive characterization of the membrane components extracted and their interactions (or lack there of) with the HAfp23 and Lck-cSH2 proteins as models for how this native bicelle scheme can be applied. The fourth chapter introduces the first detergent compatible positively charged alignment medium (pinacyanol acetate, PNA) for the measurement of residual dipolar couplings (RDCs) in membrane proteins. RDCs have been a powerful tool for protein biophysicists to accurately refine the structures of proteins. Not all proteins are compatible with all alignment media, and many media are to compatible with a lipid membrane or detergent environment necessary for measuring a membrane protein. PNA is the first positively charged medium that is compatible, and this discovery opens a new orthogonality in potential RDC data for more accurate structures of those proteins already compatible with other media, and refinement for those that have no compatible medium as of yet. The final chapter introduces a parallel technique, hybrid NMR, in the use of PNA to form partially randomly oriented samples. In a partially aligned sample, a single alignment axis of orientation produces a single degenerate RDC value per resonance in the NMR spectrum. In a partially oriented sample, a distribution of RDCs are observed co-occurring, which allows for more information to be extracted. Prior to hybrid NMR, this kind of tensor information was only found in solid state studies that require difficult sample prep. The development of hybrid NMR opens the door to rich solid state type dynamics data with the high resolution and signal-to-noise of the solution state, potentially forming an entirely new field of NMR research.