Dipolar Couplings in Biomolecular NMR Spectroscopy

It is the aim of this project to develop and implement new nuclear magnetic resonance (NMR) methodology in the study of biological macromolecules based on residual dipolar couplings (RDC), which can be observed in weakly aligned phases such as dilute liquid crystals. It has already been shown in the past that RDC provide very powerful and precise structural information and aid in obtaining extreme high-resolution protein structures. RDCs have been especially capable of precisely detecting subtle structural tertiary and quaternary rearrangements in the course of ligand binding equilibria, which has been a weakness of conventional NMR based structure determination. It is planned to apply this to the study of the Surface-layer homology domain (SLH) of a bacterial S-layer protein (SbsB of Geobacillus stearothermophilus PV72/p2), which constitutes a main component of bacterial cell walls. Because of their ability to spontaneously self-assemble into crystalline two-dimensional layers S-layer proteins are of considerable biotechnological interest. The function of this SLH domain is recognition and anchoring of the S- layer protein to the bacterial secondary cell wall polymer (SCWP), whose main component is a carbohydrate matrix. To characterise this binding process in detail a high-resolution structure of the apo-protein in solution as well as of the holo-protein with its bound ligand or a suitable model of it will be determined using current RDC methodology. With this high resolution model the binding mode and the determinants of its specificity can be understood on a molecular level. Beyond structure refinement RDCs are capable by themselves of solving a protein (backbone) structure de-novo. Alternative approaches have used RDCs to mine databases of proteins for homologous (sub-) structures, that can be assembled into protein structures even in the complete absence of NOEs. A serious limitation of the approaches has been so far that they require very complete sets of RDCs (up to five DC pre residue), which may be impossible to obtain in practice. It is therefore planned in the course of this research project to combine the discriminatory power of RDCs with methods of structure prediction based on bioinformatics to quickly verify (or falsify) predictions. It is expected that this will significantly speed up the process of structure determination without the requirement of a complete set of RDC.

It is the aim of this project to develop and implement new nuclear magnetic resonance (NMR) methodology in the study of biological macromolecules based on residual dipolar couplings (RDC), which can be observed in weakly aligned phases such as dilute liquid crystals. It has already been shown in the past that RDC provide very powerful and precise structural information and aid in obtaining extreme high-resolution protein structures. RDCs have been especially capable of precisely detecting subtle structural tertiary and quaternary rearrangements in the course of ligand binding equilibria, which has been a weakness of conventional NMR based structure determination. It is planned to apply this to the study of the Surface-layer homology domain (SLH) of a bacterial S-layer protein (SbsB of Geobacillus stearothermophilus PV72/p2), which constitutes a main component of bacterial cell walls. Because of their ability to spontaneously self-assemble into crystalline two-dimensional layers S-layer proteins are of considerable biotechnological interest. The function of this SLH domain is recognition and anchoring of the S- layer protein to the bacterial secondary cell wall polymer (SCWP), whose main component is a carbohydrate matrix. To characterise this binding process in detail a high-resolution structure of the apo-protein in solution as well as of the holo-protein with its bound ligand or a suitable model of it will be determined using current RDC methodology. With this high resolution model the binding mode and the determinants of its specificity can be understood on a molecular level. Beyond structure refinement RDCs are capable by themselves of solving a protein (backbone) structure de-novo. Alternative approaches have used RDCs to mine databases of proteins for homologous (sub-) structures, that can be assembled into protein structures even in the complete absence of NOEs. A serious limitation of the approaches has been so far that they require very complete sets of RDCs (up to five DC pre residue), which may be impossible to obtain in practice. It is therefore planned in the course of this research project to combine the discriminatory power of RDCs with methods of structure prediction based on bioinformatics to quickly verify (or falsify) predictions. It is expected that this will significantly speed up the process of structure determination without the requirement of a complete set of RDC.