NMR IN MOLECULAR BIOLOGY.

Structure, binding and molecular recognition.

July 10-15, 1999. Hotel Alixares. Granada (Spain)

 

1. INTRODUCTION

During the last two decades, nuclear magnetic resonance ( NMR) emerged as a sound alternative technique to X-Ray crystal diffraction for determining three-dimensional structures of biomacromolecules at atomic level. As compared to X-Ray crystal diffraction, NMR offers two attractive advantages. First, it resolves the three-dimensional structure of the biomacromolecule in aqueous solution, i. e., in a medium similar to the one where it performs its function "in vivo". Second, it provides invaluable dynamic information.

A grasp of the structure is central to characterise in full detail all biochemical processes in which proteins, nucleic acids, carbohydrates, and all other biomolecules are involved. Understanding these processes frequently requires to know not only the structure of a particular biomolecule, but also those of its partners and their complexes. This happens to be fundamental to comprehend how those molecules interact. Important interactive processes take place, for instance in enzymatic catalysis, the immune response, signalling traffic, the regulation of gene expression, and many others.

Structure is also of essence to rationally engineer these molecules to improve their output, make them more resistant to environmental conditions, or increase their selectivity. From a more applied point of view, the structure determination forms the basis of a rational drug design, by means of which we may be able to interfere with pathogenic agents or processes.

2. REPORT

The 23 Conference Lectures were divided into five main topics: methodologies, structure and function, membrane proteins and ligand binding, protein-protein interactions and nucleic acid-protein interactions. 83 posters were presented by the more than one hundred participants. Two keynote lectures gave balanced attention to basic academic research and oriented applications: one by Kurt Wüthrich (Institut für Molekularbiologie und Biophysik, ETH, Zürich), and another by Steve Fesik from Abbott Labs., Illinois, USA.

One of the disadvantages imputed to the NMR technique is its inability to deal with large biomacromolecules. This is due to strong signal overlapping and line broadening resulting from increased relaxation rates. Multidimensional spectroscopy and deuterium labelling were introduced in recent years in order to increase the size limit of proteins to about 40 kDa. Kurt Wüthrich described in detail the method TROSY (transverse relaxation-optimised spectroscopy) recently developed in his lab, that is based on the theoretically predicted cancellation of transverse relaxation effects for one component of the XH multiplet (X is an heteronucleus), whose narrow lines can be then properly edited and observed. Wütrich reported a novel development, the CRIPNET element (cross-relaxation-enhanced polarisation transfer), which improves the efficiency of magnetisation transfers in very large molecules. The use of TROSY and CRIPNET, together with suitably isotope-labelled systems, can yield data on proteins or nucleic acids with molecular weights of several hundred thousands daltons, which marks a major breakthrough in NMR structure determination. These developments makes it possible to tackle a long standing problem in Structural Biology: the determination of the structure of membrane proteins solubilized in micelles or lipid vesicles.

Progress on the method of weak alignment of macromolecules with the magnetic field was addressed by several speakers including Ad Bax and Marius Clore, from the NIH (Bethesda) and Ivano Bertini (Florence University). Biomolecular structure determination by NMR to date has been based on very local parameters such as coupling constants and nuclear relaxation transfer rates, more widely known by nuclear Overhauser enhancements (NOEs). In contrast, dipolar couplings for weakly aligned macromolecules provide information on individual bond vectors relative to the molecular alignment tensor and therefore have an intrinsic long-range character. Thus, residual dipolar couplings introduce constraints between parts of the structure which are not connected by NOEs. Paramagnetic metalloproteins with large magnetic anisotropy tend to align spontaneously at high magnetic field strengths. Other systems need to employ liquid crystal media like bicelles (disc-shaped phospholipid particles) or philamentous phages to properly orient the macromolecules in the sample. Improvements on the experimental control of these media and on the use of residual dipolar couplings and anisotropic shifts to determine or refine three-dimensional structures of large proteins were discussed by the above lecturers.

Two additional lectures were delivered related to novel parameters in the determination of structure and dynamics of biomolecules. The first one, by Christian Griesinger (Frankfurt University), described the use of paramagnetic tags and cross-correlated relaxation, and the second one, by Stephan Grzesiek from the Institute of Structural Biology of Jülich (Germany), reported on the use of scalar couplings to detect hydrogen bonds. Finally, Christina Redfield (Oxford University) stressed the importance of using anisotropic models for rotational diffusion in order to properly interpret nuclear relaxation measurements to get information on internal molecular dynamics.

The second Keynote Lecture dealt with the ability of NMR to determine three dimensional structures of protein/ligand complexes in solution, which provides invaluable information on binding specificity and a new way to carry out highly efficient structure-based drug design. NMR methods can be used to discover high affinity ligands to appropriate protein receptors. Then, small organic molecules that bind to proximal subsites in the protein are identified, optimized and linked together to create a high affinity ligand, which greatly reduced chemical synthesis and time. NMR can thus be considered as a new and very especial high-throughput screening method for new pharmaceuticals.

Within the section of structure-function relationships, there were several lectures addressing a wide variety of important cellular, physiological, and medical processes. Annalisa Pastore (MRC, London) and Michael Sattler (EMBL, Heiderberg) described the structure and functional implications of two different proteins involved in genetic diseases. Horscht Kessler (Technische Universität, Munich) demonstrated that the N-terminal domain of a protein of the AAA-family (ATPases Associated with a variety of cellular Activities), which functions as a chaperon, provides for most, if not all, of the sites involved in the binding of folding substrates. Gordon Roberts (University of Leicester) described the two domain structure of the protein RhoGDI, a modulator of small GTP-binding proteins, which function as molecular switches, and Josep Rizo (Dallas, Texas) used NMR to unravel some key steps of the complex protein machinery governing the release of neurotransmitters through synaptic vesicle exocytosis. A great expectation exists on the possibility that solid state NMR measurements may provide atomic-level structural information on high-molecular-weight biopolymers in noncrystalline solids and disordered membrane-bound systems. Robert Tycko (NIH, Bethesda) reported on progress in the solid state field and presented results on fibril-forming peptides of interest in relation to Alzheimer disease.

Protein-protein recognition processes are fundamentally important in a wide range of biological processes, such as cell adhesion, signalling, and the regulation of the cell cycle. Some proteins intervening in those processes have a modular construction and their function depends critically on the way how these modules are linked and how they interact between themselves. Progress in this field was addressed by Iain Campbell (Oxford University, UK). Moreover, Ernest Laue (Cambridge University, UK) presented two examples of protein/protein interactions of interest in cell cycle control, one of which providing an interesting example of how proteins that are partially unstructured in solution can recognise each other and bind specifically. Frederick Dahlquist (University of Oregon) analysed the interactions between CheY and CheA, a two component regulatory system used in bacterial chemotaxis, and Yoji Arata (Tsukuba, Japan) summarised the work of his lab on the elucidation of the structure-function relationship in the antibody system. Although the antibody molecule has a molecular weight of 150 kDa, Arata and colleagues were able, by means of selective 13C and 15N labelling of the protein backbone and the sugar moiety, to obtain key information about their respective roles in the biological function of the antibody.

The last section dealt with nucleic acid-protein recognition. Cees Hilbers (University of Nijmegen) presented a combined NMR and Molecular Dynamics study of the structure and dynamics of single stranded DNA binding proteins from bacteriophages M13 and Pf3. Martin Billeter (University of Göteborg) centred his lecture on the molecular recognition of DNA by homeodomains. New experimental data on a larger complex led to an improved view of the protein-DNA recognition site, which was complemented by Molecular Dynamics calculations in order to substantiate the nature of the fluctuations in the network of intermolecular interactions between the homeodomain, the DNA and the interfacial water. Robert Kaptein (Uthrecht University) presented two cases involving allosteric interactions in protein-DNA recognition. The first one was related to the glucocorticoid receptor, and the second with the lac repressor DNA-binding domain. He also illustrated the procedure followed in the determination of the tetrameric structure of the Mnt repressor of bacteriophage P22.

Mike Summers (Howard Hughes Institute, Baltimore) described the three-dimensional structure of the human immunodefficiency virus-type1 (HIV-1) nucleocapside protein bound to the SL3 stem-loop recognition element of the genomic y RNA packaging signal (Figure 1). Tight binding (dissociation constant, ~100nM) is mediated by specific interactions between the amino- and carboxy-terminal CCHC-type zinc knuckles of the protein and two bases in the RNA tetraloop. The structure provide insights into the mechanism of viral genome recognition and serves as a basis for designing inhibitors to interfere with genome encapsidations.

RNA-protein complexes play a central role in every aspect of the post-transcriptional regulation of gene expression. The work carried out by Gabriele Varany and colleagues (MRC Lab of Molecular Biology, Cambridge) aims at understanding the structural principles underlying RNA-protein recognition and the mechanism by which gene expression occurs and is regulated. On this line, the structure of complexes of growing size and complexity was presented. Extreme examples were provided by the human protein U1A, which recognises single stranded RNA in a highly specific fashion (Figure 2), and the 40 kDa ternary complex between two of those U1A proteins and the U1A polyadenylation regulatory element RNA.

In sum, we feel that the whole programme of the Conference, invited Lectures as well as the 83 communications presented as posters, has met if not surpassed the standards of previous meetings in the series, has enriched our view about the current state of NMR and its applications to a wide variety of important biological problems, and has set the stage for future meetings in the series. We consider this Conference rewarding and memorable for its scientific content and educational value.

PUBLICATIONS RESULTING FROM THE EVENT

A book of Abstracts was edited, which is attached to this report.

ORGANISATIONAL AND ADMINISTRATIVE ASPECTS

The administration run smoothly thank to the efficient and expert collaboration of Mrs. Corinne Le Moal, Jackie McLelland and Caroline Grimont from the European Science Foundation in Strasbourg. The hardest point was the selection process: we had to turn down more than one hundred applications, which created a great number of problems. On the other hand, this is a clear indication of the success of the "NMR in Molecular Biology" series and underlines the need for financing future editions. We would like to thank to the personnel of the Hotel Alixares for all facilities given. Finally, we thank very specially to Mrs. Joana Martinez-Flores, the ESF Secretary at the Conference, for her kindness, efficiency, and expertise.

 
ACKNOWLEDGEMENT

The support from the European Commission Euroconference activity of the Training & Mobility of Researchers (TMR) Programme is gratefully acknowledged.
 

 
Prof. Manuel Rico, Chairman