Analysis of Virus - Receptor Interactions with NMR

Human rhinoviruses (HRVs) are the main causative agents of common cold infections. Belonging to the family picornaviridae, their icosahedral capsid is composed of four proteins that encase an RNA genome of positive (messenger) sense polarity. Despite their close evolutionary relationship the 102 serotypes exhibit different specificity with respect to the receptor used for cell entry. The minor group with 10 serotypes bind to the low- density lipoprotein receptor (LDLR) and the major group, with 91serotypes, bind to intercellular adhesion molecule 1. The basis of this receptor discrimination and the principle underlying the affinity of the different serotypes for a particular receptor is unknown. In all minor group HRVs only a single surface-exposed lysine in the HI loop of VP1 is conserved that is absent from most major group HRVs. This lysine appears to be necessary but not sufficient for receptor binding. On the other hand, the LDLR contains negatively charged amino acid residues, which most probably establish salt bridges with the positively charged viral surface. To understand the interaction between virus and receptor in molecular detail, knowledge of the complementary surfaces, the principles underlying mutual recognition and discrimination, and the molecular dynamics of the attachment process are required. Recently developed nuclear magnetic resonance (NMR) methods allow the characterization of ligand binding (e.g. the attachment of (small) receptors to (large) viruses) at atomic resolution. These methods are well suited to study the interaction of a single ligand binding repeat of the receptor (~ 4 x103 D) with different minor group viruses (~ 8 x106 D). As a main result of these investigations we expect i) identification of those amino acid residues of the receptor that are most important in virus binding, ii) an extensive understanding of the molecular basis of virus - receptor recognition and discrimination, and iii) determination of the binding kinetics. The methodologies established during this work and the knowledge accumulated will be applicable to other virus- receptor systems and will eventually lead to a simple method for the characterization of interacting molecular surfaces.

Human rhinoviruses cause mild infections usually limited to the upper respiratory tract and leading to massive secretion of mucus from the nose known as the common cold. Due to the large number of serotypes virtually everybody becomes infected more than once a year during the whole life since immunity against one rhinovirus type does not protect against re-infection with another type. Seventy eight (the major receptor group) of the 99 rhinovirus serotypes use intercellular adhesion molecule 1 for cell entry whereas 12 serotypes (the minor receptor group) enter via members of the low-density lipoprotein receptor family. The recent determination of the X-ray structure of a complex between a minor group virus (HRV2) and a fragment of the very-low density lipoprotein receptor identified the amino acid residues involved in the interaction; apart from a single lysine that is present in a surface exposed loop of the capsid protein VP1 in all minor group viruses, none of the other residues is strictly conserved. Therefore, the basis of the highly specific receptor recognition is not understood. We thus started to examine whether saturation transfer difference nuclear magnetic resonance (STD-NMR) is appropriate for the investigation of the interaction of the minor group serotypes with their receptor. STD NMR allows determining the binding epitope of a small molecule attaching to a large protein at atomic resolution. The method has so far been employed mostly for (small) organic compounds binding to (large) proteins. Here, the binding molecule has a molecular weight that is larger than any of those used in STD NMR before. In a first set of experiments we had already demonstrated that the contacts between an organic antivirally active compound and the virus could be identified by this methodology. Using surface plasmon resonance techniques we then determined the affinity of various recombinant fragments of the very-low density lipoprotein receptor for HRV2 and found that the kinetic on- and off rates of a single receptor module were within the range required for STD NMR measurements. Pilot STD NMR experiments showed that this technique indeed is applicable; the receptor-fragment, although relatively large (ca. 6 kDa), can be investigated with respect to binding to the virus. The virus particles are large enough to place the on resonance frequency in a spectral region where no receptor- fragment resonance lines are located. The first spectra also indicated that it will be possible to analyze the ligand binding epitope once an assignment of all resonance lines is achieved. Constructs to express receptor modules without the his6-tag introduced to simplify the purification but complicating the interpretation of the NMR spectra were made. For the definitive assignment of the protons recombinant receptor modules labeled with 15N and 13C now need now to be produced.