Structural studies on foot-and-mouth disease virus
Foot-and-mouth disease viruses (FMDVs) constitute the aphthovirus genus of the Picornaviridae. The structures of Oi subtype viruses OiK and G67 have been solved and comparisons reveal the structural basis of monoclonal antibody escape mutations in G67. Escape mutations are seen to occur at surface-exposed residues and to provoke structural changes limited to the altered side chains. Comparisons of the structures of O1 and O1BFS (Acharya et al., Nature 337, 709-716 (1989)) suggest that changes occurring 'in-the-field' in response to polyclonal antibody pressure may be subtly different from mutations produced by monoclonal antibody pressure in vitro. Field mutations are seen to alter less exposed residues and to have more far-reaching structural effects than the in vitro, monoclonal provoked mutations. Crystals of G67 are seen to be 'intimately twinned', the data possessing extra symmetry due to a mis-packing of the crystals. A protocol, based on current real-space averaging procedures with a novel constraint imposed, has been used successfully to deconvolute these data. This method might be more generally applied to deconvolute the wavelength overlaps that occur when using the Laue method. The structures of C-S8cl and mutant SD6-6 have been solved at a resolution of 3.5Å. These structures enable comparisons between members of different FMDV serotypes to be made for the first time, namely: serotype 0 (O1BFS) and serotype C (C-S8cl). Flexibility of the Arg-Gly-Asp containing G-H loop of VP1 is seen to be amongst the most conserved structural features. This loop is implicated in receptor binding and possible roles for the observed flexibility are discussed. The CS8cl structure also reveals more detail in previously disordered regions of the capsid, namely: the N-terminal residues of VP2 and potential myristate density under the 5-fold axis of the virion. Analysis of structures from the Protein Data Bank reveals different patterns of amino acid use in proteins involved in the two halves of the immune recognition event i.e. immunoglobulins and viruses. These patterns seem to be based not only on the characteristics of the most used amino acids but also on characteristics of the nucleotide codons used to code for them.