No description available.

No description available.

No description available.

No description available.

No description available.

No description available.

The influenza BM2 proton channel is a small, tetrameric α-helical membrane protein, which is vital in multiple stages of the life cycle of the virus. The transmembrane (TM) domain is responsible for modulating pH to facilitate release of viral genetic material whilst the cytoplasmic domain is purported to play a role in virus budding via interactions with another protein, BM 1. Coarse-grained molecular dynamics (CGMD) simulations have been employed in several studies of a-helical membrane protein folding with success. This methodology is applied to the BM2 TM domain to generate a structure of the tetramer based on sequence alone. A converged tetrameric bundle with left-handed packing was generated and stable following conversion to atomistic representation and conventional atomistic MD simulation. Sequential protonation of His19 residues within the putative HxxxW gating motif led to subtle rearrangements in the packing of the helices, allowing formation of a water wire once three of the His19 residues were protonated. The influence of solubilising detergents on membrane protein structure is a topic of debate in the literature. The recent solution NMR structure of the BM2 TM domain was used as the basis of a study on the influence of two types of detergent micelle environment compared to a dipalmitoylphosphatidylcholine (DPPC) bilayer. The detergents chosen were zwitterionic dihexanoylphosphatidylcholine (DHPC) and non- ionic β-dodecylmaltoside (DDM). The results of these simulations suggest that BM2 is able to modulate its helix packing to compensate for hydrophobic mismatch in the DPPC bilayer. The protein-detergent complexes appeared to retain some of the features of detergent-only micelles. DDM detergent molecules replicated the binding mode of DPPC lipids (parallel to the protein bundle axis) whereas the DHPC detergent molecules adopted a binding mode perpendicular to the protein bundle axis. This appeared to influence the binding on/off rates of each detergent. Recent developments in the field of mass spectrometry have allowed intact membrane protein complexes in detergent micelles to be stabilised in the gas phase. Questions remain over the structure of these gas phase complexes compared to solution phase. A combination of non-equilibrium and equilibrium MD simulations was used to study the effects of transfer from solution to vacuum of protein-detergent complexes. The DHPC detergent micelle formed destabilising interactions with BM2 upon dehydration whereas the presence of the DDM micelle stabilised the solution phase protein conformation. In the final Chapter the multiscale simulation approach is used to study the interactions between BM2 and BM1. Protein-protein interactions formed via CGMD simulation were conserved upon atomistic simulation. The results of this thesis highlight the worth of multiscale simulation approaches in observing a greater range of timescales and processes.

The family Picornaviridae includes many important human pathogens. RNA structures play important roles in picornavirus molecular biology and recent evidence suggests that these are more extensive than previously thought. In this project we identified a number of potential RNA structures in picornavirus genomes and started to analyse one of these structures. The work focussed on human parechoviruses (HPe V). The structure of the HPe V 5' untranslated region (UTR) was analysed by obtaining several new sequences and using an alignment of 60 sequences to identify covariant changes. This allowed the previously predicted structure to be confirmed and refined. Aligned sequences representing most picornavirus species were then analysed for suppression of synonymous codon variation (SSCV). Strong SSCV was seen in several cases and this was often related to the presence of RNA structures including the Cre and novel potential structures. Patterns of conserved dinucleotides were also used to identify regions of importance in the picornavirus genome. A new program, Dinucleotider (1.0) was developed and used, which allows a graphical output of conserved dinucleotides in aligned sequences. CG was found to be the most informative dinucleotide and could be used to identify regions of the picornavirus genome, which corresponded to the 5'UTR, 3'UTR and Cre, as well as further new structures. Genetic analysis of a predicted structure in the 3D-encoding region of HPe V s, was carried out by making two mutants, with 3 or 6 mutations in one of the structural domains. Both sets of mutations had little effect on virus growth in cultured cells, suggesting that the structure does not play a critical role in replication and other possible roles need to be identified. Overall, this project has allowed several RNA structures to be identified in picornaviruses. These are conserved between related viruses and presumably play important roles in the biology of picornaviruses. They need to be studied further in order to improve understanding of how picornaviruses infect cells, which is required to improve diagnosis and control of these pathogens.

Caliciviruses cause importarit diseases of humans and animals; noroviruses are responsible for outbreaks of gastroenteritis in man, and feline calicivirus (FCV) causes an upper respiratory tract infection in cats. Since 2003, murine norovirus (MNV) has been used as a model for norovirus infection since human noroviruses are unable to replicate in tissue culture. Caliciviruses have a positive-sense RNA genome of about 7 to 8 kb, with a viral protein, VPg, covalently linked to the 5' end of the RNA. The VPg acts as a novel proteinaceous 'cap substitute' and interacts with the cap-binding protein, eIF4E. To further analyse the requirement of translation initiation factors for calicivirus translation, I have investigated the effect of FCV and MNV infection on eIF4E and its regulator, 4E-BPl. I found that FCV and MNV infection have different effects on the phosphorylation status of eIF4E and its regulator, 4E-BPl. For both viruses, eIF4E is phosphorylated and 4E-BPl is dephosphorylated during the course of infection. This would be expected to inhibit host cell protein synthesis. This has led me to investigate the signalling pathways involved in mediating these effects. Using specific inhibitors I have shown that phosphorylation of eIF4E during FCV and MNV infection occurs through activation of the MEKlMAPK pathway, while 4E-BPl phosphorylation occurs through PI3KJmTOR activation. When I use specific inhibitors are used to block either the p38 or MEK pathways, replication of MNV is inhibited. However, FCV replication is only inhibited by MEK inhibition. In addition, eIF4E phosphorylation is prevented by blocking p38 but not MEK. This suggests that the changes in eIF4E phosphorylation may not affect the interaction between eIF4E and VPg or be important for Fev and MNV replication. These, changes may therefore only be involved in host-cell. shut-off. In contrast, inhibition of the mTOR pathway significantly inhibited both FeV and MNV replication, suggesting that the phosphorylation status of 4E-BPl during infection is important for virus replication. I hypothesise that the change in phosphorylation status of 4E-BPl may lead to the inhibition of eIF4F complex formation and therefore calicivirus translation; this may affect the switch from translation to replication of the viral RNA. I propose that the mTOR pathway could be a target for antiviral therapy. I have also further investigated the exact requirements for components of the eIF4F cap-binding complex for translation initiation on FeV and MNV mRNA. I have shown that a small-molecule inhibitor of the eIF4E-eIF4G interaction has a greater inhibitory effect on FeV than MNV replication, demonstrating a requirement by FeV for eIF4E and an intact eIF4G for viral RNA translation.

The genomes of viruses must be packaged into a protective protein shell known as the capsid. For simple viruses with a positive-sense, single-stranded RNA genome, this packaging is specific, despite the presence of a large number of competing cellular transcripts. Interactions between the genome and capsid proteins must at least in part determine the selectivity of packaging. and the organization of the packaged RNA. This organization is difficult to visualize as genomic RNAs are large complex molecules that can in principle adopt many different conformations and orientations within the viral capsid. This thesis describes the use of electron microscopy techniques to determine the structure of simple positive-sense RNA icosahedral viruses. The principal aim has been to study the genome organization of two such viruses using methods based on electron cryo-microscopy (cryo-EM). Single-particle electron microscopy and icosahedral 3D reconstruction has been used to probe the RNA structure and organization of the picorna-like Cowpea mosaic virus (CPMV). Sub-nanometer resolution 3D reconstruction, has allowed the secondary structure elements of the viral capsid to be resolved. We show that the CPMV genomic RNA is organized as a dodecahedral caged structure, and identify RNA binding sites on the capsid inner-surface. The binding sites do not appear to enforce a particular bound conformation on the RNA, as has been observed for members of the Nodaviridae. We also image, for the first time, a C-terminal 24 amino acid segment that is implicated in RNA packaging and virus assembly but missing from previous structural studies. We present a pseudo-atomic model for this sequence. Electron cryo-tomography and sub-tomographic averaging has been used to carry out the asymmetric 3D reconstruction of the bacteriophage MS2:F-pili (virus-receptor) complex to - 40 A resolution. The resulting structure, the first such, provides unique insights into the biology of a single-stranded RNA virus. We describe for the first time, the asymmetric distribution of genomic RNA within an asymmetrically averaged virus capsid, confirming previous ideas about RNA packaging. We show that MS2 is not truly icosahedral, as the viral maturation protein appears to replace one of the capsid protein dimers in the protein shell.