Human papillomavirus type 16 (HPV16) is a causative agent of cervical cancer. In many HPV induced cancers the HPV genome is present integrated into the cellular chromosomes. Integration of HPV DNA into that of the host promotes genomic instability and progression to cancer. Factors that promote integration remain to be fully identified. HPV replication during DNA damage is a factor that may promote viral DNA integration. HPV origin replication is investigated in the presence of DNA damaging agents. HPV replication produces double strand DNA breaks in the viral genome and DNA damaging agents such as etoposide will result in double strand breaks in genomic DNA therefore creating substrates for viral integration. HPV origin replication during DNA damage is not inhibited in vivo and in vitro. In contradiction to these results, replication initiated by the SV40 helicase, Large T antigen, is arrested in response to DNA damage and ATR is the candidate kinase for mediating the arrest. In order to carry out replication of the viral origin, the HPV genome encodes for the E1 protein. E1 forms a di-hexameric helicase complex that replicates the viral genome. The failure of E1 to be targeted by ATR/ATM allows HPV replication in the presence of DNA damaging agents. Such replication will result in double strand breaks in the viral genome ultimately promoting viral integration and progression of cervical cancer. Upon activation of the DNA damage response cellular DNA replication is stalled. The targets of the DNA damage pathways at the replication fork that result in the inhibition of DNA replication are not clearly defined. Evidence has shown that the MCM2-7 cellular helicase complex is targeted for phosphorylation by the ATR/ATM kinases and therefore LTAg helicase may also be targeted in a similar manner in response to DNA damage. The results presented in this thesis support the hypothesis that the replicative helicase is a direct target for phosphorylation by the DNA damaging signalling kinases. Using SV40 replication as a model for eukaryotic DNA replication the results in this thesis show that LTAg is targeted for phosphorylation by the DNA damage signalling kinases ATR/ATM in response to DNA damage. Large T Antigen protein levels decrease after prolonged treatment with etoposide suggesting that the degradation may inhibit the replication functions of this viral helicase. To carry out viral replication HPV encodes two proteins, E1 and E2, which interact with cellular factors to replicate the viral genome. E2 forms homodimers and binds to 12bp palindromic sequences adjacent to the viral origin and recruits the viral helicase, E1, to the origin. The regulatory consequences of the E1.E2 interaction have been controversial. The data in this thesis demonstrates the ability of E1 to stabilise E2 increasing E2 half life. Analysis using a mutant E2 that binds weakly to E1 demonstrates that the stabilisation is a consequence of a direct E1.E2 interaction. Furthermore this thesis also demonstrates that the presence of E1 results in the redistribution of HPV16 E2 by enhancing its affinity for chromatin. The E2 protein tethers the viral genome onto the host chromatin during cellular mitosis thereby ensuring the distribution of viral episomes into both daughter cells. The redistribution of E2 onto chromatin suggests a role for E1 in enhancing HPV genome segregation functions by stabilising the association of E2 with mitotic chromosomes. E2 also regulates transcription from the viral genome. The data in this thesis shows that the E1 protein enhances E2 transcription function in a manner that suggests the E1 protein itself can contribute to transcriptional regulation not simply by E2 stabilisation but by a direct stimulation of E2-mediated transcription. E1 regulation of E2 function is again dependent on a direct protein-protein interaction. Taken together these results suggest that co-expression of E1 with E2 can increase E2 stability, enhance its affinity with chromatin and enhance E2-mediated transcription. These consequences are discussed with relation to the virus life cycle.

Nonmelanoma skin cancer (NMSC) is the commonest cancer worldwide and is a significant and increasing burden on health care resources. NMSC aetiology is fundamentally linked to sun exposure although infection with oncogenic viruses including Human Papillomaviruses (HPV) is thought to be a cofactor. Fair skinned populations residing in geographical areas with high sun exposure have an increased incidence of NMSC and an excess of NMSC is also observed in certain patient groups including immunosuppressed organ transplant recipients. This thesis describes the epidemiology of NMSC in the immunosuppressed renal transplant population in the West of Scotland. Clinical samples from this population were investigated for the presence of HPV utilising a PCR reverse hybridisation technique. This assay specifically examined for HPV of the genus Beta that have been previously linked to NMSC. The effect of presence of HPV in these samples was evaluated through an investigation of the expression of cellular biomarkers. A biomarker expression pattern specific to HPV infected lesions would add further support to the link between HPV and NMSC. Samples were probed for Ki67, p16, p53, MCM2 and MCM5 antigens in addition to novel antigen, Topoisomerase II Beta Binding Protein 1 (TopBP1). TopBP1 is a host cellular protein that is involved in the DNA damage response and is an interacting partner for HPV thus making it a likely candidate for involvement in cutaneous carcinogenesis. A biomarker expression pattern specific to HPV infected NMSC was not identified although aberrant expression of TopBP1 was identified in a subset of skin cancers. To investigate this novel observation, a molecular investigation of the interaction between TopBP1 and the viral replication factor E2 was carried out, firstly in HPV-16 before extending the work to the cutaneous virus HPV-8. A mutant of HPV-16E2 that failed to bind TopBP1 was generated. Failure to bind TopBP1 resulted in a phenotype exhibiting compromised viral replication. The equivalent mutation generated in HPV-8E2 resulted in an even more compromised replicative phenotype. In addition to this work, gene targets of TopBP1 were also identified by a microarray analysis. This was carried out using MCF7 cells untreated or damaged by ultraviolet radiation (UVR) with endogenous TopBP1 depleted by SiRNA treatment. These studies showed that TopBP1 is involved in a number of cellular processes and is likely to be involved in controlling DNA damage targets following UVR induced DNA damage. The work described in this thesis provides further evidence supporting the role of HPV in the aetiology of nonmelanoma skin cancer. Generation of an E2 mutant that fails to bind TopBP1 and exhibits a compromised replication phenotype provides further evidence to support the hypothesis that the E2-TopBP1 interaction is essential for viral replication and therefore the viral life cycle. This work may facilitate the development of anti-viral therapies for HPV-associated disease by targeting HPV DNA replication through disruption of the E2/TopBP1 interaction.

Recombination is an important biological process in a diverse range of viruses, particularly those with single-stranded ribonucleic acid (RNA) genomes including the enteroviruses. Mutations caused by the error-prone RNA-dependent RNA polymerase (RdRp) and the vast population size of these virus populations are evolutionary mechanisms that generate genetic diversity – this allows viruses to survive under changing environmental pressures (e.g. adaptive host immunity). Ribonucleic acid recombination has been identified as another contributing mechanism involved in diversification, by removing interfering or lethal mutations from a virus genome, and by establishing new viruses. Virus RNA recombination is well documented and is identified in several virus families including picornaviruses. However, recombination is a rare event and the study of the molecular mechanisms behind virus recombination is complicated by our ability to isolate and analyse recombinants from a mixed virus population including the parental viruses. The objectives of this study were to firstly devise a method for generating populations of natural recombinant viruses, and secondly, to study the molecular processes that determine where and when recombination occurred in the enteroviral genome. During this project, an in vitro system was developed to allow the recovery of recombinant enteroviruses in the absence of their parental viruses. Two virus RNA molecules containing “lesions” rendering them unable to generate viable virus on their own were co-transfected into mouse L929 cells. The method required two parental virus RNA molecules to be present in a single cell to produce a viable recombinant virus. Reverse Transcription-Polymerase Chain Reaction (RT-PCR) and sequencing analysis confirmed the recombinant nature of progeny virus genomes. Experimental data confirmed the effectiveness of the method and provided evidence that recombination occurs in at least two phases. Initial template switching, referring to the transfer of RdRp from one RNA template to another mid-replication, occurred apparently indiscriminately and with the addition of extra virus and non-virus sequence at the junction sites. In the second phase, any additional sequence was lost during subsequent rounds of replication and selection. The approach was expanded to incorporate viruses from different enterovirus species to investigate intra- and interspecies recombination events.

The oceans cover ca. 70% of the Earth’s surface and due to their depth encompass around 300 times the habitable volume of the terrestrial environment. The exact proportion of life on Earth that exists in the oceans is unknown as many ocean species remain undiscovered; in particular this holds true for the viruses that infect marine bacterioplankton. It is currently thought that viruses that infect bacteria, bacteriophages or phages, can numerically exceed their hosts by a factor of ten, however, this abundant and diverse group of organisms is still poorly understood. This is especially true of phages that infect members of the Roseobacter clade. Globally, members of the Roseobacter lineage can comprise up to a quarter of the marine microbial community and often dominate the alga-associated bacterial community. In this study phages capable of infecting species of Roseobacter were isolated and characterised. Two Roseovarius-specific phages, RLP1 and RPP1, were isolated from UK coastal waters; morphological and sequence data identified them as belonging to the N4-like genus of Podoviridae. Comparative genomic analysis of both Roseovarius phages to other N4-like phages such as Escherichia coli phage N4 and Sulfitobacter sp. EE-36 phage EE36Φ1, revealed a number of conserved core genes involved in DNA metabolism, transcription control and virion structure. Comparison of N4-like Roseobacter phages (RLP1, RPP1, EE36Φ1 and Ruegeria pomeroyi DSS-3 phage DSS3Φ2) also revealed a number of peripheral genes which are likely to interact directly with host proteins/machinery specific to the Roseobacter group. Unusually, both RLP1 and RPP1 appeared to only infect host cells when in semi-solid agar matrix, but not in liquid culture. Comparison of the outer surface of agar-embedded and planktonic cells revealed different outer-membrane protein and lipopolysaccharide expression profiles. This suggests that some Roseobacter species (spp.) change components of their bacterial cell surface according to their physiological state: agar-embedded/sessile or planktonic and RLP1 and RPP1 exploit this by binding to (a) receptor(s) only expressed during sessile conditions. A number of prophage-like elements were also induced from three Roseobacter spp. by exposure of growing cultures to the DNA-damaging chemical Mitomycin C. These were identified by electron microscopy as belonging to the Siphoviridae family. The results of this project suggest that within the marine environment there remain many uncharacterised phages with peculiar biochemical properties and a wealth of genomic information.

Adenoviruses (Ads) have many attractive characteristics for use as agents of genebased vaccines and therapies. The most frequently used Ad vectors in preclinical research are based on Ad5. However, in the clinical setting Ad5 vectors have severe limitations. About 90% of the population have neutralising antibodies against Ad5 and infection requires expression of the viral receptor CAR, which is not present on important cell types. Previous data from this laboratory suggested that the species D adenovirus, Ad19a, may overcome some of these limitations. Most relevant for vaccination is its high efficiency of infection of human dendritic cells (DCs), the most important antigen presenting cells. This highly effective DC targeting was retained in Ad19aGFP vectors. To investigate the potential of Ad19a vectors for vaccination further, two transgenes, the nucleocapsid gene from pneumovirus of mice (PVM-N), and a HIV polyprotein cassette (HIVA), were inserted into replication-deficient Ad5 and Ad19a vectors using recombineering. rAd19aPVM and rAd19aHIVA expressed a significantly higher amount of transgene compared with their Ad5 homologues. Encouraging results were obtained when the ability of rAd5PVM and rAd19aPVM to protect mice from lethal PVM challenge was examined using various prime/boost vaccinations. A dose of 106 pfu of rAd19aPVM, but not rAd5PVM, provided protection. rAd5PVM did, however, protect mice at the same dose when combined with rAd19aPVM in a heterologous prime boost schedule. Vaccination-induced IgG responses to PVM-N did not correlate with protection, implicating cell-mediated immune responses in protection. Utilising rAd19aGFP, evidence is also provided that Ad19a may use CD46 and to some extent CAR as a receptor on CHO cells, expanding our knowledge of the basic biology of this virus.

Respiratory Syncytial Virus (RSV) is a member of the pneumovirus genus (family Paramyxoviridae, subfamily Pneumovirinae). RSV is an important respiratory virus of both infants and the elderly, representing an underappreciated burden on health care systems. In addition, re-infections can occur despite the presence of pre-existing immunity, suggesting that immunological memory to RSV is incomplete. To date, treatment of RSV infection is limited to the provision of supportive care and no effective vaccine is available. Although several are currently under investigation, these candidates focus upon the delivery of the F and G antigens of RSV to stimulate the immune system, rather than the internal antigens, which may provide cross protection between different subtypes of RSV. Vaccine development has been greatly hindered by the lack of an appropriate animal model in which to study vaccine efficacy and pneumovirus pathogenesis. Pneumonia virus of mice (PVM) is also a member of the Pneumovirus genus and, like RSV infection of humans, causes a bronchiolitis and fatal pneumonia in its natural host, the mouse. PVM has been proposed as an appropriate model system in which to both study pneumovirus pathogenesis and vaccine efficacy. The PVM model system was adapted to investigate a potential vaccination strategy to address the lack of an available RSV vaccine. Replication deficient recombinant adenovirus serotype 5 (rAd5) vectors were constructed which expressed the F, M and N genes of PVM J3666, in addition to a control construct, which expressed the LacZ gene of E. coli. The constructs were administered via the intranasal route to BALB/c mice and were able to elicit complete protection against a lethal dose of pathogenic PVM J3666, in both short-term experiments and in a long-term experiment, up to 20 weeks post immunisation. The protection effect elicited by the constructs was observed when administered in a single dose, and in alternative mouse strains, C3H/He-mg and C57BL/6, which had differing immunity haplotypes. The rAd5 vectors generated a PVM specific IgG humoral response to PVM and Ad5 antigen which did not correlate as the primary mediator of protection. The rAd5 candidate expressing the N gene of PVM was shown to induce IFNγ secreting T-cells. The use of a peptide library of PVM N protein determined that a specific response could be identified towards the amino acids N41-90, N81-130, N161-210 and N281-330. Thus, the PVM infection model of BALB/c mice provides an immunological platform to facilitate the study of RSV and PVM pathogenesis, immunology and vaccine development.

Rotavirus encodes six structural and six non-structural proteins. In contrast to the structural proteins, the functional roles of the non-structural proteins are not well defined beyond a realisation that they must have a role in the viral replication cycle. A fuller understanding of the replication cycle must therefore rest on determining the specific roles played by the non-structural proteins. Non-structural protein NSP1 shows high levels of sequence divergence. A generally well conserved cysteine-rich region at the amino-terminus may form a zinc finger structure. It has been shown to possess non-specific RNA-binding activity, and has been found associated with the smallest of three replication intermediates (RIs) found in infected cells, together with the viral proteins VP1, VP3 and NSP3. VP2 and VP6 are added sequentially to the pre-core RI to form the core RI and single-shelled RI respectively. The function of NSP1 in the replication cycle and the importance of its presence in early replication complexes has not been determined. The intermolecular interactions that occur between the components of the RIs have not been defined. Protein-protein interactions between NSP1 and VP1, VP2, VP3, and NSP3, from the UKtc strain of bovine rotavirus, were investigated using a variety of approaches, the first of which was the yeast two-hybrid system. In this assay a self-interaction of NSP1 was not detected. Protein-protein interactions between NSPl and VPl, VP2, VP3, and NSP3, were also not detected. Both the full-length protein and a truncated NSPl, consisting of only the amino terminal third of the protein, were tested. A direct self-interaction of NSP3 was shown and quantified. Radio-immunoprecipitation analysis of in vitro translated viral proteins using specific anti-NSP1 serum was also employed. However, it failed to detect direct protein-protein interactions between NSP1 and VPI, VP2, and VP3. Immunoprecipitation of UKtc rotavirus-infected celllysates with anti-NSP1 serum showed the co-precipitation of viral proteins VPl, VP2, VP3NP4, VP6 and NSP3, with NSP1. It was proposed that NSP1 formed a previously unrecognised complex with these proteins. Immunoprecipitation of nuclease-treated infected cell lysates showed a reduction in the co-precipitation of VP2, VP3NP4 and NSP3 with NSP1. No reduction in the co-precipitation of VP6 was seen. The association of the complex proteins may be mediated by RNA binding. Immunoprecipitation with an anti-VP6 monoclonal antibody reciprocally precipitated small amounts of NSP1, VP2, VP3/VP4, and NSP3, with VP6.

The structure of the herpesviruses particle is characterised by an icosahedral capsid surrounded by a proteinaceous tegument layer and is enclosed by a lipid envelope. The understanding of the structure of the capsid, primarily through the use of cyro-electron microscopy, is greater to than of the tegument, due to the typically amorphous nature of the tegument. The interaction between the capsid and tegument has been well studied, unveiling interactions limited to the capsid vertices involving two minor capsid proteins, pUL17 and pUL25, and the large tegument protein pUL36. In herpes simplex virus – 1 (HSV-1), pUL17 and pUL25 form the capsid vertex-specific component (CVSC), a heterodimeric structure which resides in top of triplexes between peripentonal hexons. pUL36 has been suggested to connect the CVSC to the penton and to the rest of the tegument proteins. Recent studies on the gammaherpesvirus Karposi’s sarcoma-associated herpesvirus (KSHV) and the alphaherpesvirus pseudorabies virus (PrV) have questioned both the protein content of the CVSC and the organization of pUL17 and pUL25. As well as the composition of the CVSC, these studies have provided further insight as to the location of tegument assembly to the capsid, a subject that remains a highly contested. In order to clarify the content of the CVSC, virus mutants with deletions of the large inner tegument protein pUL36, and a second inner tegument protein, pUL37, were analysed using cryo-electron microscopy and icosahedral reconstructions. The examination and comparisons of these mutants with wild-type HSV-1 revealed that the CVSC is not only formed by pUL17 and pUL25 as originally reported, but also pUL36, as suggested in the most recent studies. In addition, comparisons of the capsid structure of the pUL36 deletion mutant with a mutant with a full deletion of pUL34, a protein implicated in nuclear egress as part of the nuclear envelopment complex (NEC) and therefore cannot exit the nucleus, suggest that at least part of pUL36 is present on nuclear capsids. Further analyses of the mutant virus with pUL36 deleted using immunofluorescence and Western blotting, along with the capsid reconstruction of a pUL36 deletion mutant which retains only the N-terminal 361 codons suggest that the C-terminal end of pUL36 is present in the nucleus. The work presented here offers additional evidence to clarify the roles of pUL17, pUL25 and pUL36 in tegument assembly. In particular, structural analysis has implied that the contributions of pUL36 to the nuclear capsid stabilizes the CVSC structure from a structural stand-point, emphasizing it’s importance as a multifunctioning protein which acts as a bridge between the capsid and tegument.

There are many RNA virus pathogens of humans including influenza, parainfluenza, enteroviruses, parechoviruses, coronaviruses, pneumoviruses, and metapneumoviruses. These commonly invade and infect the respiratory and gastrointestinal tracts, giving rise to acute and chronic respiratory tract infections, and some may also reach the central nervous system (CNS) either via haematogenous or neural routes resulting in a variety of clinical presentations, (e.g. meningitis, encephalitis) which may lead to severe irreparable damage such as poliomyelitis especially in young children. Early correct diagnosis of viral infections is indispensable in order to prevent potential outbreaks which threaten the public health worldwide and might lead to high morbidity and some with significant mortality. Several laboratory techniques such as virus isolation, direct visualization of viral particles, detection of viral antigens and/or nucleic acids, and detection of host immune response (e.g. anti-viral antibodies) to infection, are available for diagnosis, but may have significant drawbacks such as time-inefficiency, cost and certain individual limitations. Microarray technology offers one way to overcome some of these limitations. In this study, a microarray chip containing 7967 oligonucleotides (probes) covering the whole genomes of human enteroviruses, rhinoviruses, respiratory syncytial viruses, metapneumoviruses and influenza viruses was designed and constructed using both OligoArray and Agilent eArray software, to allow simultaneous detection of any of the above viruses present in any clinical specimen. This virochip was tested against positive controls and clinical samples known to contain RNA nucleic acids of these viruses. Viral RNA was reverse transcribed, and amplified. Considerable effort was expended in trying to optimise a multiple displacement amplification (MDA) protocol for whole genome amplification, and in addition, long-range PCR was also utilised. Amplification products were fragmented, labelled and loaded into a hybridization reaction with the designed viral probes printed on the virochip. The results revealed (i) a number of technical problems associated with MDA; (ii) that some probes either failed to recognise their intended targets, or produced cross-reactive signals with non-intended targets; (iii) that many of the designed probes hybridized to their relevant viral nucleic acids and generated hybridization signals of high fluorescent intensity offering an opportunity to develop this probe array in order to be used for the identification of a wide variety of virus species up to the serogroup level or beyond (if required) specifically those causing CNS and respiratory tract infections.

Numerous endogenous retroviruses (ERVs) are found in all mammalian genomes and represent retroviruses which have, by chance, integrated into the germline and are transmitted vertically from parents to offspring. In many non-human primates these insertions have not been well-studied. ERVs provide a snapshot of the retroviruses a host has been exposed to during its evolutionary history, including retroviruses which are no longer circulating. Accurate annotation and characterisation of ERV regions is an important step in interpreting the huge amount of genetic information available for increasing numbers of organisms. This project represents an extensive study into the diversity of ERVs in the genomes of primates and related ERVs in rodents, lagomorphs and tree shrews. The focus is on groups of ERVs for which previous analyses are patchy or outdated, particularly in terms of their evolutionary history and possible transmission routes. A pipeline has been developed to comprehensively and rapidly screen genomes for ERVs and phylogenetic analysis has been performed in order to characterise these ERVs. Laboratory study was used to complement the bioinformatics analysis. Almost 200,000 ERV fragments, many of which have not previously been characterised, were identified. A novel endogenous member of the lentivirus genus of retroviruses, which are rarely found in an endogenous form, was identified in the bushbaby Galago moholi. This ERV may represent an ancient ancestor of modern human immunodeficiency virus (HIV). Another retrovirus, gibbon ape leukaemia virus, previously thought to be a common pathogen in gibbons, was found to not exist in contemporary gibbons and a route through which a single cross species transmission event may have resulted in all known cases of this disease worldwide was identified. Endogenous epsilonretroviruses, usually considered to be viruses of fish and amphibians, were identified in all screened species of primates.