- Volume 1, Issue 1A, 2019

Histomonas meleagridis is a protozoan parasite that causes mortality and morbidity in a wide range of gallinaceous fowl. It most notably affects turkeys, causing 80–100 % mortality in a flock. There is no commercial treatment for this parasite at the moment and attempts at a vaccine have failed. In this study we investigated the potential efficacy of novel rumen and chicken gastrointestinal tract microbiome-derived antimicrobial peptides (AMPs) against H.meleagridis. H. meleagridis was cultivated in growth flasks in Dwyer media. The concentration of the parasite was determined using a Haemocytometer, the cells were counted under the microscope. The AMP challenges were carried out in 96-well plates. The starting concentration of the AMPs were 1024 µg ml−1 down to 2 µg ml−1. The cell densities of H. meleagridis were checked at 24 h. The protozoal cell densities were checked by extracting DNA and performing qPCRs targeted against FeHyD and Rpb1 genes. Fluorescent microscopy was also used to check parasite densities as described above. Five Chicken and three Rumen AMPs resulted in decreases in protozoal cell densities following microscopy, suggesting that the peptides show potential therapeutic application for blackhead disease. Nonetheless, the qPCR primers were non-specific and resulted in bacterial DNA amplification. Consequently, in order to understand the biological function of H. meleagridis and design new qPCR primers allowing quantification, we isolated the macronuclei of a H. meleagridis strain followed by sequencing using the illumine Hi-Seq 2500 and paired ends. Future work will focus on confirming how well the AMPs work agent H. meleagridis by employing electron microscopy to observe how the AMP affect the cells.

Amoebic Gill Disease (AGD) is a major problem in the aquaculture industry, as it is responsible for substantial losses of farmed Atlantic salmon in various worldwide locations. The disease is caused by the usually free-living Paramoeba perurans compromising the gills through the resulting development of hyperplastic lesions and lamellar fusion. These structural changes result in a reduction in the functional surface area of the gill tissues. Recent research has focused on identifying bacteria present within a culture of P. perurans, through performing isolation and identification of bacteria present in the cultures using 16S sequencing. Further NGS sequencing was performed from various culture conditions to provide insight into the changes of the bacterial microbiome during amoeba culture. As attempts to isolate the amoeba from the bacterial contamination has been unsuccessful, consideration into a possible symbiotic relationship between the amoeba and bacteria was considered. A filtering method was used to attempt to identify the genera of bacteria present within the amoeba. The isolation and 16S sequencing identified the presence of various marine bacteria, including those of the Pseudoalteromonas, Halomonas, Cellulophaga and Mesonia genera. The NGS sequencing identified a substantial proportion of sequences to match the Vibrio genus and suggests an association between this genus and the amoeba. If symbiotic relationships between specific bacteria and amoeba can be confirmed, the bacteria could potentially be used as an indicator organism for the risk of AGD outbreak. It may also provide an indirect target for the control and treatment of AGD.

The gastro-intestinal tract hosts a complex microbial ecosystem that helps regulate the physiological, immunological and nutritional functions of the pig so disturbances within this microbiota can have profound effects on porcine health and disease. The gut microbiota is shaped by the environment, immune pressures and diet. The weaning transition represents a critical time-point that can interfere with intestinal development and cause dramatic shifts in the gut microbiota. Previously, in-feed antibiotics were used to counteract the adverse effects of weaning but an EU-wide ban since 2006 has propelled the search for safe and sustainable alternatives within the livestock industry. Studies have indicated that dietary fatty acids are efficient in terms of minimising weaning disorders including elevated incidence of enteric disease and immunodepression and thus represent a promising alternative to antibiotics. Furthermore, previous research has indicated that including fatty acid mixtures in the porcine diet has a positive effect on overall animal performance, increasing growth rate and enhancing feed conversion ratios. It is thought that these health benefits and production gains stem from the ability of fatty acids to mediate the effects of gut microbiota on intestinal immune function through modifying the proportions of microorganisms present. The aim of this research was to use deep sequencing and metagenomic methods to explicitly explore the influence fatty acid supplementation has on shaping the intestinal microbiome during weaning. Results have indicated that fatty acids play a role in modifying the intestinal microflora, establishing an environment that favours the growth of commensal species such as Lactobacillus.

Hantaviruses are a diverse group of single-stranded, negative sense RNA viruses, belonging to the Bunyaviridae family. They are primarily rodent-borne and transmitted into humans through the inhalation of aerosolised excreta of infected animals. Each hantavirus is typically associated with a single reservoir species, resulting in a persistent, yet asymptomatic infection. In humans, however, in addition to asymptomatic infection two disease syndromes are associated with hantavirus infection, hantavirus haemorrhagic fever with renal syndrome (HFRS) and hantavirus pulmonary syndrome (HPS). Tatenale Hantavirus (TATV) was first isolated from field voles in NW England in 2013 and has since been detected at other sites in Northern England. However, only two small fragments of the L and S segments have been sequenced, precluding full phylogenetic characterisation of the virus. Field voles were sampled at two sites in Leicestershire (N=104) and Cheshire (N=12), and their lungs subjected to a degenerate pan-hantavirus RT-PCR assay, targeting a section of the polymerase gene. Eleven Cheshire and one Leicestershire sample were TATV positive, and were 94 % homologous across the amplicon. Illumina Hi-Seq sequencing revealed the complete coding regions of the Leicestershire isolate. Analysis of the sequence showed that TATV is most closely related to Khabarovsk virus, with a homology of 91%/88%/89 % (AA) across the L/M/S Segments, respectively. ICTV species demarcation requires an AA difference of >7 % across the S and M, TATV meets the requirement of a novel species. Retrieval of full coding sequence will allow further investigation into TATV, and its potential to infect humans through pseudotyping of the glycoproteins.

Enteric fever is major health issue in developing countries and it is becoming progressively untreatable due to increase of antimicrobial resistance. The causative agent, Salmonella enterica, replicates in host phagocytes in various organs and regulates expression of hundreds of genes in response to host signals. PhoPQ is one of the key regulators and essential for virulence in humans and in a mouse typhoid fever model. The sensor kinase PhoQ responds to diverse stimuli (Mg2+ limitation, low pH, cationic antimicrobial peptides, high osmolarity, and, indirectly, to reducing conditions). However, which signals are predominant in vivo remains unclear. To address this issue, we determined the activity of the PhoPQ regulon using a chromosomally encoded fusion of the PhoP-dependent phoNp promoter to gfp-ova and we quantified single-cell fluorescence levels of Salmonella in spleen of infected mice using flow cytometry. The results show extensive heterogeneity in PhoP-activities in the Salmonella population. Comparison of mutants with sensing defects suggests that, in vivo, PhoQ responds to a combination of antimicrobial peptides, acidic pH and low Mg2+, but not to reducing conditions. Negative feedbacks have also only a minor impact. Single-cell analysis of phoNp and asrp promoters suggests differential environmental pH as a major driver of heterogeneous PhoP activities. A combination of immunohistochemistry and proteomics of Salmonella from infected samples was used to validate this hypothesis. Together, our data show how Salmonella uses the PhoPQ system to integrate various host signals in order to tune expression of virulence factors to the diverse tissue microenvironments that this pathogen inhabits.

Mycobacterium bovis, a member of the M. tuberculosis complex, causes bovine tuberculosis, one of the most important veterinary health problems in the UK. In the absence of improved control the projected economic burden in the UK over the next decade is predicted to be £1bn. Although transmission of M. bovis occurs primarily between infected mammalian hosts, M. bovis has been demonstrated to persist in soil, suggesting an environmental infectious route. M. bovis is likely predated on by environmental amoebae such as Dictyostelium discoideum and as such may have evolved mechanisms to modulate the interaction with amoebae. In this study we have investigated M. bovis interactions in vitro with D. discoideum. We demonstrate that virulent M. bovis evades destruction by D. discoideum. Using a genome-wide M. bovis transposon mutant library, we selected for mutants that failed to escape D. discoideum after 48 h of infection. Mutants of genes encoding the MCE4 transport system, genes involved in sulpholipid synthesis/transport, and genes encoding PPE and PE-PGRS protein families remained associated with D. discoideum. Most strikingly, mutations in 11 genes of the major mycobacterial virulence locus ESX-1, which encodes a T7 secretion system implicated in bacterial transit through host cell membranes, were significantly enriched in D. discoideum. Our data demonstrate that known virulence factors involved in host-pathogen interactions in mammalian hosts also play a role in D. discoideum-M. bovis interactions. Our data further suggest that M. bovis has evolved to actively transit bacteriverous D. discoideum rather that to use it as a replicative niche.

Within a large-scale report commissioned by the Wellcome Trust [1], primary school teachers were described as facing barriers in teaching science. The top barriers described were the lack of budget and resources, lack of time and curricular importance as well as other issues such as a lack of subject knowledge or confidence and concerns relating to space and resource access. Teaching science is just one part of a primary teacher’s complex role and is a subject in which most primary teachers do not have a degree or A level qualification [2,3]. There is little Microbiology content within the National Curriculum, however, schools can introduce additional scientific content within the Primary Key Stages. Within this context, a Microbiology, Genomics and Bioinformatics researcher in association with Key Stage 2 classes in a Norfolk Junior School carried out a joint project for Microbiology-related science in conjunction with Norwich Research Park facilities in May 2017 and the Microbiology Society. Here we report on the findings from teacher and pupil’s perspectives and consider how Microbiology/Hygiene could be presented to this age group in a classroom setting.

1. ‘State of the Nation’ report of UK primary science education. S. Leonardi et al. CFE Research, Leicester LE1 5TE. September 2017.

2. The Royal Society (2010) Science and mathematics education 5–14. A ‘state of the nation’ report. London: the Royal Society

3. ASE Guide to Primary Science Education (Serret and Earle, 2018), reviewed in https://tdtrust.org/cpd-primaryscience

Antibiotic production and cellular development in bacteria are intimately linked to the extracellular environment. One key mechanism by which bacteria recognise and respond to these external cues is through two-component regulatory systems (2CS). Consisting of a membrane-bound sensor kinase and a cognate DNA-binding response regulator, these 2CSs are essential in the response to a myriad of signals including antibiotic attack, microbial interaction and nutrient availability. The soil-associated filamentous actinobacteria Streptomyces spp. are prolific antibiotic producers with a large number of 2CS. In addition to their complex life cycle a new developmental stage has recently been described termed exploration. When exploring, the streptomycetes rapidly expand via non-branching vegetative hyphae when contacted by fungi, a juxtaposition to the canonical lifecycle which ends in sporulation. Fifteen of the fifty-six 2CS in the model organism Streptomyces venezuelae are highly conserved throughout Streptomycetaceae. Having developed and screened a 2CS operon deletion library within S. venezuelae we determined that one of these, cutRS, is involved in co-ordinating exploration and antibiotic production. The cutRS deletion mutant displays unrepressed exploration and overproduction of chloramphenicol. With a greater understanding of how Streptomyces spp. identify and respond to external signals we can imitate and subvert these systems. Using this we aim to activate cryptic biosynthetic gene clusters enabling the discovery of novel antimicrobial products which may prove beneficial in the clinical setting.

Actinobacteria are ubiquitous in soil and well-known for producing antimicrobial compounds. Increasingly, members of this phylum are found to form symbiotic relationships, for example with plants and insects, and are thought to provide protection against host infection. However, it remains poorly understood how Actinobacteria are recruited to microbiomes and whether secondary metabolites are produced in vivo. Acromyrmex echinatior leaf cutter ants transmit Pseudonocardia bacteria between generations and also recruit Streptomyces to their cuticular microbiome. We show that Pseudonocardia species isolated from the ant cuticle inhibit the fungal nest pathogen Escovopsisweberi and dual RNA-sequencing confirmed that Pseudonocardia secondary metabolite gene clusters are expressed in vivo on the ant cuticle. RNA stable isotope probing showed that ants supply cuticular resources to their microbiome which may fuel interference competition and select for antibiotic-producing bacteria. Similar to leaf cutter ants we also show that plant roots recruit growth-promoting and antibiotic-producing Streptomyces bacteria, but appear not to transmit them via their seeds. Root exudates are hypothesized to play a major role in root microbiome recruitment and DNA stable isotope probing coupled with Illumina sequencing showed that these were actively metabolized by many bacterial genera. However, Streptomyces appeared to be outcompeted by more abundant Proteobacteria, despite the fact that isolates could grow on purified exudates in the absence of competition. Streptomyces root exudate preferences are now being evaluated using comparative metabolomics. Defining the factors that influence the competitiveness of protective bacteria when colonizing microbiomes has implications for the development of more consistent biocontrol strategies and prebiotic techniques.

Streptomyces species are important producers of bioactive compounds such as antibiotics, antitumor and immunosuppressant drugs. Around two-thirds of all known natural antibiotics are produced by these bacteria and antibiotic production is linked to sporulation. The discovery of new bioactive compounds has declined since the 1960s but genome sequencing has revealed the potential to identify many more bioactive compounds. They are many secondary metabolite gene clusters which are inactive (silent) under laboratory conditions. We characterize the highly conserved actinobacterial two component system MtrAB which coordinates sporulation with secondary metabolite production in the two model organisms Streptomyces venezuelae and S. coelicolor. Deletion of the histidine kinase gene mtrB resulted in increased production of the antibiotic chloramphenicol in S. venezuelae and actinorhodin and undecylprodigiosin in S. coelicolor. Chloramphenicol is not usually produced under laboratory condition which suggests that deleting mtrB can activate silent antibiotic clusters. Additionally, we introduced point mutations at the D56 phosphorylation site of MtrA by CRISPR-Cas9 to abolish phosphorylation. This mutant shows the same phenotype as the ΔmtrA strain and chloramphenicol production is increased. Chromatin immunoprecipitation and sequencing (ChIP-seq) was used to identify MtrA targets and revealed that MtrA likely controls secondary metabolite production by binding to the promoter regions of cluster situated regulators in both model organisms. Additionally, MtrA binds upstream of genes involved in DNA replication and cell division. To our knowledge this is the first evidence of the connection of development and secondary metabolite production by a two component system.

Aurodox, a specialised metabolite from the soil bacterium Streptomyces goldiniensis, has been shown to inhibit the Enteropathogenic Escherichia coli (EPEC) Type III Secretion System (T3SS). To further assess the utility of this molecule as an anti-virulence compound, a better understanding of its mechanism of action is required. We used whole transcriptome analysis, cell infection and GFP-reporter assays to show that Aurodox transcriptionally downregulates the expression of the Locus of Enterocyte Effacement (LEE) pathogenicity island-which encodes for the T3SS, acting via its master regulator, Ler. We have also observed similar effects across other enteric pathogens carrying a homologous T3SS such as Enterohemorrhagic Escherichia coli (EHEC). These properties suggest Aurodox may have potential for the treatment of E. coli infections of the gut. Despite the recent interest in the compound, the biosynthesis of Aurodox by Streptomyces goldiniensis is still poorly understood. To gain insight in to this, we have sequenced the whole genome of S. goldiniensis and identified a putative Aurodox biosynthetic gene cluster (BGC) which shares a high level of functional homology with the BGC encoding Kirromycin, a non-methylated Aurodox derivative. In-depth analysis of the BGC supports a model where a unique polyketide synthase pathway involving a combination of both Cis and Trans-Acyltransferases synthesise the Aurodox polyketide backbone, followed by decoration and finally the addition of a methyl group. Future work will include the heterologous expression this BGC to confirm its role in Aurodox biosynthesis, with the ultimate aim to produce novel Aurodox derivatives.

Finding new antimicrobial compounds is vital to combat the growing threat of resistance. Most currently used antibiotics originate from actinomycetes discovered more than half a century ago. We recently reported the new species Streptomyces formicae, isolated from the African fungus-farming plant-ant, Tetraponera penzigi. S. formicae produces a novel family of polyketide antibiotics, the formicamycins, that have potent activity against resistant pathogens including MRSA and vancomycin-resistant Enterococci (VRE). Using CRISPR/Cas9, we have identified and characterised the genes responsible for formicamycin biosynthesis in the native producer. In addition, we used cappable RNA- and ChIP-sequencing to determine the transcriptional organisation of the pathway. We exploited this information to generate multiple mutants of S. formicae that overproduce formicamycins and their biosynthetic intermediates, some of which also have bioactivity. Furthermore, the potential for novel chemistry from S. formicae is not limited to the formicamycin pathway; antiSMASH analysis shows this talented strain contains at least 45 secondary metabolite biosynthetic gene clusters (BGCs). Under standard laboratory conditions, wild-type S. formicae also exhibits antifungal activity against the drug resistant Lamentospora prolificans, and when the formicamycin BGC is deleted, the strain produces even more potent antibacterial activity against MRSA. To identify the biosynthetic pathways for these metabolites, entire BGCs up to 208 kbp were deleted using CRISPR. Overall, this work demonstrates that searching under-explored environments for new species combined with genome editing is a promising route towards finding novel anti-invectives.

Microorganisms have been increasingly exploited for their remarkable ability to produce diverse natural products, which display bioactivities ranging from antimicrobial and anticancer to signalling and developmental. Genome sequencing has revealed that bacteria harbour many more biosynthetic gene clusters (BGCs) for natural products than are currently characterised. Whilst several genome mining tools have been developed to aid discovery, some classes of BGCs remain difficult to identify. One such class is ribosomally-synthesised and post-translationally modified peptides (RiPPs). RiPP BGCs are small with few regions of homology, and the short precursor peptides are rarely annotated in genomes. After developing a targeted genome mining approach for RiPPs, we identified a novel family of BGCs spanning over 200 actinobacterial genomes. The presence of diverse biosynthetic enzymes and sequence variation of the precursor peptides suggest that these BGCs may produce several structurally diverse molecules. We have successfully TAR cloned one such BGC from Streptomycesand expressed it heterologously, and metabolomic analysis led to identification of the pathway product. Gene deletion experiments have also confirmed the involvement of individual biosynthetic enzymes. The compound has been purified and the structure elucidated by NMR. Further work will focus on exploring RiPP BGCs from other species and investigating the biological role of these molecules. The discovery of such a diverse and highly conserved group of BGCs highlights that we are still scratching the surface of the huge biosynthetic capacity of microorganisms, and the use of more sophisticated genome mining tools could help unveil many more important molecules in the future.

The large amount of data from inexpensive sequencing means that the number of putative biosynthetic gene clusters (BGCs) far exceeds our ability to experimentally characterize them. This necessitates the need for development of further tools to analyze putative BGCs to flag those of interest for further characterization. clusterTools implements a framework to aid in the in silico characterization of BGCs by identifying regions of the DNA, containing homologous proteins, or coding sequences containing specific functional domain compositions using user-built HMM rules, in close proximity, reporting results in an easy to visualize manner. clusterTools complements existing software for BGC analysis in two ways. First, by running clusterTools on databases of genomic sequences in an exploratory mode, the user can identify and download regions of interest in the DNA for further processing and annotation in programs such as antiSMASH. Second, if clusterTools is run on databases constructed from putative gene clusters generated by antiSMASH, one can rapidly identify clusters on interest from the group that warrant further analysis and experimental characterization. We demonstrate the use of clusterTools as part of our workflow to identify BGCs of specific classes of natural products that would be difficult to identify with existing methods, particularly clusters containing assembly line domains as components, including those involved in bacterial polyketide alkaloid biosynthesis. clusterTools can also be used to identify novel BGCs by incorporating regulatory and antibiotic resistance elements. Standalone versions of clusterTools are available for Macintosh, Windows, and Linux.

The natural product, prodigiosin, caught researchers’ interest more than a century ago because of its bright red color. Today, interest in this tripyrrolic secondary metabolite remains strong due to its biological effects, including potent antibiotic activity against various Gram-positive bacteria. Further exploration of the potential of this class of molecules requires libraries of analogues. Yet, the total synthesis of prodigiosin-like compounds (prodiginines) proves challenging. This can be overcome by highjacking the bacterial biosynthetic machinery via mutasynthesis. Although a number of different bacteria produce prodigiosin, its biosynthetic pathway is well conserved. The final precursors, MAP and MBC, are always produced independently, before a terminal condensation reaction. Our work focuses on the last step of the formation of MAP: the oxidation of H2MAP. We were able to isolate and characterize HapB, the enzyme catalyzing this reaction in Hahella chejuensis. In addition, we showed that some modifications of alkyl substituents on the C2 and C3 positions of H2MAP did not alter HapB activity significantly. We then fed these analogues of H2MAP into Serratia ΔpigD, a mutant of Serratia ATCC sp.39006 which does not produce any endogenous H2MAP. As expected from the in vitro testing, chain elongation past two carbons on the C2 position could not be accepted whereas all substrates with a modification on the C3 position restored pigmentation, leading to the formation of eight novel prodiginines.

Non-ribosomal peptide synthetases (NRPSs) are responsible for the natural in vivo production of a large number of therapeutically relevant compounds. The non-ribosomal peptides (NRPs) formed by their action are the product of the long modular assembly lines, whilst the terminal step in the biosynthesis is the hydrolytic release, and frequently, macrocyclisation of the aminoacyl-S-thioester by an embedded thioesterase (TE). The surugamide biosynthetic pathway is composed of two NRPS assembly lines. One produces surugamide A, which is a cyclic octapeptide, and the other produces surugamide F, a linear decapeptide. The terminal module of each system lacks an embedded TE, which led us to question how the peptides are released from the assembly line (and cyclised in the case of surugamide A). We characterised an alpha/beta hydrolase and β-lactamase in vivo and established that the former is a type II TE for surugamide A, but not surugamide F, and that the latter is a trans-acting release factor for both compounds. In vitro substrate utilisation assays unambiguously established that the β-lactamase can produce mature surugamides A and F from N-acetylcysteamine (SNAC) thioester mimics of the cognate terminal biosynthetic intermediates. Using bioinformatics, we estimate that ∼12 % of filamentous Actinobacteria harbor an NRPS system lacking an embedded TE and instead use a trans-acting β-lactamase release strategy. This expands the paradigmatic understanding of how non-ribosomal peptides are released from the terminal NRPS module and adds a new dimension to the synthetic biology toolkit, potentially useful in the search for novel antibiotics.

Microbially-derived antimicrobial compounds are a rich source of clinical antibiotic leads. However, discovery rates have declined over the past 40 years due, in part, to high rediscovery rates of known compounds from traditional soil-based screening approaches. In this study, an ancient hot-spring water source was tested for the presence of antimicrobial-producing bacteria using culture techniques which led to isolation of two organisms capable of inhibiting the growth of multiple bacterial species. Oxford Nanopore whole genome sequencing was used to identify these two isolates as being in one of two genera; Streptomyces and Paenibacillus. Bioinformatic analysis revealed both isolates to have multiple novel secondary metabolite gene clusters. Investigations of the Streptomyces sp. by natural product chemistry techniques showed the organism to produce multiple antimicrobial compounds, these were effective methicillin-resistant Staphylococcus aureus and vancomycin-resistant Enterococcus. This study underlines the value of investigating non-traditional habitats in the search for novel antibiotic-producing organisms.

Herein we report the identification and characterisation of two linear antimicrobial peptides (AMPs), HG2 and HG4, with activity against a wide range of multi-drug resistant (MDR) bacteria, especially methicillin resistant Staphylococcus aureus (MRSA) strains, a highly problematic group of Gram-positive bacteria in the hospital and community environment. To identify the novel AMPs presented here, we employed the classifier model design, a feature extraction method using molecular descriptors for amino acids for the analysis, visualization, and interpretation of AMP activitiesfrom a rumen metagenomic dataset. This allowed for the in silicodiscrimination of active and inactive peptides in order to define a small number of promising novel lead AMP test candidates for chemical synthesis and experimental evaluation. In vitrodata suggest that the chosen AMPs are fast acting, show strong biofilm inhibition and dispersal activity and are efficacious in an in vivomodel of MRSA USA 300 infection, whilst showing little toxicity to human erythrocytes and human primary cell lines ex vivo. Observations from biophysical AMP-lipid-interactions and electron microscopy suggest that the newly identified peptides interact with the cell membrane and may be involved in the inhibition of other cellular processes. Amphiphilic conformations associated with membrane disruption are also observed in 3D molecular modelling of the peptides. HG2 and HG4 both preferentially bind to MRSA total lipids rather than with human cell lipids indicating that HG4 may form superior templates for safer therapeutic candidates for MDR bacterial infections.

Streptomyces species are a major source of antibiotics but are often grown under restrictive conditions that limit biosynthetic gene expression. As a result, the vast majority of secondary metabolites within a single species remain unexpressed in the lab along with the huge variety in chemical structures and bioactivities. The overarching aim of this project was to identify conditions that produce novel antibiotics, specifically against Gram-negative pathogens. The culture collection NCIMB contains hundreds of Streptomyces isolated from around the world. Strains were selected from the collection guided by preliminary bioactivity studies and available literature. These were then grown in an extensive variety of conditions designed to stimulate production of a wide variety of secondary metabolites, detected by UPLC-MS and analysed using the freely available metabolomic tools MZmine, MetaboAnalyst, and GNPS. Conditions included various carbon and nitrogen sources, temperatures, stresses, epigenetic inhibitors, and other microbes. Compounds active against the Gram negative multidrug resistant pathogen Acinetobacter baumannii were detected in the scaleup culture supernatant fractions. Metabolite identification through GNPS did not detect any previously discovered compounds active against A. baumannii, indicating a potentially novel antibiotic against one of the WHO’s priority pathogens.

Streptomyces species are well known for harbouring a large number of extracytoplasmic function (ECF) RNA polymerase sigma (σ) factors; nearly all of regulated genes required for morphological differentiation and/or response to environmental stress. The activity of ECF σ factors is typically modulated by a cognate anti-σ factor protein co-encoded at the same locus. In previous work, we identified σAntA as a cluster-situated regulator of starter unit biosynthesis in the production of antimycin, an anticancer compound. Unlike a canonical ECF σ factor, whose activity is controlled by a cognate anti-σ factor, σAntA is an orphan, which raises intriguing questions about how its activity is controlled. Inspection of the σAntA amino acid sequence revealed a carboxyterminal di-alanine, which is the only motif of the Clp-protease recognition signal obligately required for proteolysis. Here, we show by immunoblotting that the abundance of 3xFLAG-σAntA in vivo is enhanced by alteration of the C-terminal Ala-Ala to Asp-Asp and that abundance of both variants is elevated in the absence of genes encoding the Clp-protease and its regulatory subunits ClpX and ClpA. We also show that (His)6-SUMO-σAntA, but not a variant lacking the C-terminal di-alanine motif, is degraded by the ClpXP protease in vitro. These data unambiguously establish direct proteolysis as an alternative control strategy for ECF RNA polymerase σ factors and expands the paradigmatic understanding microbial signal transduction regulation.

Non-tuberculous mycobacteria (NTMs), such as Mycobacterium abscessus and chimaera, can cause high mortality and morbidity amongst patients who are immunocompromised or have chronic lung diseases, such as cystic fibrosis. Mycobacterial biofilms can form in the alveolar walls of such patients following inhalation from environmental reservoirs. Recently, pathogenic strains have been isolated from shower heads and hospital air conditioning units. Biofilms are notoriously difficult to eradicate and are associated with the development of increased antimicrobial resistance (AMR). Treatment for M. abscessus and chimaera infections often requires 2–3 antibiotics over 2 years. The question we want to address is whether the increased AMR and treatment time in NTM biofilm formation is due to lack of antibiotic penetration, resulting in non-therapeutic and AMR-generative levels, or the development of phenotypic and/or genetic resistance. In this project we will use Nano-SIMS (nano-scale secondary ion mass spectrometry) to measure penetration of the antibiotic bedaquiline (BDQ), used to treat NTM infections, into individual cells of NTM biofilms (M. abscessus and M. chimaera). Nano-SIMS maps the relative abundance of different ions down to the nano-scale and can be used to measure in profile through the biofilm. In addition, the minimum inhibitory concentration (MIC) and minimum duration for killing (MDK) of BDK will be measured to determine antibiotic susceptibility in biofilms and a planktonic growth control. Understanding the AMR generation and prolonged treatment in NTM biofilms could lead to improved mortality and morbidity. The development of novel treatment strategies could enhance treatment efficacy, reduce treatment duration and AMR development.

Bacterial antibiotic resistance is widely regarded to be one of the most pressing threats facing humanity. Finding new antibiotics is a vital research area and can now be supported by a vast reservoir of readily available ‘omic data on the back of the explosion of low cost sequencing technologies. Antimicrobial Peptides (AMPs): endogenous peptides that provide a fast and effective means of defence against pathogens as part of the innate immune response. The detection of AMPs in metagenomic data is a tantalising low-hanging fruit for computational biologists. Large reservoirs of existing sequences exist and are well annotated and understood. Post-computational wet-lab work is relatively cheap with spot synthesis of peptides cheaply available from a wide array of third party companies. A well organised screening program can screen 100 s of prospects a day against model bacterial organisms to test for activity and is one of the few areas of biological science that can scale to meet the data output from computational prediction toolkits. AMPLY, an in-house tool designed at Aberystwyth University, supported by Life Science Wales and working in collaboration with Tika Diagnostics at St. George’s Hospital (London) and Queen’s University (Belfast) is part of a next wave of computational drug discovery platforms and is already uncovering a treasure trove of novel AMPs in diverse microbial environments. However, AMPLY’s abilities extend beyond the analysis of ‘omic data alone and its predictive modelling has extracted a therapeutically viable novel AMP built from a string of ancient Welsh poetry.

Klebsiella pneumoniae (Kp) is an infamous cause of multi-drug resistant (MDR) healthcare-associated infections and several MDR clones are globally distributed. A small number of drug-susceptible clones have also become globally distributed, causing severe community-acquired infections. These ‘hypervirulent’ clones are distinguished by expression of highly serum-resistant K1/K2 capsules, plus high prevalence of acquired virulence determinants. While hypervirulence and drug resistance are usually mutually exclusive, there are now increasing reports of convergent strains that are both highly virulent and MDR – a potentially disastrous combination. To better understand the risks of MDR-virulence convergence, we leveraged a collection of >2200 Kp genomes to identify 28 common clones (n≥10 genomes each), and performed a genomic evolutionary comparison. Eight MDR and 6 hypervirulent clones were identified by acquired resistance and virulence gene prevalence. Chromosomal recombination, capsule locus diversity, pan-genome, plasmid and phage dynamics were compared. MDR clones were highly diverse, with frequent chromosomal recombination generating extensive capsule locus diversity. Additional pan-genome diversity was driven by frequent acquisition/loss of both plasmids and phage. In contrast, chromosomal recombination was rare in the hypervirulent clones, which were each associated with only a single capsule locus and showed significant reduction in pan-genome diversity, largely driven by a reduction in plasmid diversity. These data suggest that hypervirulent clones are subject to some sort of constraint for horizontal gene-transfer. Hence we predict MDR clones pose the greatest risk for MDR-virulence convergence because they are more likely to acquire virulence genes than hypervirulent clones are to acquire resistance genes.

Enteropathogenic E. coli (EPEC) were the first E. coli strains linked to human disease (1945), and pose a serious health threat. The burden for public health changed over the past century as it did for other diarrheal diseases; whilst it is still endemic in large parts of South America, cases in Europe are commonly associated with recent travel. Despite its long-standing importance for global health and its significant impact on child mortality in Low- and Middle Income Countries (LMICs), we still have very limited understanding of what defines an EPEC beyond the diagnostic LEE island, or indeed if there are any shared characteristics, as most in-depth analyses of their pathogenic determinants are confined to few model strains. We present a study analysing ∼1300 whole-genome sequences combining published and newly sequenced datasets of EPEC and non-EPEC strains to map the evolutionary history and molecular determinants. Importantly, we have expanded the published data with whole-genome sequences of 300 historical and contemporary EPEC strains from a large collection of clinical isolates, mainly from Brazil and England, which enables us to compare the epidemiology in a high-income with a LMIC over an extended time frame. We furthermore present a molecular definition of EPEC, including mapping of phage islands and de-novo prediction of effectors in this large-scale dataset, as well as investigating the patterns of adhesins and other secretion systems, thus characterising different EPEC lineages which have emerged numerous times during the evolution of E. coli.

Throughout evolution, evolving objects (domains, genes, operons etc.) have continuously combined, forming new proteins, gene clusters, and genomes. Horizontal gene transfer, particularly among prokaryotes, has facilitated this combinatorial process. Thus, evolving objects that interact positively or synergistically with each other are expected to co-occur more often than by chance; conversely, evolving objects may avoid co-occurrence, indicating an antagonistic or redundant functionality between objects. In this work, we use methods adapted from graph theory to understand patterns of co-occurrence and exclusion in prokaryotes. We have implemented multi-level graph models in which each node (vertex) is a gene or species connected by an edge (relationship) to another node to display these coincidence relationships. Our method incorporates the phylogenetic distribution and synthenic distances of these genes, and we demonstrate how these concepts can be used to identify conserved clusters of vertical and horizontally inherited units of selection. We apply these multi-level graph models to a variety of datasets including prokaryotic pangenomes, and metagenomic sequencing datasets from human-associated microbial communities. We find evidence for genes that significantly co-occur with each other within each of these datasets; these genetic clusters include objects from characterized biological pathways but also include genes with unknown functions. Further, we identify genes that exclude each other, indicating evolving objects with antagonistic or redundant biological functions. This work represents a different approach to understanding the evolution of prokaryotes and allows us to draw novel hypotheses as to the potential role of these genetic clusters in prokaryote biology.

Pangenome analysis can help reveal functional and physiological differences between organisms by linking genes present in their genomes with phenotypic characteristics. While this works well for closely-related organisms and large numbers of isolates, analysis of species with a large pangenome may be problematic as optimal sample size increases with intraspecies diversity. Organisms such as E. coli exhibit high intraspecies diversity, resulting in an exceptionally large pangenome, pushing the limits of what constitutes a species. However, the phylogeny of the mobile elements can differ strongly from the phylogeny of the rest of the genome, reflecting how rapidly these elements can be gained or lost. We suggest when studying adaptations on a short genetic time-scale (ie, within species or strain), the mobile regions of the genome may be more informative than the more stable parts of the genome. To this end, we developed a pipeline, Horizontally Acquired Partial Pangenome of Inserted Elements (or ‘happie’), which allows researchers to study the mobile pangenome. Thus, happie could be used as a proxy to detect genes associated with a particular trait amidst the noise of a diverse genetic background. We aim to use happie to compare a panel of known soil-persistent E. coli isolates and a curated panel of non-soil-persistent isolates. This will allow us to determine whether specific mobile elements (or genes carried on those elements) are associated with soil persistence, which would indicate that the cost of harbouring the mobile element was less than the advantage conferred.

Systemic salmonellosis encompasses typhoid and paratyphoid fever, and invasive non-typhoidal salmonellosis, with high mortality and morbidity amongst children and the immunocompromised in low-resource settings. Immunisation efforts remain hampered by the unavailability of safe vaccines with cross-protectivity against causative Salmonella strains. Characterisation of the within-host Salmonella dynamics in the naïve and immunised host can elucidate the mechanisms by which different vaccine types exert their protective effect, and help in vaccine selection and design. Experimental data tracking the changes in bacterial population composition in the different tissues of the host at different timepoints can be coupled with mechanistic mathematical models to estimate the parameters governing the processes of bacterial replication, killing and inter-organ migration. Using a recently described minimisation-divergence estimation approach, we extend a three-compartmental mechanistic model and re-analyse existing datasets to better characterise the bacterial migratory processes between the blood, liver and spleen in the early stages of infection, and the overall Salmonella dynamics in the later phases in the naïve host. We apply the same model to published data from mice immunised with either a live-attenuated or killed whole-cell vaccine to identify their in vivo differential impacts on Salmonella migration, replication and death. Finally, we identify alternative experimental designs to improve the statistical qualities of the mathematical model and allow better inference of parameters governing the unobserved processes of bacterial dynamics.

Infections account for >20 % of all deaths worldwide, and are particularly problematic in low-middle income countries (LMICs), with bacterial infections killing approximately 5 million people annually. The crisis of antimicrobial resistance means our options for controlling infections are narrowing. Vaccines are a cost-effective approach to prevent infectious disease. However, there are many bacterial infections against which we lack any licensed vaccine. The BactiVac Network (birmingham.ac.uk/bactivac), was launched in August 2017 and is led by Profs Cal MacLennan and Adam Cunningham. BactiVac is a global bacterial vaccinology network established to accelerate the development of vaccines against bacterial infections, particularly those relevant to LMICs. BactiVac brings together academic, industrial and other partners involved in vaccine research against human and animal bacterial infections from the UK and LMICs. The Network fosters partnership and provides catalyst pump-priming project and training funding to encourage cross-collaboration across our membership. Full details are available on our website. BactiVac has >600 members across >60 countries with 39 % based in LMICs and 10 % in industry. Membership is free – join us at bit.ly/apply BactiVac Benefits of membership include: Access to catalyst funding – third round for pump-priming projects closes 5 May 2019, open call for training awards Invitation to subsidised Annual Network Meetings–next meeting will be held in March 2020 Access to Members’ Directory – to develop collaborations and access distinct expertise. We encourage you to become a member, be involved in this initiative, and to be part of its growing success now and in the future.

Reverse vaccinology 2.0 (RV 2.0), in which the cloning and recombinant expression of antigen-specific antibodies is followed by determination of their functional activity, is a valuable approach that can unravel novel vaccine antigens, or reinforce the vaccine candidacy of already known antigens. Invasive meningococcal disease (IMD) remains a serious source of concern even with the availability of vaccines. Incomplete strain coverage is a limitation of current vaccines, hence efforts to identify candidate antigens that will compose supplementary or replacement vaccines are necessitated. In this proof-of-principle study, we sought to assess the applicability of RV 2.0 to anti-meningococcal vaccine antigen discovery. Antibody-secreting cells (ASCs) obtained from a patient convalescing from serogroup B (MenB) IMD, were isolated and sorted singly using FACS. The specificity and functionality of each antibody produced by individual ASCs were assessed in ELISA and bactericidal assays, respectively. Eight cross-reactive anti-meningococcal antibodies were successfully cloned; three of these mediated complement-dependent killing of antigenically-heterologous MenB strains. Western blot data shows binding of these three bactericidal antibodies to a ∼35 kDa antigen. None of the three bactericidal antibodies were reactive with a target in current vaccine formulations strongly suggesting that the ∼35 kDa antigen does not compose available vaccines. Unequivocal determination of the identity of the ∼35 kDa antigen is ongoing. Given the need for antigens that would compose improved or novel anti-meningococcal vaccines, this study shows that the RV 2.0 approach has the potential to be a powerful tool in the identification of functionally-immunogenic anti-meningococcal antigens.

Extra-intestinal pathogenic E. coli (ExPEC) is one of the leading causes of bacteraemia and urinary tract infections (UTI) worldwide. At present, there is no licensed vaccine available and with the increasing incidence of multidrug resistance (MDR) the demand for developing alternatives to antibiotics is paramount. We have developed in vitro functional antibody assays to fast track screening of patient convalescent sera from ExPEC infections to identify potential vaccine antigens. A functional immunological assay will also facilitate the assessment of vaccine candidates against the many disease-causing serotypes circulating in the population. Outer membrane vesicles (OMVs) purified from a mutant (ΔlpxM) MDR O25 (ST131) strain were used to immunise mice to raise antiserum against the bacterial strain. Immunised mice were protected in a mouse model of E. coli bacteraemia. The antiserum from protected animals was used to develop an antibody-mediated bacterial killing assay and a complement deposition assay (CDA). The antiserum mediated killing with human IgG and IgM-depleted plasma as the complement source, and also showed deposition of C3b and C5b-9 onto the bacterial surface determined by flow cytometry. This may be a useful screening tool for mining for new antigens and in assessing the antisera raised against the candidates for cross serotype protection.

Campylobacter is the leading cause of food-borne gastroenteritis worldwide and the most common cause of infection in humans have been linked to the consumption of contaminated poultry meat. Previous poultry vaccines have provided short term reduction in Campylobacter intestinal load in chickens but have limited commercial efficacy because of high diversity and rapid evolution of strains. Genome-wide association study (GWAS) of isolates from humans and chicken have identified genes associated with survival of Campylobacter through the poultry processing chain. These genes represent targets for vaccine design that would be selectively neutral within the chicken gut and therefore could be sustained in the population. We designed an autogenous vaccine based on isolates with survival-associated genes and monitored the efficacy in the chicken and through processing. First, we sampled for Campylobacter across five broiler farms feeding into one abattoir in Norfolk and characterized genomic diversity using next generation sequencing (NGS), including the identification of survival genes. These data were then used in a predictive model to identify the minimum number of strains to be used in vaccine design based on known immunogenicity of the strains. Second, the vaccine was developed and given at two time-points to a whole farm of breeder chickens. Third, vaccine efficacy was monitored at three time points post-vaccination in vaccinated breeders and their progeny. The cfu/g reduction in the ceacum and neck skin were determined. This study demonstrates the potential for NGS in autogenous vaccine design to reduce harmful strains of bacteria that are carried commensally in livestock.

This study determined the in-ovo antiviral effect of crude Methanolic leaf extract of Cymbopogon citratus on Newcastle disease virus. Cold extraction was carried out using analytical grade methanol. Phytochemical screening of the crude extract was carried out using standard procedures. Antiviral assay was carried out in nine-day old specific pathogen free embryonated hens’ eggs in three designs made of five eggs per group (Virus with extract, virus only and un-inoculated groups) with concentrations ranging from 12.5 to 100 mg ml−1. Egg toxicity of the extract was determined for concentrations of 12.5, 25, 50, 100, 200, 300 and 400 mg ml−1. Inoculated eggs were incubated at 37 °C and observed daily for 96 h for embryo survival and mortality. Spot haemagglutination was carried out on bacteria-free allantoic fluid from the embryonated eggs to detect the presence of the virus. Phytochemical assay revealed the presence of saponins, flavonoids, steroids, terpenes, phlebotannins and terpenoids. Mild toxicity was observed at concentrations of 100 mg ml−1 and above. There was no haemagglutination of fluid from the eggs inoculated with a combination of Virus and extract at concentrations of 50 and 100 mg ml−1. The current findings demonstrated that leaf extract of Cymbopogon citratus has potential medicinal value as well as antiviral activity against Newcastle disease virus in-vivo. The specific mechanism of action remains to be studied to further elucidate on its potential as a therapeutic product for the treatment of Newcastle Disease.

This was a cross sectional and observational study, conducted in the Department of Microbiology, Sylhet MAG Osmani Medical College, Sylhet. The study period was from July 2010 to June 2011 in objective to explore the Varicella-Zoster virus immune status (VZV-IgG) in pregnant mothers and neonates up to six months post birth. For this purpose, 60 pregnant women, 60 new born babies and 60 infants aged six months were selected. The mean age of the pregnant woman was 28.8 (SD±4.7) years. The sex of new born babies and six months aged infants was identical [29 (48.3 %) male vs 31 (51.7 %) male; P=0.715]. Seroprevalence rate of IgG antibody of Varicella-Zoster Virus in Pregnant, mothers 81.7%, in new born babies 78.3 % and in infants aged six months were 10.0 %. The seropositivity of VZV IgG level in both pregnant mothers and newborn babies were almost similar but the infant aged six months were significantly lower than that of new born babies (p This showed that a significant proportion of Bangladeshi pregnant mother is susceptible to varicella and infant aged six months is highly susceptible to varicella. Any vaccination strategy must have to take into account these epidemiological variability of the country.

The lyssavirus genus is a diverse group of viruses all capable of causing an invariably fatal disease known as rabies, most commonly caused by the prototype species rabies virus (RABV). Alongside RABV the lyssavirus genus currently contains 15 other viruses capable of causing rabies. These viruses are broadly categorised into phylogroups according to the predicted level of vaccine protection, with protection from current vaccines and therapeutics afforded against phylogroup I but not II or III. Current evidence suggests that for a protective neutralising antibody response against RABV a neutralising antibody titre of 0.5 IU ml−1 is sufficient. This arbitrary value has been developed and promoted as a serological cut-off based on the reactivity of defined sera with a standardised dose of RABV. Studies using cross protection assays, have suggested that for protection against more divergent members of the genus, even those in phylogroup I, 10-fold or greater than the 0.5 IU ml−1 antibody titres are required. The continued discovery of novel lyssaviruses globally warrants an in-depth assessment of the protective titre required to protect against all the lyssaviruses to inform occupationally high-risk groups (e.g., scientists, bat workers and speleologists). Based on live virus neutralization assays, a minimum of 7 distinct lyssavirus glycoprotein antigens would have to be included in any pan-lyssavirus vaccine. Certainly, representative immunogens from all lyssavirus species characterized in phylogroups II and III are required to stimulate a pan-lyssavirus response.

Avian influenza is highly contagious poultry disease and the mortality can reach 100 %, leading to huge economic burden and threat to food security. Vaccination is the most effective strategy for prevention and control of influenza. Recombinant vector vaccines are effective promising vaccines capable of immunizing against multiple pathogens. Traditional methods for constructing recombinant vaccines involve homologous recombination or bacterial artificial chromosomes but these methods can be time-consuming and labour-intensive. The clustered regularly interspaced palindromic repeats (CRISPR)/Cas9 system is a recently developed gene editing technology which has proven beneficial to gene modification and offers an alternative for constructing recombinant vaccines. There are mainly two methods used for gene insertion; error-prone non-homologous end joining (NHEJ) and the high-fidelity homology-directed repair (HDR) pathway. Owing to its high fidelity, most studies focus on using HDR for vaccine development but NHEJ offers some attractive advantages through its high efficiency. In our study, both HDR and NHEJ dependent CRISPR/Cas9 systems were explored for the rapid generation of recombinant influenza vaccines using duck enteritis virus and herpesvirus of turkeys as vectors.

Arbidol (arb, umifenovir) is used clinically in several countries as an anti-influenza virus drug. We have previously shown that arb inhibits many viruses including hepatitis C virus, Ebola and Zika, and that the primary mode of action appears to be via inhibition of virus entry and/or fusion of viral membranes with intracellular endosomal membranes. We have also shown that arb is a good inhibitor of (non-enveloped) poliovirus types 1 and 3. Here, we evaluate the antiviral potential of arb against another picornavirus, foot-and-mouth disease virus (FMDV), an important veterinary pathogen. Sub-cytotoxic concentrations of arb inhibited the replication of FMDV replicon RNA. arb inhibition of FMDV RNA replication was not a result of generalised inhibition of uptake of cargo (e.g. plasmid DNA or RNA), nor did arb inhibit FMDV replication when added at 4 h post-transfection of FMDV replicon RNA. FMDV replication was blocked by the replication inhibitor guanidium hydrochloride (GuHCl). Upon GuHCl removal, FMDV replication was restored, and arb suppressed this recovery of virus replication. For other picornaviruses, recovery of virus replication upon GuHCl removal has been shown to require translation. However, arb did not suppress cap- or internal ribosome entry site (IRES)-dependent translation. arb also inhibited infectious equine rhinitis associated virus (ERAV), a close relative of FMDV. Testing of arb against infectious FMDV is in progress. Collectively, the data demonstrate that arb inhibits certain picornaviruses by a mechanism that appears to be independent of effects on virus entry but involves inhibition of genome replication.

Nipah virus (NiV) is a zoonotic paramyxovirus that causes severe and often fatal respiratory and neurological disease in humans. Since 1998, NiV outbreaks have occurred in Malaysia, Bangladesh and India. NiV poses a significant pandemic threat due to its broad host range and the widespread distribution of its natural host, the Pteropus fruit bat. Despite this, there are currently no licenced therapeutics or vaccines for use in either humans or livestock. Aiming to develop a safe and inexpensive vaccine to protect livestock in future outbreaks and to prevent onward transmission to humans, we are evaluating the immunogenicity and efficacy of three candidate vaccines in pigs. These vaccines are based on the NiV G or F surface glycoproteins, proteins which are essential for mediating virus-cell or cell-cell entry. The candidate vaccines include: (i) a recombinant soluble NiV G protein subunit, (ii) a recombinant molecular clamp stabilised NiV F protein subunit and (iii) a replication deficient, adenoviral vectored NiV G protein. We have developed low biocontainment assays to characterise the antibody responses to NiV and to aid identification of immune correlates of protection through quantification of both antigen-specific and neutralising antibody responses to our vaccine candidates. This includes anti-NiV F/G indirect ELISAs, a microneutralisation test using pseudotyped particles, and a microfusion inhibition test using a quantifiable cell-cell fusion assay. Using these assays we have demonstrated that all three of our novel vaccines are immunogenic in pigs and capable of generating a robust antibody response, with evidence for neutralising antibodies.

Constituent proteins that reside within promyelocytic leukaemia nuclear bodies (PML-NBs) are known to play an important role in cellular restriction of herpesvirus gene expression as a component of the intrinsic antiviral immune response to DNA virus infection. However, the precise mechanism(s) of PML-NB genome detection, entrapment, and silencing remain unknown. We have recently reported a sensitive method of viral genome detection that enables the visualisation of infecting EdC (5-Ethynyl-2’-deoxycytidine) labelled HSV-1 genomes (HSV-1EdC). We report that ultraviolet light (UV) irradiation of HSV-1EdC virions inhibits the nuclear recognition of viral DNA (vDNA) by core PML-NB constituent proteins (PML, Sp100, and Daxx). This impairment of recruitment was independent of UV source, but dependent on dose and time course of UV irradiation that correlated with a loss in viral genome decompaction upon nuclear entry. Moreover, UV treatment promoted premature uncoating of viral genomes within the cytoplasm of infected cells in a dose-dependent manner. Our data highlights the importance of genome decompaction in PML-NB sensing of infecting viral genomes. Additionally, it uncovers a previously undescribed mechanism of action for the induction of innate immune defences during herpesvirus infection that have historically utilised UV inactivation to promote the activation of cellular pattern recognition receptors (PPRs) and induction of interferon stimulated gene (ISG) expression.

The Flavivirus small integral membrane (M) protein resides within mature infectious particles yet, unlike the envelope (E) protein which mediates membrane fusion upon encountering low pH within the acidifying endosome, the function of M within this context is unknown. We are investigating whether Zika virus (ZIKV) M protein exhibits channel activity, acting as a viroporin playing a role mediating virus entry and uncoating. Importantly, ZIKV entry was blocked in a dose-dependent fashion by the prototypic channel blocker rimantadine and the drug also prevented virus spread in mouse models of ZIKV infection. Molecular dynamics simulations supported that M protein is able to oligomerise into a hexameric viroporin channel, opening of which is promoted in an acidified environment via protonation of a conserved histidine residue. Rimantadine is predicted to bind onto this structure in silico at a lumenal binding site, against which we are currently designing improved small molecule inhibitors that could form the basis of novel M protein targeted drug discovery. Importantly, drugs targeting M might either prevent or reduce the severity of ZIKV infections, including those crossing the placenta, and could also be translated for use against other Flaviviruses.

Influenza A Virus (IAV) is a zoonotic pathogen which causes seasonal epidemics worldwide, resulting in significant morbidity and mortality. Due to rapid virus evolution, available vaccines/antiviral drugs must be re-formulated and re-administered in most years. There is therefore an unmet need for a ‘universal’ vaccine to provide long-lasting, adaptive immunity to multiple strains of IAV. The aim of this study was to assess outer membrane vesicles (OMVs) produced by Bacteroides thetaiotaomicron (Bt) containing IAV antigens as a candidate universal mucosal vaccine. First, using intranasally-delivered fluorescently-labelled OMVs and FACS we demonstrated good uptake of OMVs by dendritic cells in both secondary and tertiary lymphoid tissue, showing OMVs are efficiently trafficked from the mucosa to the lymphatic system. Immunohistology confirmed this and showed expansion of lymphoid tissues and lymphoid follicles after OMV delivery. The stem/stalk of IAV HA is less variable and will generate antibodies that are cross-protective. Thus, we produced OMVs (termed H5F) that contained the stalk region of HA from strain A/VietNam/1203/04 (H5N1). We inoculated mice intranasally with H5F expressing OMVs or wild type OMVs and assessed antibody production and protection from challenge using a lethal dose of a different strain of influenza (PR8; H1N1). OMV-H5F-inoculted mice produced IgG and IgA in the respiratory tract that recognised both H5 HA and H1 HA. OMV-H5F inoculated mice were also protected from lethal challenge and supported lower virus titres. Our results indicate a strong potential for this approach to generate a universal IAV vaccine.

The vast majority of infectious agents enter the body via mucosal sites yet there are very few licensed mucosal vaccines able to generate protective immunity at the sites of pathogen entry. A major obstacle in developing mucosal vaccine is delivering biologically active vaccine antigens (Ag) to mucosa-associated lymphoid tissues to prime protective immune responses. To address these issues we have developed a drug delivery technology platform to deliver intact antigens to the respiratory and gastrointestinal tract using outer membrane vesicles (OMV) naturally produced by Bacteroides thetaiotaomicron (Bt), a non-pathogenic human commensal gut bacterium. We have developed the capability of engineering Bt to express antigens of interest in their OMVs which we have shown are stable for long periods of time across a wide temperatures range. They also have inherent adjuvanticity as shown by the ability of native OMVs to elicit the formation of organised lymphoid follicles comprising dendritic cells and T and B cells in both the upper and lower respiratory tract after intranasal administration. The intransal administration of Bt OMVs expressing the pre-fusion headless hemagglutinin mini-stem protein of influenza type A virus (IAV) subtype, H5N1, induced high titre antigen-specific antibodies in the respiratory mucosa (IgA) and serum (IgG) that conferred heterotypic protection to infection by a H1N1 IAV. Collectively, our data demonstrates the feasibility of using Bt OMVs in mucosal vaccine formulations for respiratory infections.

Chikungunya (CHIKV), an arbovirus that belongs to the Alphavirus genus of the Togaviridae family, causes a disease characterized by acute onset of fever accompanied by arthralgia. CHIKV also has been associated with cases of meningoencephalitis (primarily in neonates), Guillain–Barré syndrome and hemorrhagic disease. The clinical similarities, cross-reactivity and cocirculation of arboviruses in Brazil have complicated their differentiation, highlighting the need for new diagnostic tools. A serologic test can be useful for acute detections as well as for surveillance and epidemiological studies. Point of care tests currently available to detect CHIKV infections have been associated with low accuracy. The aim of this work was to design a synthetic gene and generate a recombinant protein to be used as antigens in diagnostic assays. Computational methods were used to predict its structure and antigenic potential. To confirm predictions, the gene coding for the recombinant CHIKV protein (rCHIKV) was synthetized and the protein was purified by affinity chromatography. Antigenicity of the protein was initially confirmed by western-blot using sera from CHIKV infected mice. Additionally, the seroreactivity of r-CHIKV protein was evaluated using a panel of sera samples from human patients, CHIKV seropositive or not, by indirect IgG ELISA. The r-CHIKV protein showed sensitivity of 95 % and specificity of 96 %. No cross-reactivity was found against sera of Zika and Dengue positive patients. These results indicate that this proteins maybe useful antigen to detect CHIKV infections in ELISA assays. Our next step includes the development of a rapid test kit.

Identifying transient low-level hepatitis B (HBV) surface antigen (sAg) post-vaccination is a well-established phenomenon in haemodialysis patients. These results create a clinical conundrum; should patients be deemed potentially infectious, requiring further testing and heightened infection control precautions, or should negligible risk be assumed? National guidance covering this scenario is lacking; therefore we conducted a survey of the clinical practice of UK virology departments. Thirty-six laboratories were contacted and 17 responses received (47 %). Eleven responders had observed sAg positivity post-vaccination, for a maximum duration of 3 weeks. Clinical management was highly variable. Only 3 hospitals had specific written policies, with others following an internal consensus or a standard protocol for sAg positive patients without recent vaccine. Further testing consisted of HBV serology (3/11), HBV PCR (1/11) or PCR and serology (7/11). Three departments were highly risk averse, recommending universal isolation with dedicated dialysis machines. Two departments did not advise additional precautions and 5 based decisions on risk factors e.g. foreign travel. Heightened precautions were stopped on the basis of negative serology (1/11), PCR (5/11) or serology and PCR (2/11). To summarise, vaccine-derived sAg positivity is commonly encountered on haemodialysis units. Recommendations on the interval between vaccination and testing would alleviate the issue, with this study suggesting a minimum of 3 weeks. Infection control precautions varied considerably. In most cases, patients are risk assessed for isolation as a minimum; however this may depend on local expertise and the availability of isolation facilities. It is clear that national guidance would standardise patient care.

Molecular diagnostics have superseded cell culture and neutralisation as the gold standard for enterovirus (EV) diagnosis and serotyping in resource rich countries. EV positive isolates from Nottingham University Hospitals NHS Trust are referred to the Enteric Virus Unit (EVU) for typing; performed through amplification and sequencing of viral capsid protein-1 (VP1). An audit of local EVU typing data (2013–2017 showed a significant (P<0.005) increase in untypable EV strains isolated from clinical specimens. Following a literature review of published EV typing assays and in silico analyses of 2201 EV genomes, an end point RT-snPCR VP1-typing assay was derived from a widely cited protocol and successfully employed to produce a simple, reliable and cost-effective typing assay. Twenty-three previously untypable isolates were identified, with phylogenetic analyses of the VP1 region showing broad distribution across EV-A and B species. This assay detected serotypes across EV species A-D, including EV-D68, responsible for severe neurological and respiratory disease. Following an increase in severe EV infections, including six cases of suspected acute flaccid myelitis, this RT-snPCR was employed locally as a rapid typing assay. In combination with 5’UTR PCR and an EV-D68 specific PCR, designed to amplify the complete VP1 region, EV-D68 was detected in respiratory and neurological specimens, including the CSF of a patient with brainstem encephalitis.

Nottingham University Hospitals Trust receives circa 13 000 samples for diagnosis of respiratory and neurological viruses per annum, however positive results are achieved in approximately 50 % of respiratory and 20 % of neurological investigations. We therefore aimed to retrospectively extend the diagnostic spectrum for these samples by applying a battery of degenerate PCR assays to surplus diagnostic nucleic acids.218 previously negative respiratory specimens collected in 2016 from children under 5 years old, identified positivity of 11 % for Human Bocavirus, 5 % for Human Enteroviruses (including 3 cases of Enterovirus D68), 4 % for Human Coronaviruses and one individual positive for Trichodysplasia Spinulosa Polyomavirus. Complementary investigation of 1730 previously negative specimens from children and adults with neurological symptoms yielded positive results for Hepatitis E, BK Polyomavirus and Astroviruses in addition to Entero- and Parechoviruses apparently missed by standard diagnostic assays. Our extensive archive further allowed us to investigate relatively rare viral infections in significant numbers. Therefore we also studied the genetic and clinical epidemiology of the human Rubulavirus pathogens Parainfluenza 2 and 4 in 121 and 237 patients respectively between 2013 and 2017. This indicated co-circulation of three clusters of Parainfluenza 4 in Nottingham with greater presentation of subtype 4b than 4a and 2 clades of Parainfluenza 2. 5 fatalities were recorded in the Parainfluenza 2 cohort. In summary, surplus nucleic acid from viral diagnostic laboratories represents a valuable resource for both service development and clinical research. Coronavirus, Bocavirus and Enterovirus testing have since been implemented in the routine diagnostic panel for respiratory investigations.

Herpes simplex virus type 1 (HSV-1) and 2 (HSV-2) infections may be life-threatening in immunocompromised patients. Antiviral treatment is available but resistance can develop. The gold standard for resistance testing is cell culture-based phenotyping which is slow (3–4 weeks) and has a high (50 %) failure rate. Faster (<10 working days) genotypic testing of viral thymidine kinase (TK) and DNA polymerase (DNApol) can be used with >95 % success rate. However, a reference database for interpreting drug susceptibility from genetic data is lacking. We developed an HSV genotypic test and genotype-to-phenotype drug resistance database. We evaluated the service using 325 clinical samples, previously characterised by phenotypic susceptibility testing, from 248 treatment-experienced patients. The median age was 42.5 years [IQR30.0–51.0] and 50.8 % (n=126) were female. Clinical details were as follows: 42.3 % (n=105) haemato-oncology patients; 12.1 % (n=30) HIV-infected; 4.4 % (n=11) unspecified immunosuppression; 2.4 % (n=6) congenital infection; 1.2 % (n=3) solid organ transplant; 37.5 % (n=93) unknown. HSV-1 was identified in 58.2 % (n=189) samples. Phenotypic testing identified resistance in 63.7 % (n=207) samples. Genotypic testing identified a resistance-associated mutation (RAM) in TK and/or DNApol in 200/207 samples, a positive percent agreement of 96.6 %, whereas a RAM was detected in 1/118 susceptible samples, a negative percent agreement of 99.2 %. Most RAM occurred in TK (n=195; 97.0 %) with few in DNApol (n=32; 15.9 %). In summary, through herPHEgen we have developed a robust HSV resistance testing service, providing clinicians with timely and accurate results. This will improve clinical decision-making, optimising treatment efficacy and minimising toxicity in immunocompromised patients with HSV.

Current guidelines indicate that baseline INI resistance testing is not required; however, increased prescribing may lead to increased resistance mutation prevalence. NGS allows the detection of minority variants (MVs) not be detected by conventional methods. Barts Health NHS Trust patient records were reviewed to identify INI naïve patients prescribed INI between March 2014 and March 2017. Baseline plasma samples were extracted for nucleic acid and amplified in a single RT-PCR. Libraries were prepared using the Nextera XT library preparation kit and sequencing performed on an Illumina Miseq. Sequencing reads were assembled into consensus sequences and resistances determined using the Stanford HIV resistance database (HIVdb). MVs were variants <20 % of the overall population. Of 248 patients prescribed INI, 128 patients were excluded from the study due to an undetectable viral load or lack of sample. Of the remaining 121 patients, integrase sequences were successfully obtained from 91. No major INI mutations were detected. Six patients were identified with an HIVdb INI score ≥10: E157Q (4 patients) and T97A (2 patients). Only 1 MV accessory mutations was observed (A128T). Of patients with resistance mutations, 5 received raltegravir- and 2 received dolutegravir-based regimes. No treatment failures were observed. Prevalence of baseline INI mutations increased in recent years: 0/12 (0 %) patients in 2014–2015, 2/32 (6.3 %) patients in 2015–2016 and 5/47 (10.6 %) patients in 2016–2017. Resistance MVs to INI were extremely rare. Although this study supports current baseline INI resistance testing guidelines, increasing mutation prevalence’s may require future revision of guidelines.

Routine clinical diagnosis of patients with a suspected viral infection involves screening with multiple assays, often limited to only a single genus or species. This approach, however, may fail to detect novel species, atypically presenting viruses (i.e a ‘respiratory’ virus causing neurological symptoms) and viruses that are imported from other countries; often due to clinicians focusing on the ‘likeliest’ candidates. High-throughput sequencing (HTS) allows for the identification viruses present within a sample, by sequencing all viral genomes present. Without the inherent bias of limiting screening targeted assays, these unusual viruses are more likely to be detected. Samples taken from patients with an illness of unknown etiology, were grouped into 5 pools; 2 Respiratory, 2 CSF and an EDTA blood pool. The CSF Pools each contained 200 samples, the Respiratory pools 100 and the EDTA blood pool 80. HTS libraries were created from each of these pools and an additional CSF sample from a single patient with encephalitis and were then sequenced using an Illumina HiSeq platform. Human Pegivirus was detected in both CSF Pools, the EDTA pool and a single respiratory pool. Picobirnavirus was detected in a respiratory pool. RT-PCR was used to screen individual samples compromising these pools. BK Polyomavirus and Mastadenovirus C were detected in the CSF of a patient who had presented with encephalitis. Coinfection of these viruses typically cause neurological symptoms only in immune-compromised patients, so this exemplifies the advantage of using HTS for the detection of atypically presenting viruses.

To date, models of human cytomegalovirus (HCMV) latency and reactivation have depended on the use of primary myeloid cells, which have limited availability, are difficult to culture and are challenging to genetically modify. We now show that induced pluripotent stem cells (iPSCs) derived from circulating late outgrowth endothelial progenitors (EPC) can be differentiated down the myeloid lineage, where HCMV latent carriage and reactivation is known to occur in vivo, and act as a model to allow the interrogation of viral and cellular factors involved in latency and reactivation of this persistent human pathogen. In contrast, monocytes generated from iPSCs derived from de-differentiated fibroblasts failed to support HCMV latent carriage. These iPSCs derived from EPCs may also be suitable for in depth genetic interrogation of other viruses which also infect cells of the myeloid lineage, such as HIV and Zika.

As an obligate intracellular parasite, human cytomegalovirus (HCMV) completely relies on host machinery to replicate. Understanding which host factors are required for virus replication contributes to our understanding of virus biology and cell biology, identification of potential targets for antiviral therapy. High-throughput small interfering RNA (siRNA) screens are a powerful approach to identify novel host-virus interactions. Conventional screens often use reporter genes as a proxy for virus replication, rather than measuring production of infectious virus. We developed a two-step siRNA screen that independently measured primary replication and virus production. Screening with a library targeting almost 7000 genes, we identified 37 genes involved in early stages of HCMV replication and 15 genes specifically involved in later aspects, such as late gene expression, assembly and egress. These include factors in ubiquitin-dependent protein degradation pathway, and components of the mediator complex. Furthermore, we showed that the induction of SIN3A, a transcriptional regulator that forms a repressor complex with histone deacetylase 1 and 2, is essential for late gene expression and virus production. This study demonstrates a powerful two-step high throughput approach which identifies key host factors underpinning HCMV replication and informs our understanding of how the virus interacts with its host.

Genome segmentation offers certain evolutionary benefits to a number of pathogenic RNA viruses, including rotaviruses and influenza viruses. However, as the number of RNA segments per virion increases, the task of a non-random selection of a full set of distinct genomic RNAs poses a formidable challenge to maintaining the integrity of segmented genomes. Recently we have identified sequence-specific inter-segment interactions between rotavirus (+)ssRNA genome segment precursors. We have shown that binding of the rotavirus-encoded non-structural protein NSP2 to viral ssRNAs results in the remodeling of RNA, which is conducive to formation of inter-segment contacts. These protein-RNA interactions result in the stabilisation of extended intermolecular RNA-RNA contacts, potentially underpinning transient inter-segment interactions prior to genome encapsidation and replication. Using this approach, we have identified a number of RNA-RNA interaction sites in the rotavirus genome, which are likely to be involved in genome segment assortment process. Having established the role of NSP2 in promoting inter-segment RNA-RNA contacts, we have developed multiplexed imaging tools for directvisualization of the RNA assortment process in rotavirus-infected cells by employing single-molecule RNA FISH. To unravel the mechanisms, by which NSP2 controls the formation of inter-molecular RNA helices, we have applied RNA structure probing methods that allowed us to monitor conformational rearrangements, which are prerequisite for theformation of the RNA assortment complex. Our findings open up unique avenues for understanding the challenges for further improvement of the recently developed fully plasmid-based reverse genetics systems for rotaviruses.

Infectious bronchitis virus (IBV), a gammacoronavirus, causes the economically important poultry disease, infectious bronchitis, resulting in reduced weight gain and egg quality. As observed for many viruses, during replication, IBV shuts off translation of host proteins, preventing synthesis of important products of the innate immunity, which are pivotal in fighting viral infection. This work investigates the role of stress granules in IBV translational control. Stress granules are membranes-less aggregations of stalled translation initiation complexes comprising translation initiation factors, 40S ribosome and RNA binding proteins. These structures serve as sites of storage and sequestration of translational machinery and cellular mRNA while simultaneously enabling intracellular signalling and antiviral responses. It is shown here by immunofluorescence that IBV induces stress granules in only a proportion of infected cells. These stress granules occur late in the virus life cycle and appear canonical, containing multiple stress granule markers and showing mRNA exchange with ribosomes. In addition, stress granule markers are not diverted to sites of virus replication, as seen during replication of some other viruses. Interestingly, IBV infection results in resistance to chemicals that induce stress granules via eukaryotic initiation factor 2α (eIF2α). Consistent with this, eIF2α is not phosphorylated at any time during IBV infection. This also indicates a non-canonical signalling pathway for IBV-induced stress granules. Significantly, stress granule formation is uncoupled from translational arrest as visualised using ribopuromycylation. Therefore, IBV replication both induces and inhibits cellular stress granule signalling in a process that is uncoupled from shut off of host translation.

Chikungunya virus (CHIKV) causes fever and debilitating joint pain, with frequent long-term health implications and cumulating fatalities worldwide. There are no specific antivirals and vaccines, therefore understanding CHIKV replication is essential to establish treatments and preventative measures. Cellular ion channels are druggable targets and are known to facilitate replication of RNA viruses. To determine if the activities of cellular chloride channels (Cl--channels) are required during CHIKV replication, we applied broad-ranging inhibitors and siRNA to mammalian and invertebrate cells. The Cl--channel inhibitors DIDS, 9-ACA and NPPB significantly reduced the titre of released CHIKV progeny at 12 h post-infection in a dose-dependent manner suggesting that Cl--channels are pro-viral factors. Analysis of viral protein expression and time-of-inhibitor-addition studies indicated that CHIKV requires Cl--channels at post-entry and pre-egress stages. Replication of a sub-genomic replicon was restricted and genome copy numbers reduced by Cl--channel inhibition, implying that Cl--channels are involved in genome replication. siRNA knock-down identified the chloride intracellular channels (CLIC) 1 and 4 to be required for the CHIKV infectious cycle with CLIC1 interacting with the viral protein nsP3. We hypothesise that the channels play a role in formation or maintenance of the membranous, viral replication-complexes and that this important role is conservt amongst the mammalian and invertebrate hosts. These findings advance our understanding of CHIKV replication in the two host environments and help to identify drugs/druggable targets for treatment and prevention of CHIKV disease.

Influenza virus encodes a heterotrimeric RNA-dependent RNA polymerase (RdRP), composed of subunits PB1, PB2 and PA, that carries out both transcription and replication of the viral RNA genome segments in the context of ribonucleoproteins. Replication of negative-sense viral RNA (vRNA) is a two-step process, progressing via a positive-sense complementary RNA (cRNA) intermediate. The mechanism of viral genome replication is mostly unknown, though there are multiple reports indicating RdRP dimerisation may be central for the process. Purified RdRPs from human and avian influenza A viruses both form dimers of heterotrimers in solution. Using a combination of X-ray crystallography, SAXS and cryo-EM, we identify the interface involved in RdRP dimerization, which is primarily located on the PA C-terminal domain. We use bimolecular fluorescence complementation (BiFC) to show that influenza RdRP forms dimers in mammalian cells through the interface identified in solution. Using a combination of cell-based and in vitro assays, we show that influenza RdRP dimerisation via the PA-C terminal domain is necessary for copying cRNA back into vRNA during viral genome replication. In addition, we show that a nanobody (a small-domain antibody) that interferes with dimerisation attenuates influenza A virus growth in cell culture. These data provide insight into the mechanism of influenza viral genome replication, and identify a potential novel drug target against influenza A virus.

To elucidate linkage between replication and encapsidation in Picornavirales, we have taken advantage of the bipartite nature of the plant-infecting member of the order, cowpea mosaic virus (CPMV), to decouple the two processes. RNA-free virus-like particles (eVLPs) can be generated by transiently co-expressing the RNA-2-encoded coat protein precursor (VP60) with the RNA-1-encoded 24K protease, in the absence of the replication machinery (Saunders et al., 2009). We have made use of the ability to produce assembled capsids of CPMV in the absence of replication to examine the putative linkage between RNA replication and packaging in the Picornavirales. We show here that the remarkable specificity of packaging observed in CPMV is due to a functional linking between the two processes of viral replication and encapsidation. We have created a series of mutant RNA-1 and RNA-2 molecules and have assessed the effect of the mutations on both the replication and packaging of the viral RNAs. We demonstrate that mutations that affect replication have a concomitant impact on encapsidation, and that RNA-1-mediated replication is required for encapsidation of both RNA-1 and RNA-2. This close coupling between replication and encapsidation provides a means for the specific packaging of viral RNAs. Moreover, we demonstrate that this feature of CPMV can be used to specifically encapsidate custom RNA by placing a sequence of choice between the RNA-2 sequences required for replication, which opens the door to novel research and therapeutic applications in the field of custom RNA packaging and delivery technologies.

Infectious bronchitis virus (IBV), an avian gammacoronavirus, is an important pathogen causing significant animal welfare problems and economic losses to the global poultry industry. Positive-strand RNA viruses, including coronaviruses, induce cellular membrane rearrangements during replication forming replication organelles, which are thought to support efficient viral RNA synthesis. IBV replication has been shown to induce the formation of double membrane vesicles (DMVs), zippered ER and tethered vesicles, known as spherules. Although these are proposed to be the site of viral RNA synthesis, this is as yet unconfirmed and is therefore the focus of these studies. Historically, dsRNA has been used as a marker for sites of coronavirus RNA synthesis, however IBV-associated dsRNA and nsp12 (the viral RNA-dependent RNA polymerase) do not colocalise in infected cells. We have determined the cellular location of the viral genome using Fluorescence In Situ Hybridisation (FISH). By comparing the immunofluorescence labelling of cells permeabilised with different detergents, we have demonstrated that dsRNA can be found within membrane-protected compartments, while nsp12 is not, indicating that the virus could be isolating the dsRNA in DMVs or spherules, affording protection from the host immune response. By incorporating uridine analogues over the course of infection with IBV, we have visualised sites of nascent viral RNA synthesis using super-resolution microscopy. This has shown that dsRNA appears to colocalise more strongly with nascent RNA than nsp12. Using these methods as well as looking at the ultrastructural level we are able to begin to discover the location of sites of IBV RNA synthesis.

Influenza A virus has a genome consisting of 8 segments of negative sense RNA. When two influenza A virus strains infect the same cell, there is potential for the progeny to package segments from both strains. This process is termed reassortment and can lead to rapid genetic shifts that have previously generated strains of influenza responsible for pandemic events. Recent evidence suggests that assembly of the eight influenza genomic segments for packaging into a virion is mediated by RNA-RNA interactions between the segments. These interactions are likely to contribute to the varying compatibilities for reassortment observed between segments from different strains of influenza. We have captured complete RNA-RNA interaction maps for several influenza A viruses using a high-throughput sequencing approach and identify extensive, redundant, networks of RNA-RNA interactions between the genomic viral RNA segments. We extended this analysis to H1N1 and H3N2 reassortants, and found that by manipulating these interactions, we can drive preferential co-segregation of segments during reassortment. This work provides the first direct evidence that RNA-RNA interactions between the influenza virus genomic segments are a key factor in driving reassortment between viral strains.

Herpes simplex virus 1 (HSV-1) establishes latency in sensory neurons, allowing it to persist for the lifetime of the host. During latency, the only abundantly transcribed HSV-1 gene is the latency-associated transcript (LAT), which is processed into the 2.0 kb major LAT intron and several microRNAs. These non-coding RNAs (ncRNAs) have been reported to influence latency, possibly through limiting apoptosis of infected cells. As these studies have used animal models or non-neuronal cell culture, we have developed a differentiated human neuroblastoma (SH-SY5Y cells) model to examine their effect in human neuronal cells. We have infected these neuronal cultures with replication-defective HSV-1, which establishes a quiescent infection and strongly expresses the latency ncRNAs. We show that quiescent HSV-1 infection reproducibly protects differentiated SH-SY5Y from etoposide-induced apoptosis. We are also further defining the contribution of different LAT ncRNAs using recombinant lentiviruses to drive expression of the LAT intron or microRNAs. Furthermore, we are also currently exploring the mechanisms of this anti-apoptosis effect, and broader virus-neuron interactions by characterising whether the human neuronal transcriptome is altered by LAT RNA expression. Improving our understanding of the molecular interactions underpinning HSV-1 latency in neurons could help develop novel therapies to target HSV-1 latency.

High-risk human papillomavirus (HR-HPV) infects epithelial cells and is the major cause of anogenital and oropharyngeal cancers. HR-HPV oncogenic activity is through E6 control of p53, but E6 binds and degrades PDZ proteins such as the tumour suppress protein hDlg (human homologue of DrosophilaDiscs Large). The E6/hDlg complex also contains Connexin 43 (Cx43) (MacDonald, Sun et al. 2012), the major building block of gap junctions that allow intercellular molecular communication. In HPV16-positive non-tumour cervical keratinocytes (W12NT: low E6 levels) Cx43/hDlg is on the plasma membrane but in the cytoplasm in W12T tumour epithelial cells (W12T: high E6 levels) correlating with loss of gap junction cell-cell communication. E6 siRNA depletion in W12T cells restored Cx43 to the cell membrane, while overexpressing E6 in HPV-negative cervical cancer cells C33a resulted in Cx43 moving to the cytoplasm. E6 could control Cx43 trafficking through controlling hDlg or by altering cell signalling. In the absence of E6 in HEK293, HaCaT and normal immortalised keratinocytes (NIKS), Cx43 and hDlg could be co-immunoprecipitated and they co-localised on the plasma membrane. Thus, the Cx43/hDlg interaction is not carcinoma cell-specific, is not dependent on HR-HPV E6, and may have a functional role in non-cancer cells. siRNA depletion of hDlg, led to reduction in Cx43 protein levels and some relocation to the cytoplasm. This indicates that HPV E6 controls Cx43 through interaction with and degradation of hDlg. However, we cannot discount that E6 itself may have additional effects on Cx43 levels and trafficking besides via hDlg.

Events controlling herpesvirus nuclear genome uncoating, nuclear transport, and the onset of transcription remain poorly understood. We have now developed procedures to examine these processes within individual cells and at the single molecule level for both the genome and the transcripts produced from it. We have combined two novel techniques of, firstly, bioorthogonal chemistry to visualise genomes which incorporate an alkyne-nucleoside analogue (ethynyl deoxycytidine, EdC) and secondly, single molecule RNA in-situ hybridisation (smFISH) which allows detection of individual mRNA transcripts. Using these techniques simultaneously, we can now qualitatively and quantitatively analyse individual transcript abundances and their intracellular localisation, in relation to the genome itself at single molecule resolution during the progression of infection. Moreover, we are able to examine these parameters when a single genome infects a cell. We have examined the transcripts of the immediate-early mRNA for ICP0, and features revealed from this work include; transcriptional ‘bursting’ with clustered transcripts around individual genomes; mean mRNA transcript number, variance, and intracellular localisation produced from a single genome; the progressive abundant ICP0 transcription occurring selectively from replicated genomes; an increasing bottleneck in cytoplasmic transport of transcripts emanating from replicated genomes; and increased transcription bursts from virtually every uncoated genome when protein synthesis is suppressed. Further, by multiplexing probes, we can simultaneously analyse distinct transcription outputs from different genes of the same or classes, and genomes, in the same individual cell. Our results reveal completely new perspectives on the very early events of genome presentation and transcription from those genomes.

Epstein-Barr virus (EBV), is an oncogenic gamma-herpesvirus, which is associated with malignant diseases of B cells, T cells, and epithelial cells. EB viruses have large DNA genomes of more than 170 kb that are difficult to clone and manipulate. Here we describe 2 different approaches for cloning whole EBV genomes of diverse strains for reverse genetics studies. The first approach used CRISPR/Cas9-mediated cloning of the entire EBV genome into a bacterial artificial chromosome (BAC) vector using homologous recombination in B cells. This method allowed the cloning of the type 2 EBV strain Jijoye for the first time, but the BAC-clones are unstable. This strategy is being modified by recoding the homology regions to make the clones more stable. The second approach involves transformation-associated recombination (TAR) cloning of EBV fragments and their assembly in yeast, which will allow for mixing and matching DNA regions from different EBV strains for functional studies. This approach is based on TAR cloning of the EBV genome as 10 overlapping fragments, which average 17 kilobases long, using the natural homologous recombination processes of the yeast. Subsequent assembly of all the overlapping fragments is undertaken in yeast or by Gibson assembly to reconstitute the infectious EBV clone. Two fragments from EBV strains B95-8 and AG876 were captured and isolated successfully, but at low efficiency. We are currently improving the TAR cloning efficiency by increasing the size of the capture homology regions to approximately 500 bp coupled with CRISPR/Cas-9-mediated fragmentation of the EBV genome.

Hepatitis B virus causes chronic liver infection in 257 million people worldwide. Current treatments against HBV can control, but not cure HBV infection. Therefore, new treatments for chronic HBV infection need to be developed. The APOBEC3 (apolipoprotein B mRNA editing enzyme catalytic polypeptide-like 3) proteins are cellular restriction factors, which have been shown to restrict viral replication for a number of viruses (e.g. HIV) and for retrotransposons. The aim of this study is to elucidate the role of APOBEC3 proteins in inhibiting Hepatitis B virus replication. Quantitative polymerase chain reaction (qPCR) was used to evaluate the impact of APOBEC3 family members on the Hepatitis B virus replication in HepG2.2.15 cells. The highest inhibition of intracellular capsid associated HBV DNA, extracellular virion associated HBV DNA was induced by APOBEC3DE, APOBEC3F, and APOBEC3G as compared to the other APOBEC3 proteins. However, APOBEC3DE showed no inhibition of HBV total RNA, whereas the highest inhibition of HBV total RNA was induced by APOBEC3F and APOBEC3G. The sub-cellular localisation of APOBEC3 proteins was determined by immunofluorescence using confocal microscopy. It was found that APOBEC3DE, APOBEC3F, and APOBEC3G localise to the cytoplasm, suggesting a crucial role of these proteins in HBV replication in the cytoplasm. Nevertheless, APOBEC3A and APOBEC3B localise to the nucleus. The expression and co-localisation of APOBEC3 proteins and viral and host proteins (Uracil DNA Glycosylase UNG and ATP-dependent RNA helicase DDX3) in HepG2.2.15 and HEK293T cells are being investigated in order to determine the interaction between these proteins.

To evade innate immunity, many viruses express interferon-antagonists that target STATs, critical mediators of immune signalling. Virus-STAT interfaces may provide new therapeutic targets but progress is hindered by a lack of direct structural data, owing to poor tractability of antagonists/full-length STATs for structural/biophysical approaches. By applying cross-saturation transfer NMR, we report the first direct structural analysis of binding of full-length STAT1 to an interferon-antagonist of a human pathogenic virus, the first such study of the virus-host interface. Analysis using mutation of the interface, biophysical characterization, immune signalling/protein-protein interaction assays including PCA, reverse genetics and animal infection demonstrated the significance of this interface in immune signaling suppression, and in disease caused by a pathogenic field-strain lyssavirus. Importantly, NMR/mutagenesis also revealed that the interface comprises multiple surfaces/domains in both the viral and cellular partners, indicating that antagonism involves extensive interactions consistent with a multifaceted inhibitory mechanism, distinct from ‘simple’ mechanisms such as tethering. Furthermore, by elucidating the spatial relationship of interactions critical to immune evasion and replication, the data provide insight into how ostensibly simple viruses can regulate these central functions via a single multifunctional protein. These data provide novel insights into fundamental viral biology, and potential exploitation of these mechanisms as new targets for antivirals and vaccine development. The study also demonstrates the power of biophysical/NMR approaches to elucidate the atomic interface of full-length STATs with regulatory proteins, providing a framework for studies to reveal immune evasion mechanisms of other pathogens in their full complexity.

The accessory protein Vpr of Human Immunodeficiency Virus type 1 (HIV-1) enhances replication of the virus in macrophages. Virus particle packaged Vpr is released in target cells shortly after entry, suggesting it is required early in infection. Why it is required for infection of macrophages and not cycling T-cells and why it induces G2/M arrest in cycling cells are unknown. Here we observe, by co-immunoprecipitation assay, an interaction between Vpr and endogenous REAF (RNA-associated Early-stage Antiviral Factor, RPRD2), a protein shown previously to potently restrict HIV infection. After HIV-1 infects macrophages, within 30 min of viral entry, Vpr induces the degradation of REAF. Subsequently, as replication continues, REAF expression is upregulated – a response which is curtailed by Vpr. REAF is more highly expressed in differentiated macrophages than in cycling T-cells. Expression in cycling cells is cell-cycle dependent and knockdown induces cell-cycle perturbation. Therefore, our results support the long held hypothesis that Vpr induces the degradation of a factor involved in the cell cycle that impedes HIV infection in macrophages.

Tegument protein pUL83 is the most abundant component of human cytomegalovirus (hCMV) particles. The viral protein is predicted to be composed of three domains: a pyrin association domain (PAD), a carboxy-terminal domain (CTD), and an intrinsically disordered linker domain (amino acids 388–479) located between the PAD and CTD. Although pUL83 has been shown to antagonize interferon (IFN) responses, it has not been fully elucidated how the viral protein may contribute to hCMV replication. In this study we demonstrate that pUL83 associates broadly with viral and host chromatin including condensed chromosomes during mitosis. We further show that the linker domain in pUL83 is both required and sufficient for host chromatin targeting, and that this interaction depends on two evolutionary conserved arginine residues (R453 and R455) in the viral protein. Our data indicate that the pUL83 linker domain specifically associates with human core histones (but not linker histones). Furthermore, pUL83 inhibits IFN-beta and IFN-lambda gene induction, but not expression of other cytokine genes, via a mechanism that largely depends on the linker domain including R453/455. Although earlier studies suggested that pUL83 is dispensable for productive hCMV infection in fibroblasts, we find that the viral protein is necessary for efficient plaque formation in these cells, specifically in the presence of IFN. Finally, the pUL83 linker domain including R453/455 contributes significantly to the plaque size in hCMV-infected fibroblasts. Overall, we propose that pUL83 promotes spread of hCMV by selectively inhibiting induction of IFN gene expression via a novel chromatin-based molecular mechanism involving core histones.

Type I interferons (IFN) are potent inducers of an anti-viral state in response to infection and have been demonstrated to inhibit cytomegalovirus (CMV) replication both in vitro and in vivo. CMV, like all herpes viruses, has the capacity to establish lifelong infections of host through the establishment of latency. As the very early stages of viral entry can trigger IFN responses we investigated the impact of IFN on the establishment of latent human CMV (HCMV) in myeloid progenitor cells. Here we show that priming of myeloid THP1 cells with type I IFN prior to infection skews infection towards a more efficient establishment of latency. This is evidenced by detection of reduced lytic gene expression, increased latent gene expression, and increased levels of reactivation following differentiation. Blockade of IFN signalling with neutralising antibodies antagonised the latent phenotype suggesting that endogenous IFN production upon infection contributed to the effect observed. Intriguingly, whilst both IFNα2 and IFNβ can drive latent infection individually, their effects were dose-dependent and demonstrated a biphasic impact on the establishment of latency, with the highest doses of IFN preventing both lytic and latent infection. These data demonstrate that the HCMV derives an unexpected benefit from IFN production. They support a hypothesis that, although anti-viral in nature, concentration-specific effects of IFN may be evident in the cells which can modulate different outcomes post infection in persistent viruses such as HCMV. Future work is identifying the IFN concentration-specific effects responsible for a cellular environment that favours the establishment of latency.

DNA sensors including cGAS, STING and IFI16 are key components of the innate immune response to infection. However, the precise mechanisms of action, in particular the relative importance of direct suppression of replication versus paracrine signalling of an antiviral state to susceptible cells remains unclear. We examined the kinetics of herpes simplex virus infection and spread in a relevant cell type, human keratinocytes, lacking one or other of these DNA sensors using time-lapse microscopy. We also examine transcriptional induction of interferon from the native locus, at single cell and single molecule level using highly sensitive RNA FISH. Our results reveal distinct aspects of the roles of these factors and reveal outcomes not appreciated by other methods. Cells lacking either of these factors showed increased susceptibility to initial infection (prior to any downstream paracrine signalling) but with quite different outcomes. Lack of cGAS resulted in increased cellular migration and cell density at the infection focus. On the other hand, cells lacking STING showed lower cell density and significantly increased cytopathic effect likely curtailing virus yield. Initial results demonstrate that we can analyse interferon transcription at single cell level with exquisite sensitivity down to a few transcripts per cell and reveal profound spatial heterogeneity in responses to induction by PAMP ligands. Altogether, our results reveal new insight into the spatial landscape of the initiation and spread of HSV and key cellular responses which likely integrate pathways including innate immunity, apoptosis and cell migration.

African horse sickness is a major infectious disease of equids and is caused by African horse sickness virus (AHSV), a dsRNA virus with 10 genome segments encoding for 7 structural and 4/5 non-structural proteins. Here, we focused on the characterisation of the AHSV NS4, the latest protein found to be expressed by this virus. In silico analysis of available sequences confirmed the existence of two phylogenetically distinct AHSV clades: NS4-I and NS4-II. NS4-II is further divided into three subtypes (a, b and g). Confocal microscopy demonstrated that all AHSV NS4 types localised in the cytoplasm of infected cells, unlike the BTV NS4 which, has a strong nucleolar localisation. The replication kinetics of reverse genetics derived AHSV NS4 deletion mutants (AHSVDNS4) were similar to their wild type counterparts in insect (Kc) or interferon incompetent (BSR) cells. However, replication of AHSVDNS4 mutants in primary horse endothelial cells was restricted, in comparison to wild-type viruses. Importantly, primary cells restriction to AHSV replication was dependent on the JAK/STAT pathway. Furthermore, AHSVDNS4 mutants were not able to efficiently suppress the secretion of anti-viral cytokines from primary cells, while the wild-type viruses suppressed this response to varying degrees. Importantly, AHSVDNS4 mutants were less virulent than their wild type counterparts in a murine model of AHSV infection. These results indicate that AHSV NS4 has a role in interferon IFN antagonism and a determinant of viral virulence. We are currently carrying out mass spectrometry analyses to identify the cellular proteins interacting with the AHSV NS4.

Paramyxoviruses (PVs) are negative-sense RNA viruses that are important in human and animal health, and can cause severe zoonotic diseases. We screened a wide range of animal and human PV matrix proteins for host interactions, and found that the long (L) isoform of the host protein ZAP (zinc-finger antiviral protein) interacted with all the tested viral matrixes. ZAP-L is constitutively expressed, has a prenlyation motif for membrane-localization, and primarily mediates antiviral activity by binding and degrading viral RNAs through the exosome complex. However, ZAP restriction of PVs has not been demonstrated in the literature. We found that knockdown of ZAP-L results in increased replication of a panel of five human and animal PVs. Overexpression of ZAP-L (but not ZAP-short) restricts replication of PVs in a reciprocal pattern – with the notable exception of Sendai virus (SeV) – and as with other viruses, mutating the prenylation motif of ZAP-L abolishes restriction. RT-qPCR of PV RNAs and pulled-down RNAs does not indicate specific targeting of a viral transcript by ZAP-L, although overall genome abundance is reduced. Finally, immunoprecipitation of ZAP shows an additional RNA-independent interaction between ZAP-L and SeV-nucleocapsid not found with HPIV3-nucleocapsid, a closely-related PV. Thus, we have observed that ZAP-L interacts with the matrix of, and restricts replication of, a wide range of paramyxoviruses. A PV that is not restricted (SeV), has an additional interaction with ZAP via its nucleocapsid protein that may ameliorate ZAP restriction. Investigating ZAP-related restriction differences between closely-related PVs may shed light on anti-viral mechanisms of ZAP.

Louping ill virus (LIV; Flavivirus, Flaviviridae) is an important—but poorly characterized—animal pathogen of significant economic concern within the UK. Transmitted by ticks, LIV predominantly causes disease in ruminants and grouse, resulting in heavy losses. LIV is closely related to another flavivirus—tick-borne encephalitis virus (TBEV) which, unlike LIV, is a significant human pathogen. The molecular mechanisms that underpin host restriction in these viruses are poorly understood however, previously it has been shown that the TBEV non-structural (NS) proteins do not act as type-I interferon (IFN) antagonists, unlike many other Flavivirus NS proteins. Therefore, to facilitate comparison with LIV we investigated the possible antagonistic actions of the LIV NS proteins using a luciferase-based IFN reporter assay. Utilising this assay we identified six LIV NS proteins that function as antagonists throughout the IFN induction cascade. We also identified and modelled a subgenomic flavivirus RNA (sfRNA) that is produced during LIV infection and is similar in structure to TBEV sfRNA. We found that the LIV and TBEV sfRNAs antagonise RIG-I, indicating that the IFN antagonistic ability of sfRNA is not limited to mosquito-borne flaviviruses. Finally, we established the first LIV reverse genetics system using circular polymerase extension reaction (CPER). This powerful tool can be used to produce chimeric viruses which will allow further investigation into the factors governing host restriction and virulence in tick-borne flaviviruses. Investigating the mechanisms that underlie LIV infection aids our understanding of interferon antagonism in flaviviruses and the molecular determinants of host restriction.

The Birnaviridae family is comprised of non-enveloped viruses with a double-stranded RNA genome that is divided into two segments, A and B. Birnaviruses are responsible for major economic losses to the poultry and aquaculture industries, and reassortment complicates their epidemiology and control. However, little is known about the nature of theirreplication in cells, or the molecular mechanism underpinning reassortment. In order to address this, we rescued two recombinant infectious bursal disease (IBD) viruses, with either a GFP11 or Tetracysteine (TC) tag at the 3’ end of segment B (IBDV-GFP11 and IBDV-TC, respectively). DF-1 cells were either transfected with GFP1-10 prior to IBDV-GFP11 infection, or stained with ReAsH following IBDV-TC infection, which led to the apprearance of green or red foci in the cytoplasm, respectively. Foci co-localised with VP3 and dsRNA, suggesting these were virus factories (VFs). The average number of VFs significantly decreased from 60 to 5 per cell between 10 and 24 h post infection (P<0.01), while the average area significantly increased from 1.24 µm2 to 45.01 µm2 (P<0.01), suggesting VFs coalesce in the cytoplasm over time. Red, green and yellow foci were observed in the cytoplasm of co-infected cells, suggesting that VFs are initially derived from distinct input viruses prior to coalescence. Live cell imaging revealed that larger VFs were more static while smaller VFs were more mobile, and fusion events were observed. We speculate that VF coalescence is required for birnavirus reassortment, and current work is aimed at determining the cellular factors that drive coalescence.

The pneumoviruses human and bovine respiratory syncytial virus (RSV) are significant global respiratory viral pathogens responsible for causing lower respiratory tract infections in humans and cattle, respectively. The formation of progeny virions at the cell surface requires the coordinated assembly of glycoproteins trafficked to the apical surface of cells via the host secretary pathway, other viral protein complexes assembled in cytoplasm inclusion bodies and lastly viral ribonucleoproteins. Although some host proteins and pathways have been implicated in pneumovirus budding and assembly (e.g. HSP90, Rab-11 and apical recycling endosomes), the full spectrum of virus-host interactions has not been fully elucidated. Using a siRNA library targeting membrane trafficking and the Incucyte® Live Cell Imaging System, we have developed and optimised a high-throughput siRNA protocol for characterising RSV egress. The siRNA library targets human proteins which are known, or predicted, to be involved in membrane trafficking or remodelling, while the Incucyte® System allows near real-time imaging over an extended time-course, generating high quality data that allows the monitoring and quantification of multiple parameters such as cell viability, viral replication and syncytia formation. Using a recombinant human RSV expressing a GFP reporter we have identified a number of proteins involved in pneumovirus trafficking in infected cells. Our techniques provide a robust and sensitive mechanism for genetic screening and the identification of pneumovirus-protein interactions.

Bluetongue virus (BTV), in family Reoviridae, is an insect-borne, double capsid virus causing haemorrhagic disease in livestock around the world. Here, we elucidate how outer capsid proteins VP2 and VP5 coordinate cell entry of BTV. The recently solved high-resolution structures reveal unique features of BTV VP2 and VP5. To identify key functional residues, we used atomic-level structural data to guide mutagenesis of VP2 and VP5 and a series of biological and biochemical approaches, including site-directed mutagenesis, reverse genetics-based virus recovery, expression and characterization of individual recombinant mutant proteins, and various in vitro and in vivo assays. We demonstrate the dynamic nature of the conformational change process, revealing that a unique zinc finger (CCCH) in VP2 acts as the major low pH sensor, coordinating VP2 detachment, subsequently allowing VP5 to sense low pH via specific histidine residues at key positions. We show that single substitution of only certain histidine residues has a lethal effect, indicating that the location of histidine in VP5 is critical to inducing changes in VP5 conformation that facilitates membrane penetration. Further, we show that the VP5 anchoring domain alone recapitulates sensing of low pH. Our data reveal a novel, multiconformational process that overcomes entry barriers faced by this multicapsid nonenveloped virus.

BK polyomavirus (BKPyV) is a small, non-enveloped dsDNA virus that can establish a lifelong, silently persistent infection in the kidney and is estimated to infect 70–90 % of the world’s population. In immunocompromised individuals, particularly bone marrow and kidney transplant patients, increases in BKPyV replication can result in significant pathological conditions. In the case of kidney transplant patients, this can result in nephropathy, which in severe cases can result in the deterioration of allograft function and loss of the transplanted organ. There are currently no antiviral treatments with strong evidence of clinical efficacy against BKPyV. Though little is known about BKPyV egress from infected cells, we have evidence showing that BKPyV can be released in a non-lytic manner by an unconventional cellular secretory pathway that bypasses the Golgi apparatus. Here, we investigate the mechanisms behind BKPyV non-lytic egress through studying the effects of knocking out candidate host proteins involved in unconventional secretory pathways on BKPyV release, examining the effects of BKPyV infection on host cell protein secretion, and ascertaining which host proteins are essential for the BKPyV life cycle via a whole-genome CRISPR screen. These experiments are uncovering novel virus-host interactions that, when targeted, could lead to antiviral effects.

Bacteriophages are viral parasites of bacteria. A successful infection starts with the adsorption of the bacteriophage to a specific receptor on the host cell surface. Most bacteriophages are thought to have a narrow host range but this can be extended in certain cases. One strategy forextending host range is to first express a known functional bacteriophage receptor protein in bacteria previously non-susceptible to the bacteriophage, thereby enabling adsorption and potential infection by viruses that target the specific receptor. To investigate the feasibility of this approach, a plasmid (pMUT13) encoding the Escherichia coli LamB porin, the receptor for bacteriophage Lambda, was transferred into three different enterobacterial genera, namely Citrobacter, Yersinia, and Serratia. Over 100 environmental bacteriophages were isolated that infected these pMUT13-containing strains, and some bacteriophages were shown to infect their respective hosts in a LamB-dependent way. The host ranges of the environmental bacteriophages were cross-tested across the heterologous genera and surface adsorption kinetics investigated. Unlike bacteriophage Lambda, which is a member of the Siphoviridae, these newly-isolated LamB-dependent bacteriophages were more commonly members of the Myoviridae, based on transmission electron microscopy and whole genome sequences. Furthermore, an interesting selection of evolved bacteriophage mutants with broader host range were isolated, and the key mutations involved in their evolution to adapt to new hosts were investigated by genome analysis.

The 2014–2016 Ebola outbreak in West Africa highlighted the need for improved diagnostics, surveillance and therapeutics for filoviruses. The need for high containment virus handling facilities creates a bottleneck hindering research efforts. A safe alternative to working with native viruses are pseudotyped viruses (PV) which are non-replicating particles bearing surface glycoprotein(s) that can be used for antibody detection. The aim of this study was to create a diagnostic tool to distinguish between genera and species of pathogenic filoviruses (e.g. neutralization tests and ELISA), avoiding the cross reactivity currently seen. High titre PVs bearing the receptor glycoprotein (GP) of different filovirus species, plus specific epitope chimeras, were successfully generated. Next, lyophilisation studies to assess particle stability/degradation transportation and long-term storage were conducted. Filoviruses maintained their titres for at least 1.5 years after lyophilisation when kept in temperatures of up to 4 °C, with all filovirus genera following a similar trend. At higher temperatures, PVs degraded to unworkable titres. Reconstituted PVs also performed well in neutralisation assays. A chimeric cuevavirus GP bearing ebolavirus (Zaire sp.) epitopes KZ52 and 1 H3 retained infectivity, with average titres of approximately 1×10 7 RLU ml−1, similar to wild type, indicating its structure was not compromised. These chimeras are now being assessed in neutralisation tests using specific monoclonal antibodies and incorporated into ELISA with PVs as antigens. The data suggests lyophilised PVs are amenable to long-term storage, and their GPs can be modified to create artificial antigens for diagnostics and serosurveillance.

The Bunyavirales order are the largest group of negative stranded RNA viruses, infecting humans as well as a bewildering array of animals and plants, in which select members cause severe or fatal disease. To enter host cells, bunyaviruses undergo endosomal transport to specific cellular destinations and exploit the changing environment of maturing endocytic vesicles to mediate genome release. Several virus-endosome fusion triggers have previously been identified, including endosomal potassium (K+) recently identified by our group. Specifically, we demonstrated a role for K+ channels and endosomal K+ concentration ([K+]) in the ‘priming’ of virions for fusion and uncoating events. Interestingly for Bunyamwera virus (BUNV), both a reduced pH and elevated [K+] were required to permit endosomal escape of the virus. To understand the molecular basis for this requirement we have used cryo-electron tomography to study the changes in virion structure upon K+ and pH treatment. These studies reveal why endosomal [K+] and K+ channels are required for bunyavirus entry, highlighting the potential of K+channels as druggable anti-viral targets.

The Bunyavirales order of segmented negative sense RNA viruses includes over 500 isolates that infect insects, animals, and plants, and are often associated with severe and fatal disease in humans. To multiply and cause disease, bunyaviruses must transport their genomes from outside the cell into the cytosol, achieved by transit through the endocytic network. We have previously shown that the model bunyaviruses Bunyamwera virus (BUNV) and Hazara virus (HAZV) exploit the changing potassium concentration ([K+]) of maturing endosomes to release their genomes at the appropriate endosomal location. K+ was identified as a biochemical cue to activate the viral fusion machinery, promoting fusion between viral and cellular membranes, consequently permitting genome release. In this study, we further define the biochemical prerequisites for BUNV and HAZV entry and their K+ dependence. We report four major findings: (1) BUNV and HAZV require cellular cholesterol during virus infection; (2) cholesterol is required during BUNV endosomal escape; (3) cholesterol depletion from host cells impairs their ability to accumulate K+ in maturing endosomes, revealing new insights into endosomal K+ homeostasis; (4) ‘priming’ BUNV virions with K+ prior to infection alleviates BUNV cholesterol requirement, revealing the mechanism of cholesterol dependence. Taken together, we provide a new model in which cholesterol abundance influences K+ endosomal homeostasis and consequently the efficiency of bunyavirus infection. The ability to inhibit bunyaviruses with existing cholesterol lowering drugs offers new options for future anti-bunyavirus interventions for pathogenic family members.

Influenza viruses have continuously evolved into multiple mutant strains from several regions, resulting in aggravated endemic or epidemic outbreak conditions. In the 2000s, several outbreaks of inter-species transmission were reported, such as, the avian H3N2 influenza virus that crossed the host barrier to dogs. The inter-species transmission gave rise to the H3N2 canine influenza virus (CIV) that spread from East Asia to North America. The newly emerged H3N2 CIV was likely to infect to cats; however, ferrets, which had a SA receptor-binding pattern similar to that of humans, were not suitable natural hosts. In addition to avian-to-dog transmission, the infectivity of pdm H1N1 and seasonal H3N2 viruses in dogs was proven when artificial inoculation of the viruses with active viral shedding in dogs caused pathologic changes in the lungs. Studies on sero-prevalence and artificial infection suggested the possibility of co-infection of and reassortment between the two viruses in dogs; later, H3N1 and variants of M-variant H3N2 reassortants between pandemic H1N1/2009 and prototype H3N2 CIV were isolated. Notably, the H3N2 CIV with the matrix gene of the pdm H1N1 virus showed more efficient transmission in ferrets than the classic H3N2 CIV. These results implied that this primary companion animal, which lives in closer proximity to humans than pigs, might act as a mixing vessel or a source of novel influenza A virus in humans. Our findings emphasized the necessity of intensive monitoring for influenza infection in companion animals for investigating the potential for the emergence of novel human influenza strains.

Bluetongue is a vector-borne disease of ruminants caused by bluetongue virus (BTV). BTV can infect essentially all domestic and wild ruminants but the clinical outcome of infection differs substantially between host species. Clinical disease induced by BTV, including haemorrhagic fever in severe cases, is normally evident only in sheep. Conversely, cattle are more resilient to BTV infection, as they develop high levels of viremia and can be reservoirs of infection, but rarely show clinical signs. Here, we concentrated on BTV-host cell interactions using primary cells as an experimental system. First, we determined that BTV reaches higher titres in ovine cells, compared to bovine cells although it induces comparable levels of antiviral cytokines in both cell types. Importantly, these differences are abolished by inhibiting the Jak/Stat pathway. In addition, pre-treatment with interferon (IFN) severely hampers BTV replication in bovine, but not in ovine, primary cells. These data suggest that bovine, unlike ovine, IFN-stimulated genes (ISGs) are effective in controlling BTV replication. Using a high-throughput flow cytometry approach, we screened an expression library of over 300 bovine ISGs to identify genes with antiviral properties against BTV. We have identified ∼10 bovine ISGs that negatively impact BTV replication (by at least 50%). Currently, we are assessing the sheep orthologues to the bovine ISGs of interest in order to investigate host-species differences. Our study provides novel insights on how bovine cells restrict BTV replication and could provide an intellectual framework to understand the host determinants involved in disease severity.

The avian coronavirus infectious bronchitis virus (IBV) is the most economically important disease of chickens in the UK, causing significant losses as a result of poor weight gain and reduced egg quality in infected birds. IBV expresses a large spike (S) glycoprotein on the surface of the virion which is responsible for attachment to host cells and is the main antigenic target for neutralising antibodies during infection. Previous work has also demonstrated that the S protein determines cell tropism in vitro. In order to investigate the involvement of the S gene in IBV pathogenesis and explore the potential for vaccine propagation in cell culture, recombinant viruses were generated using vaccinia virus based reverse genetics. Two isolates of the pathogenic M41 strain were mutated to include the S gene from a non-pathogenic lab strain with extended tropism (Beau-R) or a heterologous pathogenic field strain with restricted tropism (4/91), resulting in two recombinant IBVs termed M41K-BeauR(S) and M41K-4/91(S), respectively. These viruses were characterised in vitro and in vivo to determine the involvement of the S gene in IBV replication and pathogenicity. M41K-BeauR(S) was attenuated in vivo but exhibited the extended host tropism of the S donor strain. M41K-4/91(S) remained pathogenic and also adopted the restricted in vitro tropism of 4/91. This indicates that the S gene not only determines the cellular tropism of the virus but also plays a key role during in vivo infections, and that replacing the ectodomain of IBV S can significantly alter the pathogenicity of the resulting virus.

HLA class I (HLA-I) molecules play a crucial role in cell-mediated immunity by presenting peptide antigens to cytotoxic CD8+T cells. A pathological hallmark of Type 1 diabetes (T1D) is the hyperexpression of HLA-I in pancreatic islets that contain residual insulin producing beta cells. The expression of HLA-I can be induced following exposure to interferons (IFN). Previous studies have implicated enteroviruses (EV) as major players in triggering an autoimmune response against beta cells and it seems likely that the hyperexpression of HLA-I contributes to the recognition and targeting of beta cells by CD8+T cells. Whilst HLA-I molecules are expressed on the surface of nucleated cells, there is increasing evidence that HLA-I can also be found in a soluble form in the plasma, including in patients with viral infections. Intriguingly, soluble HLA-I (sHLA-I) levels are significantly elevated in the serum of patients with autoimmune diseases such as systemic lupus erythematosus, rheumatoid arthritis and T1D, yet the function of sHLA-I is still unclear. In this project, we are investigating the impact of interferons and poly I:C, a viral dsRNA mimetic, on the expression of surface and soluble HLA-I in the human pancreatic beta cell line EndoC-βH1, the human pancreatic ductal cell line PANC1 and in HeLa cells. We show that surface HLA-I is upregulated in response to these stimuli in each of the cell lines, with greatest magnitude in EndoC-βH1. We also show for the first time that release of sHLA-I is significantly increased in response to IFNγ in EndoC-βH1 cell line.

Epstein-Barr virus (EBV) is one of 9 Human herpesviruses that can develop lifelong persistence. These viruses provide an antigenically complex challenge that induce strong CD8+T cell immunity during primary infection and continue to shape this immunity through recurrent lytic reactivation. Here is described the first lytic proteome-wide analysis of CD8+T cell responses to EBV and the first to compare primary vs memory CD8+T cell responses to any human herpesvirus. Primary CD8+T cells were mitogenically expanded directly from the blood of infectious mononucleosis (IM) patients. Comparatively, memory CD8+T cells required pre-enrichment using autologous dendritic cells loaded with a lytically-infected EBV cell lysate and FACS selection based upon the activation marker 4-1BB. Enriched cells were then expanded in vitro as for IM cells. Preparations from 7 IM patients and 7 healthy carriers were screened against each of the 70 EBV lytic cycle proteins in combination with the donors’ HLA-I alleles. Multiple reactivities were identified across the full lytic cycle with 146 responses identified amongst the 7 IM patients and 96 amongst the 7 healthy carriers. However the distribution of responses varied between the 2 cohorts with primary responses targeting IE and a small group of E proteins whereas memory responses targeted all phases but with some prominent responses against L proteins. This infers that responses in primary infection therefore appear to be shaped by presentation on the infected cell surface prior to the activity of viral evasins. However long-term carriage appears to re-shape the virus-specific response.

Infectious bursal disease virus (IBDV) targets B cells in the bursa of Fabricius (BF), causing immunosuppression in chickens and mortality. Susceptibility differs between inbred chickens, with 0 % mortality in ‘resistant’ lines and up to 80 % mortality in ‘susceptible’ lines. However, the mechanism of disease resistance is not understood. In order to address this, chickens (n=18) from three ‘resistant’ lines (15, C and O) and one ‘susceptible’ line (W) were infected with the very virulent IBDV strain, UK661. Clinical scores were recorded and tissues harvested at necropsy on day one, two and three post-infection for RNA extraction and virus titration, compared to non-infected controls. Interestingly, within a given line, we observed a range of symptoms, with some individuals experiencing more severe disease than others, despite no difference in viral replication. Line 15 was the least susceptible to disease based on the average clinical scores (3.2 (15), 5.7 (C), 4.8 (O) and 4.7 (W)) and the percentage of birds with a clinical score of 2 or above (17 % (15), 100 % (C), 83 % (O) and 83 % (W)). The average peak virus replication was also significantly lower in line 15 birds (6.3 log10 fold change) compared to lines C or O (7.0 and 6.8 log10 fold change) (P<0.01). RNA-sequence analysis will be performed using BF samples to understand the biological pathways that confer IBDV resistance. Moreover, primary bursal cells harvested from resistant and susceptible lines were infected with IBDV ex vivo and ongoing work aims to quantify differentially expressed genes in these cells.

Chikungunya virus (CHIKV) is a member of the Alphavirus genus, transmitted to humans by mosquitoes of the Aedesgenera. Infection with CHIKV causes chikungunya fever, which in many cases can lead to chronic joint disease, leaving patients with reduced ambulation. Despite its rising potential as a threat to global health, no effective vaccine or antiviral agent for protection or treatment are available. The CHIKV non-structural protein 3 (nsP3) is essential to the virus lifecycle and is believed to be a component of the genome replication complex. However, to date, the exact role of this protein has yet been determined. Although a conserved polyproline motif in the C-terminal hypervariable domain of nsP3 has been reported to interact with cellular SH3 domains, the function of this motif remains enigmatic. To address this question we generated a panel of mutations in this motif and tested the phenotype in the context of both a subgenomic replicon and full-length infectious virus, in both mammalian and mosquito-derived cell lines. Most of the mutations were well tolerated in the sub-genomic replicon, however, a subset either attenuated or completely abolished production of infectious CHIKV. These results suggest that as well as its role in genome replication, nsP3 also functions during assembly and release of infectious virus particles and that the C-terminal polyproline motif is a critical determinant of this function.

The gammacoronavirus infectious bronchitis virus (IBV) is responsible for an acute respiratory disease in domestic fowl, which has high economic impact and welfare implications in the poultry industry. The IBV non-structural protein, nsp3, is a multifunctional protein containing several putative domains, including an ADP-ribose-1’-monophosphatase (ADRP) domain conserved among coronaviruses. Inactivation of the ADRP domain in alpha- and betacoronaviruses is associated with reduced pathogenicity in vivo and altered interferon response and cytokine profiles in the host, without affecting viral replication in vitro. Therefore, recombinant viruses lacking ADRP functions have been proposed as ideal candidates for live attenuated vaccines. A recombinant IBV (rIBV) was generated in the backbone of the pathogenic M41-K strain containing a mutation in the ADRP domain catalytic core, known to abolish ADRP function in other coronaviruses. The ADRP-defective rIBV was characterised in vitro and in vivo; conversely to previously described ADRP-defective coronaviruses, in vitro analysis showed a reduction of viral replication, and the rIBV displayed a distinctive plaque phenotype. No reversion of the mutation occurred after serial passages of the virus in primary avian cell culture, nor in ex vivotracheal organ cultures which wereutilisedas a surrogate for in vivo stability testing. Pathogenicity experiments conducted in vivo resulted in a reduction in clinical signs in comparison to M41-K-infected birds, and tracheal ciliary activity, a marker for pathogenicity, was comparable to mock infected birds. These data support the role of ADRP as a pathogenic determinant and demonstrate the potential of ADRP-defective rIBV as a promising candidate vaccine.

The epithelial layer of the respiratory system has a critical role in the defense against microbes and secretes a number of proteins that function in host defense. BPIFA1 is secreted by the epithelium of the respiratory tract and we have shown previously that it inhibits binding and entry of Influenza A Virus (IAV) into respiratory epithelium (Akram et al. 2018), Mucosal Immunol (11, 71). However, its precise biological functions remain unclear. The aim of this study was to assess the influence of BPIFA1 in antibody production during IAV infection. BPIFA1 KO and wild type C57BL/6J mice were infected with IAV using different virus doses and blood, broncho-alveolar lavage and nasal washes were collected at several time points for analysis of IAV-specific antibodies. The results showed that BPIFA1 has role in the efficient generation of virus-specific IgA in the respiratory tract. Thus, BPIFA1 has an important role not only in innate defense but also in the adaptive immune response against IAV.

St. Abb’s Head virus (SAHV), a member of the genus Phlebovirus (family Phenuiviridae, order Bunyavirales), belongs to the largest group of negative strand RNA viruses. All phleboviruses share a genome structure that comprises three segments of negative-sense or ambi-sense RNA. The viral genome is composed of the small (S), medium (M) and large (L) RNA segments. The S segment encodes the nucleocapsid (N) protein, the M segment encodes the precursor for the viral glycoproteins (Gn and Gc) and the L segment encodes the viral RNA-dependent RNA polymerase (RdRp). Some viruses within the genus also encode non-structural proteins within their S or M segments. SAHV was isolated from a pool of seabird ticks (Ixodes uriae) collected at a seabird colony in St. Abb’s Head National Nature Reserve, Berwickshire, Scotland in 1979. Antigenically, SAHV appeared to be related to the Uukuniemi serogroup of phleboviruses. Similarly, the proteins of SAHV shared similar biochemical properties to Uukuniemi phlebovirus. Here, we describe an in depth molecular characterisation of SAHV. Using next generation sequencing technology, we demonstrate that SAHV is very closely related to the Uukuniemi phlebovirus (UUKV). We examine the growth of SAHV in mammalian, avian and tick celllines and define its target cell tropism.

Human Cytomegalovirus (HCMV) is associated with significant morbidity and mortality in the immunocompromised, and is a leading cause of congenital infection. Only three drugs are available for treatment, all with significant toxicities. New therapies and a vaccine are urgently required. Susceptibility to viral infection and disease is determined in part by antiviral restriction factors (ARFs) and the viral proteins that have evolved to degrade them. Small-molecule disruption of the interaction between an ARF and a viral antagonist can inhibit viral replication and may be utilised for antiviral therapies. To identify novel restriction factors against HCMV, we previously developed a multiplexed proteomic approach to identify proteins that are actively degraded early during HCMV infection. We reasoned that these would be enriched in known and novel ARFs that the virus must degrade in order to replicate. 35 proteins were shown to be degraded according to stringent statistical criteria, which included the known anti-HCMV restriction factors Sp100 and MORC3, and a novel ARF, HLTF. Here, we present preliminary results from a combination of viral replication assays to identify other novel ARFs. These include a new ‘two-colour’ approach to characterise HCMV restriction, which aims to eliminate variability in cell density by mixing populations of cells in a single cell culture well. Preliminary studies have identified a number of proteins that inhibit virus replication, as well as a novel dependency factor which the virus may degrade in order to attenuate cellular immune signalling or help establish latency.

Previously we successful infected turkeys with the China-origin H7N9 low pathogenicity avian influenza virus (LPAIV, A/Anhui/1/13, referred to as ‘wild-type’ (Wt)) which successfully transmitted to contact turkeys with virulent outcomes, highly unusual for LPAIV infection, particularly as the LPAIV cleavage site remained unchanged in all experiments. Sequencing of progeny viruses revealed consistent emergence of the L226Q polymorphism in the HA gene, termed the ‘turkey-adapted’ (ty-ad) virus. Ty-ad and Wt were used to compare the epizootic risk posed by both H7N9 LPAIVs in turkeys and to explore the mechanisms which underpins any differences. The Wt and ty-ad viruses robustly infected inoculated and contact turkeys, producing similar shedding titres. However, the ty-ad virus was more pathogenic than the Wt virus in directly-infected and contact turkeys, causing 100 % (Wt) compared to 16 % (ty-ad) survival. The ty-ad virus was detected in broader range of turkey organs, and at higher titre, compared to the Wt variant. This contrasted with pathogenicity and tissue-tropism observations for both viruses in chickens. The wt and ty-ad viruses did not replicate without trypsin in vitro, affirming a typical LPAIV phenotype. The L226Q polymorphism is known to alter receptor binding, with key differences in receptor distribution between turkey and chicken tissues observed. Replication kinetics differences in a range of avian cells will be reported for both viruses. Consequently, if this ty-ad variant were to arise more frequently in nature, it would pose an increased virulent risk to turkeys. It is therefore important to maintain surveillance and understanding of China-origin H7N9 viruses.

African Horse Sickness (AHS) is a highly lethal, vector-borne viral disease of equids, endemic to sub-Saharan Africa but with a history of outbreaks into Europe. The pathogenesis of the virus is not fully understood; some studies have shown that certain proteins are linked to virulence, and that the outcome of infection is highly dependent on viral factors. Disease can also manifest with several different presentations in the equine host (fever form, cardiac form, pulmonary form and mixed form) but the mechanism underlying this variation is not fully understood. Pathogenicity and virulence studies of AHSV are difficult to perform in horses for logistical, ethical and financial reasons. As an alternative, Interferon-alpha receptor knockout (IFNAR-/-) mice have previously been used in AHSV vaccinology studies. Full pathology characterization of the AHSV infection in this model is the primary objective of our work, which will address questions regarding pathogenesis of AHSV. Here we present data collected from experimental infection of IFNAR (-/-) mice with a strain of AHSV serotype 4, including: clinical signs; histopathology; immuno-histochemical analysis of immune response to infection; antigen detection in tissues; and transmission electron microscopy of AHSV infected tissues. In addition, we will show data from experimental infection in this animal model comparing the pathogenicity of different AHSV strains. Results obtained indicate AHSV-4 infection is correlated with oedema and pneumonia in the lungs, inflammation in the liver and meningitis plus perivascular cuffing in the cerebrum. Other data shows that different strains of AHSV differ in terms of their pathogenicity and tropism.

Lumpy skin disease (LSD) is an emerging poxviral disease of cattle caused by the Capripoxvirus lumpy skin disease virus (LSDV) which generates widespread cutaneous lesions in affected animals. LSD is recognised as a transboundary high consequence disease in Africa where it contributes to rural poverty and food insecurity. In 2015 LSDV spread to southeastern Europe and currently poses a threat to cattle in neighbouring regions. Previous research indicates that LSDV is most likely transmitted by insect vectors however details of transmission pathways are unclear. This study was designed to identify the risk of transmission of LSDV posed by different insect vectors. A bovine experimental model of LSD was established in the high containment facilities at The Pirbright Institute. Potential insect vectors (Aedes aegypti, Culex quinquefasciatus, Stomoxys calcitrans and Culicoides nubeculosus) were fed on LSDV-infected cattle then incubated for up to eight days. Cattle and insects were regularly sampled to quantify LSDV present in different tissues and vector species. This data was then used to model the dynamics of LSDV infection and transmission. All four species of vector successfully acquired LSDV from infected cattle and maintained the virus up to eight days post feeding. The outputs of this research will now be used to design more effective LSD control programmes.

Swine influenza A virus (swIAV) causes respiratory disease and productivity loss in pigs. Swine ‘flu viruses have been known to be both zoonotic and reverse zoonotic and they contain genes of swine, avian(av) and human(hu) origin. Surveillance of swIAV subtypes is important as genotypes/phenotypes are fluid and impact with respect to epidemiology, vaccination, pig welfare, veterinary and public health. Three sub-types (H1avN1, H1N1pdm09, H1huN2) are currently found in pigs from Great Britain (GB), plus H3huN2 in Europe and their reassortants. Screening of candidate samples is carried out by RRT-PCR assays – generic detection of swIAV (M gene) followed by a specific RRT-PCR for H1N1pdm09 (HA gene), a suite of RRT-PCR assays for sub-typing (HA and NA genes) and a (differential) RRT-PCR to specifically identify reassortant swIAVs that incorporate the pandemic 2009 internal gene cassette (NP gene). Subtyping assays, conventional and/or molecular, are carried out on virus isolation-positive and –negative (RNA only) samples from clinical material (respiratory tissue and/or nasal swabs). Since 2009, the number of swIAV has expanded with the H1N1pdm09 isolates reassorting with the traditional subtypes. Many European variants arose (>25) of which some have become established – in GB including H1huN2/pdm (since 2010), and H1avN1/pdm (since 2012), and in Belgium the traditional isolates were detected plus H1pdmN1/pdm and H3huN2/pdm reassortants. PCR subtyping (2012 onwards ∼130 from GB and ∼40 from BE/NL), wholegenome sequencing and bioinformatics analysis of these isolates facilitate further diagnostic improvements and assessment of zoonotic pandemic potential (in silico and in vivo).

The low pathogenic H9N2 influenza viruses are a threat to poultry as well as global public health due to their ability to reassort with other avian influenza viruses leading to the emergence of novel reassortant viruses having pandemic potential. The continued inter-subtypic reassortment events between influenza viruses in the Indian sub-continent have led to the replacement of the already existing G1 lineage of H9N2 viruses with the UDL genotype-like (A/chicken/Pakistan UDL-01/08/H9N2) viruses, which are triple reassortants between H9N2 virus (G1 lineage), HPAI H5N1 virus (clade 2.2) and HPAI H7N3 viruses. G1 lineage of H9N2 viruses in China has also been replaced with a fitter G57 lineage which donated internal genes to novel H7N9 viruses in 2013. We assessed and compared the replication, transmission and pathogenic potential of UDL01/2008/H9N2 virus and A/Ck/Vietnam/H7F-14-BN4-315/2015 H9N2 virus of G57 lineage isolated from Vietnam in 2015. Vietnam H9N2 virus was found to be relatively more virulent compared to the UDL genotype-like H9N2 in Chickens. Our in-vitro and in-ovo infection studies also showed that Vietnam/BN4-315/H9N2 virus has greater replication fitness compared to UDL-01/08/H9N2 virus. The UDL-01/08 H9N2 reassortants carrying internal genes of Vietnam/BN4-315 virus also showed improved replication fitness in MDCK cells. It is, therefore predicted that genetic reassortment between dominant strains in the Far East and the Indian subcontinent/Middle East may generate more virulent H9N2 viruses.