Hepatitis C Virus (HCV) is a clinically important infection that leads to chronic liver disease and Human Immunodeficiency Virus (HIV) co-infected patients have more rapid progression to severe liver disease and show higher rates of HCV vertical transmission. Hepatocytes are a highly differentiated cell type and support low level HCV replication. Most studies of the viral life cycle use de-differentiated hepatoma cell lines, which are highly permissive. The mechanism behind this difference is poorly understood. We show that dimethylsulfoxide (DMSO) differentiated Huh-7 cells have a 100-fold reduction in permissivity to HCV infection. We confirm that these cells are differentiated and upregulate key liver specific markers including miR122. They are metabolically active and have intact innate signaling pathways in response to infection. We observed a 10-fold reduction in the initiation of replication and a 10-fold loss in extra-cellular particle infectivity. In contrast cell-to-cell dissemination rates were comparable and cell-contact dependent infection of differentiated cells can overcome the restrictions seen in cell-free infection. HCV cell-to-cell transmission can also be mediated by other cell types. T cells are the primary cell supporting HIV-1 infection. We have shown that HCV can bind primary and immortalized T cells and trans-infect hepatoma cells. This requires replicating HIV but is independent of co-receptor engagement. HIV-1 infection of CD4+ T cells induces a significant increase in HCV trans-infection by increased viral binding. T cells provide a vehicle for HIV-1 to promote HCV infectivity, transmission and persistence.

Germinal centres (GCs) are the sites where V-gene hypermutation and B cell selection are taking place. Testing specificity and affinity of GC B cell receptor by interaction with antigen on follicular dendritic cells (FDCs) may be an important selection process to select high affinity B cell clone. As antigen on FDC is present in the form of antigen-antibody immune complex, GC B cells are expected to have to compete with antibody to get access antigen. Initially this antibody will be of low affinity. However, during the course of an immune response, this affinity may increase. We have tested this competitive selection model by following the replacement of antibodies in the GC over the course of an immune response. The speed of this replacement is dependent on affinity. Antibody added during an ongoing GC reaction can replace antibody in the GC, but only, if it is of high enough affinity. Presence of high or low affinity antibodies on FDC influences centrocyte selection, leading to variations in apoptosis within the GC, serum affinity, and plasma cell output. Parallel in silico experiments support the idea that a dynamic GC selection threshold, dependent on the affinity of GC output cells increases affinity maturation, because it enhances selection efficiency over a longer period during the course of a GC reaction. A dynamic selection threshold may explain the termination of the GC reaction, when affinity of new B cell variants is not sufficient to overcome the affinity of antibodies produced outside the GC. IRF4 is essential for the plasma cell differentiation and Ig class switch. IRF4 mRNA and protein rapidly upregulate within one hour after naive B cells get stimulation with NP-Ficoll in QM×C57BL/6 mice, and then activated B blasts expressing intermediate level of IRF4 either go into the red pulp to form the early extrafollicular response by upregulating high level of IRF4, or travel into the follicle to differentiate into GC founding cells. IRF4 completely shuts down when cells becomes proliferating centrocytes. But IRF4 expresses again in centorcytes, which have been committed to differentiate into the plasmablasts. Its high level expression shows the GC emigrants. Here IRF4 is selected as the marker for the early plasmablats appearance on the GC-T zone interface at the beginning of the GC reaction. And further experiments by using cytokines such as IL-6, IL-10, IL-21, costimulatory signals OX40, CD30 deficient mice show that these signals can affect the development of these early IRF4\(^+\) plasmablasts on the GC-T zone interface in the TD antigen response.

The role of the alternative NF-\(\kappa\)B pathway in the development of secondary lymphoid organs (SLOs) has been well described. Tertiary lymphoid organs (TLOs) include intestinal cryptopatches (CPs) and isolated lymphoid follicles (ILFs). This thesis investigates the role of the alternative NF-\(\kappa\)B pathway in the development of colonic ILFs, using the p100\(\Delta\) mouse model, where the alternative pathway is constitutively active. We present compelling data that p100\(\Delta\) mice develop significantly more lymphoid aggregates in the colon, compared with WT littermates, and provide several lines of evidence showing that these aggregates are analogous to ILFs. Additionally, we demonstrate that in the p100\(\Delta\) a significant increase in the numbers of B and T cells, and DCs in the colon and show alterations in the subsets of colonic T cells and DCs. We also present evidence that constitutively active NF-\(\kappa\)B2 p100 in LT\(\alpha\) deficient mice induces recovery of B and T cell segregation in the spleen. Most strikingly, we show recovery of iLNs and mLN development in one of five \(p100\)\(\Delta\)\(LT\)\(\alpha^{-/-}\) mice generated. These findings demonstrate that the alternative NF-\(\kappa\)B pathway plays an important role in not only SLO development and splenic organisation, but also in the development of TLOs, specifically colonic ILFs.

This thesis explored the hypothesis that cytomegalovirus (CMV) infection and its reactivation may be a shared mechanism linking psychosocial and behavioral factors with the age-associated decline in immunity, known as immunosenescence. The first empirical chapter (Chapter 3) showed that psychological stress factors were positively associated with CMV reactivation, as measured by increased CMV-specific IgG antibodies (CMV-IgG) among those infected, while socioeconomic and lifestyle factors were associated with CMV infection rates. Chapter 4 investigated personality traits and revealed that increased neuroticism predicted elevated odds of CMV infection and higher conscientiousness was associated with lower CMV-IgG levels. Chapter 5 demonstrated that more frequent physical activity was associated with lower levels of highly-differentiated T-cells, but this association was reduced to non-significance by adjustment for CMV infection. Chapter 6 showed that dysregulated glucose metabolism, measured as higher glycated hemoglobin levels, was associated with increased highly-differentiated T-cells in CMV-infected individuals. Furthermore, hyperglycemia interacted with CMV infection for a further increased accumulation of these cells. In sum, these results suggest that CMV and psychosocial and behavioral factors co-determine the progression of immunosenescence, and that CMV reactivation may reflect imbalance among these factors. Thus, CMV reactivation is proposed as a common pathway in psychobiological relationships with immunosenescence.

Drosophila nucleosome remodelling factor (NURF) is one of the founding members of the ISWI family of ATP-dependent chromatin remodelling enzymes and mediates energy-dependent nucleosome sliding leading to transcription regulation. In previous work (Wysocka et al., 2006), NURF was shown to be recruited to gene targets by binding specific histone modifications. The largest subunit of NURF, NURF301, contains a bromodomain and three PHD finger domains that have the ability to recognize specific histone modifications. Here we determine the histone binding-specificities of these domains, and how NURF histone binding is influenced by histone modification "cross-talk". This has been analyzed by histone peptide library array assays and our study shows that the PHD2 domain specifically recognizes the histone H3K4me3 mark. This binding can be inhibited by phosphorylation of H3 Thr 3, while enhanced by acetylation of H3 Lys 9 and phosphorylation of Ser 10. The binding specificities of bromodomain, PHD and PHD1 domains were also determined. These data were confirmed by peptide pull-down, Biacore and immunofluorescence microscopy assays. Moreover, two different NURF301-A/B and NURF301-C isoforms were CTAP-tagged by recombineering, and we used chromatin immunoprecipitation coupled sequencing (ChIP-Seq) to profile the genome-wide distribution of NURF in vivo. Therefore, our results identify regulatory mechanisms of histone modifications directing recruitment of ATP-dependent chromatin remodelling enzymes.

The host/pathogen interaction is pivotal in defining the outcome of an infection. For cattle with bovine tuberculosis this occurs between antigen presenting cells and the lipid-rich surface of \(Mycobacterium\) \(bovis\). Mycobacterial lipids have been shown to modulate immune responses, however previous work used model systems, or lipids from avirulent bacteria. This study aimed to assess the bovine immunomodulatory ability of lipids extracted from virulent \(M\). \(bovis\). \(Mycobacterium\) \(bovis\) lipids were extracted and characterised. Polar and apolar fractions from different bacterial strains were similar in their composition although quantitative differences were noted. Lipopeptide was identified in the polar fractions. Polar lipid stimulation of bovine antigen presenting cells increased IL - 10 and IL - 12 production and reduced expression of MHCII and CD1b. Further investigation of the polar lipids was performed by subfractionation but no individual lipid could be found responsible for these effects. The ability of the lipid fractions to activate bovine lymphocytes was assessed. Lipopeptide in the polar fraction was found to play a role in the generation of lymphocyte responses. Screening of highly purified individual lipid molecules led to the identification of AcPIM6 which was found to be capable of driving antigen specific proliferation of bovine NKT cells.

Type 1 diabetes (T1D) is a T cell-mediated autoimmune disease in which the immune system specifically targets and destroys the pancreatic insulin-producing beta-cells. We are interested in defining whether adipose tissue-derived cytokines (adipokines) such as adiponectin (anti-inflammatory) and leptin (pro-inflammatory) could influence T1D progression. We demonstrate the expression of the leptin receptor (LEPR) on peripheral blood mononuclear cells (PBMC) and observed higher expression of LEPR on PBMC from patients with T1D. However, we found no significant functional relevance for this observation. On the other hand, we show lower expression of the adiponectin receptors on lymphocytes from patients with T1D. This was associated with a reduced capacity of adiponectin to inhibit lymphocyte trans-endothelial migration in T1D. We show that adiponectin strongly inhibited lymphocyte migration by action on the endothelium or directly on the lymphocytes. We have now established that adiponectin action is not directly targetting the lumphocytes but involves accessory cells that express higher level of the adiponectin receptors. These findings were validated \(in\) \(vivo\) using a peritonal model of inflammation and led to the discovery a newly idnetified agent able to control the transmigration of T cells. These observations underline the importance of adiponectin in the control of lymphocyte transmigration during an inflammatory response and offer a potential therapeutic agent for T cell mediated diseases.

Autoimmunity can occur when self-reactive lymphocytes of the adaptive immune system are activated upon encounter with antigen. This can lead to the development of debilitating and potentially life-threatening autoimmune diseases such as type-1 diabetes, rheumatoid arthritis and multiple sclerosis. Regulatory T cells (Tregs) are a subset of CD4+ T cells that express the lineage-specific transcription factor Foxp3 and exert dominant peripheral tolerance to maintain immune homeostasis. It is therefore important to fully understand the underlying mechanisms of Treg development, homeostasis and function due to the positive and negative effects that therapeutic manipulation could have on this essential T lymphocyte population. Many effector molecules have been proposed to have a central role in regulatory T cell function, and it is now clear that Tregs are equipped with multiple mechanisms by which to exert suppressive function. It has been reported that the cytotoxic T lymphocyte antigen-4 (CTLA-4) receptor is constitutively expressed by regulatory T cells and a role for this molecule in Treg suppression has been suggested. This investigation revealed a role for CTLA-4 in maintaining homeostasis of the peripheral regulatory T cell compartment. In addition, using a transgenic mouse model that permitted the development of antigen-specific Ctla-4-deficient Tregs, a role for the CTLA-4 receptor in Treg suppressive function was identified. The data obtained suggest that the CTLA-4 receptor may function on regulatory T cells by modulating CD80/CD86 co-stimulatory molecule expression by antigen-presenting cells, and hence their capacity to activate conventional T cells to generate effector T cells and instigate an effective immune response.

The thymus is essential for T-cell development and is histologically divided into two specialized microenvironments: cortex and medulla. The cortex selects functional T cells and the medulla removes potentially auto-reactive T cells. Both selection events are mainly driven by thymic epithelial cells (TECs). However, ageing causes a loss of TECs, consequently resulting in a reduction of thymic function. Several studies have sought to reconstitute thymus function by regenerating TECs, with limited success. We aim to generate thymic tissues from defined sources that may ultimately be transplanted to improve immune function. Moreover, this study seeks to identify TEC progenitors and define distinct TEC developmental pathways, which will ultimately be helpful in understanding both thymic atrophy and thymic regeneration. We show that induced pluripotent stem (iPS) cells can generate functional TECs in vivo. Moreover, our data suggest a serial link between cTEC and mTEC lineage development. Specifically, we find that Aire+ mTEC can develop from progenitors expressing the cTEC marker CD205, and CD205+ progenitors acquire the expression of mTEC regulator Receptor Activator of NF-κB (RANK) before losing their cTEC phenotype. Collectively, our data help clarify potential cellular targets for the re-establishment of functional thymus tissue.

Introduction: GBS has a highly diverse clinical course and outcome. Currently available literature suggests that despite treatment about 20 % of patients remain disabled at one year and about 5 % patients die. These data come from clinical trials conducted between 1984 and 2006. Most of these studies included severe GBS cases. We conducted a multicentre prospective observational study looking at clinical and biological determinants of prognosis of GBS. As part of this study, I had an opportunity to analyse the data collected from 15 UK centres; looking at clinical and treatment patterns, various outcomes including ability to walk at 12 months, pain and quality of life. We also analysed Electrophysiological data from our local centre (Glasgow); compared newly published electrophysiological diagnostic criteria with existing criteria to determine whether serial studies are required for final electrophysiological diagnosis. Finally, to identify the patients with poor prognosis early in the disease course, we attempted to validate the currently available clinical prognostic models. Method: We conducted a multicentre prospective observational study named IGOS (International GBS Outcome Study) with a web-based entry system. It aimed to study at least 1000 patients over 3 years. The study included two modules: 1) core module which consist of a) acute clinical data collection at 0, 1, 2, 4 weeks and follow up data at 6 and 12 months b) serum samples collection at each clinical data entry point c) electrophysiology studies within 2 weeks 2) optional modules included additional electrophysiology studies at 4 weeks, CSF studies and long term outcome data at 2 and 3 years. As the study still ongoing, I analysed the data of 122 GBS patients recruited from 15 UK centres between May 2012 and Jan 2015. Results: In our cohort about 20 % patients remained disabled at 1 year, 18% required mechanical ventilation (MV), 5 % died. Pain continued to remain a major disabling symptom in more than half of the patients however unable to perform usual activity was the most disabling QoL domain affected at 12 months and was an important contributing factor affecting quality of life. Intravenous immunoglobulin was the most commonly prescribed treatment followed by plasma exchange. Immunotherapy was not beneficial in mildly affected GBS patients. Currently available electro diagnostic criteria are not very sensitive in identifying final EP subtypes and newly published Rajabally's criteria potentially addresses this issue and should be used in clinical practice to establish final EP diagnosis. Existing prognostic models EGOS and mEGOS performed well in our cohort and showed good discriminatory capacity. Discussion: Despite wider availability of immunotherapy prognosis of GBS has not changed in last 20years, which highlights the urgent need of more effective treatments in these patients. However new therapy can be expensive and can be only beneficial if the patients with poor prognosis are identified early in course. This can only be achieved by developing good prognostic models. Our results show that existing available models EGOS/MEGOS validates well and provides a proof of the concept that prognostic model can be used to identify patients with poor prognosis when the treatment is most beneficial. GBS continues to remain a clinical diagnosis. While there are drawbacks of existing EP criteria, newly developed Rajabally's criteria are sufficient to establish final EP diagnosis.