Electronic diaries were used to collect data on self-reported physical symptoms and psychological variables from 26 patients with multiple medically unexplained physical symptoms who were currently managed in primary care. Data was collected twice daily for 12 weeks using visual analogue scales on handheld personal computers. The diaries were acceptable and generated data with generally good compliance and high reliability. Recordings showed strong short term autocorrelation. Correlations within individual time series were compared by meta-analysis which showed moderately strong correlations between symptom severity and low mood but weaker correlations with anxiety and illness concern. There was significant heterogeneity between individuals. Granger causality analysis showed only weak evidence for sequential cause either in a psychosomatic or somato-psychic direction. A non-linear dynamic analysis (sample entropy) showed significantly reduced inherent complexity in the time series data. Interview data confirmed the presence of chaotic narratives and a struggle for meaning and coherence through the symptom experience which contradicted conventional simple medical explanations. Patients’ daily experience refutes simple associations between day to day stress and current symptoms; however changes in mood are more strongly associated with changes in symptom. The study is the first to apply a measure of complexity to physical symptom diaries and shows it to be reduced in keeping with a “loss of complexity” hypothesis. The collection of diary data appeared to encourage constructive reflection on the contents. Participants saw the diaries as worthwhile and were generally open to explanations linking mind and body. Simple stress-illness models were neither popular nor supported by the data.

Excised human skin can be used as a model to assess the potency, immunogenicity and contact sensitivity of potential therapeutics or cosmetics via the assessment of histological damage. The current method of assessing the damage uses traditional manual histological assessment, which is inherently subjective, time consuming and prone to intra-observer variability. Computer aided analysis has the potential to address issues surrounding traditional histological techniques through the application of quantitative analysis. This thesis describes the development of a computer aided process to assess the immune-mediated structural breakdown of human skin tissue. Research presented includes assessment and optimisation of image acquisition methodologies, development of an image processing and segmentation algorithm, identification and extraction of a novel set of descriptive image features and the evaluation of a selected subset of these features in a classification model. A new segmentation method is presented to identify epidermis tissue from skin with varying degrees of histopathological damage. Combining enhanced colour information with general image intensity information, the fully automated methodology segments the epidermis with a mean specificity of 97.7%, a mean sensitivity of 89.4% and a mean accuracy of 96.5% and segments effectively for different severities of tissue damage. A set of 140 feature measurements containing information about the tissue changes associated with different grades of histopathological skin damage were identified and a wrapper algorithm employed to select a subset of the extracted features, evaluating feature subsets based their prediction error for an independent test set in a Naïve Bayes Classifier. The final classification algorithm classified a 169 image set with an accuracy of 94.1%, of these images 20 were an unseen validation set for which the accuracy was 85.0%. The final classification method has a comparable accuracy to the existing manual method, improved repeatability and reproducibility and does not require an experienced histopathologist.

Tolerogenic dendritic cells are monocyte-derived dendritic cells (DC) cultured such that they adopt an immunoregulatory phenotype. In vitro, these cells are able to induce and maintain T cell tolerance through deviation of naive T cells to an anti-inflammatory phenotype and induction of anergy in memory T cells. Equivalent cells suppress established arthritis in murine models and tolerogenic DC are presently the subject of a phase I safety and efficacy trial at Newcastle University as part of the AutoDeCRA study. However, in spite of these promising data, we are yet to rigorously explore the basis of the phenotype of tolerogenic DC and lack markers to unequivocally distinguish them from other types of DC. This body of work is concerned with the development of a workflow to enable these questions to be addressed using mass spectrometry-based quantitative proteomics. Specifically, methods have been optimized and validated to enable a) enrichment and proteolytic digestion of membrane proteins, favouring their detection over more abundant cytoplasmic and nuclear proteins in LC/MS; b) differential stable isotope labelling of peptide N- and C-termini, enabling ‘isobaric peptide termini labelling’-based relative quantitation at the MS2 level; c) pipette-tip based anion exchange fractionation of IPTL-labelled peptides prior to LC/MS analysis, broadening depth of proteome coverage. Efforts to apply aspects of the workflow to perform quantitative comparisons of the whole cell proteomes and qualitative profiling of the membrane proteomes of mature and tolerogenic DC are also documented. It is envisaged that future application of this optimized workflow as a whole will enable the identification and relative quantitation of significant numbers of mature and tolerogenic DC plasma membrane proteins. Differentially expressed proteins of interest identified through this approach may then be further investigated for putative roles in tolerance induction.

The human mononuclear phagocyte system of monocytes, macrophages and dendritic cells participates in both innate and adaptive immune responses. However, the accurate identities, functions and inter-relationships of these crucial immune cells during inflammation are poorly defined. Two inflammatory settings were examined in this work: the skin in acute Graft-versus-Host Disease and the lung during experimental inflammation induced by LPS inhalation. The purpose of this enquiry was to characterize inflammatory mononuclear phagocytes in tissue, investigate their origins and explore their contribution to disease pathogenesis. In the skin GvHD study, shave biopsies were obtained from 73 individuals on presentation with acute rash following bone marrow transplantation (BMT). Controls were obtained from 19 BMT recipients at matched time points without rash and 26 healthy individuals undergoing plastic surgery. Tissue was digested and the leukocyte composition analysed by flow cytometry/sorting. Sorted populations were used in functional assays or in gene expression experiments performed using NanoString technology. An in vitro equivalent of CD14-expressing mononuclear phagocytes (MPs) was developed using HLA-matched mixed leukocyte reactions. Gene expression was measured and the function of these equivalents in a skin explant model of GvHD was tested. GvHD lesional skin was characterized by expansion of CD14-expressing MPs (GVH14) and a reduction in CD1c-expressing MPs. GVH14 were identified as donor monocyte-derived macrophages. Functionally, GVH14 could produce chemokines to recruit T lymphocytes to lesions. They were capable of activating and expanding T lymphocytes in vitro. GVH14 equivalents could damage basal keratinocytes of the epidermis without the presence of T cells. This characterization has identified a novel pathogeneic role for macrophages in acute GvHD. In the LPS inhalation study, 13 healthy individuals received saline (0.9% sodium chloride) and 13 received LPS (0.9% sodium chloride with 2mg LPS from E.coli 026:B6) by dosimeter nebulizer. Blood samples were obtained at 2, 4, 6 and 24 hours following inhalation. Between 7 and 8 hours post-inhalation, bronchoalveolar lavage (BAL) of a sub-segment of the right middle lobe was performed. BAL fluid supernatant chemokines and cytokines were analysed by multiplexed ELISA. The cellular component of BAL was analysed by flow cytometry/sorting. Sorted populations were used in functional assays or in gene expression experiments performed using NanoString technology. Seven distinct MPs were identified in steady state (i.e. following saline inhalation). Following LPS inhalation, neutrophils, CD14-expressing MPs and CD1c-expressing MPs were expanded. Phenotypically, CD1c expressing MPs resembled blood cDC2. Both subsets of blood cDC2 were recruited to the airspace but their distinct functions and gene expression profiles converged upon recruitment. This analysis provides the first detailed characterization of BAL fluid MPs. As such it provides a foundation for studying MPs in human lung diseases, including the Idiopathic Pneumonia Syndrome occurring after BMT. It detailed the surprising observation that blood cDC2 can be recruited to tissue in inflammation, challenging the dogma that inflammatory MPs must be monocyte-derived.

NK cells are known to respond in vitro to P. falciparum-infected red blood cells (iRBCs), although responses are highly heterogeneous between donors. Although their role during malaria infection is not fully understood, they may play a role in cytokine production during early infection, and furthermore may interact with and kill iRBCs. The work described in this thesis examines the role of NK cell receptors in determining the functional outcomes of NK cell activation by iRBCs, focusing on NK cell responses to exogenous cytokines and the phenotypic and functional profiles of NK cells from malaria naive (LSHTM) and malaria exposed (Ugandan) subjects. In a model of early activation of NK cells by accessory cell-derived cytokines, I have shown a key role for IL-18 in mediating NK cell responses during both primary and secondary immune responses as IL-18 synergises with cytokines from the common gamma chain family. NK cells from LSHTM donors showed low background expression of IFN- and CD25, but responded to iRBCs by secretion of IFN-, which was potentiated by exogenous IL-15. By contrast, NK cells from Ugandan donors showed higher background CD25 expression and signs of in vivo/ex vivo preactivation and enhanced responsiveness to IL-15, but did not make any appreciable response to iRBCs. Potential explanations for these findings are explored and discussed. KIR genotype and KIR expression also varied between LSHTM and Ugandan donors. Specifically, expansions of KIR2DL1+ CD57+ NKG2C+ NK cell populations (possibly driven by human cytomegalovirus (HCMV) infection) were observed in the Ugandan donors. Conversely, percentages of KIR2DL3+ and 2DS4+ NK cells were higher among LSHTM donors, indicating that HLA genotype or allelic KIR polymorphisms may influence KIR expression. Finally, the formation of NK-iRBC conjugates, which may be a precursor to NK cellmediated killing of iRBCs, was observed in cells from nearly all donors, but did not correlate with other functional responses. Analysis of KIR expression and NK cell functional responses indicated that donors expressing inhibitory KIR2DL5 had reduced numbers of conjugates. Further experiments indicated that KIR2DL5 might be specifically upregulated after incubation with iRBCs, and that individuals carrying a normally non-expressed KIR2DL5 gene may be able to express this gene under certain circumstances. This tentatively suggests a role for KIR2DL5 during NK cell responses to malaria infection, and suggests a possible function for a common but frequently non-expressed gene. In summary, my work suggests that NK cell responses are strongly influenced by cytokine receptor and KIR expression, which in turn depend on NK cell maturation status. KIR expression patterns may in part explain differential NK responses to iRBCs between LSHTM and Ugandan donors. I also propose a possible role for KIR2DL5 in malaria infection, and a reason for the low expression of this gene in African populations.

Point-of-care testing (POCT) is an emerging technology which can lead to an eruptive change of lifestyle and medication of population against the traditional medical laboratory. Since living organisms are intrinsically flexible and malleable, the flexible substrate is a necessity for successful integration of electronics in biological systems that do not cause discomfort during prolonged use. Isotropic conductive adhesives (ICAs) are attractive to wearable POCT devices because ICAs are environmentally friendly and allow a lower processing temperature than soldering which protects heat-sensitive components. Vertical interconnections and optical interconnections are considered as the technologies to realise the miniaturised high-performance devices for the future applications. This thesis focused on the multifunctional integration to enable both electrical and optical vertical interconnections through one via hole that can be fabricated in flexible substrates. The functional properties of the via and their response to the external loadings which are likely encountered in the POCT devices are the primary concerns of this PhD project. In this thesis, the research of curing effect on via performance was first conducted by studying the relationship between curing conditions and material properties. Based on differential scanning calorimetry (DSC) analysis results, two-parameter autocatalytic model (Sestak-Berggren model) was established as the most suitable curing process description of our typical ICA composed of epoxy-based binders and Ag filler particles. A link between curing conditions and the mechanical properties of ICAs was established based on the DMA experiments. A series of test vehicles containing vias filled with ICAs were cured under varying conditions. The electrical resistance of the ICA filled vias were measured before testing and in real time during thermal cycling tests, damp heat tests and bending tests. A simplified model was derived to represent rivet-shaped vias in the flexible printed circuit boards (FPCBs) based on the assumption of homogenous ICAs. An equation was thus proposed to evaluate the resistance of the model. Vias with different cap sizes were also tested, and the equation was validated. Those samples were divided into three groups for thermal cycling test, damp heat ageing test and bending test. Finite element analysis (FEA) was used to aid better understanding of the electrical conduction mechanisms. Based on theoretical equation and simulation model, the fistula-shape ICA via was fabricated in flexible PCB. Its hollow nature provides the space for integrations of optical or fluidic circuits. Resistance measurements and reliability tests proved that carefully designed and manufactured small bores in vias did not comprise the performance. Test vehicles with optoelectrical vias were made through two different approaches to prove the feasibility of multifunctional vertical interconnections in flexible substrates. A case study was carried out on reflection Photoplethysmography (rPPG) sensors manufacturing, using a specially designed optoelectronic system. ICA-based low-temperature manufacture processes were developed to enable the integration of these flexible but delicate substrates and components. In the manufacturing routes, a modified stencil printing setup, which merges two printing-curing steps (vias forming and components bonding) into one step, was developed to save both time and energy. The assembled probes showed the outstanding performance in functional and physiological tests. The results from this thesis are anticipated to facilitate the understanding of ICA via conduction mechanism and provide an applicable tool to optimise the design and manufacturing of optoelectrical vias.

Sugar oligomers which are linked to proteins and lipids play important roles in a large number of biological processes. These sugars are referred to as glycans. In the immune system, almost all key proteins are glycosylated. Glycans regulate the migration, recognition, activation, and apoptosis of immune cells, as well as the activities of antibodies. Owing to glycan complexity, the study of glycosylation is challenging. Mass spectrometry (MS) is a state-of-the-art technology which is ideally suited to investigating glycosylation, because of its ultra-high sensitivity and resolution, as well as its ability to analyze individual molecules in a complex, heterogeneous mixture. In this thesis, mass spectrometry was used to investigate the abnormal glycosylation of two newly discovered immune disorders: a hyper IgE syndrome and a congenital disorder of glycosylation (CDG). In the hyper IgE syndrome, a total set of glycans on leukocytes was analysed (glycomic studies). A reduction in tri- and tetra-antennary glycans was observed in the patients. In addition, substantially increased levels of hybrid glycans were detected in a patient with more severe symptoms, and decreased fucosylation was found on their neutrophils. Site specific glycosylation analysis (glycoproteomic studies) was done on IgE, and 6 of 7 potential N-glycosylation sites on this antibody were mapped, which did not show significant glycosylation changes. In the CDG, tri-glucosylated high mannose glycans were observed, which helped the identification of an ER glucosidase I defect. Mass spectrometry was also used to investigate engineered antibodies which are designed to treat immune disorders. A dramatic increase in sialylation was observd in an IgG1 after introducing point mutations in the Fc region, and a considerable amount of high mannose glycans were detected in an IgG1 hexamer.

This thesis describes the synthesis of iron oxide nanoparticles for use as contrast agents in biomedical imaging, specifically for MRI. The limitations of single imaging modalities can be overcome by the synergistic combination of two or more imaging techniques, e.g. the low sensitivity but high resolution of MRI complements the high sensitivity but low resolution of PET. The large surface area of superparamagnetic iron oxide nanoparticles (SPION) allows relatively simple functionalisation. The large size of a single combined nanoparticle MRI/PET probe would slow down in vivo movement, diminishing radioactivity before reaching its target. Pre-targeting using a magnetic nanoparticle followed by the injection of the radio-labelled molecule at the correct time will ensure radioactivity remains sufficiently high. Thus, the investigation of dual-modality probe development is also a focus of the thesis. Chapter 2 discusses the preparation of iron oxide nanoparticles with a core diameter of 6 nm via the high temperature thermal decomposition of iron salts. Direct modification to the surface of the nanoparticles was carried out using various small molecules with differing anchoring groups, the most successful being sodium alendronate, a bisphosphonate ligand. Chapter 3 describes the further functionalisation of the nanoparticles. One way this was achieved was by the incorporation of PEG chains of different lengths to increase water solubility and biocompatibility. Functionalisation with a strained alkyne for eventual in vitro/in vivo copper-free cycloaddition with an azide group was also achieved. The PET moiety was designed to be a 68Ga-azido-DOTA complex. Prior to radiolabelling with gallium-68, the copper-free cyclised resultant nanoparticles were characterised by the use of lanthanide analogues (Eu, Tb and Gd). Eu and Tb allowed for fluorescence spectroscopy, while the Gd allowed for relaxivity measurements to be carried out. Unexpected fluorescence results were observed for the Eu and Tb analogues. The Gd-NP conjugates are further investigated in Chapter 4. Combination of both a T1 and T2 moiety results in changes to the relaxivity of the resultant nanoparticle which can act as a dual-weighted MRI probe. The relaxivities are found to vary with modifications to the nanoparticle construct. Finally, preliminary in vitro experiments with macrophages were carried out to investigate whether there was significant preferential uptake between M1 and M2 macrophages. A single-chain variable fragment (scFv) specific to Fractalkine, a chemokine important in the progression of atherosclerosis was prepared, for use as a targeting moiety towards the imaging of vulnerable plaque.

Conventional 1H MRI relies on protons in bodily tissues which are abundant in most organs of the human body whilst the lungs consist mainly of air spaces with a resulting low density of 1H that can be imaged. Hyperpolarised gas (3He and 129Xe) MRI provides solutions to these challenges by imaging the ventilated airspaces and provides clinical sensitivity to lung disease pathophysiology such as ventilation defects, obstructive lung disease, emphysematous alveolar damage, intrapulmonary oxygen quantification, ventilation-perfusion mismatch and interstitial diseases. In addition, xenon is lipophilic, dissolves into the pulmonary bloodstream and is transported to distal organs (such as heart, brain, kidney and liver). Upon reaching the cerebral vasculature, xenon passively diffuses into the brain tissues. 129Xe exhibits a wide chemical shift range, providing contrast for different compartments of the brain (such as grey matter, white matter, cerebrospinal fluid and blood). Using LASER spin exchange optical pumping, the NMR sensitivity of hyperpolarised gases is dramatically increased allowing the detection at very low concentrations. In this thesis, I have developed original instrumentation for new applications of hyperpolarised gas MRI in lungs and brain and make novel research contributions in order to; (a) Improve the signal-to-noise ratio of 1H signal from the lungs in multi-nuclear same-breath lung imaging, by establishing an accurate analytical method to design a 1H receiver coil array that functions nested inside a 3He or 129Xe transmit-receive coil. (b) Demonstrate triple-nuclear (129Xe-3He-1H) same-breath lung imaging (structure-ventilation) and dual-nuclear (129Xe-3He) same-breath apparent diffusion coefficient mapping with a new dual tuned RF transceiver coil design. (c) Demonstrate the feasibility of using hyperpolarized 129Xe for in vivo evaluation of cerebral perfusion and, using tracer kinetic analysis, enable quantitative measurement of the intrinsic properties of the blood-brain barrier.

Pixelated Cd(Zn)Te radiation detectors are a promising technology for X-ray imaging applications but their areas are limited to 5 cm2. Active-edge sensors, without guard bands,are explored and characterised in this work to produce a large panel pixelated Cd(Zn)Te detector built of tiled modules with minimal gaps between them. The characterisation of an active-edge sensor fabricated using present processing technologies showed that 87 % of all edge pixels had excellent spectroscopic characteristics for X-ray imaging. However non-uniformities in the charge collection were observed in 34 % of the pixels. These were attributed to regions with poor charge collection efficiency up to 200 µm from the edge due to a low electric field strength near the edge that was caused by the high edge surface leakage currents. New techniques for the processing of the crystal edges were investigated with the aim of improving the sensitivity of the detectors up to the edge of the crystal. The leakage current was significantly reduced when the diced edges of CdTe sensors were lapped with a 3 µm alumina slurry followed by a polish with a 0.3 µm alumina slurry. This resulted in 60 % of edge pixels with excellent characteristics for X-ray imaging. The remaining 40 % presented poor spectroscopy due to damaged pixels, as a consequence of the difficulties in handling the 1 mm thick crystal whilst manually processing the edge surfaces. The polished and diced surfaces were illuminated edge-on, between the cathode and the anode, for the first time ever in CdTe. Poor detection and charge collection efficiency were observed within 12 µm from the polished edge surface and 80 µm from the diced edge surface. This was attributed to a high density of electron traps at the crystal edge due to dicing and processing that originated multiple trapping and de-trapping of charge carriers. This work concludes that active-edge CdTe radiation detectors are a promising technology for the production of a large Cd(Zn)Te radiation detector for X-ray imaging. The nonuniformities seen in the edge pixels are related to the high edge surface leakage currents due to the introduction of trap states during dicing. These are reduced by edge processing which creates active-edge pixels sensitive to radiation within 12 µm from the edge surface.