Beta cell replacement is the only therapy currently able to restore insulin independence in C-peptide negative Type 1 diabetes. This can be achieved through either vascularised whole pancreas transplantation or isolated islet transplantation. Whole pancreas transplantation can deliver long-term normalisation of blood glucose but nevertheless is associated with significant peri-operative morbidity and mortality. Islet transplantation is a much less invasive procedure which can provide complete resolution of disabling hypoglycaemia with potential insulin independence. However, at the outset of this research, this procedure was restricted to UK centres with islet isolation facilities. Two main hurdles prevent wide application of this treatment modality. First, a limit to the number of isolation facilities that can be established. These are sophisticated and very expensive laboratories and require extensive staff training and commitment. Second, there is significantly limited supply of suitable donor pancreases meeting less than 1% of the potential clinical need. The goal of this PhD project was to address, from the UK perspective, the limitations in transplantation sites and scarcity of suitable islets for clinical transplantation maintaining a clinically relevant focus through the study of primary human islets. The overall aims were: 1. To demonstrate safe and effective islet transport despite relatively long journey time in the UK without need for complete revalidation / repreparation for transplant at satellite site. 2. To evaluate potential for enhancing mass and function in intact human islets including the potential of in vitro incubation with pregnancy-related hormones 3. To evaluate potential for in vitro expansion of human pancreatic cells with determination of whether a functional beta-cell phenotype can be maintained over passage and whether putative pancreatic stem cells can be identified in culture. 4. To determine potential for redifferentiation of expanded human pancreatic cells towards a functional beta-cell phenotype by in vitro pseudo-islet formation. III A safe, practical and clinically viable islet transportation system was successfully established to suit the UK setting. This system enabled efficient utilisation of central islet isolation facilities for production of clinical grade high quality islets for transportation to several islet transplantation centres. Cooling a relatively small volume of high density islets in transport bags enabled transported islets to be transplanted at the satellite centres without any further manipulation and repreparation. This work underpinned government funding of a national islet transplantation program and led to the first successful UK islet transplantation of transported islets. Validated standard operating procedures (SOPs) created in this project were adopted nationally by all islet receiving centres. Augmentation of islet mass and function in intact primary human islets was attempted following treatment with pregnancy related hormones and a panel of other growth factors. In contrast to previously published reports on rodent islets or adherent human islets, these studies confirmed that growth factor treatment can only maintain as opposed to increasing islet mass or function in intact human islets. Nevertheless, prolactin induced a significant increase in insulin expression and potentiated insulin secretion in response to elevated glucose concentrations. Proliferative cultures from pancreatic islets were established and characterised achieving high proliferative capacity with potential to produce transplantable mass sufficient for several recipients from a single donor. Despite maintenance of differentiated phenotypic markers in early passages there was an accelerated loss of β-cell function. Nevertheless expression of pluripotency-associated markers was identified for first time suggesting stem cells residing in adult pancreas. Potential for enhanced β-cell function in islet proliferative culture with minimal manipulation was demonstrated through pseudo-islet formation. This was associated with down-regulation of pluripotency-associated markers and enhanced β-cell function in vitro IV and limited, but confirmed activity, in vivo. However, currently this remains limited and insufficient for clinical impact. In conclusion, validation of an effective and safe islet transport system was achieved underpinning unique National Health Service funding of a UK national clinical programme of islet transplantation and enabling the first successful UK transplantation of transported islets. Definitive studies to demonstrate functional enhancement in human islets in response to pregnancy related hormones and/or other factors would require toxic pre-conditioning. Significant islet mass expansion was attained in vitro with potential functional enhancement with pseudo-islet formation. However, further differentiation studies and validation of a practical large scale culture system remain an important requirement for any potential future clinical application.

Adolescence has been shown to be a critical time for effective self-management of insulin-dependent diabetes mellitus (IDDM) with many individuals struggling to achieve optimal levels of self-care. This study explored the experience of adolescent life with IDDM via the use of qualitative interviews, the transcripts of which were analysed using Interpretative Phenomenological Analysis (JP A). Ten adolescents between the ages of 16 and 19 years were interviewed, using'a semi-structured interview technique. Three recurrent themes were elicited across the data set and are presented within this thesis. These are: the view of IDDM as an illness, the emotional cost of IDDM, and the relationship with health care professionals. These themes allow a dialogue to continue between health-based research and that of service provision. The first theme, 'View of IDDM as an Illness', demonstrated that IDDM was often perceived as being a complex, unpredictable and at times difficult to manage illness that demanded routine. It was also perceived as incurring self-management challenges including those that were more social in nature (for example, restricting certain aspects of adolescent life). The second theme, 'The emotional cost of IDDM', indicated that experience of negative emotions, anxiety, fear and frustration in particular, was commonplace for a number of the adolescents. The source of these emotions was widespread and included peer interactions, how IDDM was perceived as an illness, difficulties inherent in achieving and maintaining good metabolic control and also those that related to the third theme, namely 'Relationship with Health Care Professionals' . Within this third theme, factors such as trust and entitlement to 3 information regarding IDDM were core features, with negative emotional states being induced when these needs were not met. Inherent within all of these themes and sub-themes was the role of emotions. They were found to be central determinants of not only the ways in which IDDM was represented but they were also seen to guide coping in terms of self-management decisions and behaviours. When considered in terms of the current literature a number of similarities and distinctions were found. Current predictive models commonly treat each construct as independent units. The present results show that complex relationships may exist, often with considerable emotional associations. Crucially, it is proposed that the metabolic implications of such emotional experience are not given sufficient attention in the current literature. With regards to ways in which the adolescents coped with the emotional experience of IDDM, the present fmdings demonstrate that aspects from a number of models of stress and coping could be applied but that one model on its own was insufficient. Any future model has to appreciate the impact of emotions in not only initial appraisals of health threats but also their contribution to subsequent appraisals of coping dependent on the behavioural outcome, in addition to the emotional outcomes and the impact of these on IDDM-specific cognitions, self-management and metabolic stability.

This thesis describes research into telemedicine systems and associated decision support algorithms for the self-management of Type 1 diabetes. It considers how a patient can improve or maintain their blood glucose control, and how technology (with appropriate support from healthcare professionals) can assist them.

Background New Onset Diabetes After Transplantation (NODAT) is an important complication of renal transplantation associated with decreased graft and patient survival. This study aimed to validate established physiological measures of glucose homeostasis in the renal transplant population and use these techniques to establish the range of glucose handling in patients pre-transplant and the impact of transplantation on glucose handling. The main impact of NODAT is an increase in cardiovascular risk. Cardiovascular assessments were therefore performed in all patients in parallel with assessments of glucose homeostasis. Patients and Methods The study involved two distinct cohorts of patients: Patients identified retrospectively by an electronic database and a prospective cohort of patients on the live kidney donor waiting list whom were in the final stages of work-up for listing for a date for surgery. The retrospective cohort were analysed to establish epidemiological factors that predict NODAT and determine the natural history of NODAT in patients treated at our unit. A sub-group of this cohort was those who appeared to have recovered from NODAT on the basis of measurement of fasting plasma glucose, random plasma glucose or HbA1c. These patients underwent an oral glucose tolerance test (OGTT) to verify their diabetes status and perform the Homeostasis Model Assessment (HOMA) to estimate their beta cell function (%8) and insulin sensitivity (%S). The prospective cohort underwent an OGTT before transplant (n=18). 12 of the cohort also underwent a euglycaemic hyperinsulinaemic clamp study and 5 an arginine intravenous glucose tolerance test. Repeat testing was performed at 3 months. Finally cardiovascular assessment was performed in all patients by measurement of augmentation index and mean arterial blood pressure. Results The retrospective analysis of 306 consecutive renal transplants over 3 years demonstrated age at time of transplant and family history of diabetes to be predictors of development of NODA T. Also 10 of 12 individuals thought to have resolved from NODAT as judged by random and fasting glucose measurement and off hypoglycaemic and insulin treatment were shown to still have abnormal glucose handling, 8 were still diabetic as judged by a OGTT. Within the LKD cohort (n=18) the pre-transplant OGTT demonstrated 4 participants to have IGT and 2 participants to have diabetes. Glucose disposal rate (GDR) fell Significantly from pre to post transplant. Discussion This study has shown that insulin resistance is increased at three months post live donor kidney transplant. There was a trend also for pancreatic beta cell function to decline. Cardiovascular risk factors were shown to have not changed by three months. Our long-term objective is to create an algorithm that will enable clinicians to define the overall individual risk of developing NODAT. This will result in more rigorous surveillance of the individual and, with development of novel regimens, tailoring of post-transplant immunosuppression to ensure optimum protection from rejection and avoid the significant cardiovascular morbidity and mortality associated with developing diabetes. Resolution of NODAT should be based solely on an OGTT and long term surveillance offered as part of the graft follow-up. ii.

Maternal microchimerism (MMc) results from transfer of maternal cells to the fetus during pregnancy. These cells persist into adulthood in peripheral blood as well as multiple parenchymal organs and high levels of MMc have been associated with Type 1 diabetes mellitus (TID). To understand the potential importance of MMc in T1D the aims of this PhD were 1. to develop strategies to detect and phenotype MMc in healthy human pancreas 2. to compare MMc frequency and phenotype in healthy and T1D pancreas 3. to develop strategies to confirm the 'maternal ' origin of MMc at the molecular level MMc positive for markers of fully differentiated endocrine, acinar, and ductal cells were identified in healthy human pancreas. The frequencies of insulin+ MMc varied from 0.05% to 0.8%. Some MMc in neonatal pancreas were positive for Ki67, indicating their potential for replication. MMc were detected in fresh islets, as well as in islet-derived cells. In male T1D pancreas, no female CD45+ cells were observed in insulitis, indicating that MMc are not effector cells of the autoimmune response. Laser capture microdissection of FISH-labelled candidate cells from frozen sections was successful. FISH did not interfere with the subsequent genotyping. PCR efficiencics of pooled cells (10) vs. single cells were 100% vs. 23.1%. Additionally, a separate aim of this project was to determine whether adult human islets were amenable to immortalisation using an SV40lhTERT retroviral system and characterisation of an islet derived cell line is described. These studies indicate that some maternal cells have multilineage potential. Increased levels of MMc in TlD may represent a failure of anti-maternal tolerance or contribute to attempted tissue repair but they do not facilitate host beta cell attack. Further studies are required using the strategies developed in this study to better define the role of MMc in T1D.

In spite of major developments in insulin production and methods of delivery, problems remain in the day to day management of insulin treated diabetes. One striking difference between currently available insulin treatments is the relative difference in weight gain and hypoglycaemia frequency. With the use of stable isotope technology the differential effects ofO.5units!kg of subcutaneous insulin detemir and NPH insulin on glucose flux, lipid metabolism and brain function during hyperglycaemia were investigated. Two protocols were designed, a protocol to look at the action of insulin following a period of insulin withdrawal and a second protocol using the gold standard euglycaernic and hyperglycaemic hyperinsulinaemic clamp methodology.

In recent times the relationship between insulin resistance and endothelial dysfunction has been closely examined and is thought to be pivotal in the pathophysiology of atherosclerosis in patients with diabetes. Interestingly, epidemiological data seems to associate low IGF-1 levels with adverse cardiovascular outcomes. Given the known structural and functional similarities between insulin and IGF-1 pathways, I investigated the effects of IGF-1 on endothelial function, by concentrating on the role of its receptor. During the course of this project, four major mouse models were used. A model of diet-induced obesity allowed in vivo and ex vivo studies 10 explore metabolic and vascular effects of IGF-1 in the lean and high-fat state. Following a key finding of IGF-1 R depletion in these obese mice, similar studies using a no obese mouse model of whole-body and endothelial-cell specific IGF-1 R insufficiency were conducted, to separate the effects of IGF-1 R down regulation from those of obesity. These studies revealed that a reduction of IGF-1 R numbers, at a whole-body and endothelial-cell specific manner leads to enhanced basal and insulin-stimulated nitric oxide (NO) bioavailability, with . evidence of a gene dose effect. Furthermore reducing IGF-1 R numbers leads to a reduction of hybrid receptors in vivo and in vitro, thereby pointing to a possible mechanism for the observed findings. Finally to assess whether modulating IGF-1R would rescue vascular insulin sensitivity, mice with haploinsufficiency of the IR were bred with mice with haploinsufficiency of IGF-1R. This showed that reducing IGF-1R and hybrid receptors in vivo, restored vascular insulin sensitivity in a murine model of vascular insulin resistance. Therefore these studies have identified IGF-1 R as a negative regulator of NO bioavailability and insulin sensitivity in the endothelium.

Diabetes mellitus has reached epidemic proportions and affects more than 200 million people world-wide. It is a heterogenous metabolic disorder, defined by the presence of hyperglycaemia. This study sought to use the application of proteomics to elucidate the mechanisms involved in glucose stimulated insulin secretion in MIN6 pancreatic p--cells, by determining and identifying differential expression of proteins in response to glucose stimulation (i) in a global context, and (ii) at the insulin-containing vesicle level. Optimisation of 2-dimensional gel electrophoresis (2DGE) resulted in the determination of the MIN6 p-cell proteome and the identification of 84 different proteins. Many proteins were found to be differentially expressed as a result of glucose stimulation, reflecting the enhancement of insulin exocytosis. These include the up-regulation of molecular chaperones involved in protein biosynthesis, such as T complex protein, which mediates folding of actin and tubulin, and heat shock cognate protein 70, which is involved in uncoating of clathrin coated vesicles. Proteins known to be involved in the transport of insulin- containing vesicles from the storage pool to the readily releasable pool, such as actin, tubulin B5 and tubulin alpha were also identified to be up regulated. For an increase in insulin biosynthesis to manifest itself as a rise in glucose-stimulated insulin secretion, translocation of insulin-containing vesicles from the Golgi region to the plasma membrane, and maturation of vesicles are required. Proteomic analysis of the insulin-containing vesicle was undertaken to obtain a better understanding of vesicle regulation. To date 24 proteins have been identified including the heat shock cognate protein - involved in uncoating of clathrin- coated vesicles; vacuolar ATP synthase subunits, which maintain a low intravesicular pH, important for proinsulin conversion and storage of zinc-insulin hexamers, and glucose-regulated proteins 54, 78 and 94. Several proteins not directly attached to vesicles, but involved in vesicle cycle regulation co purified with vesicles for example actin and tubulin. Both actin and tubulin form core complex with synapsin proteins, synaptotagmin and SNAP-25 during vesicle docking thus explaining their presence on insulin vesicles. Proteomic analysis of the MIN6 pancreatic p-cell also identified translationally controlled tumour protein (TCTP) to be significantly up-regulated in response to glucose stimulation, leading to further investigation into its role and function. This is the first study (i) to identify the presence of TCTP in the p-cell by proteomic analysis and (ii) to investigate the regulation and function of TCTP in the pancreatic p-cell. From the data presented TCTP has been identified as a novel glucose-regulated protein, whose regulation involves post-translational modification by both phosphorylation and glycosylation. These post-translational modifications are regulated and induced by glucose, and have been demonstrated to affect its cellular localisation, with an increase in the nuclear localisation of TCTP in response to glucose stimulation, as well as increased localisation with the cytoskeletal protein actin. Additionally, reducing TCTP protein expression using siRNA resulted in enhanced p-cell death. Preliminary studies also suggest that TCTP may be involved in the process of insulin secretion, with a significant decrease in insulin secretion observed as a result of reduced TCTP protein expression. Data presented in this study illustrate that TCTP is not only vital to p-cell survival, but potentially is a key player in GSIS, at multiple levels; from the regulation of gene transcription, to the trafficking and release of the insulin-containing vesicles.

Islet transplantation holds vast potential as a treatment for type 1 diabetes mellitus and provides recipients with short term insulin independence. A major limitation of this treatment is the lack of donor beta (β) cells available for transplantation. Significant progress has been made with stem cell differentiation protocols; current methods have generated cell populations which possess a functioning β cell phenotype. However, these cells are not suitable for clinical transplantation. Two dimensional cell culture systems do not accurately mimic the complexity of the in vivo pancreatic environment, reducing the effectiveness of current β-cell differentiation protocols. The paradigm shift into three dimensional tissue culture provides an attractive area of investigation, and the use of three dimensional culture methods has improved growth in a variety of cell types ex vivo. The mass culture of β cell analogues on a 3D biomimetic environment is now necessary, and may offer a new platform on which an alternative source of transplantable cell populations can be differentiated and cultured successfully. This thesis aims to develop and condition BioVyon™, a high density polyethylene (HDPE) based biomaterial, for use as a mass β cell cultivation system In order to achieve this, a number of objectives will need to be met: (I) Complete characterisation and assessment of all properties of Biovyon™ that have a direct influence on cell culture, (ii) modification of the BioVyon™ surface chemistry to promote cell adhesion and growth, (iii) absorbance of proteins to mimic the pancreatic environment and aid proliferation of the Min-6 cell line, (iv) assessment of the Min-6 cell line phenotype after extended period of culture on the modified BioVyon ™ environment. Scanning electron microscopy and Atomic Force Microscopy were used to characterise the surface of the HDPE material post plasma etching. Advancing and receding (ARCA) and Fourier Transform Infrared Spectroscopy were used to analyse the elemental changes to the polymer surface. The HDPE biomaterial was conditioned using plasma etching, subsequent adhesion and growth of Min-6 cells was quantified using a lactate dehydrogenase assay. Min-6 populations were seeded at a density of lxl0/6i on BioVyon™ and tissue culture plastic of comparable surface area, and analysed after extended periods of growth. An insulin EUSA was used to quantify insulin released by populations of β cells at different time points witin the BioVyon™ in response to fluctuating glucose concentrations. The results obtained in this thesis indicate that BioVyon™ offers an appropriate structural environment for cell culture. Pore size and frit dimensions allow for cell infiltration and the effective diffusion of oxygen and nutrients. Plasma etching incorporated oxygen groups and a novel surface topography that improved cell adhesion and growth. β-cell phenotype was protected and sustained in cell populations cultured within the BioVyon™ environment. In conclusion, BioVyon™ can be conditioned to function as an effective 30 cell culture system. Modification of the surface chemistry has enabled BioVyonTl • to harbour and sustain large populations of the Min-6 cell line. The protected β-cell phenotype in the Min-6 populations suggests BioVyon™ could hold potential as a stem cell culture and differentiation platform.

Exposure to microorganisms is considered an environmental factor which can contribute to diabetes mellitus via cytotoxicity or autoimmune responses against pancreatic cells. Firstly, the effects on rat insulinoma pancreatic β-cell line of secondary metabolites pyrrolnitrin (Burkholderia spp), phenazine compounds (Pseudomonas aeruginosa and Burkholderia spp) were investigated. Both compounds separately showed significant cytotoxicity after 24 h and at concentrations of 10 & 100 ng/ml potentiated insulin gene transcription, Ca2+ content and glucose-stimulated insulin secretion (GSIS). Furthermore, the outward membrane current was inhibited by phenazine (100 ng/ml) or pyrrolnitrin (10 or 100 ng/ml). Secondly, the capacity of 45 microbial species to bind insulin was screened in order to assess how common insulin binding was amongst microorganisms Burkholderia multivorans, B. cenocepacia and Aeromonas salmonicida bound insulin. A genomic library of B. multivorans was constructed in λ Zap Express and screened successfully for insulin binding recombinants. Recombinant phagemids p1 & p2 were excised, p1, encoded an insulin binding protein (IBP1 30 kDa) with homology to the iron complex siderophore receptor. For p2, two IBPs were detected at 20 & 30 kDa (IBP2 & IBP3), representing an intracellular and outer membrane peptide transporter. Comparison of IBP1 and human insulin receptor (HIR) produced 6 linear epitopes, and for IBP2 & IBP3 produced 3 epitopes. Thirdly, glutamic acid decarboxylase GAD65 is a major pancreatic autoantigen contributing to autoimmune diabetes. To assess the likelihood that microorganisms possess epitopes that mimic regions on GAD, 45 microbial species were tested for homology. This was facilitated by purifying recombinant GAD protein which was used to produce GAD antiserum. Four E. coli cross-reacting proteins were identified using mass spectrometry, outer-membrane protein A, formate dehydrogenase, superoxide dismutase and DNA starvation protein. Epitopes occurred at the C-terminal region of GAD65 (residues 419–565), a region previously reported to be targeted by autoantibodies. This study suggests that pyrrolnitrin and phenazine are cytotoxic to pancreatic β-cells and B. multivorans IBPs linear epitopes may be diabetogenic, particularly in patients with cystic fibrosis related diabetes (CFRD) who suffer a long term infections with Pseudomonas and Burkholderia species. Furthermore, microbial GAD epitopes could potentially induce an autoimmune response leading to diabetes.