Insulin resistance is a characteristic feature of type 2 diabetes mellitus (DM ) and is associated with increased cardiovascular disease. This thesis presents one randomised controlled trial and two cross-sectional analyses, examining the influence of fruit and vegetable consumption and endocrine factors (se~ hormone binding globulin and vitamin D) on insulin resistance. After a 4 week diet of 1-2 portions fruit and vegetables per day, 105 overweight (8MI 27 - 35kg/m2), nondiabetic subjects at elevated cardiovascular disease risk (>20% 1 O-year CVD risk), were randomised to follow a diet of 1-2,4 or 7 or more portions fruit and vegetables per day for the 12-week intervention. Insulin resistance was assessed pre and post intervention using a two-step euglycaemic hyperinsulinaemic clamp. In the cross-sectional analyses, baseline clamp assessments were correlated with sex hormone binding globulin concentration and with vitamin D concentration respectively. Increasing fruit and vegetable consumption was not associated with a change in insulin resistance. Sex hormone binding globulin concentration was inversely correlated with insulin resistance, independent of androgen concentrations and adiposity in the subgroup of 28 postmenopausal women. No association was demonstrated in men. Vitamin D was not associated with insulin resistance. In overweight people at high risk of cardiovascular disease, increased fruit and vegetable intake has no effect on insulin resistance. Weight was maintained in our study and it is possible that alterations in weight have a greater impact on insulin resistance than dietary composition. We suggest that beneficial effects of fruit and vegetable consumption are not mediated by change in insulin resistance. We demonstrate an inverse association between sex hormone binding globulin concentration and insulin resistance, primarily peripheral insulin resistance. We demonstrate no association between vitamin D and insulin resistance and suggest that an association between vitamin D and type 2 DM may be mediated via effects on insulin secretion.

The impact of dietary interventions on the functional aspects of HDL has been controversial and poorly studied; therefore, this thesis focused on the differing functions of HDL in subjects with diabetes, and if dietary interventions influenced this function in these subjects. An in vitro investigation into the effect of freezing on HDL stability found that SAA and PON-1 were stable when frozen in serum up to 12 months, although stability was reduced in HDL·. The opposite was true for CETP, suggesting that when an analyte is unstable in its matrix, analysis should be carried out on freshly isolated samples where possible or error minimised with a batch analysis. We also identified in the fasting state that subjects with TID displayed increased SAA- inflammation, while the augmentation of the HDL-associated enzymes to a more proatherogenic phenotype was influenced by both 'hyperglycaemia and obesity. Conversely, in the T2D pilot study, we found that both T2D and obesity did not influence SAAinflammation in these subjects, although obesity may have been responsible for the proatherogenic changes to the HDL-associated enzymes. Additionally, postprandial lipaemia did not further augment the proatherogenic phenotype in either cohort. We also identified antiatherogenic changes in HDL of a T2D cohort following a high F&V diet, suggesting that HDL in this group had increased antioxidant and RCT capabilities, which would ultimately reduce their risk of atherosclerosis and CVD. Overall, we have demonstrated that heightened inflammation can cause HDL dysfunctionality, reducing its ability to participate in RCT. Furthermore, this thesis has provided evidence for the importance of a high F&V diet to reduce oxidative stress and improve HDL functionality. Careful consideration of these results and implementing a revised diet could reduce the risk of CVD, and improve quality of life in those affected by diabetes.

This thesis examined the incidence and temporal trends of Type 1 diabetes diagnosed in Northern Ireland children using data from the Northern Ireland Childhood Diabetes Register (NICDR). Overall, there was evidence of a departure from linearity in incidence with indications that rates are levelling off in recent years. Further analyses also indicated that birth cohort effects were evident in the incidence rates suggesting that exposures in early life may play an aetiological role in this condition. A systematic review and meta-analyses was performed in this thesis to investigate the association of childhood vaccinations and subsequent risk of Type 1 diabetes. Twelve studies investigating a range of vaccinations were included. Results provided no evidence to suggest an association between childhood vaccinations and risk of Type 1 diabetes. A study using data from the Clinical Practice Research Datalink (CPRD) was included in this thesis to report findings on depression and other complications in young people diagnosed with Type 1 diabetes. This study found that rates of depression were significantly higher in cases with diabetes compared to controls without diabetes. Results also showed elevated rates of microvascular complications and significantly higher rates of cardiovascular disease compared to matched controls. Another focus of this thesis was on mortality in individuals with Type 1 diabetes. Population-based studies reporting relative mortality in Type 1 diabetes diagnosed in young people were systematically reviewed. In total, 23 independent studies were included. Associations between relative mortality and study/ country characteristics were explored. In addition to this review, a further two UK-based studies were performed to investigate mortality, one using data from the NICDR and the other using the CPRD. Both studies found excess mortality rates in individuals with Type 1 diabetes when compared, respectively, to the general population and to a group of controls without diabetes.

Allogeneic islet transplantation offers the possibility to treat selected patients with brittle Type 1 Diabetes Mellitus (T1DM). Limited availability of islets is a major obstacle to the more widespread use of islet transplantation as a therapy for the majority of patients with T1DM. This is exacerbated by extensive islet cell death during the early post transplantation period, which increases the number of islets required to achieve insulin independence. Additionally, disturbances to normal islet architecture and morphology, as well as suboptimal vascular engraftment during the post transplantation period, contribute to the long term decline in graft function. As well as substantial stresses to the islets during the post transplantation period, functional p-cell mass is lost during pre-transplant culture, further contributing to the inefficient use of valuable donor islets. Mesenchymal Stem Cells (MSCs) secrete a number of soluble trophic factors to affect neighbouring cells, making them excellent candidates for improving the survival of islet cells during culture and after transplantation. The overall aims of the studies described in this thesis were to investigate strategies to improve islet transplantation outcome. Using a syngeneic minimal islet mass model, it was demonstrated that MSC co-transplantation improved the rate and number of streptozotocin-induced diabetic mice attaining normoglycaemia by one month. The beneficial influence of MSCs was attributed to the maintenance of normal islet size and morphology, as well as increased rate and overall extent of islet revascularisation. Additional studies demonstrated that dispersing islets beneath the kidney capsule of diabetic mice produced superior transplantation outcome to that of islets which were implanted as a single pellet, confirming the importance of maintaining normal islet size and morphology at the implantation site.

The widespread use of islet transplantation as a treatment for type 1 diabetes is limited in part by the necessity for long term immunosuppression. The principle of immunoisolation is to create a physical barrier surrounding the islets to prevent immune destruction after implantation, therefore removing the requirement for immunosuppression. Alginate microcapsules are the most commonly used immunoisolation device which are typically implanted intraperitoneally. Intraperitoneal microencapsulated islet grafts are suboptimal, requiring at least double the number of islets to reverse hyperglycaemia compared to non-encapsulated islet grafts. The large size of microencapsulated islet grafts also restricts the selection of implantation site. We aimed to determine if the subcutaneous site is a suitable alternative site to the intraperitoneal site for microencapsulated islet grafts. It was discovered that equivalent microencapsulated islet grafts that were efficacious at the intraperitoneal site were ineffective at the subcutaneous site. Helper cells can be used to improve islet transplantation by secreting beneficial factors. L cells produce glucagon-like peptide-1 which is known to have several positive effects on islets. Co-encapsulation of islets with L cells increased islet insulin secretion but did not improve graft outcome. Mesenchymal stem cells (MSCs) can be co-transplanted to improve the graft outcome of non-encapsulated grafts. Islets co-encapsulated with MSCs had improved insulin secretion and also improved graft outcome. Immunoisolation of islets by conformal coating has the potential to maximise the diffusion of vital molecules and minimise the graft volume, enabling transplantation to preferred sites. Using allogeneic islets implanted at the kidney subcapsular site it was found that non-encapsulated grafts rejected whereas a novel conformal coating protected grafts from rejection in 5/7 of recipients. In summary, immunoisolated graft efficacy can be optimised by the selection of site, the use of helper cells, and with a novel conformal coating approach.

Plant-derived extracts have been used as folk remedies for Type 2 diabetes mellitus (T2DM) for many centuries, and offer the potential of cheap and readily available alternatives to conventional pharmaceuticals in developing countries. Extracts of Gymnema sylvestre (GS) and Costus pictus (CP) are reported to have antidiabetic activity in vivo. The exact molecular mode of action(s) of GS and CP is unclear but the antihyperglycemic effect seen in animal studies was associated with dramatic increases in insulin secretion so in my thesis I have examined the effect of alcoholic aqueous GS extract, named OSA®, on blood glucose and plasma insulin levels in vivo from animals and humans with T2DM and I have measured the effect of GS aqueous alcoholic and CP methanolic extracts on insulin secretion in vitro from the MIN6 β-cell line and isolated mouse and human islets. The in vivo data showed that OSA® increased insulin levels and reduced blood glucose in patients with T2DM and ameliorated glucose intolerance in animal model of diabetes. The in vitro data demonstrated that OSA® and CP have a direct stimulatory effect on insulin secretion which was not associated with compromised membrane integrity or decreased β-cell viability. OSA®-, but not CP-, induced insulin secretion was coupled with elevations in insulin gene expression suggesting that GS may reserve β-cell insulin store following chronic stimulation of insulin. Single cell calcium microfluorimetry measurements showed that OSA® and CP elevated intracellular Ca2+ concentrations ([Ca2+]i) in Fura-2-loaded β-cell, an effect which was completely abolished by the removal of extracellular Ca2+ or blockade of voltage-gated Ca2+ channels (VGCC). These in vitro observations suggest that one mode of action of OSA® and CP is through stimulating insulin secretion which may be mediated, in part, by the ability of OSA® and CP to increase [Ca2+]i levels through VGCC. In addition, OSA®-induced insulin secretion was partially associated with protein kinase activation and was independent of classical PKC, CaMKII or cAMP activation. OSA® was also found to protect β-cells from cytokine-induced apoptosis. Our measurement of caspase-3 and -7 production showed partial reduction in OSA®-treated MIN6 cells and mouse islets following cytokines exposure. These data were further supported by our microarray analysis of mouse islets when challenged with cytokines in the presence of OSA®. Enrichment analysis indicated that OSA® reduced caspase-3 gene expression most likely through activation of PI3K/AKT pathway and subsequent downstream effectors including MnSOD.

This thesis explores three broad areas of interest in the pathophysiology of type 2 diabetes (T2D). The first part of the thesis examines the relative contributions of body fat measurements on T2D and related morbidities in a large cohort of twins. A proxy measure of visceral fat was constructed from anthropometric and dual-energy X-ray absorptiometry, its heritability was estimated at 58% using the classical twin model and its influence on morbidity was compared to total abdominal fat and the body mass index. The findings from this work show that intra-abdominal adiposity confers the greatest independent risk on morbidity and appears to almost entirely mediate the observed association between morbidity and all the other measures of adiposity investigated. The second part of this thesis is a candidate gene study of the PARL/ABCC5 gene region motivated by prior evidence suggesting a role for PARL in T2D susceptibility. Using a single marker test of association, SNPs in and around the PARL gene showed no evidence of association with T2D. However, analysis based upon SNPs in the entire gene region (184,743-185,548Kb, build 36) using a multi-marker test of association, provided strong evidence that the neighbouring gene (ABCC5) is associated with T2D in both European and African American samples. In addition, ABCC5 expression in subcutaneous adipose tissue was strongly associated with fasting insulin and glucose serum levels, visceral fat accumulation, and T2D with evidence that the disease susceptibility variant(s) is a regulatory element (an expression quantitative trait locus) located at intron 26 in ABCC5. The third component of this thesis is a comprehensive investigation into the potential role of nuclear-encoded mitochondrial (NEM) genes in the aetiology of T2D. A pathway analysis approach is used to test for enrichment of T2D association signals across the genome in defined NEM gene sets. From this analysis, the biological pathways of glycolysis, the tricarboxylic acid cycle, and mitochondrial translation all show evidence of pathway enrichment. These findings demonstrate for the first time, potential associations between these pathways and T2D susceptibility in European and African American samples.

Peroxisome Proliferator-activated Receptor-y (PPARy) is a ligand-activated transcription factor responsible for controlling genes involved in lipid and glucose metabolism. The thiazolidinediones (TZDs) are a class of synthetic PPARy ligands used as anti-hyperglycaemic agents in the treatment of type 2 diabetes. Recently it has been shown that these agonists have additional anti-inflammatory properties beyond their metabolic functions. This study evaluated the role of TZDs in monocytic cell activation and endothelial function. Human monocytic cell lines and peripheral monocytes were stimulated with various advanced glycation end-products (AGEs) to mimic the chronic inflammatory hyperglycaemic state typical of diabetes. Release of cytokines, such as TNF-α, was determined as indicators of inflammation. Endothelial function was studied using rabbit aortic rings and primary human aortic cells PPARy agonists, GW7845 and rosiglitazone, were observed to significantly reduce AGE-induced TNF-α expression and release. It was demonstrated that this reduction was not completely dependent on PPARy transcriptional activity, as PPARy antagonists did not negate the actions of the agonist. The anti-inflammatory properties of PPARy agonists appeared to be (i). time-dependent, (ii) dose-dependent (iii) cell type specific and (iv) involve two distinct pathways: PPARy-dependent and PPARy-independent. To elucidate the molecular mechanisms controlling the observed inhibition of TNF-α release the effect of PPARy agonists on the activation of ERK and NF-кB was investigated. It was found that rosiglitazone reduced gBSA-induced ERK activity, although basal levels remained unaffected. Surprisingly, rosiglitazone did not appear to modulate glycated BSA-stimulated NF-кB DNA binding activity. Conversely, rosiglitazone increase basal levels of NF-кB DNA binding activity but this did not result in NF-кB-dependent TNF-α gene expression. It was postulated that this up regulation of NF-кB might interfere with signalling processes necessary for inflammatory responses. The effect of PPARy agonist on receptor of AGE (RAGE) expression was studied at both transcriptional and translational level. PPARy agonist did not increase mRNA expression for RAGE, however the protein expression was increased in the presence of the agonist after 24hrs, indicating that the regulation of RAGE by the agonist may be at the post-transcriptional and/or post-translational levels. PPARy is also expressed in endothelium and vascular smooth muscle cells. Therefore this study also investigated the role of agonists on the vascular tone of the vessel wall in rabbit aortic rings. It was determined that GW7845 (>20µM) inhibited both phenylephrine-induced contraction and acetylcholine-induced relaxation, whereas sodium nitroprusside (SNP)-induced relaxation was slightly improved. This enhancement of SNP induced relaxation may indicate an increased sensitivity of VSMCs to the exogenous nitric oxide. GW7845 did not influence the NO release and eNOS expression in human primary endothelial cells. It was postulated that these direct vasomodulatory effects involve cGMP signalling and calcium trafficking in smooth muscle cells. Taken together these findings indicate that PPARy agonists can reduce the inflammatory processes of monocytic cells. However the mechanisms involved appear to be influenced by both PPRE-dependent and PPRE-independent pathways and the net anti-inflammatory and anti-atherogenic clinical benefits of TZD therapy are likely to be governed by the interactions and balance between these two pathways.

Graves’ disease (GD) is an antibody-mediated autoimmune disease caused by thyroid stimulating antibodies (TSAbs) activating the thyrotropin receptor (TSHR). Both genetic and environmental factors contribute to pathogenesis and one such factor is the infectious agent, Yersinia enterocolitica. Two high affinity TSAbs, (KSAbl, lgG2t>/K; KSAb2, IgGaa/K), developed from an experimental murine model of GD, share common germline Ig genes and thus are derived from the same precursor B cell clone which diversified through somatic hypermutation. The shared germline genes were expressed as recombinant (r) Fab germline and did not display measurable binding to TSHR in three assay systems: radioreceptor assay, flow cytometry and cAMP stimulation bioassay. Hence the precursor B cell clone was not endowed with autoreactive potential. To assess the individual contribution of KSAbl’s heavy and light chains to TSHR reactivity chimeric rFab constructs were created. The interaction and stimulation of TSHR was dependent on the mature heavy chain. Significantly, antigen recognition studies of rFab germline, KSAbl and KSAb2 showed binding to OmpA, OmpC and OmpF of Y. enterocolitica. Our findings raise the possibility that the clonal expansion of B cell populations that acquire autoreactivity in the periphery may be associated with the porin proteins of Y. enterocolitica.

No description available.