The glycaemic environment within the pancreatic beta cell is reflective of prevailing, circulating glucose conditions. Furthermore, within the pancreatic beta cell, glucose and glucose metabolites provide a suitable environment for non-enzymatic glycation to occur. Insulin which is produced and secreted from the beta cell has been shown to readily undergo glycation and is detectable within the beta cell. Under hyperglycaemic conditions, such as in diabetes, it has been proposed that the glycation of insulin is augmented and as such will provide a novel means of identifying diabetes. This thesis examines the secretion of glycated insulin from healthy and diabetic individuals, how secretion is modulated following treatment with anti-diabetic medication using a novel immunoassay and furthermore, proposes potential mechanistic routes through which glycated insulin affects insulin secretion. Pre-clinical studies in pancreatic BRIN-BDII cells and murine islets have demonstrated an autocrine inhibitory effect of insulin release. At physiologically relevant intra-islet concentrations, insulin suppressed the insulin-secretory responses of a range of secretagogues acting at different points in the beta cell stimulus-secretion coupling pathway including tolbutamide (43% inhibition), alanine (45% inhibition), IBMX (35% inhibition), forskolin (28% inhibition), and arginine (41 % inhibition). In addition, alanine and IBMX- up-regulation of insulin gene expression was decreased in the presence of insulin. Furthermore, insulin decreased gene expression of the potassium channel pore forming proteins, KIR6.2 and SUR-I, and also expression of the insulin receptor. A similar inhibitory effect on insulin release and through tolbutamide-mediated pathway was demonstrated with glycated insulin albeit to a lesser degree. Insulin (10 urnol/l) inhibited its pancreatic secretion by 65% at high glucose concentrations compared to glycated insulin which exhibited 77% inhibition at the same concentration. Importantly, glycated insulin also inhibited the autocrine release of insulin and had diminished action through the tolbutamide stimulated pathway, which may contribute to insulin resistance. Furthermore, these studies suggest inhibition by glycated insulin and insulin may be mediated through impaired membrane depolarisation. As part of this thesis, an ELISA was optimised for the quantification of circulating glycated insulin in a number of human clinical studies to assess glycated insulin as a novel biomarker for diabetes. Following optimisation, the assay was determined to be specific (spike recovery 90-97.5%, (ECso 0.37 nmol/l) and sensitive «19.1 pmol/l) for glycated insulin detection. The assay was found to be reproducible with intra and inter-assay variations of 7.9% and 8.9%, respectively. In vitro studies in pancreatic beta-cells demonstrated augmented insulin release and suppression of glycated insulin release following treatment with metformin at hyperglycaemic conditions. Furthermore, the combination of metformin with GLP-1, further augmented metformin induced insulin release while glycated insulin release remained suppressed. These findings highlight an important supplementary benefit of metformin. In clinical studies, this thesis has for the first time, demonstrated circulating glycated insulin release to be raised 3-fold in type 2 diabetic subjects and 2-fold in pre- diabetes confirming a role for glycated insulin in identifying asymptomatic diabetes and those with impaired glucose tolerance. Importantly, in studies of pregnant women, circulating concentrations of glycated insulin were found to be augmented lO-fold and may present a novel means of monitoring the beta-cell expansion during pregnancy. In type 2 diabetes, glycated insulin exhibited a rapid release from the beta-cell following a mixed meal. Circulating glycated insulin concentrations were found to be similar between fasting or non-fasting however post-prandial sampling should be carried out at least 30 min following meal ingestion to avoid any post-prandial spikes. To further examine the drug induced changes to glycated insulin secretion in a clinical setting, the effect of anti-diabetic drugs on post-prandial glycated insulin were assessed in pre-diabetic and type 2 diabetic subjects. Metformin reduced glycated insulin excursion by 48% in pre-diabetes and 60% in diabetes. Additionally, Nateglinide inhibited the glycated insulin excursion by 65% in type 2 diabetes. Studies herein have demonstrated that glycated insulin is dynamically secreted from the beta-cell in pre-diabetes, diabetes and gestational diabetes, and is affected by anti-diabetic therapeutic agents. Collectively these studies confirm a putative role for glycated insulin as a biomarker for diabetes however further work to investigate the natural history of glycated insulin secretion, in diabetes is required.