The pancreatic islets of Langerhans play a fundamental role in the stabilisation of blood glucose levels. Pancreatic islets are spherical structures composed of multiple cell types, with each individual cell type secreting a peptide hormone, such as insulin and glucagon, which regulates whole body glucose homoeostasis. Defective hormone secretion from islet cells is a hallmark of certain metabolic diseases, including type 2 diabetes mellitus (T2D). The most abundant islet cell type is the pancreatic β-cell, a specialised cell type that secretes the hypoglycaemic hormone insulin in response to raised glucose levels. Alterations in both β-cell mass and function are the causative factor for the development of T2D. Both genetic and environmental factors are known to underlie the decline in β-cell function typical of T2D. Specifically, genome wide association studies (GWAS) have identified over 100 genomic loci that are associated with T2D risk. Among these loci, variants that lie within/near ADCY5, SLC30A8 and PAX6 show associations with both T2D and abnormal glycaemic parameters typical of a diabetic phenotype. Therefore, the aims of this thesis were to understand how variants associated with T2D manifest at the level of the pancreatic islet. The expression of these genes was therefore manipulated through the generation of tissue-specific transgenic and knockout mice and by RNA interference in human tissue. Reducing the expression of ADCY5, encoding adenylate cyclase five, in human islet tissue reduced glucose-stimulated insulin secretion. This was accompanied by impairments in the metabolic and non-metabolic parameters that govern the secretory response of islets to glucose and other secretagogue. Deleting ZnT8, a β-cell zinc transporter and the gene product of the Slc30a8 gene, in the mouse β-cell significantly impaired the ability of these animals to mount effective responses to glucose. Interestingly, the reverse phenotype, i.e. improved glucose tolerance, was seen in animal models that overexpress ZnT8 in the β-cell. Finally, deletion of Pax6 in the adult mouse resulted in a drastic diabetic phenotype accompanied with changes in the cellular architecture of the islet and alterations in β-cell glucose signalling. Therefore, ADCY5, SLC30A8/ZnT8 and PAX6 gene variants likely negatively impact upon β-cell mass and function leading to a diabetic phenotype. Furthermore, these genes highlight distinct pathways, intrinsic to the pancreatic β-cell, which could be therapeutically targeted in the treatment of T2D.