An important mechanism involved in the pathogenesis of type 2 diabetes is elevation of plasma free fatty acids which induce insulin resistance and may impair both β-cell function and mass (β-cell lipotoxicity). The objective of my thesis was to investigate the role of oxidative stress in β-cell lipotoxicity, using in vivo, ex vivo, and in vitro models. I used in vivo models of 48h i.v. oleate or olive oil infusion in Wistar rats followed by hyperglycemic clamps, or islet secretion studies ex vivo, and in vitro models of 48h exposure to oleate in isolated islets. My first study showed that 48h oleate infusion decreased the insulin response to a hyperglycemic clamp, an effect prevented by coinfusion of the antioxidants N-acetylcysteine and taurine. Similar to the findings in vivo, 48h infusion of oleate decreased glucose stimulated insulin secretion (GSIS) ex vivo, and induced oxidative stress in isolated islets, effects prevented by coinfusion of the antioxidants N-acetylcysteine, taurine, or tempol. Islets exposed to oleate or palmitate showed a decreased insulin response to high glucose and increased levels of oxidative stress, effects prevented by taurine. Therefore, my data are the first demonstration that oxidative stress plays a role in the decrease in β-cell secretory function induced by prolonged exposure to FFA, in vitro and in vivo. My second study addressed downstream effects of oxidative stress involving inflammation. A 48h infusion of oleate or olive oil decreased β-cell function during a hyperglycemic clamp, an effect prevented by coinfusion of the IKKβ inhibitor salicylate. GSIS in isolated islets was impaired by olive oil or oleate and restored by salicylate. These results suggest a potential role for both oxidative stress and inflammation in lipid-induced β-cell dysfunction. My third study addressed downstream effects of oxidative stress involving β-cell insulin signalling. A 48h infusion of oleate or olive oil decreased β-cell function during a hyperglycemic clamp, an effect prevented by coinfusion of the tyrosine phosphatase inhibitor bisperoxovanadate. GSIS in isolated islets was impaired by olive oil or oleate and restored by bisperoxovanadate, suggesting a role of FFA in decreasing β-cell function by induction of β-cell insulin resistance.

PI3K signaling in pancreatic β cells has been shown to be important in modulating β cell mass and function under basal condition. Evidence suggests that a specific group of insulin promoter-expressing neurons also modulates glucose metabolism and energy homeostasis through their PI3K signaling. Thus we hypothesize that PI3K activation via PTEN deletion under the control of rat insulin promoter (RIP) in pancreatic β cells and RIP-expressing neurons will protect against hyperglycemia and diabetes in experimentally induced mouse models of type 2 diabetes. In Chapter IV, we showed that RIP-mediated PTEN deletion in pancreatic β cells led to PI3K activation and subsequent increased β cell mass and function, thus protected the mice from high fat diet (HFD)-induced diabetes. Furthermore in the absence of global leptin signaling, β cell-specific PTEN deletion maintained β cell function in the setting of severe insulin resistance, therefore prevented diabetes development. Interestingly, RIP-mediated PTEN deletion also resulted in increased peripheral insulin sensitivity due to PI3K activation in central nervous system. In Chapter V, we showed this increased insulin sensitivity was maintained after HFD feeding, which also contributed to the protection against diabetes. These mice also showed increased visceral adipogenesis and subcutaneous adiposity on HFD, which were dramatically attenuated in the absence of leptin signaling, indicated the essential role of peripheral leptin action in mediating the insulin sensitive phenotype from neuronal RIP PTEN deletion. Finally, we demonstrated that the insulin sensitizing phenotype in these mice was not mediated through ventromedial hypothalamic nuclei (VMH), such that VMH-specific PTEN deletion did not alter energy homeostasis or glucose metabolism. Together, the data from this thesis points to an inhibitory role of PTEN in both central nervous system and pancreatic β cells in glycemic control. Therefore, PTEN may represent a potential target for diabetes prevention and treatment.

Uncoupling protein-2 (UCP2) may be linked to the attenuation of reactive oxygen species (ROS), but it is unclear whether this phenomenon pertains to the pancreatic beta cell. In this study, a UCP2-deficient mouse model was used to assess the importance of UCP2 to beta cell viability. We investigated the effect of UCP2 absence in response to a beta cell cytotoxic model of diabetes induction. In vivo treatment by the cytotoxic agent streptozotocin led to overall beta cell loss, but severity was not exacerbated by UCP2 deficiency. We also examined ROS production and cell viability in islet cells exposed to various stressors associated with oxidative stress. In vitro measurements of ROS and cell death in islet cells demonstrated that the response was not influenced by UCP2 expression. In contrast with UCP2 overexpression studies showing cytoprotection, this study reveals that beta cell survival is not compromised by the absence of UCP2.

The unfolded protein response, which is activated during ER stress, counteracts stress conditions by increasing folding capacity and by increasing the degradation of misfolded ER proteins by the ER-Associated Degradation (ERAD) system. Studies using an engineered insulinoma cell line with inducible expression of the Akita folding-deficient insulin have shown a large induction of Herp, a protein that has been implicated in the ERAD pathway. We hypothesized that Herp is an essential protein that regulates the degradation of misfolded insulin during the ER stress response. Indeed, we found that the degradation of mutant insulin is Herp-dependent and that maintaining Herp expression is vital for maintaining cell survival. We have also observed that the expression of Herp mRNA and protein is induced in various cell culture and animal models of diabetes. These results suggest that Herp is an important ER stress response protein that is induced under diabetic conditions in pancreatic β-cells.

The unfolded protein response, which is activated during ER stress, counteracts stress conditions by increasing folding capacity and by increasing the degradation of misfolded ER proteins by the ER-Associated Degradation (ERAD) system. Studies using an engineered insulinoma cell line with inducible expression of the Akita folding-deficient insulin have shown a large induction of Herp, a protein that has been implicated in the ERAD pathway. We hypothesized that Herp is an essential protein that regulates the degradation of misfolded insulin during the ER stress response. Indeed, we found that the degradation of mutant insulin is Herp-dependent and that maintaining Herp expression is vital for maintaining cell survival. We have also observed that the expression of Herp mRNA and protein is induced in various cell culture and animal models of diabetes. These results suggest that Herp is an important ER stress response protein that is induced under diabetic conditions in pancreatic β-cells.

PI3K signaling in pancreatic β cells has been shown to be important in modulating β cell mass and function under basal condition. Evidence suggests that a specific group of insulin promoter-expressing neurons also modulates glucose metabolism and energy homeostasis through their PI3K signaling. Thus we hypothesize that PI3K activation via PTEN deletion under the control of rat insulin promoter (RIP) in pancreatic β cells and RIP-expressing neurons will protect against hyperglycemia and diabetes in experimentally induced mouse models of type 2 diabetes. In Chapter IV, we showed that RIP-mediated PTEN deletion in pancreatic β cells led to PI3K activation and subsequent increased β cell mass and function, thus protected the mice from high fat diet (HFD)-induced diabetes. Furthermore in the absence of global leptin signaling, β cell-specific PTEN deletion maintained β cell function in the setting of severe insulin resistance, therefore prevented diabetes development. Interestingly, RIP-mediated PTEN deletion also resulted in increased peripheral insulin sensitivity due to PI3K activation in central nervous system. In Chapter V, we showed this increased insulin sensitivity was maintained after HFD feeding, which also contributed to the protection against diabetes. These mice also showed increased visceral adipogenesis and subcutaneous adiposity on HFD, which were dramatically attenuated in the absence of leptin signaling, indicated the essential role of peripheral leptin action in mediating the insulin sensitive phenotype from neuronal RIP PTEN deletion. Finally, we demonstrated that the insulin sensitizing phenotype in these mice was not mediated through ventromedial hypothalamic nuclei (VMH), such that VMH-specific PTEN deletion did not alter energy homeostasis or glucose metabolism. Together, the data from this thesis points to an inhibitory role of PTEN in both central nervous system and pancreatic β cells in glycemic control. Therefore, PTEN may represent a potential target for diabetes prevention and treatment.

Uncoupling protein-2 (UCP2) may be linked to the attenuation of reactive oxygen species (ROS), but it is unclear whether this phenomenon pertains to the pancreatic beta cell. In this study, a UCP2-deficient mouse model was used to assess the importance of UCP2 to beta cell viability. We investigated the effect of UCP2 absence in response to a beta cell cytotoxic model of diabetes induction. In vivo treatment by the cytotoxic agent streptozotocin led to overall beta cell loss, but severity was not exacerbated by UCP2 deficiency. We also examined ROS production and cell viability in islet cells exposed to various stressors associated with oxidative stress. In vitro measurements of ROS and cell death in islet cells demonstrated that the response was not influenced by UCP2 expression. In contrast with UCP2 overexpression studies showing cytoprotection, this study reveals that beta cell survival is not compromised by the absence of UCP2.

An important mechanism involved in the pathogenesis of type 2 diabetes is elevation of plasma free fatty acids which induce insulin resistance and may impair both β-cell function and mass (β-cell lipotoxicity). The objective of my thesis was to investigate the role of oxidative stress in β-cell lipotoxicity, using in vivo, ex vivo, and in vitro models. I used in vivo models of 48h i.v. oleate or olive oil infusion in Wistar rats followed by hyperglycemic clamps, or islet secretion studies ex vivo, and in vitro models of 48h exposure to oleate in isolated islets. My first study showed that 48h oleate infusion decreased the insulin response to a hyperglycemic clamp, an effect prevented by coinfusion of the antioxidants N-acetylcysteine and taurine. Similar to the findings in vivo, 48h infusion of oleate decreased glucose stimulated insulin secretion (GSIS) ex vivo, and induced oxidative stress in isolated islets, effects prevented by coinfusion of the antioxidants N-acetylcysteine, taurine, or tempol. Islets exposed to oleate or palmitate showed a decreased insulin response to high glucose and increased levels of oxidative stress, effects prevented by taurine. Therefore, my data are the first demonstration that oxidative stress plays a role in the decrease in β-cell secretory function induced by prolonged exposure to FFA, in vitro and in vivo. My second study addressed downstream effects of oxidative stress involving inflammation. A 48h infusion of oleate or olive oil decreased β-cell function during a hyperglycemic clamp, an effect prevented by coinfusion of the IKKβ inhibitor salicylate. GSIS in isolated islets was impaired by olive oil or oleate and restored by salicylate. These results suggest a potential role for both oxidative stress and inflammation in lipid-induced β-cell dysfunction. My third study addressed downstream effects of oxidative stress involving β-cell insulin signalling. A 48h infusion of oleate or olive oil decreased β-cell function during a hyperglycemic clamp, an effect prevented by coinfusion of the tyrosine phosphatase inhibitor bisperoxovanadate. GSIS in isolated islets was impaired by olive oil or oleate and restored by bisperoxovanadate, suggesting a role of FFA in decreasing β-cell function by induction of β-cell insulin resistance.

Type 1 and 2 Diabetes Mellitus are a growing health problem throughout the world. There is an increasing  need for methodologies, which are both reliable and non-invasive to measure the amount of insulin-producing tissue (Beta-cell mass, or BCM), as well as rapidly quantify changes in the BCM due to the onset of disease, beta-cell replacement therapy, or other treatments. Positron Emission Tomography (PET) is a non-invasive, quantitative functional imaging technique which can be used to study dynamical or static processes inside the body. In this thesis, we present a study protocol for in vivo imaging of the most common form of beta- cell replacement therapy; islet transplantation. Islets were labeled with the PET tracer, 2-deoxy-2[18F]fluoro-D-glucose ([18F]FDG), and administered intra-portally, while the recipient was monitored by PET/CT. The hepatic distribution of the islets was highly heterogeneous, and around 25% (human) or 50% (porcine) of the administered islets could not be found in the liver after completed transplantation, confirming previous reports of considerable cell injury during the procedure leading to low hepatic engraftment. Native BCM in the pancreas can potentially be quantified using a PET tracer with sufficiently high specificity, but the major obstacle is the relative low amounts of insulin producing tissue (only 1-2% of the pancreatic volume). Two tetrabenazine analogues, [18F]FE-(+)-DTBZ and [18F]FE-(+)-DTBZ-d4, are ligands to VMAT2, which is expressed in islet tissue. Both analogues were investigated and characterized as potential BCM imaging agents both in vitro and in vivo.  Both tracers exhibited high preferential binding to islet tissue compared to exocrine pancreatic tissue. However, the specificity was not high enough to overcome the obscuring exocrine signal in vivo (7-10% of the signal originating from specific islet tracer uptake). This thesis demonstrates that it is possible to quantitatively assess islet transplantation by PET imaging. In vivo determination of native pancreatic BCM is, in theory, possible with both [18F]FE-(+)-DTBZ and [18F]FE-(+)-DTBZ-d4, but tracer analogues with higher islet specificity is needed for quantification of smaller BCM changes with physiological impact.

Endoplasmic reticulum (ER) stress has been implicated as a causative factor in the development of pancreatic beta-cell dysfunction and death resulting in type 2 diabetes. This thesis examined the role of ATF6beta in the ER stress response of beta-cells. Using an ATF6beta-specific antibody, expression of full-length ATF6beta was detected in various insulinoma cell lines and rodent islets and the induction of the active form (ATF6beta-p60) under ER stress conditions. Knock-down of ATF6beta in INS-1 832/13 cells did not affect mRNA induction of known ER stress response genes in response to tunicamycin-induced ER stress, however it increased the susceptibility of beta-cells to apoptosis. Conversely, overexpression of ATF6beta-p60 reduced the apoptotic phenotype. Microarray results suggest ATF6beta functions to induce expression of adaptive genes also regulated by ATF6alpha, but also several specific targets genes. These findings have increased our understanding of the role of ATF6beta in the ER stress response of beta-cells.