OBJECTIVE: High concentrations of circulating glucose are believed to contribute to defective insulin secretion and β-cell function in diabetes and at least some of this effect appears to be caused by glucose-induced β-cell apoptosis. In mammalian cells, apoptotic cell death is controlled by the interplay of proapoptotic and antiapoptotic members of the Bcl-2 family. We investigated the apoptotic pathway induced in mouse pancreatic islet cells after exposure to high concentrations of the reducing sugars ribose and glucose as a model of β-cell death due to long-term metabolic stress.RESEARCH DESIGN AND METHODS: Islets isolated from mice lacking molecules implicated in cell death pathways were exposed to high concentrations of glucose or ribose. Apoptosis was measured by analysis of DNA fragmentation and release of mitochondrial cytochrome c.RESULTS: Deficiency of interleukin-1 receptors or Fas did not diminish apoptosis, making involvement of inflammatory cytokine receptor or death receptor signaling in glucose-induced apoptosis unlikely. In contrast, overexpression of the prosurvival protein Bcl-2 or deficiency of the apoptosis initiating BH3-only proteins Bim or Puma, or the downstream apoptosis effector Bax, markedly reduced glucose- or ribose-induced killing of islets. Loss of other BH3-only proteins Bid or Noxa, or the Bax-related effector Bak, had no impact on glucose-induced apoptosis.CONCLUSIONS: These results implicate the Bcl-2 regulated apoptotic pathway in glucose-induced islet cell killing and indicate points in the pathway at which interventional strategies can be designed.

Apoptosis of pancreatic beta cells is a feature of type 2 diabetes and its prevention may have therapeutic benefit. High glucose concentrations induce apoptosis of islet cells, and this requires the proapoptotic Bcl-2 homology domain 3 (BH3)-only proteins Bim and Puma. We studied the stress pathways induced by glucotoxicity in beta cells that result in apoptosis. High concentrations of glucose or ribose increased expression of the transcription factor CHOP (C/EBP homologous protein) but not endoplasmic reticulum (ER) chaperones, indicating activation of proapoptotic ER stress signaling. Inhibition of ER stress prevented ribose-induced upregulation of Chop and Puma mRNA, and partially protected islets from glucotoxicity. Loss of Bim or Puma partially protected islets from the canonical ER stressor thapsigargin. The antioxidant N-acetyl-cysteine also partially protected islets from glucotoxicity. Islets deficient in both Bim and Puma, but not Bim or Puma alone, were significantly protected from killing induced by the mitochondrial reactive oxygen species donor rotenone. Our data demonstrate that high concentrations of glucose induce ER and oxidative stress, which causes cell death mediated by Bim and Puma. We observed significantly higher Bim and Puma mRNA in islets of human donors with type 2 diabetes. This indicates that inhibition of Bim and Puma, or their inducers, may prevent beta-cell destruction in type 2 diabetes.

Recent advances in immunotherapeutics have not yet changed the routine management of autoimmune type 1 diabetes. There is an opportunity to repurpose therapeutics used to treat other diseases to treat type 1 diabetes, especially when there is evidence for overlapping mechanisms. Janus kinase (JAK) 1/JAK2 inhibitors are in development or clinical use for indications including rheumatoid arthritis. There is good evidence for activation of the JAK1/JAK2 and signal transducer and activator of transcription (STAT) 1 pathway in human type 1 diabetes and in mouse models, especially in β-cells. We tested the hypothesis that using these drugs to block the JAK-STAT pathway would prevent autoimmune diabetes. The JAK1/JAK2 inhibitor AZD1480 blocked the effect of cytokines on mouse and human β-cells by inhibiting MHC class I upregulation. This prevented the direct interaction between CD8+ T cells and β-cells, and reduced immune cell infiltration into islets. NOD mice treated with AZD1480 were protected from autoimmune diabetes, and diabetes was reversed in newly diagnosed NOD mice. This provides mechanistic groundwork for repurposing clinically approved JAK1/JAK2 inhibitors for type 1 diabetes.

OBJECTIVE—Apoptosis of pancreatic β-cells is critical in both diabetes development and failure of islet transplantation. The role in these processes of pro- and antiapoptotic Bcl-2 family proteins, which regulate apoptosis by controlling mitochondrial integrity, remains poorly understood. We investigated the role of the BH3-only protein Bid and the multi-BH domain proapoptotic Bax and Bak, as well as prosurvival Bcl-2, in β-cell apoptosis.RESEARCH DESIGN AND METHODS—We isolated islets from mice lacking Bid, Bax, or Bak and those overexpressing Bcl-2 and exposed them to Fas ligand, tumor necrosis factor (TNF)-α, and proinflammatory cytokines or cytotoxic stimuli that activate the mitochondrial apoptotic pathway (staurosporine, etoposide, γ-radiation, tunicamycin, and thapsigargin). Nuclear fragmentation was measured by flow cytometry.RESULTS—Development and function of islets were not affected by loss of Bid, and Bid-deficient islets were as susceptible as wild-type islets to cytotoxic stimuli that cause apoptosis via the mitochondrial pathway. In contrast, Bid-deficient islets and those overexpressing antiapoptotic Bcl-2 were protected from Fas ligand–induced apoptosis. Bid-deficient islets were also resistant to apoptosis induced by TNF-α plus cycloheximide and were partially resistant to proinflammatory cytokine-induced death. Loss of the multi-BH domain proapoptotic Bax or Bak protected islets partially from death receptor–induced apoptosis.CONCLUSIONS—These results demonstrate that Bid is essential for death receptor–induced apoptosis of islets, similar to its demonstrated role in hepatocytes. This indicates that blocking Bid activity may be useful for protection of islets from immune-mediated attack and possibly also in other pathological states in which β-cells are destroyed.