Accumulating evidence suggests that endoplasmic reticulum (ER) stress plays a role in the pathogenesis of diabetes, contributing to pancreatic beta-cell loss and insulin resistance. Components of the unfolded protein response (UPR) play a dual role in beta-cells, acting as beneficial regulators under physiological conditions or as triggers of beta-cell dysfunction and apoptosis under situations of chronic stress. Novel findings suggest that "what makes a beta-cell a beta-cell", i.e., its enormous capacity to synthesize and secrete insulin, is also its Achilles heel, rendering it vulnerable to chronic high glucose and fatty acid exposure, agents that contribute to beta-cell failure in type 2 diabetes. In this review, we address the transition from physiology to pathology, namely how and why the physiological UPR evolves to a proapoptotic ER stress response and which defenses are triggered by beta-cells against these challenges. ER stress may also link obesity and insulin resistance in type 2 diabetes. High fat feeding and obesity induce ER stress in liver, which suppresses insulin signaling via c-Jun N-terminal kinase activation. In vitro data suggest that ER stress may also contribute to cytokine-induced beta-cell death. Thus, the cytokines IL-1beta and interferon-gamma, putative mediators of beta-cell loss in type 1 diabetes, induce severe ER stress through, respectively, NO-mediated depletion of ER calcium and inhibition of ER chaperones, thus hampering beta-cell defenses and amplifying the proapoptotic pathways. A better understanding of the pathways regulating ER stress in beta-cells may be instrumental for the design of novel therapies to prevent beta-cell loss in diabetes.

Type 1 diabetes (T1D) is an autoimmune disease in which pancreatic beta cells are killed by infiltrating immune cells and by cytokines released by these cells. Signaling events occurring in the pancreatic beta cells are decisive for their survival or death in diabetes. We have used RNA sequencing (RNA–seq) to identify transcripts, including splice variants, expressed in human islets of Langerhans under control conditions or following exposure to the pro-inflammatory cytokines interleukin-1β (IL-1β) and interferon-γ (IFN-γ). Based on this unique dataset, we examined whether putative candidate genes for T1D, previously identified by GWAS, are expressed in human islets. A total of 29,776 transcripts were identified as expressed in human islets. Expression of around 20% of these transcripts was modified by pro-inflammatory cytokines, including apoptosis- and inflammation-related genes. Chemokines were among the transcripts most modified by cytokines, a finding confirmed at the protein level by ELISA. Interestingly, 35% of the genes expressed in human islets undergo alternative splicing as annotated in RefSeq, and cytokines caused substantial changes in spliced transcripts. Nova1, previously considered a brain-specific regulator of mRNA splicing, is expressed in islets and its knockdown modified splicing. 25/41 of the candidate genes for T1D are expressed in islets, and cytokines modified expression of several of these transcripts. The present study doubles the number of known genes expressed in human islets and shows that cytokines modify alternative splicing in human islet cells. Importantly, it indicates that more than half of the known T1D candidate genes are expressed in human islets. This, and the production of a large number of chemokines and cytokines by cytokine-exposed islets, reinforces the concept of a dialog between pancreatic islets and the immune system in T1D. This dialog is modulated by candidate genes for the disease at both the immune system and beta cell level. Pancreatic beta cells are destroyed by the immune system in type 1 diabetes mellitus, causing insulin dependence for life. Candidate genes for diabetes contribute to this process by acting both at the immune system and, as we suggest here, at the pancreatic beta cell level. We have utilized a novel technology, RNA sequencing, to define all transcripts expressed in human pancreatic islets under basal conditions and following exposure to cytokines, pro-inflammatory mediators that contribute to trigger diabetes. Our observations double the number of known genes present in human islets and indicate that >60% of the candidate genes for type 1 diabetes are expressed in beta cells. The data also show that pro-inflammatory cytokines modify alternative splicing in human islets, a process that may generate novel RNAs and proteins recognizable by the immune system. This, taken together with the findings that pancreatic beta cells themselves express and release many cytokines and chemokines (proteins that attract immune cells), indicates that early type 1 diabetes is characterized by a dialog between beta cells and the immune system. We suggest that candidate genes for diabetes function at least in part as “writers” for the beta cell words in this dialog.