Type 2 diabetes is characterized by increased circulating nutrients including glucose and free fatty acids (FFA).The literature clearly shows that chronic exposure of β cells to elevated glucose results in impaired β-cell function [124–126], but the data regarding cellular toxicity in response to this nutrient are more mixed. Exposure of cultured β-cell lines or islets to high glucose can, in some cases, result in increased β-cell death [127–132]. This may result from oxidative stress, activation of Fas receptor-mediated or mitochondrial apoptosis and may involve thioredoxin interacting protein (TXNIP) [127–129]. However, this effect of glucose to induce apoptosis is not a universal finding; several studies have demonstrated that the “toxic” effects of chronic hyperglycemia to impair β-cell function are reversible even after several weeks in culture [133,134]. Further in vitro and in vivo studies exposing β cell to elevated glucose have shown beneficial effects with glucose-promoting survival signals, suppressing apoptosis  or resulting in increased β-cell replication [136–138].
Exposure of islets/β cells to increased FFA levels alone, or in the presence of hyperglycemia results in impaired insulin release [139–141]. Culture of β cells in the presence of increased FFA, particularly palmitate, can also result in β-cell apoptosis [132,142–144]. This has been shown to occur via some of the same mechanisms as glucose-induced apoptosis, namely oxidative stress, ER stress and activation of the mitochondrial apoptosis pathway, and additionally may require increased ceramide or nitric oxide levels. However, similar to the observations with elevated glucose, high fat feeding or lipid infusions in vivo result in increased β-cell mass, as a result of increased β-cell replication [62,138].
Thus, taken together, the effects of nutrient excess appear to be more detrimental to β-cell secretory function rather than clearly inducing β-cell death, and some of these effects may be reversible.
As mentioned, amyloid deposition occurs in islets in the majority of subjects with T2DM, as well as in subjects with cystic fibrosis-related diabetes [17,95,110,123]. Accumulation of islet amyloid occurs due to the aggregation of the normally soluble β-cell peptide IAPP, which is then deposited in the islet extracellular matrix, between islet capillaries and β cells. This aggregation only appears to occur under conditions of diabetes or islet dysfunction, with islet amyloid being relatively rare in individuals without diabetes even in individuals with extremely high circulating levels of IAPP [110,145,146]. The underlying cause of this IAPP aggregation is unclear, but may involve impaired processing of IAPP from its precursor proIAPP [147–149] and/or interaction between IAPP and extracellular matrix components, principally heparan sulfate proteoglycans [150–153]. Human autopsy studies have yielded somewhat conflicting results, but the literature clearly demonstrates that the extent of islet amyloid deposition is associated with decreased β-cell volume [17,154] and increased β-cell apoptosis  (Figure 5.5). Studies using cultured human islets and transgenic animals expressing human IAPP (mouse and rat IAPP are not amyloidogenic) have further elucidated the mechanism(s) by which IAPP aggregation may elicit β-cell toxicity. Culture of human or transgenic mouse islets under conditions that favor amyloid formation, for example high glucose, result in amyloid-induced oxidative stress and increased β-cell apoptosis, thereby leading to a reduction in β-cell area [155–160]. This β-cell loss can occur via activation of the cell surface death receptor Fas , or cJun N-terminal kinase (JNK) and downstream activation of apoptosis . Additionally, when human IAPP aggregation and thereby amyloid formation is inhibited by Congo red  or overexpression of the enzyme neprilysin  β-cell apoptosis is reduced, suggesting IAPP aggregation is an important mediator of β-cell toxicity. Some, but not all, studies have demonstrated that expression of human IAPP results in an ER stress response [164–166]. However, this appears to be related to the magnitude of IAPP overexpression, and does not occur at physiologic levels of human IAPP, nor does it differ between human islets from individuals with T2DM who do or do not have amyloid deposits . Finally, recent data have shown that human IAPP in its aggregated form is proinflammatory, eliciting cytokine and chemokine production from macrophages/dendritic cells [167,168]. Further, islet IL-1β expression may be increased in conditions of amyloid deposition [112,161,167], suggesting a novel mechanism by which islet amyloid may result in β-cell death.
As discussed earlier, inflammation in the islet has long been established as a hallmark of T1DM. Islet infiltration and release of molecules such as proinflammatory cytokines have clearly been implicated in β-cell death in this form of diabetes . In T2DM, the concept that low-grade, chronic inflammation exists, most likely associated with insulin resistance, is a relatively new idea. As this field of research has emerged, so too has the hypothesis that inflammation in the islet may play a role in β-cell death in T2DM . However, this remains a controversial area. Evidence in favor of a role for islet inflammation includes reports of increased islet production of interleukin 1β following chronic high glucose culture of human islets . Islet interleukin 1β production has also been suggested in models of islet amyloid formation [112,161,167]. Activation of signaling pathways associated with the innate immune response (namely toll-like receptors) has been shown to occur in β cells in response to agents such as FFA or lipopolysaccharide [170,171]. This activation can lead to β-cell toxicity and death, suggesting that inflammation may play a role in the demise of the β cell in T2DM.
Summary and future directions
The morphology of the pancreas and pancreatic islet is complex, and disturbances in pancreas and islet volume/arrangement that occur in diabetes are multifactorial. Loss of β cells is a common feature of type 1-, type 2-, and cystic fibrosis-related diabetes. However, the mechanisms that underlie this pathology differ significantly among the various types of diabetes. Our understanding of how β-cell destruction occurs in type 1 and type 2 diabetes has been improved by a large number of studies, but we still have much to learn about how this occurs. Emerging areas of interest include understanding how changes in islet vasculature, innervation, and extracellular matrix contribute to derangements in islet morphology, which may in turn shed new light on the causes of β-cell loss in diabetes.
This work was supported by the Department of Veterans Affairs (Seattle Division VA Puget Sound Health Care System, Seattle, WA, USA) and National Institutes of Health grants DK088082 (RLH), DK017047 and DK075998 (SEK).