Loss of ?-cell functional identity
The dominance of β-cell functional impairment in T2DM implies that β-cells have lost, at least in part, their normal insulin secretory phenotype. The associated molecular features have been discussed in a number of insightful reviews and research articles, and the role of genetic, epigenetic, transcriptomic and proteomic changes has been described extensively [151 – 166]. At the cellular level, β-cell insulin degranulation and the recently hypothesized β-cell dedifferentiation phenomenon could play key roles. Insulin granules can be easily identified by electron microscopy on the basis of their typical morphology, characterized by a dense core and a more or less clearly visible halo [15,36,167,168]. In addition, secretory granules can be subdivided into mature and immature, based on distinct ultrastructural, biochemical, and functional characteristics [167 – 171], with a relative ratio of 6 to 10 in normal human β-cells [18,172]. Notably, this ratio may be under genetic control, as documented by the finding that β-cells from nondiabetic carriers of the Gly(972)→Arg IRS-1 polymorphism contain a several-fold greater number of immature secretory granules and a lower number of mature granules compared to control β-cells . Also, depending on the technique used, it has been estimated that the number of insulin granules in rodent β-cells ranges from 5000 to more than 10,000 per cell [168,174], of which around 10% are in close proximity to the plasma membrane (“docked” granules) . A few studies have suggested that docked granules could be mainly involved in the first phase of insulin release, but this issue remains controversial [161 – 171,176 – 178].
The amount of insulin granules in β-cells from human subjects with T2DM has been investigated by electron microscopy, showing an average 30 – 40% reduction versus control samples, due to a lower number of mature granules [18,172]. Interestingly, the amount of docked granules has been reported to be similar in type 2 diabetic and nondiabetic β-cells, but with a much higher proportion of immature granules (around fourfold compared to the control cases) . In addition to these changes, decreased gene and/or protein expression of molecules involved in vesicle exocytosis (such as syntaxin-1A, SNAP-25, VAMP-2, Munc-18, Munc 13-1, synaptotagmin V and synaptophysin) has been found in islets from type 2 diabetic patients . Clearly, these changes can conceivably contribute to impair insulin turnover in β-cells from diabetic subjects . At the same time, the insulin granule density of human type 2 diabetic β-cells can be restored (Figure 24.12). In fact, when islets prepared from type 2 diabetic donors were studied after a 24-hour incubation in culture medium containing 5.5 mM glucose and therapeutical concentration of metformin, the volume of mature insulin granules increased from 1.9 ± 0.5 to 3.4 ± 0.3%, which was similar to the value in nondiabetic β-cells (3.2 ± 0.7%) . Although further work is needed, these latter results are in agreement with observations showing replenishment of β-cell granules after reduction of blood glucose in the diabetic Psammomys obesus , and with ex vivo and in vivo experiments demonstrating recovery of insulin secretory function of human type 2 diabetic β-cells [26,88,165,182 – 184].
Degranulation can also lead to underestimation of β-cell number. In a recent article , pancreas samples from non- diabetic and matched type 2 diabetic subjects were studied in parallel by immunocytochemistry and electron microscopy. In addition, morphologic, ultrastructural and glucose-stimulated insulin secretion experiments were performed with nondiabetic islets after 24 hours in culture in the presence of 22.2mM glucose . By immunocytochemistry, the fractional islet insulin-positive area was 30–40% lower in type 2 diabetic islets, as expected. However, electron microscopy showed that the amount of β-cells in the diabetic islets was only marginally decreased, but marked β-cell degranulation was evident. These findings were reproduced after exposing nondiabetic islets to high glucose. It is therefore possible that a proportion of β-cells in T2DM islets may not be detectable by standard immunohistochemistry staining due to insulin degranulation, potentially leading to an overestimation of β-cell loss. At the same time, the study confirmed reduced glucose-stimulated insulin secretion from both type 2 diabetic islets and nondiabetic islets exposed to high glucose, indicating that major β-cell secretory impairment may indeed occur without any actual loss of β-cells .
Interestingly, some islet cells expressing insulin may also show glucagon immunoreactivity [12,149]. In a study, it has been reported that the percentage of insulin-positive cells that were also glucagon immunoreactive is greater in human T2DM (3.2 ± 1.4%) than in nondiabetic samples (0.4 ± 0.1%), and that this proportion may be much higher (16.8±5.0%) depending on the type of antidiabetic therapy . The presence of cells expressing both insulin and glucagon has been also confirmed by gene expression studies on “β-cells” obtained by the laser capture microdissection technique , and could be interpreted as a potential β-cell transdifferentiation to α cells in response to insulin resistance . The secretory phenotype of these dual hormone-positive cells is currently unknown. How- ever, cells containing both insulin and glucagon may also be considered as β-cells at some stage of differentiation. In rodent experiments deleting the key transcription factor FOXO1 in combination with permanent genetic labeling of β-cells, it has been shown that the apparent loss of β-cells was not due to their death, but to the missed expression of fundamental β-cell genes, including insulin, glucose transporter 2, and glucokinase . Furthermore, a proportion of FOXO1-deficient β-cells became able to express glucagon , all this suggesting that under given conditions insulin degranulation could be a step towards more profound β-cell changes, such as dedifferentiation and transdifferentiation [186 – 188] (see Figure 24.13). Although some evidence is available to suggest that β-cells dedifferentiation may also occur in human T2DM [188,189], clearly more work is needed to evaluate if and to what extent dedifferentiation has a role in the loss of β-cell functional identity in the disease. If this was the case, new possibilities could be envisaged to protect and/or rescue the β-cell. For instance, resveratrol administration prevented β-cell dedifferentiation in the Rhesus monkey on a diabetogenic diet , and insulin therapy with restoration of near-normal glycemia was associated with β-cell redifferentiation in a mouse model of insulin-secretory deficiency by β-cell inexcitability .
β-cell failure is central to the onset and progression of human T2DM. Both decreased β-cell mass and impaired insulin secretory function have been documented ex vivo and in vivo, due to inherited and acquired factors, leading to varying combinations of molecular and cellular alterations typical of the disease. Growing evidence indicates that functional defects are likely to play the dominant role in most cases of diabetes, with a proportion of β-cells escaping death at the expense of the acquisition of a non- functional phenotype. Protection and rescue of the β-cell insulin secretory function should represent the primary objective of any intervention aimed at the prevention and treatment of T2DM.