We examined the influence of two K ATP channel openers, diazoxide and an analog (NNC 55-0118), on experimental ␤-cell damage induced by streptozotocin (STZ; 0.5 mmol/l). Rat pancreatic islets were exposed to diazoxide or NNC 55-0118 for 30 min and were further incubated for 30 min after the addition of STZ. The islets were then washed and cultured for 24 h. Islets exposed to STZ alone showed extensive morphological damage, reduced glucose oxidation, low insulin content, and severely impaired glucose-stimulated insulin secretion and proinsulin biosynthesis. Islets treated with STZ in the presence of the channel openers (0.03-0.30 mmol/l) showed dose-dependent preservation of the morphology and improved glucose oxidation rates, insulin content, and secretion. NNC 55-0118 was capable of fully counteracting the STZ impairment, whereas diazoxide had a less protective effect. NNC 55-0118 did not counteract STZ-induced depression of islet NAD levels when examined 2 h after STZ exposure, which suggests that the mechanism of action by NNC 55-0118 is not through an inhibition of poly(ADP-ribose) polymerase. The results illustrate that K ATP channel openers can protect insulinproducing cells against toxic damage, an effect that may be of use in subjects with ongoing insulitis. Diabetes 49:1131-1136, 2000 I n human autoimmune diabetes, the insulin-producing ␤-cells are generally thought to be destroyed by cytotoxic T-cells that are driven by an antigen-specific process (1,2). In experiments with isolated islets, we have found that glucose increases the expression of ␤-cell autoantigens (3) and that diazoxide lowers the antigen amount (4). Against this background, we conducted a clinical trial and observed that a 3-month supplementary treatment with diazoxide had a beneficial effect on residual ␤-cell function in patients with new-onset type 1 diabetes (5). "␤-cell rest" may thus protect islet cells from autoimmune destruction (6) and may help to explain the remission phenomenon often noted subsequent to the initiation of insulin treatment.Diazoxide and other K ATP channel openers such as cromakalim and pinacidil have been used in experimental studies of the ischemic heart, and beneficial cardioprotective effects have been observed (7-9). Although the exact mechanism of this phenomenon is not understood, an opening of mitochondrial K + channels seems to be involved that results in dissipation of the inner mitochondrial membrane potential. This leads to net oxidation of the mitochondria with an apparent reduction in energy wastage (7,9,10).Diazoxide is known to act on K ATP channels in the plasma membrane of ␤-cells (11). It hyperpolarizes the membrane and inhibits the energy-consuming process of insulin secretion (12). Recently, exposure of ␤-cells to diazoxide has also been found to engage mitochondrial K ATP channels (13). In the present study, we examined the influence of diazoxide and NNC 55-0118, a new K ATP channel opener analog, on experimental ␤-cell damage induced by streptozotocin (STZ), an agent known to cause ene...

Aims/hypothesis. We aimed to study the effects of two K ATP channel openers (KCO), diazoxide and the more potent compound NNC 55-0118, on beta-cell suppression and/or toxicity induced by alloxan, sodium nitroprusside and IL-1β. Methods. Islets from rats were exposed to 0.3 mmol/l diazoxide or NNC 55-0118 for 30 min and either alloxan (0.5 mmol/l), sodium nitroprusside (0.5 mmol/l) or IL-1β (12.5 or 25 U/ml) were added and the incubation continued for 30 min. Islets were then washed and incubated for 24 h before examination. Results. After exposure to alloxan, islets showed reduced glucose oxidation rate and impaired glucosestimulated insulin release. NNC 55-0118 counteracted the effects of alloxan, while diazoxide was less effective. After treatment with sodium nitroprusside, islet glucose oxidation rates were reduced and this was prevented by pretreatment with NNC 55-0118. In shortterm experiments the potassium channel openers (KCOs) did not influence the IL-1β effect on insulin secretion. However, long-term addition (24 h) of NNC 55-0118 counteracted IL-1β induced inhibition of the glucose oxidation rate. It was shown, using the fluorescent probe JC-1, that the mitochondrial membrane potential was reduced by the potassium channel openers (KCOs), most strongly by NNC 55-0118. Nevertheless culture with KCOs for 72 h did not cause irreversible damage to the islets. Conclusion/interpretation. Potassium channel openers (KCOs), in particular NNC 55-0118, prevented the toxic effects of alloxan and sodium nitroprusside. IL-1β mediated suppression was reduced by NNC 55-0118 provided the long-term addition of the potassium channel opener (KCO). The protective mechanism of potassium channel openers (KCOs) might involve a decrease of the mitochondrial membrane potential. [Diabetologia (2003) 46:80-88] Keywords Alloxan, interleukin-1β, mitochondria, nitric oxide, pancreatic islets, potassium channel opener, sodium nitroprusside. Environmental as well as immunological events are thought to be of importance for the destruction of the insulin-producing beta-cells in the islets of Langerhans in Type 1 diabetes [1]. The mechanisms and mediators of this process are not fully understood. There is evidence suggesting that a period of reduced activity following the start of insulin treatment is beneficial in new onset diabetes [2,3] and that the cellular activity affects the susceptibility to damage in vitro [4,5,6]. Inhibition of insulin secretion by the use of potassium channel openers (KCOs) provides a means of inducing 'beta-cell rest'. The drugs open the ATP-sensitive potassium channel (K ATP channel) and hyperpol-

The pancreatic B-cell GLUT2 transporter and glucose metabolism were examined in isolated rat islets subjected to treatments affecting insulin secretion. Diazoxide was used to inhibit, while glipizide or depolarization of the plasma membrane with a high extracellular K + concentration were used to stimulate insulin release in short-term experiments. Islet GLUT2 and insulin were determined by quantitative immunohistochemistry and GLUT2 was also determined by Western blot analysis. Islet net glucose uptake and glucose oxidation were measured using radioactively labelled glucose. Exposure of the islets to diazoxide was associated with a marked increase in the B-cell plasma membrane staining for GLUT2 and increased net glucose uptake. Glucose oxidation was not changed, which may reflect a lowered energy requirement. Conversely, islets subjected to a stimulated insulin secretion with glipizide or a high extracellular K + concentration showed a reduced staining of the GLUT2 transporter. The net glucose uptake and glucose oxidation were also reduced. In islets exposed to the high K + concentration no change in the molecular weight or phosphorylation of GLUT2 was observed but a lesser amount of the transporter was found by Western blot analysis. Thus, GLUT2 and glucose uptake in the pancreatic B-cell are modified by the secretory process, which suggests that changes in the glucose transporter have a functional role in normal B-cell physiology.

No abstract

K ATP channels are important for insulin secretion and depolarization of vascular smooth muscle. In view of the importance of drugs affecting K ATP channels in the treatment of diabetes, we investigated the effects of these channels on splanchnic blood perfusion in general and pancreatic islet blood flow in particular. We treated anesthetized Sprague-Dawley rats with the K ATP channel openers diazoxide or NNC 55-0118 or the K ATP channel closer glipizide. Both diazoxide and NNC 55-0118 dose-dependently increased total pancreatic and islet blood flow in the presence of moderate hyperglycemia, but had no effects on the blood perfusion of other splanchnic organs. Diazoxide markedly lowered the mean arterial blood pressure and thus increased vascular conductance in all organs studied. NNC 55-0118 had much smaller effects on the blood pressure. Glipizide did not affect total pancreatic blood flow, but decreased islet blood flow by 50% in the presence of hypoglycemia. We conclude that K ATP channels actively participate in the blood flow regulation of the pancreatic islets and that substances affecting such channels may also influence islet blood flow.