Purpose: To investigate the association between fruit and vegetable intake (FVI) and the risk of developing prediabetes and type 2 diabetes (T2D) in a Swedish prospective cohort study. Methods: Subjects were 6961 men and women aged 35–56 years old at baseline, participating in the Stockholm Diabetes Prevention Program cohort. By design, the cohort was enriched by 50% with subjects that had family history of diabetes. Anthropometric measurements, oral glucose tolerance tests and questionnaires on lifestyle and dietary factors were carried out at baseline and two follow-up occasions. Cox proportional hazard models were used to estimate hazard ratios with 95% CIs. Results: During a mean follow-up time of 20 ± 4 years, 1024 subjects developed T2D and 870 prediabetes. After adjustments for confounders, the highest tertile of total FVI was associated with a lower risk of developing T2D in men (HR 0.76, 95% CI 0.60–0.96). There was also an inverse association between total fruit intake and prediabetes risk in men, with the HR for the highest tertile being 0.76 (95% CI 0.58–1.00). As for subtypes, higher intake of apples/pears was inversely associated with T2D risk in both sexes, whereas higher intakes of banana, cabbage and tomato were positively associated with T2D or prediabetes risk in either men or women. Conclusion: We found an inverse association between higher total FVI and T2D risk and between higher fruit intake and prediabetes risk, in men but not in women. Certain fruit and vegetable subtypes showed varying results and require further investigation.

Beta cells of pancreatic islets are susceptible to functional deficits and damage by hypoxia. Here we aimed to characterize such effects and to test for and pharmacological means to alleviate a negative impact of hypoxia. Rat and human pancreatic islets were subjected to 5.5 h of hypoxia after which functional and viability parameters were measured subsequent to the hypoxic period and/or following a 22 h re-oxygenation period. Preconditioning with diazoxide or other agents was usually done during a 22 h period prior to hypoxia. Insulin contents decreased by 23% after 5.5 h of hypoxia and by 61% after a re-oxygenation period. Preconditioning with diazoxide time-dependently alleviated these hypoxia effects in rat and human islets. Hypoxia reduced proinsulin biosynthesis (³H-leucine incorporation into proinsulin) by 35%. Preconditioning counteracted this decrease by 91%. Preconditioning reduced hypoxia-induced necrosis by 40%, attenuated lowering of proteins of mitochondrial complexes I–IV and enhanced stimulation of HIF-1-alpha and phosphorylated AMPK proteins. Preconditioning by diazoxide was abolished by co-exposure to tolbutamide or elevated potassium (i.e. conditions which increase Ca²⁺ inflow). Preconditioning with nifedipine, a calcium channel blocker, partly reproduced effects of diazoxide. Both diazoxide and nifedipine moderately reduced basal glucose oxidation whereas glucose-induced oxygen consumption (tested with diazoxide) was unaffected. Preconditioning with diaxoxide enhanced insulin contents in transplants of rat islets to non-diabetic rats and lowered hyperglycemia vs. non-preconditioned islets in streptozotocin-diabetic rats. Preconditioning of human islet transplants lowered hyperglycemia in streptozotocin-diabetic nude mice. 1) Prior blocking of Ca²⁺ inflow associates with lesser hypoxia-induced damage, 2) preconditioning affects basal mitochondrial metabolism and accelerates activation of hypoxia-reactive and potentially protective factors, 3) results indicate that preconditioning by K⁺-ATP-channel openers has therapeutic potential for islet transplantations.

We investigated the impact of a diabetic state with hyperglycemia on morphometry of β cell mitochondria and modifying influence of a K+-ATP channel opener and we related in vivo findings with gluco...

Diazoxide enhances glucose-induced insulin secretion from β-cells through mechanisms that are not fully elucidated. Here, we used microarray analysis (Affymetrix) to investigate effects of diazoxide. Pancreatic islets were cultured overnight at 27, 11, or 5.5 mmol/l glucose with or without diazoxide. Inclusion of diazoxide upregulated altogether 211 genes (signal log₂ ratio ≥0.5) and downregulated 200 genes (signal log₂ ratio −0.5 or lower), and 92% of diazoxide's effects (up- and downregulation) were observed only after coculture with 11 or 27 mmol/l glucose. We found that 11 mmol/l diazoxide upregulated 97 genes and downregulated 21 genes. Increasing the glucose concentration to 27 mmol/l markedly shifted these proportions toward downregulation (101 genes upregulated and 160 genes downregulated). At 27 mmol/l glucose, most genes downregulated by diazoxide were oppositely affected by glucose (80%). Diazoxide influenced expression of several genes central to β-cell metabolism. Diazoxide downregulated genes of fatty acid oxidation, upregulated genes of fatty acid synthesis, and downregulated uncoupling protein 2 and lactic acid dehydrogenase. Diazoxide upregulated certain genes known to support β-cell functionality, such as NKX6.1 and PDX1. Long-term elevated glucose is permissive for most of diazoxide's effects on gene expression, the proportion of effects shifting to downregulation with increasing glucose concentration. Effects of diazoxide on gene expression could serve to enhance β-cell functionality during continuous hyperglycemia.

Voltage-dependent K⁺ channels (Kv) mediate repolarisation of β-cell action potentials, and thereby abrogate insulin secretion. The role of the Kv1.1 K⁺ channel in this process is however unclear. We tested for presence of Kv1.1 in different species and tested for a functional role of Kv1.1 by assessing pancreatic islet function in BALB/cByJ (wild-type) and megencephaly (mceph/mceph) mice, the latter having a deletion in the Kv1.1 gene. Kv1.1 expression was detected in islets from wild-type mice, SD rats and humans, and expression of truncated Kv1.1 was detected in mceph/mceph islets. Full-length Kv1.1 protein was present in islets from wild-type mice, but, as expected, not in those from mceph/mceph mice. Kv1.1 expression was localized to the β-cell population and also to α- and δ-cells, with evidence of over-expression of truncated Kv1.1 in mceph/mceph islets. Blood glucose, insulin content, and islet morphology were normal in mceph/mceph mice, but glucose-induced insulin release from batch-incubated islets was (moderately) higher than that from wild-type islets. Reciprocal blocking of Kv1.1 by dendrotoxin-K increased insulin secretion from wild-type but not mceph/mceph islets. Glucose-induced action potential duration, as well as firing frequency, was increased in mceph/mceph mouse β-cells. This duration effect on action potential in β-cells from mceph/mceph mice was mimicked by dendrotoxin-K in β-cells from wild-type mice. Observations concerning the effects of both the mceph mutation, and of dendrotoxin-K, on glucose-induced insulin release were confirmed in pancreatic islets from Kv1.1 null mice. Kv1.1 channels are expressed in the β-cells of several species, and these channels can influence glucose-stimulated insulin release.