Mitochondrial reactive oxygen species reduce insulin secretion by pancreatic beta-cells.

Pancreatic beta-cells exposed to hyperglycemia produce reactive oxygen species (ROS). Because beta-cells are sensitive to oxidative stress, excessive ROS may cause dysfunction of beta-cells. Here we demonstrate that mitochondrial ROS suppress glucose-induced insulin secretion (GIIS) from beta-cells. Intracellular ROS increased 15min after exposure to high glucose and this effect was blunted by inhibitors of the mitochondrial function. GIIS was also suppressed by H(2)O(2), a chemical substitute for ROS. Interestingly, the first-phase of GIIS could be suppressed by 50 microM H(2)O(2). H(2)O(2) or high glucose suppressed the activity of glyceraldehyde 3-phosphate dehydrogenase (GAPDH), a glycolytic enzyme, and inhibitors of the mitochondrial function abolished the latter effects. Our data suggested that high glucose induced mitochondrial ROS, which suppressed first-phase of GIIS, at least in part, through the suppression of GAPDH activity. We propose that mitochondrial overwork is a potential mechanism causing impaired first-phase of GIIS in the early stages of diabetes mellitus.

Sudden onset of diabetes with ketoacidosis in a patient treated with FK506/tacrolimus.

We report a 43-year-old man who presented diabetic ketoacidosis 1 year after receiving kidney transplantation. He was a recipient of renal transplantation treated with metyl-prednisolone and tacrolimus regimen. The serum level of tacrolimus was 12.4 ng/ml, and he showed hyperphagia before a month of admission. A week before admission, he was aware of polydipsia, polyuria, and general fatigue. He visited our hospital and was found to have severe hyperglycemia (925 mg/dl), significant ketosis and mild metabolic acidosis (pH 7.341), although he had not been diagnosed as diabetes mellitus. He administrated in our hospital, and was treated with insulin for 5 weeks. He was not obese (BMI = 18.2 kg/m(2)) and had no family history of type 2 diabetes. He was finally treated with diet therapy alone. The 24 h urine C-peptide secretion on the third hospital day was low (8.4 microg per day). However, no autoantibodies against pancreatic islets were positive, and his insulin secretion was recovered at discharge suggesting that he was not type 1 diabetes. Although, tacrolimus has been reported to cause or worsen diabetes mellitus, the present case suggests that it could cause severe decrease in insulin secretion which leading to diabetic ketoacidosis in lean subject without previous history of diabetes mellitus.