Permanent neonatal diabetes mellitus (PNDM) is a rare disorder usually presenting within 6 months of birth.Although several genes have been linked to this disorder, in almost half the cases documented in Italy, the genetic cause remains unknown. Because the Akita mouse bearing a mutation in the Ins2 gene exhibits PNDM associated with pancreatic β cell apoptosis, we sequenced the human insulin gene in PNDM subjects with unidentified mutations. We discovered 7 heterozygous mutations in 10 unrelated probands. In 8 of these patients, insulin secretion was detectable at diabetes onset, but rapidly declined over time. When these mutant proinsulins were expressed in HEK293 cells, we observed defects in insulin protein folding and secretion. In these experiments, expression of the mutant proinsulins was also associated with increased Grp78 protein expression and XBP1 mRNA splicing, 2 markers of endoplasmic reticulum stress, and with increased apoptosis. Similarly transfected INS-1E insulinoma cells had diminished viability compared with those expressing WT proinsulin. In conclusion, we find that mutations in the insulin gene that promote proinsulin misfolding may cause PNDM.

We studied the effects of genetic background on the phenotype of ob/ob mice, a model of severe obesity, insulin resistance, and diabetes caused by leptin deficiency. Despite a comparable degree of obesity and hyperinsulinemia, C57BL/6J ob/ob mice had much milder hyperglycemia and, surprisingly, normal circulating adiponectin levels despite still-prominent signs of insulin resistance. Hyperinsulinemic-euglycemic clamp revealed relatively less whole-body and muscle insulin resistance in C57BL/6J ob/ob mice, whereas liver insulin resistance tended to be more severe than in FVB/N ob/ob mice. C57BL/6J ob/ob mice had also more rapid clearance of circulating triglycerides and more severe hepatic steatosis. We suggest that strain-related distinction in lipid handling is the most important player in the differences in diabetic phenotype and insulin sensitivity, whereas the impact of circulating adiponectin levels on the overall phenotype of ob/ob mice is less important.

Mutations in the pancreatic ATP-sensitive K؉ channel (K ATP channel) cause permanent neonatal diabetes mellitus (PNDM) in humans. All of the K ATP channel mutations examined result in decreased ATP inhibition, which in turn is predicted to suppress insulin secretion. Here we describe a patient with severe PNDM, which includes developmental delay and epilepsy, in addition to neonatal diabetes (developmental delay, epilepsy, and neonatal diabetes [DEND]), due to a G334D mutation in the Kir6.2 subunit of K ATP channel. The patient was wholly unresponsive to sulfonylurea therapy (up to 1.14 mg ⅐ kg ؊1 ⅐ day ؊1 ) and remained insulin dependent. Consistent with the putative role of G334 as an ATP-binding residue, reconstituted homomeric and mixed WT؉G334D channels exhibit absent or reduced ATP sensitivity but normal gating behavior in the absence of ATP. In disagreement with the sulfonylurea insensitivity of the affected patient, the G334D mutation has no effect on the sulfonylurea inhibition of reconstituted channels in excised patches. However, in macroscopic rubidium-efflux assays in intact cells, reconstituted mutant channels do exhibit a decreased, but still present, sulfonylurea response. The results demonstrate that ATP-binding site mutations can indeed cause DEND and suggest the possibility that sulfonylurea insensitivity of such patients may be a secondary reflection of the presence of DEND rather than a simple reflection of the underlying molecular basis.