We investigated the effect of 24 h of exposure to various glucose concentrations on insulin secretion by isolated rat pancreatic islets and purified rat beta-cells. Compared with islets cultured with standard medium (5.5 mM glucose), islets cultured with 16.7 mM glucose showed a higher basal insulin release (means +/- SE, 3.0 +/- 0.5 vs. 0.7 +/- 0.2%, n = 8, P less than .005) and reduced glucose-stimulated insulin secretion (2.4 +/- 0.3 vs. 5.8 +/- 0.4%, n = 8, P less than .005). Similar results were also obtained with purified beta-cells. The effect of high glucose was time dependent (present after 12 h, maximal after 24 h) and reversible: when islets cultured with high glucose were transferred to standard medium, normal responsiveness to glucose was restored within 8 h and normal basal release within 24 h. Mannitol, 3-O-methylglucose, and 2-deoxyglucose were not able to mimic the effects of glucose. Islets or purified beta-cells cultured in the presence of high glucose had a normal response when stimulated with glyburide, dibutyryl cyclic AMP, and isobutylmethylxanthine. Tunicamycin, an inhibitor of N-terminal glycosylation, prevented glucose-induced desensitization when added during 24 h of islet culture with 16.7 mM glucose. Swainsonine, another agent that influences glycosylation, had a similar effect. Our study indicates 1) that 24 h of exposure to high glucose induces a specific and reversible impairment of insulin secretion in response to glucose, 2) that this is a direct effect of glucose on beta-cells, and 3) that islet glucose metabolism and glycosylation processes may play a critical role in determining glucose desensitization.

We investigated the effect of 24 h exposure to 100 nmol/l glibenclamide on insulin secretion in isolated rat pancreatic islets. The insulin content was similar in control islets and in islets preincubated with 100 nmol/l glibenclamide for 24 h. In islets preexposed to glibenclamide: 1) the subsequent response to a maximal glibenclamide stimulatory concentration (10 mumol/l, 1 h at 37 C) was greatly reduced in comparison to control islets (0.69 +/- 0.20% vs 2.16 +/- 0.41%; mean +/- SE; n = 14; p less than 0.001); 2) the response to 100 mumol/l tolbutamide stimulation was also reduced (0.55 +/- 0.15% vs 2.38 +/- 0.44%; n = 8; p less than 0.001); 3) the response to 16.7 mmo/l glucose, both in the presence or in the absence of 1 mmol/l IBMX, a phosphodiesterase inhibitor, was also diminished by about 50% (1.79 +/- 0.39% vs. 3.22 +/- 0.42%; n = 14, p less than 0.001). In glibenclamide pretreated islets, blunted responses to stimuli were confirmed also by dynamic studies using a perifusion system. The effect of glibenclamide preincubation was fully reversible: when islets cultured in the presence of glibenclamide were transferred to a glibenclamide-free medium for further 24 h, insulin release in response to glibenclamide stimulation returned to control values. We conclude that prolonged exposure of rat pancreatic islets to glibenclamide induces a reversible desensitization to a variety of metabolic stimuli. The inhibition by prolonged glibenclamide exposure of a common pathway in the mechanism of insulin release is one possible explanation for these results.

No abstract

We studied the effect of chronic exposure to high glucose on the glucose transport regulation in hamster pancreatic Beta cells in permanent culture (HIT). Cells were exposed to either 5.5 mmol/l or 16.7 mmol/l glucose for 48 h and then glucose transport was studied by measuring the (3H)-2-deoxyglucose uptake for 5 and 10 min at 37 degrees C. The 2-deoxyglucose uptake was lower in cells pre-exposed to glucose 16.7 mmol/l for 48 h compared to cells pre-exposed to 5.5 (12.0 +/- 1.6 vs 19.1 +/- 1.2 nmol/0.1 mg after 5 min, and 22.2 +/- 2.6 vs 39.0 +/- 2.9 after 10 min respectively, mean +/- SEM, n = 5, p less than 0.01). In order to investigate the mechanism(s) for glucose impairment of glucose transport, we studied the glucose carrier gene expression in the same cells by Northern and slot-blot analysis. When total RNA was extracted from HIT cells cultured at either 5.5 or 16.7 mmol/l glucose and then hybridized to 32P-labelled cDNA probes for the glucose transporter 1, the glucose transporter 2 and beta-actin, a significant reduction of both glucose transporter 1 (-63.9 +/- 4.1%, mean +/- SEM, n = 3) and glucose transporter 2 (-48.9 +/- 3.2%) mRNA was observed in HIT cells cultured with high glucose. In the same experiments no change of beta-actin mRNA was observed, suggesting that the effect of high glucose was specific on the glucose-transporter mRNAs.(ABSTRACT TRUNCATED AT 250 WORDS)