Lars-Åke Idahl

The pancreatic β-cell recognition of insulin secretagogues: Comparisons of glucose with glyceraldehyde isomers and dihydroxyacetone

d-Glyceraldehyde stimulated the release of insulin from pancreatic islets of Umea-obob-mice whether or not glucose was present in the medium. Like the action of glucose, that of d-glyceraldehyde was biphasic in time, exhibited a sigmoidal dose-response relationship, was potentiated by theophylline, arginine, iodoacetamide, or l-glyceraldehyde, and was inhibited by epinephrine, 2,4-dinitrophenol, or Ca2+ deficiency. Half-maximum and maximum stimulations were produced by about 3 mm and 10 mm d-glyceraldehyde. Positive interactions were observed between 5 mm d-glyceraldehyde and 5 mm glucose and between 10 mm d-glyceraldehyde and 10 mm leucine. Mannoheptulose (10 mm) or glucosamine (10 mm) did not inhibit but potentiated the effect of 10 mm d-glyceraldehyde. Dihydroxyacetone (2.5–20 mm) also initiated insulin release in the absence of glucose. On the other hand, 5–10 mm l-glyceraldehyde did not initiate secretion but potentiated the effects of 5 mm glucose or 5 mm d-glyceraldehyde. d-Glyceraldehyde or dihydroxyacetone reduced the production of 14CO2 from d-[U-14C]glucose; l-glyceraldehyde had a smaller and statistically insignificant effect. The results suggest that by being phosphorylated and entering glycolysis in the β-cells, d-glyceraldehyde and dihydroxyacetone act as functional analogues of glucose as secretory stimulus. Initiation of insulin release by glucose, d-glyceraldehyde, or dihydroxyacetone may thus depend on the production of a metabolic signal at or below the triose phosphate level.

Adenosine Triphosphate Levels of Mammalian Pancreatic B Cells after Stimulation with Glucose and Hypoglycemic Sulfonylureas

A microchemical technic was applied to elucidate the possible role of ATP in insulin secretion by measuring the levels of this metabolite in pancreatic islets from obese hyperglycemic mice. The β cell content of ATP was markedly reduced within the first minute after interruption of the blood supply. A steady-state level of about 5 mmoles of ATP was noted in islets incubated in the absence of glucose. The corresponding ATP level was twice as high when at least 1 mg./ml. of glucose was present in the incubation medium. While the contents of ATP and glycogen remained constant when the pancreatic islets were incubated with diazoxide, the amounts of both these metabolites were significantly reduced by concentrations of sulfonylurea compounds known to stimulate the insulin release. The sulfonylurea effect on the islet ATP content implies a change in the “phosphate potential” of the β cells, which might stimulate glycogenolysis and increase the glycolytic flux. The sulfonylurea stimulation of insulin release might thus be related to a product of glucose degradation beyond the level of glucose-6-phosphate.

A micro perifusion device for pancreatic islets allowing concomitant recordings of intermediate metabolites and insulin release.

Abstract A method for continuous nonrecycled perifusion (suprafusion) of isolated pancreatic islets is described. The small size of the tissue chamber (4.5 μl) ensures good resolution of multiphasic secretory response. In addition to allowing dynamic recordings of insulin release, the design of the apparatus facilitates studies of intermediary metabolism by the freeze-stop technique.

CHARACTERIZATION OF THE RAT PAROTID β‐ADRENOCEPTOR

1 The effects of various β‐adrenoceptor agonists on amylase secretion from the rat parotid gland were studied by means of two different in vitro techniques. 2 The dose‐response relation for each agonist was established, as also were the ED50 values. 3 All drugs appeared to act directly on the acinar cells, as reserpine‐treatment did not abolish their secretagogic effects. 4 Two groups of agonists could be distinguished: one group consisting of adrenaline, noradrenaline and the β1‐selective agonist prenalterol (H133/22) with a high enzyme discharge potency and a second group consisting of the β2‐agonists, terbutaline and salbutamol, with a markedly lower effect. 5 The present data further support the theory that rat parotid acinar cells are supplied mainly with β‐adrenoceptors of the β1‐subtype, similar to those present in heart and adipose tissue.

Function of microdissected pancreatic islets cultured in a chemically defined medium. I. Insulin content and release.

SummaryMicrodissected pancreatic islets from non-inbredob/ob-mice, were cultured for 6 or 7 days in serum-free tissue culture medium 199. The insulin content of the islets decreased 60% during culture in 17 mM or 28 mM glucose and about 70% in the presence of 3.3 mM or 5.6 mM glucose. At the end of a culture period in high glucose, the sum of the insulin in the islet plus that in the culture medium was almost twice as high as the insulin content of fresh islets, indicating an active insulin biosynthesis. The maximal insulin response to glucose after culture in 17 mM or 28 mM glucose was about 40% of that in fresh islets; after culture in 3.3 mM glucose it was 10%. Half-maximal stimulation was observed at a glucose concentration of 5 mM for islets cultured with high glucose as compared to 9 mM for fresh islets. Like glucose, glibenclamide was a more effective insulin stimulator after culture with a high glucose concentration than with a low one. However, leucine-induced insulin release was not affected by the glucose concentration in the preceding culture medium. Whereas potentiation of glucose-stimulated release by arginine or dibutyryl-cAMP was independent of glucose concentration during the culture, theophylline released three times more insulin when the islets had been cultured with high glucose.

Dynamics of pancreatic β-cell responses to glucose

SummaryMicroperifusion was used to study the dynamics of insulin release and metabolic changes in pancreatic islets microdissected fromob/ob-mice. When the islets were suddenly exposed to a high glucose concentration, their content of glucose-6-phosphate and fructose-1, 6-diphosphate started to rise immediately, whereas a secretory response was not observed until after about 80 sec. This is consistent with the hypothesis that glucose metabolism initiates insulin release. Fasting the donor animals for 18 h reduced the initial phase of glucose-stimulated release but left the rise of glucose-6-phosphate, as well as the late phase of sustained release, unaffected. The use of a tissue culture medium (TCM 199) instead of Krebs-Ringer bicarbonate buffer as the basal medium increased the secretory responses to glucose, particularly the initial phase, both in islets from fed and fasted mice. However, the difference between fed and fasted mice remained. Theophylline did not restore the impaired initial secretory response after fasting. The responses of islet fructose-1,6-diphosphate and adenosine-5-triphosphate to glucose depended on the nutritional state of the animals and on the type of basal medium. No consistent correlation was observed between the changes of these substrates and the dynamics of insulin release.

Measurements of serum glucose using the luciferin/luciferase system and a liquid scintillation spectrometer

A single-step assay for serum glucose measurements is described. The assay is based on the phosphorylation of D-glucose by glucokinase and the measurement of ATP consumption by firefly luciferase. The luminescence is recorded in an ordinary liquid scintillation spectrometer. The use of stable reagents and a stable final signal (light emission) makes it possible to analyze a large number of samples in each assay run. The assay is of particular value when repeated serum glucose determinations are performed on samples from small laboratory animals.

DISSOCIATION OF β‐ADRENOCEPTOR‐INDUCED EFFECTS ON AMYLASE SECRETION AND CYCLIC ADENOSINE 3′,5′‐MONOPHOSPHATE ACCUMULATION

1 By using a multi‐channel microperifusion system the effects of noradrenaline, the β1‐adrenoceptor agonist prenalterol, and the β2‐selective agonist terbutaline were studied on amylase release and cyclic adenosine 3′,5′‐monophosphate (cyclic AMP) content in rat parotid and guinea‐pig submandibular glands. 2 Noradrenaline caused significant amylase discharge and cyclic AMP accumulation. 3 Prenalterol was as effective as noradrenaline in causing amylase release but did not significantly affect the cyclic AMP content. 4 Terbutaline stimulated cyclic AMP accumulation, but had little effect on amylase secretion. 5 The present study reveals that there is a dissociation of the β‐adrenoceptor‐induced amylase release and cyclic AMP formation, and that this dissociation may be due to different β‐adrenoceptor subtypes.

Influence of anoxia on glucose metabolism in pancreatic islets: Lack of correlation between fructose-1,6-diphosphate and apparent glycolytic flux

SummaryWhen equilibrated with O2-CO2 (95:5), pancreatic islets of non-inbredob/ob-mice exhibited a sigmoidal dependence of3H2O production on D-(5-3H)-glucose concentration; the rate was most sensitive to changes of glucose concentration around 5mM and tended to be maximum above about 15 mM glucose.3H2O production from more than 5 mM D-(5-3H)-glucose was about twice as fast as the production of14CO2 from equimolar D-(U-14C)-glucose. Islets equilibrated with N2-CO2 (95:5) did not exhibit a sigmoidal dose-response curve for3H2O production, the process being inhibited by anoxia at glucose concentrations above 5mM. Pieces of exocrine pancreas had a slower aerobic3H2O production than the islets and showed a clear enhancement of the process during anoxia. In comparison with oxygenated islets, anoxic islets exhibited decreased concentrations of glucose-6-phosphate and increased concentrations of fructose-l,6-diphosphate. The concomitant inhibition of glycolytic flux may be due to a low lactate dehydrogenase activity in islets yielding a slow reoxidation of NADH and a slow phosphoglyceraldehyde oxidation under anaerobic conditions.

Methyl pyruvate initiates membrane depolarization and insulin release by metabolic factors other than ATP.

The role of mitochondria in stimulus-secretion coupling of pancreatic beta-cells was examined using methyl pyruvate (MP). MP stimulated insulin secretion in the absence of glucose, with maximal effect at 5 mM. K+ (30 mM) alone, or in combination with diazoxide (100 microM), failed to enhance MP-induced secretion. Diazoxide (100 microM) inhibited MP-induced insulin secretion. MP depolarized the beta-cell in a concentration-dependent manner (5-20 mM). The sustained depolarization induced by 20 mM MP was not influenced by 100 microM diazoxide, but the continuous spiking activity was suppressed by 500 microM diazoxide. Pyruvate failed to initiate insulin release (5-20 mM) or to depolarize the membrane potential. ATP production in isolated beta-cell mitochondria was detected as accumulation of ATP in the medium during incubation in the presence of malate or glutamate in combination with pyruvate or MP. There was no difference in ATP production induced by pyruvate/malate or MP/malate in isolated beta-cell mitochondria. ATP production by MP/glutamate was higher than that induced by pyruvate/glutamate, but it was much lower than that induced by alpha-ketoisocaproate/glutamate. Pyruvate (5 mM) or MP (5 mM) had no effect on the ATP/ADP ratio in whole islets, whereas glucose (20 mM) significantly increased the whole islet ATP/ADP ratio. It is concluded that MP-induced beta-cell membrane depolarization or insulin release does not relate directly to mitochondrial ATP production. Instead MP may exert a direct extramitochondrial effect, or it may stimulate beta-cell mitochondria to produce coupling factors different from ATP to initiate insulin release.