Imidazolines and the pancreatic B-cell. Actions and binding sites.

Stimulation of insulin secretion by imidazoline compounds displays variable characteristics. Phentolamine (10-100 microM) increased secretion of perifused mouse islets at nonstimulatory glucose concentrations (5 mM) and even in the absence of glucose. Idazoxan (20-100 microM) elicited a moderate increase in insulin secretion, which required the presence of a stimulatory glucose concentration (10 mM). Phentolamine is therefore a stimulator of secretion in its own right, whereas idazoxan may be termed an enhancer of secretion. Both compounds inhibited the activity of ATP-dependent K+ channels in inside-out patches from B-cells; however, idazoxan achieved only an incomplete block. Both compounds depolarized the B-cell plasma membrane to an extent that permitted the opening of voltage-dependent Ca2+ channels (-40 to -30 mV). An increase in cytoplasmic Ca2+ concentration was induced by phentolamine and much less so by idazoxan. Activation of protein kinase C, a possible mechanism to amplify Ca(2+)-induced secretion, could not be verified for phentolamine. It thus appears that stimulation of insulin secretion by phentolamine is due to its blocking effect on KATP channels, which may be the correlate of non-adrenergic imidazoline binding sites which were characterized in insulin-secreting HIT cells. Whether incomplete closure of KATP channels by idazoxan or additional effects are responsible for the requirement of high glucose to stimulate secretion remains to be clarified.

Heterogeneous characteristics of imidazoline-induced insulin secretion.

Imidazolines are regarded as a pharmacological group of insulin secretagogues with one uniform mechanism of action, namely closure of ATP-dependent K+ channels (K(ATP) channels) and, in consequence, depolarization of the plasma membrane, Ca2+ influx and stimulation of secretion. This assumption was investigated by measuring insulin secretion from perifused pancreatic islets in response to three imidazoline compounds and comparing the characteristics of secretion with changes in membrane potential and cytosolic Ca2+ concentration [Ca2+]i of single beta-cells. Phentolamine (32 microM) stimulated insulin secretion from perifused mouse islets in the presence of stimulatory (10 mM and 30 mM) and substimulatory (5 mM) glucose concentrations and even in the absence of glucose. Idazoxan in concentrations up to 500 microM was virtually ineffective in the presence of 5 mM glucose. At 10 mM glucose, there was a moderate but significant increase of secretion by idazoxan, 20 microM being nearly as effective as 100 microM. The effect of phentolamine was of slow onset and irreversible in the time frame of the experiments, while the effect of idazoxan was of fast onset and reversible. Alinidine also stimulated secretion in the presence of 10 mM glucose with fast and reversible kinetics, but in contrast to idazoxan, 100 microM was clearly more effective than 20 microM. These heterogeneous characteristics of secretion were reflected by changes of [Ca2+]i: the increase of [Ca2+]i by phentolamine was slow and only partially reversible, whereas idazoxan led to a smaller, but faster and reversible response. The increase of [Ca2+]i by phentolamine and idazoxan was abolished by the Ca2+ channel blocker D 600. Surprisingly, all three compounds depolarized the beta-cell plasma membrane from a resting potential of -71 mV to about -36 mV. Again, the effect of phentolamine was slow and that of idazoxan and alinidine fast. Thus, the characteristics of phentolamine-induced secretion appear to be attributable to the consequences of K(ATP) channel closure. It is unclear, however, why all three test compounds achieved the same degree of depolarization in spite of their known different efficiency to close K(ATP) channels. Apparently, there are additional mechanisms involved in the action of idazoxan and alinidine, which may contribute to the obvious differences in the characteristics of secretion.

No evidence for PKC activation in stimulation of insulin secretion by phentolamine.

The possible role of protein kinase C (PKC) activation in the course of insulin secretion induced by the imidazoline phentolamine was investigated by measuring the insulin secretion of perifused mouse islets and of insulin-secreting HIT cells and by measuring the PKC activity of HIT cells. When normal mouse islets were perifused with the imidazoline phentolamine (32 microM) or the sulfonylurea glibenclamide (1 microM), neither phentolamine nor glibenclamide could produce a stimulation of secretion which was stronger than that elicited by a strong depolarization. Under the same conditions, tetradecanoylphorbolacetate (TPA, 50 nM), a known activator of PKC activity in pancreatic islets, markedly enhanced the secretion induced by K+ depolarization. Phentolamine also stimulated insulin secretion of superfused HIT cells. When PKC activity in HIT cells was down-regulated to 15% of the initial value by overnight exposure to TPA (50 nM), the stimulatory effect of TPA on secretion was virtually abolished, while phentolamine was still able to elicit a monophasic secretion. TPA (50 nM) induced the typical redistribution of PKC activity in HIT cells: within 2 min, the share of membrane-bound PKC activity rose from 26% to 87% of the total PKC activity, which remained unchanged. In contrast, phentolamine (32 microM) had no effect on PKC distribution, did not down-regulate PKC and had no effect on PKC activity once it was down-regulated by TPA. Thus, the recent suggestion that the insulinotropic effect of imidazolines involves an activation of PKC could not be verified for phentolamine.

Imidazoline/guanidinium binding sites and their relation to inhibition of K(ATP) channels in pancreatic B-cells.

To elucidate the beta-cytotropic effect of imidazoline compounds their inhibitory effect on ATP-dependent K+ channels (K(ATP) channels) in pancreatic B-cells was compared with their binding to membranes from insulin-secreting HIT T15 cells. K(ATP) channels in inside-out patches from B-cells were closed with the following rank order of efficacy at 10 microM: guanabenz > phentolamine = alinidine > clonidine > idazoxan > rilmenidine = amiloride. The last four compounds achieved an incomplete inhibition only. In contrast to sulfonylureas, the inhibitory action of imidazolines was not enhanced by ADP. With intact cells the site which mediates inhibition is less easily accessible for protonated compounds, suggesting a location at the inner face of the plasma membrane. Competition binding experiments were performed by masking alpha-adrenoceptors and using [3H]clonidine as ligand. Homologous displacement of [3H]clonidine revealed two distinct binding sites in HIT cell membranes characterized by dissociation constants of 38 nM and 4,911 nM and maximal binding capacities of 118 fmol/mg protein and 18 pmol/mg protein. Generally, ligands for I2 imidazoline receptors were more potent than ligands for I1 imidazoline receptors to displace [3H]clonidine from the high affinity site, which does not fit into the current classification of imidazoline receptors. Binding to the second site had affinities in the micromolar range, similar to the concentrations necessary to inhibit K(ATP) channels in B-cells. However, alinidine and phentolamine inhibited K(ATP) channels already at concentrations at which they displaced [3H]clonidine only from the high affinity site, but not yet from the low affinity site. Since the proportion of the low and high affinity site varied in dependence of the competitor, the imidazoline binding sites in HIT cells may not be independent, but may rather represent two interacting or interconvertible sites both of which may be involved in K(ATP) channel closure.

Effects of imidazoline compounds on cytoplasmic Ca2+ concentration and ATP-sensitive K+ channels in pancreatic B-cells.

Phentolamine, an alpha-adrenoceptor-blocking agent with an imidazoline structure, induces an increase in the cytoplasmic Ca2+ concentration of pancreatic B-cells. This effect occurs at a concentration (32 microM) at which phentolamine is able to enhance glucose-induced insulin secretion. The increase in cytoplasmic Ca2+ concentration caused by phentolamine is additive to the one elicited by a maximally effective concentration of tolbutamide (100 microM). Imidazoline-binding sites in insulin-secreting HIT cells can also be occupied by the guanidinium compound guanabenz, which was found to be a potent and reversible blocker of ATP-dependent K(+)-channels in B-cells. In contrast to phentolamine, guanabenz blocks the ATP-dependent K(+)-channels only in the inside-out mode, but not in the cell-attached mode of the patch-clamp technique. In conclusion, imidazolines and structurally related compounds block ATP-dependent K(+)-channels by binding to the cytoplasmic face of the plasma membrane, and may have effects on other ion channels which contribute to the elevation of cytoplasmic Ca2+ concentration and, consequently, to insulin release.

Dual action of clonidine on insulin release: suppression, but stimulation when alpha 2-adrenoceptors are blocked.

As shown previously clonidine reduces glucose-stimulated insulin release and this effect is mediated by inhibitory postsynaptic alpha 2-adrenoceptors. The present experiments demonstrate that clonidine has the additional property to also stimulate insulin release. This became evident when the alpha 2-adrenoceptors of isolated islets were blocked by benextramine, and thus protected from being stimulated by clonidine. In the presence of benextramine, clonidine no longer reduced, but on the contrary enhanced, the release of insulin in response to glucose. In control experiments benextramine by itself failed to affect insulin release from isolated islets. These results show that the imidazoline derivative clonidine shares the property of other imidazoline compounds to enhance the insulin response to glucose. All of these agents may stimulate insulin by binding to "imidazoline-preferring" sites, that clearly differ from alpha-adrenoceptors.

An insulin-releasing property of imidazoline derivatives is not limited to compounds that block alpha-adrenoceptors.

As we have demonstrated previously phentolamine stimulates the release of additional insulin from isolated mouse islets and raises plasma insulin levels in the whole rat. This effect was independent of the well known property of phentolamine to block alpha-adrenoceptors. In experiments on isolated pancreatic islets from mice we now demonstrate that tolazoline and antazoline which are chemically closely related to phentolamine, share its ability to potentiate insulin release. The following results were taken as evidence that this effect does not result from an alpha-adrenoceptor blocking action of imidazoline compounds. More than 10 times higher concentrations of phentolamine were required to liberate additional insulin from isolated islets than were effective in counteracting the inhibitory effect of clonidine on insulin release. The newly introduced alpha 2-adrenoceptor antagonist BDF 8933, which is an imidazoline derivative, stimulates insulin release as well, while the irreversible alpha-adrenoceptor blocking agent benextramine of different structure failed to do so, even when being present in concentrations blocking the alpha 2-adrenoceptor-mediated effects of clonidine. Antazoline shared the ability of phentolamine to stimulate insulin release despite having no or only very little alpha-adrenoceptor blocking activity. When used under our conditions, it almost entirely failed to alleviate the inhibition of insulin release induced by clonidine. We conclude that the response of the islet cells to imidazoline derivatives is not limited to those capable of blocking alpha-adrenoceptors. On the other hand, alpha-adrenoceptor blocking agents of different chemical structure fail to induce the release of additional insulin.(ABSTRACT TRUNCATED AT 250 WORDS)

Phentolamine, a deceptive tool to investigate sympathetic nervous control of insulin release.

We investigated the effects of phentolamine and another more selective alpha 2-adrenoceptor antagonist, rauwolscine, on insulin release in vivo (in female Wistar-rats) and in vitro (in perfused rat pancreas and in isolated perifused mouse islets). Phentolamine was found to significantly increase glucose-induced insulin release. On the other hand, rauwolscine failed to do so, when applied in a concentration that effectively antagonized the inhibitory effect of clonidine. These results demonstrate that phentolamine is capable of directly stimulating insulin release. This effect is thus not mediated by alpha-adrenoceptors. For this reason phentolamine is not an appropriate tool to study possible inhibitory effects of the sympathetic nervous system on insulin release. An enhanced insulin response as may be observed in animals and in man in the presence of phentolamine does not furnish evidence for a tonic inhibitory control of the islet cells by the sympathetic nervous system.

Effects of pimozide and apomorphine on insulin secretion from the perfused rat pancreas.

The effects of apomorphine, a partial agonist of dopamine, and of pimozide, an agent commonly used as an antagonist of dopamine, on insulin secretion from the perfused rat pancreas were studied. Apomorphine (10 and 100 microM) significantly reduced the insulin released in response to a stimulating glucose load. As judged from the perfusion pressure vasoconstriction could not account for the inhibitory action. The effect of apomorphine was not reversed by phentolamine infused to block alpha-receptors. Pimozide (1 microM) slightly weakened the inhibitory effect of apomorphine (10 microM) but failed to completely restore a normal secretion. An increase of the pimozide concentration (10 microM) revealed, however, that the agent itself reduced the glucose-induced insulin release. Thus a differentiation of dopamine agonistic and antagonistic effects could not be achieved with the aid of pimozide.

Measurement of blood flow in pancreatic islets of the rat: effect of isoproterenol and norepinephrine.

Blood flow to the pancreatic islets of the anesthetized rat has been measured by application of microspheres, intravital staining of the endocrine tissue by Dithizon, and induction of organ transparency by incubation in glycerol. Thus the microspheres could be counted separately in the islet tissue and in the remaining organ. The total pancreatic blood flow in the rat amounted to 0.48 +/- 0.04 ml.min-1.g-1 and the flow fraction of the islet tissue to 1.22 +/- 0.09%, corresponding to a flow rate through the pancreatic islets of 5.42 +/- 0.63 microliters/min. When isoproterenol was applied (0.1 and 1.0 micrograms.kg-1.min-1 iv for 15 min), total pancreatic blood flow rose to 0.60 +/- 0.08 and 0.98 +/- 0.10 (P less than 0.01) ml.min-1.g-1, whereas the flow fraction to the islets decreased to 0.82 +/- 0.08 (P less than 0.01) and 0.61 +/- 0.07% (P less than 0.01). The absolute islet perfusion remained nearly unchanged. Norepinephrine (2.0 and 5.0 micrograms.kg-1.min-1 for 15 min) similarly increased total blood flow to 0.78 +/- 0.08 (P less than 0.01) and 0.64 +/- 0.08 ml.min-1.g-1) while reducing the islet flow fraction to 1.15 +/- 0.13 and 0.83 +/- 0.05% (P less than 0.01). The absolute flow rate through the islet tissue did not change significantly. The conclusion may be drawn from these experiments that changes in total pancreatic blood flow are not necessarily accompanied by corresponding changes in the perfusion rate of the endocrine islet tissue.

Insulin binding and response to insulin of adipocytes from thyroxine-treated rats.

Insulin binding to isolated fat cells from rats rendered hyperthyroid by daily injections of T4 (1 mg/kg) for 5 days was approximately doubled. The Scatchard curves reflected a large increase in receptor number, as well as an elevation in affinity of the high affinity binding sites. The response to insulin of the fat cells, however, was not increased accordingly: glucose incorporation into lipid in the presence of insulin did not differ significantly from that observed in the control group, whereas the effect of insulin on the lipolytic response to isoprenaline (isoproterenol) was even reduced in the T4-treated animals. T4 treatment had thus dissociated insulin binding from the metabolic effects of insulin, since the increase in membrane receptors was not paralleled by an enhanced effect of the hormone. Since levels of serum insulin were increased in the treated animals, the increase in number of insulin receptors was not mediated by reduced exposure to insulin. Propranolol failed to fully antagonize the effect of T4 on insulin binding, and reserpine treatment even enhanced it. It seems unlikely, therefore, that the increase in insulin receptors of adipocytes results from an augmented response to endogenous catecholamines in T4-treated rats.

Insulin release by tolbutamide and glibenclamide. A comparative study on the perfused rat pancreas.

Glibenclamide, a "second generation" sulfonylurea, produced the same pattern of insulin release from the perfused rat pancreas as did tolbutamide. The stimulatory effect was closely dependent on the glucose concentration present. Both agents enhanced insulin secretion at 5--10 mM glucose, whereas no additional insulin was released when maximally stimulating levels of glucose (20 and 30 mM) were present. The concentrations of glibenclamide stimulating insulin release were 100--400 times lower than equieffective levels of tolbutamide. At glucose levels of 3 or 8 mM, however, glibenclamide did not liberate significantly more insulin from the pancreas than did tolbutamide. Thus the differences of tolbutamide and glibenclamide were quantitative rather than qualitative. Although the active concentrations differed the effects produced were comparable.

Effects of polymer-linked sulfonylurea derivatives on insulin release.

In order to elucidate whether polymerlinked sulfonylurea derivatives may serve as tools to locate the sulfonylurea receptor site, an easily detectable sulfonylurea was synthetized and coupled to cyanogen-bromide activated dextran. The conjugate proved to be unstable and to release sufficient amounts of free agent to explain its insulinotropic activity. Based on these findings previous results claiming that dextran-linked sulfonylureas retain biological activity may be questioned. A stable conjugate was obtained when the acrylyl derivative of the sulfonylurea was co-polymerized with acrylamide. Both a high (greater than 50000) and a low (1500) molecular weight fraction of this conjugate failed to stimulate insulin release from the perfused rat pancreas, questioning the ability of macromolecular conjugates to combine with the sulfonylurea receptor site.

The inhibition of insulin secretion from the perfused rat pancreas after thyroxine treatment.

Thyroxine treatment did not significantly affect the immediate insulin secretory response of the perfused rat pancreas, but it inhibited the late phase of D-glucose-induced insulin secretion. Thyroxine treatment did not inhibit D-glyceraldehyde-, D-mannose-, and tolbutamide-induced insulin release from the perfused pancreas. An increase in the D-glucose concentration of the perfusion medium as well as feeding of the rats did not restore insulin secretion after thyroxine treatment. The inhibition of D-glucose-induced insulin release in response to thyroxine treatment was reversed after addition of either D-glyceraldehyde, dihydroxyacetone, DL-glyceric acid, pyruvate, or alpha-ketobutyrate to the perfusion medium. Tolbutamide, L-glucose, D-fructose, D-mannose, L-lactate, and propionic acid were not able to overcome the inhibition of D-glucose-induced insulin secretion. Except for alpha-ketobutyrate all substances which were effective in reversing the inhibition of D-glucose-induced insulin release were glycolytic intermediates. Comparing the glycolytic alpha-ketoacid pyruvate and the non-glycolytic ketoacid alpha-ketobutyrate, the only part common to both substances was the ketoacid moiety. It is concluded from these findings that the ketoacid moiety of the alpha-ketoacids plays an important role in reversing the effect of thyroxine on D-glucose-induced insulin release.

Inhibition of insulin and glucagon release from the perfused rat pancreas by cyproheptadine (Periactinol, Nuran).

The tricyclic compound cyproheptadine (Periactinol, Nuran) inhibited glucose-induced insulin release from the perfused rat pancreas. Tolbutamide-stimulated insulin release was significantly reduced in the presence and completely suppressed in the absence of a substimulatory glucose concentration (5 mM). Arginine produced a slow rise of insulin release, which was completely abolished by cyproheptadine. Furthermore the biphasic glucagon release due to the stimulus was inhibited. Oxidation of 14C-glucose in isolated islets was unaltered in the presence of cyproheptadine, and pyruvate added to the perfusion medium failed to reverse the inhibitory effect on glucose induced insulin release, indicating that impaired glucose metabolism is not responsible for the inhibition. In addition, the inhibition remained unchanged when phentolamine was present, suggesting that the effect is not mediated by inhibitory adrenergic alpha receptors. Theophylline, in contrast, partly overcame the inhibition. When the calcium concentration of the medium was enhanced, the inhibitory effect of cyproheptadine was still visible, although the relative inhibition had become smaller. The results suggest that cyproheptadine blocks insulin release by affecting a fundamental step of the stimulus-secretion coupling common to peptide hormones. A participation of a calcium-antagonizing effect in the inhibition is discussed.

Thyroid function and insulin secretion from the perfused pancreas in the rat.

The influence of thyroid function on the kinetics of glucose-induced insulin secretion from the isolated perfused rat pancreas has been studied. L-Thyroxine (L-T4) administration did not modify the immediate insulin secretory response of the perfused pancreas to glucose. L-Triiodothyronine (L-T3) treatment as well as propylthiouracil (PTU) treatment decreased the immediate insulin secretory response of the pancreas slightly. Only thyroidectomy (Tx) reduced the immediate secretory response of the pancreas significantly. L-T4 and L-T3 treatment inhibited the late phase of glucose-induced insulin secretion from the isolated perfused rat pancreas, whereas TX and PTU treatment resulted in increased insulin secretion. D-Thyroxine (D-T4) did not affect glucose-induced insulin release from the pancreas. Concomitantly, several parameters indicative of thyroid function were determined in these animals. When changes in body weight, rectal temperature, plasma glucose, plasma cholesterol, and plasma butanol-extractable iodine (BEI) in these rats were compared with the insulin secretory responses, it was evident that experimental hyperthyroidism results in decreased insulin release, whereas experimental hypothyroidism induces increased insulin secretion from the pancreas. The transitions from hypothyroid to euthyroid to hyperthyroid states are accompanied by a steady decrease in glucose-induced insulin release from the rat pancreas. Inhibition of glucose-induced insulin secretion from the pancreas is therefore a specific effect of thyroid hormones.

Inhibition o'f insulin release by cyproheptadine: effects on 3',5'-cyclic-AMP-content and 45Ca-accumulation of incubated mouse islets.

Cyproheptadine (1, 10 and 100 muM) significantly reduced insulin release from isolated mouse islets in response to glucose. In contrast, 1 mM cyproheptadine induced a large release of insulin into the incubation medium probably due to islet cell damage, since the islets had lost a considerable amount of their protein content. 3',5'-cyclic-AMP-levels of the islets were not significantly affected by 10 muM cyproheptadine in the presence as well as in the absence of theophylline (10 mM). As the inhibitory effect of cyproheptadine on insulin release was correlated with reduced accumulation of calcium-45, the agent may inhibit insulin release by interfering with the calcium handling of the beta-cell.