Glucagon secretion from the pancreas is primarily governed by the prevailing arterial glucose concentration, and hypoglycemia is a major stimulant of the secretion. Glucagon activates adneylate cyclase, resulting in an increase in cyclic AMP levels and subsequently in the activity of cyclic AMP-dependent protein kinase. The activated protein kinase in turn activates phosphorylase, causing glycogenolysis in the liver with release of glucose into the blood. Hypoglycemia also causes the release of catecholamines (CAs) from the adrenal medulla, and released CAs stimulate glycogenolysis in the liver through β-adrenoceptors by activating phosphorylase through the same cyclic AMP-initiated cascade as is produced by glucagon. With regard to the release of glucagon and CAs, the possible interaction of both hormones has been suggested, which provides information important to the understanding of the sequence of their release under hypoglycemic condition. Existence of α- and β-adrenoceptors enhancing glucagon secretion in the pancreas (1,2) suggests that CAs released from the adrenal medulla act to facilitate glucagon secretion. As for the effect of glucagon on adrenal CA release, earlier works provided evidence for the releasing action of glucagon in dogs (3,4), suggesting that released glucagon stimulates the release of adrenal CAs. On the other hand, increase in cyclic AMP level by various adenylate cyclase activators, phosphodiesterase (PDE) inhibitors and cyclic AMP derivatives did not evoke adrenal CA release or caused only a slight increase in the release, but markedly potentiated the stimulation-evoked release (5,6,7). These findings suggest that glucagon acts on the adrenal medulla primarily as a modulator rather than a stimulator, since physiological and pharmacological actions of glucagon are mediated by cyclic AMP. However, data regarding effects of glucagon on the stimulation-evoked adrenal CA release are lacking.
In the present study, we examined the effects of glucagon on the release of adrenal CAs in response to splanchnic nerve stimulation (SNS), acetylcholine (ACh), 1,1-dimethyl-4-phenyl-piperazinium iodide (DMPP), and muscarine in anesthetized dogs to elucidate whether glucagon modulates the CA release mediated by nicotinic and muscarinic receptors. The releasing effect of glucagon on adrenal CAs was also examined to determine which effect, modulatory or stimulatory, is the primary action of glucagon.
Mongrel dogs of either sex weighing 7-14 kg were anesthetized with 30 mg/kg sodium pentobarbital intravenously (I.V.); a constant level of anesthesia was then maintained by an intravenous infusion of sodium pentobarbital at a rate of 4-6 mg/kg/h with an infusion pump (201B, Atom, Tokyo, Japan). Artificial respiration was achieved with a respiration pump (model 607, Harvard Apparatus, Mills, MA, U.S.A.), with room air administered at 18 strokes/min (20 ml/kg tidal volume). The surgical procedure used in the present study was described previously (8). The left adrenal gland was exposed by a retroperitoneal flank incision, and a polyethylene cannula was inserted into the left adrenolumbar vein for collection of the venous effluent blood from the adrenal gland. A thread was placed around the juncture of the adrenolumbar vein and the abdominal vena cava. Adrenal blood samples were obtained by pulling the thread, thus occluding the adrenolumbar vein and causing a retrograde flow of blood to ensue. The 1- or 2-ml blood samples were collected in chilled test tubes containing disodium EDTA. When not being sampled, adrenal venous blood was returned directly to the vena cava. Coagulation of blood was prevented by an initial intravenous injection of sodium heparin (250 U/kg). Systemic blood pressure (BP) and heart rate (HR) were measured by a pressure transducer (MPU-0.5, Nihon Kohden, Tokyo, Japan) and a cardiotachometer (RT-5, Nihon Kohden), respectively, and recorded on a heat-writing oscillograph (RJG-4128, Nihon Kohden).
Administration of drugs into the adrenal gland
The procedure for intraarterial (i.a.) administration of drugs into the adrenal gland was reported previously (9). The left phrenicoabdominal artery was dissected to expose its origin from the abdominal aorta. A needle connected to a Y-shaped polyethylene catheter was inserted in the phrenicoabdominal artery at its origin for intraarterial infusion of 0.9% saline solution (as a vehicle) and glucagon and for intraarterial injection of ACh, muscarine, and DMPP. These cholinergic drugs were injected for 3 s.
The left splanchnic nerves were dissected free from surrounding tissue and cut. A bipolar platinum electrode was placed in contact with the distal end of the splanchnic nerves. The splanchnic nerves were stimulated for 6 min with rectangular pulses of 1 ms and 10 V (supramaximal voltage) delivered by an electronic stimulator (SEN-1101, Nihon Kohden) and an isolation unit (SS-101J, Nihon Kohden). Stimuli were applied at 1 Hz for 3 min, and subsequently at 3 Hz for 3 min more during a 6-min stimulus period.
The dogs were divided into five groups. In group 1 (n = 5), the effect of glucagon on the SNS-induced increase in CA output was examined. SNS was repeated four times at 30-min intervals. The first SNS trial during saline infusion into the adrenal gland was regarded as a control. Glucagon infusions (0.1, 0.3, and 1 μg/min) were started 5 min before the start of the second, third, and fourth SNS, respectively. In group 2 (n = 7), the effect of glucagon on the ACh-induced increase in CA output was examined. A series of ACh injections (1.5 and 3 μg) into the adrenal gland was repeated four times at 30-min intervals. The interval between two doses of ACh was 10 min. The first series of ACh injections during saline infusion was regarded as a control. Glucagon infusion (0.1, 0.3, and 1 μg/min) was started 5 min before the second, third, and fourth series of ACh injections, respectively. In groups 3 (n = 6) and 4 (n = 6), the effects of glucagon on increases in CA output induced by DMPP (1 and 2 μg) and muscarine (1 and 2 μg) were examined, respectively, with the same protocol used in group 2. In group 5 (n = 6), the effect of glucagon on cyclic AMP overflow determined from adrenal venous blood was examined as an index of cyclic AMP production in adrenal medullary cells. Glucagon infusion (0.1, 0.3, and 1 μg/min) for 15 min was repeated three times at 30-min intervals in an increasing dose order.
Blood sampling and determination of adrenal CA output
In groups 1-4, adrenal venous blood was sampled before and during SNS, ACh, DMPP, and muscarine injections to determine basal CA output and stimuli-induced increases in CA output, respectively. The sampling during basal state (during saline or glucagon infusion) was performed 2 min before SNS or series of drug injections. The time required to collect 1 ml (during basal state or SNS) or 2 ml (during cholinergic drug injection) blood allowed estimation of adrenal venous flow rate. In group 5, adrenal venous blood was sampled before and during each dose of glucagon infusion (3 and 10 min after the start of infusion).
Adrenal blood samples were centrifuged to obtain plasma samples. CAs were extracted from plasma by alumina adsorption method, and plasma epinephrine (EPI) and norepinephrine (NE) concentrations were determined by high-performance liquid chromatography with electrochemical detection (LC-304, Bioanalytical Systems, West Lafayette, IN, U.S.A.), as described previously (8). Adrenal EPI and NE output (nanograms per minute) was calculated by multiplying plasma CA concentration (nanograms per milliliter) by adrenal plasma flow rate (milliliters per minute). Adrenal plasma flow rate was determined from the adrenal venous blood flow rate and the hematocrit of adrenal venous blood. The basal CA output was determined from samples collected before SNS or injections of the cholinergic drugs. The SNS-, ACh-, muscarine-, or DMPP-induced increase in CA output was calculated by subtracting the basal CA output from that obtained during the stimulus state.
Determination of cyclic AMP overflow
Adrenal venous plasma cyclic AMP concentration was determined by cyclic AMP radioimmunoassay kit. Adrenal cyclic AMP overflow (picomoles per minute) was calculated by multiplying plasma cyclic AMP concentration (picomoles per milliliter) by adrenal plasma flow rate (milliliters per minute).
The results are mean ± SEM. Analysis of variance and Dunnett's test were used for statistical analysis of multiple comparisons of data; p < 0.05 was considered statistically significant.
The drugs used were glucagon, muscarine chloride (Sigma Chemical, St. Louis, MO, U.S.A.), ACh chloride (Daiichi Seiyaku, Tokyo, Japan), DMPP (Aldrich Chemical, Milwaukee, WI, U.S.A.). A cyclic AMP radioimmunoassay kit was used (Yamasa Shoyu, Cyoshi, Japan). All drugs were dissolved in 0.9% saline solution.
Effects of glucagon on SNS-, ACh-, DMPP-, and muscarine-induced increases in CA output
SNS (1 and 3 Hz) or intraarterial injections of ACh (1.5 and 3 μg), DMPP (1 and 2 μg) and muscarine (1 and 2 μg) into the adrenal gland produced a frequency- or dose-dependent increase in adrenal venous plasma CA concentration (data not shown). The ACh- and muscarine-induced increases in CA concentration were accompanied by increases in adrenal plasma flow rate (Table 1). SNS and DMPP had no effect on adrenal plasma flow rate. As a result, SNS, ACh, DMPP, and muscarine increased both EPI and NE output from the adrenal gland. The increase in CA output induced by SNS (3 Hz), ACh (3 μg), DMPP (2 μg), and muscarine (2 μg) were 467 ± 111, 409 ± 49, 424 ± 35, and 311 ± 60 ng/min in EPI output and 54 ± 8, 74 ± 6, 121 ± 31, and 65 ± 13 ng/min in NE output, respectively, which increases were > 100 times basal output.
Infusions of glucagon (0.1, 0.3, and 1 μg/min) into the adrenal gland enhanced the SNS-induced increases in CA output in a dose-dependent manner: significantly at 0.3, and 1 μg/min in EPI output and in NE output (1 Hz SNS) (Fig. 1). Glucagon produced a dose-dependent enhancement of the ACh- and DMPP-induced increases in CA output. Significant enhancement was observed at 1 μg/min glucagon infusion in EPI output and at 0.1, 0.3, and 1 μg/min in NE output in ACh experiments (Fig. 2) and at 0.3 and 1 μg/min in EPI output and at 1 μg/min in NE output in DMPP experiments (Fig. 3). The muscarine-induced increases in CA output were not affected even by the highest dose (1 μg/min) of glucagon infusion (Fig. 4). To compare the effects of glucagon on the increases in CA output induced by SNS, ACh, DMPP, and muscarine, we calculated percent increases (Fig. 5). These data indicate clearly that glucagon has a facilitatory effect on the increases in CA output induced by SNS, ACh, and DMPP, but not on that induced by muscarine.
Glucagon did not affect basal CA output at 0.1 and 0.3 μg/min, but increased it slightly at 1 μg/min. In groups 1 to 4 (n = 24), values of basal CA output before and during 0.1-, 0.3-, and 1-μg/min glucagon infusion were 1.9 ± 0.3, 2.3 ± 1.1, 2.5 ± 0.6, and 3.6 ± 1.4 ng/min in EPI output and 0.2 ± 0.1, 0.5 ± 0.2, 0.2 ± 0.1, and 1.1 ± 0.1 ng/min in NE output, respectively. Adrenal plasma flow rate (during basal state and during stimulus state by SNS, ACh, DMPP, and muscarine) was not affected by glucagon (Table 1). Glucagon did not affect BP or HR: mean BP before and during 0.1-, 0.3-, and 1-μg/min glucagon infusion was 109 ± 7, 110 ± 7, 109 ± 7, and 108 ± 7 mm Hg and HR was 130 ± 7, 129 ± 6, 127 ± 5, and 128 ± 6 beats/min, respectively, in groups 1-4 (n = 24).
Effects of glucagon on cyclic AMP overflow
Figure 6 shows the effect of glucagon on cyclic AMP overflow from the adrenal gland. Infusion of glucagon did not affect the cyclic AMP overflow at 0.1 μg/min, but increased it significantly at 0.3 and 1 μg/min as compared with the corresponding control values obtained before each infusion. There were no differences in cyclic AMP overflow estimated 3 and 10 min after the start of glucagon infusion.
The aims of the present study were to investigate (a) whether glucagon modifies the release of adrenal CAs mediated by nicotinic and muscarinic receptors similarly or differently, and (b) whether glucagon stimulates the release of CAs directly. Furthermore, the effect of glucagon on cyclic AMP overflow in adrenal production of cyclic AMP overflow in adrenal venous blood was examined as an index of the production of cyclic AMP in adrenal medullary cells. The cholinergic drugs (Ach, DMPP, and muscarine) were injected intraarterially into the adrenal gland to eliminate their hemodynamic influences on adrenal CA release. Because glucose blood level may influence the release of CAs through changes in sympathetic nerve activity, glucagon was infused into the adrenal gland to minimize the increase in glucose levels. In the dog adrenal preparation used, the splanchnic nerves were cut to eliminate sympathetic neural influence. Glucagon enhanced the SNS-, ACh-, and DMPP-induced increases in CA output significantly but did not modify the muscarine-induced increases. Previously, using the same experimental conditions, we demonstrated that the SNS-evoked release of CAs is mediated mainly by nicotinic receptors (9,10). Therefore, the enhancing effects of glucagon on the SNS- and DMPP-induced CA release indicate that glucagon facilitates the release by affecting the process mediated by nicotinic receptor activation. The failure of glucagon to affect the muscarine-induced release indicates that glucagon has no effect on the process mediated by muscarine receptor activation. The muscarine-induced CA release is of muscarinic nature because it was inhibited by the muscarinic receptor antagonist pirenzepine (9). We also demonstrated that the ACh-induced CA release was inhibited by hexamethonium and by pirenzepine; therefore, ACh releases CAs by activating both nicotinic and muscarinic receptors (9). On the basis of these findings, the enhancing effect of glucagon on the ACh-induced CA release is explained by its facilitatory action on the nicotinic receptor-mediated CA release.
Physiological responses to glucagon are caused by increase in intracellular cyclic AMP level resulting from an increase in adenylate cyclase activity. In the present study, glucagon increased cyclic AMP overflow in adrenal venous blood. The increase in cyclic AMP overflow and the enhancement of SNS-, ACh-, and DMPP-induced increases in CA output were observed over the same dose range of glucagon infusion (0.3 and 1 μg/min). These results suggest that cyclic AMP plays a role, as an intracellular messenger, in the facilitatory effect of glucagon on the nicotinic receptor-mediated CA release. Although the mechanism by which cyclic AMP facilitates the release is not clear, it probably affects the release process involved in the nicotinic receptor-mediated pathway but not in the muscarinic receptor-mediated pathway. On exocytotic release of CAs from the adrenal medulla, the increase in intracellular Ca2+ is an essential step regardless of nictonic or muscarinic receptor activation. Nicotinic receptor activation promotes Ca2+ influx through voltage-dependent Ca2+ channels (11-15), and muscarinic receptor activation causes Ca2+ mobilization from intracellular storage sites (16-20). Therefore, we postulate that, as one of the possible mechanisms, cyclic AMP produced by glucagon enhances the nicotinic receptor-mediated release by facilitating Ca2+ influx into the cells. In support of this hypothesis, cyclic AMP derivatives and forskolin were reported to enhance the ACh-evoked Ca release by enhancing Ca2+ uptake into the adrenal medullary cells (7,21).
The basal EPI and NE output increased during the highest dose of glucagon (1 μg/min), but the increase was only slight. The enhancement of nicotinic receptor-mediated release was induced by lower doses (0.1 and 0.3 μg/min). These results suggest that glucagon acts on the adrenal medulla primarily as a modulator rather than a stimulator. This consideration is supported by the findings that increase in cyclic AMP level induced by various adenylate cyclase activators, PDE inhibitors, and cyclic AMP derivatives did not evoke CA release or caused only slight increase in the release, but markedly potentiated stimulation-evoked release from perfused dog adrenal glands (5,7) and cultured bovine adrenal chromaffin cells (6). Earlier studies have provided evidence of the stimulating effect of glucagon on CA release from the innervated adrenal gland in intact dogs (4) and in isolated, perfused dog adrenal gland (3) We believe that in the experiments of intact dogs glucagon enhanced the release of CAs by affecting the nicotinic receptor-dependent pathway and not by releasing CAs directly. We observed that tonic sympathetic neural activity releases CAs continuously through the nicotinic receptor mechanism in conditions in which splanchnic nerves are intact. In the experiments of the perfused adrenal gland, doses of glucagon were 10-1,000 times higher than those we used in our study, indicating that glucagon is capable of releasing CAs at higher doses, in addition to its facilitatory action on the nicotinic receptor-dependent pathway.
We demonstrated that glucagon facilitated the SNS-, ACh-, and DMPP-evoked release of CAs from the dog adrenal gland but had no effect on muscarine-evoked release. These results suggest that glucagon plays a facilitatory role in adrenal CA release mediated by nicotinic receptors but not by muscarinic receptors and that increase in cyclic AMP induced by glucagon in adrenal medullary cells may be responsible for its selective action.
Acknowledgment: This work was supported in part by Grants No. 05671804 and 06672163 for Scientific Research from The Ministry of Education, Science and Culture, Japan.
1. Samols E, Weir GC. Adrenergic modulation of pancreatic A, B and D cells: α-adrenergic suppression and β-adrenergic stimulation of somatostatin secretion, α-adrenergic stimulation of glucagon secretion in the perfused dog pancreas. J Clin Invest
2. Holst JJ, Schwartz TW, Knuhtsen S, Jensen SL, Nielsen OV. Autonomic nervous control of the endocrine secretion from the isolated perfused pig pancreas. J Auton Nerv Syst
3. Scian LF, Westermann CD, Verdesca AS, Hilton JG. Adrenocortical and medullary effects of glucagon. Am J Physiol
4. Sarcione EJ, Back N, Sokal JE, Mehlman B, Knoblock E. Elevation of plasma epinephrine levels produced by glucagon in vivo. Endocrinology
5. Morita K, Dohi T, Kitayama S, Tsujimoto A. Enhancement of acetylcholine-evoked catecholamine release from perfused dog adrenals by elevating cyclic AMP levels. Arch Int Pharmacodyn Ther
6. Morita K, Dohi T, Kitayama S, Koyama Y, Tsujimoto A. Enhancement of stimulation-evoked catecholamine release from cultured bovine adrenal chromaffin cells by forskolin. J Neurochem
7. Tsujimoto A, Morita K, Kitayama S, Dohi T. Facilitation of acetylcholine-evoked catecholamine release by cyclic AMP on isolated perfused dog adrenal glands. Arch Int Pharmacodyn
8. Kimura T, Katoh M, Satoh S. Inhibition by opioid agonists and enhancement by antagonists of the release of catecholamines from the dog adrenal gland in response to splanchnic nerve stimulation: evidence for the functional role of opioid receptors. J Pharmacol Exp Ther
9. Kimura T, Shimamura T, Satoh S. Effects of pirenzepine and hexamethonium on adrenal acetylcholine release in response to endogenous and exogenous acetylcholine in anesthetized dogs. J Cardiovasc Pharmacol
10. Shimamura T, Kimura T, Satoh S. Effects of pirenzepine, AF-DX-116 and gallamine on the release of catecholamines from the dog adrenal gland in response to splanchnic nerve stimulation: inter-action of M1 and M2 receptors with nicotinic receptors. J Pharmacol Exp Ther
11. Douglas WW, Kanno T, Sampson Sr. Effects of acetylcholine and other medullary secretagogues and antagonists on the membrane potential of adrenal chromaffin cells: an analysis employing techniques of tissue culture. J Physiol (Lond)
12. Kidokoro Y, Miyazaki S, Ozawa S. Acetylcholine-induced membrane depolarization and potential fluctuations in the rat adrenal chromaffin cell. J Physiol (Lond)
13. Corcoran JJ, Krshner N. Inhibition of calcium uptake, sodium uptake and catecholamine secretion by methoxyverapamil (D 600) in primary cultures of adrenal medulla cells. J Neurochem
14. Cena V, Nicolas GP, Sanchez GP, Kirpekar SM, Garcia AG. Pharmacological dissection of receptor-associated and voltage-sensitive ionic channels involved in catechloamine release. Neuroscience
15. Garcia AG, Sala F, Reig JA, Garcia L. Dihydropyridine BAY-K-8644 activates chromaffin cell calcium channels. Nature
16. Cheek TR, Burgoyne RD. Effect of activation of muscarinic receptors on intracellular free calcium and secretion in bovine chromaffin cells. Biochem Biophys Acta
17. Nakazato Y, Oleshansky M, Yamada Y. Voltage-independent catecholamine release mediated by the activation of muscarinic receptors in guinea-pig adrenal glands. Br J Pharmacol
18. Misbahuddin M. Muscarinic stimulation of guinea-pig adrenal chromaffin cells stimulates catecholamine secretion without significant increase in Ca2+
uptake. Neurosci Lett
19. Yamada Y, Teraoka H, Nakazato Y, Ohga A. Intracellular Ca2+
antagonist TMB-8 blocks catecholamine secretion evoked by caffeine and acetylcholine from perfused cat adrenal glands in the absence of extracellular Ca2+
. Neurosci Lett
20. O'Sullivan AJ, Burgoyne RD. A comparison of bradykinin angiotensin II and muscarinic stimulation of cultured bovine adrenal chromaffin cells. Biosci Rep
21. Morita K, Dohi T, Kitayama S, Koyama Y, Tsujimoto A. Stimulation-evoked Ca2+
fluxes in cultured bovine adrenal chromaffin cells are enhanced by forskolin. J Neurochem