Exogenous Epinephrine Research Articles

Pressor responses induced by Bay K 8644 involve both release of adrenal catecholamines and calcium channel activation

1. The dihydropyridine calcium channel activator, Bay K 8644, is believed to increase mean arterial blood pressure in several animal models, as a result of direct activation of vascular smooth muscle cells by increasing calcium influx through the voltage-dependent calcium channels. The purpose of the current study was to elucidate further the mechanism of action of Bay K 8644, by examining the possibility that the pressor response to Bay K 8644 may also be the result of indirect activation of the vascular smooth muscle cells by release of adrenal catecholamines. 2. Intravenous administration of Bay K 8644 increased mean arterial pressure in a dose-dependent manner in conscious, normotensive rats. This pressor response was blocked by calcium channel blockers (nifedipine, verapamil, and diltiazem) at doses lower than were necessary to decrease resting mean arterial pressure. 3. alpha-Adrenoceptor antagonists (phentolamine, yohimbine, and prazosin) completely blocked the Bay K 8644-induced pressor responses and converted them to depressor responses. Adrenalectomy did not alter the inhibitory effect of phentolamine on the pressor response to Bay K 8644. However, adrenalectomy or adrenal demedullectomy prevented the phentolamine-induced reversal of the Bay K 8644 pressor response to a depressor response. In addition, adrenalectomy did not affect the ability of phentolamine to reverse the pressor response to exogenous adrenaline administration to a depressor response. 4. These data suggest that the pressor response to Bay K 8644 may involve both direct activation of vascular smooth muscle cells and indirect activation of the muscle cells by release of adrenal catecholamines.

Proenkephalin—A derived peptides do not modulate cardiovascular effects of epinephrine on the isolated rat atrial preparations

The catecholamines and the opioid peptides are found to be co-localized in the adrenomedullary chromaffin cells. They are co-secreted from the chromaffin granules in response to various stimuli. The stress-induced released of epinephrine is known to exert its effect on the cardiovascular system resulting in the changes in heart rate and blood pressure. However, the role of the co-released proenkephalin—A derived peptides has not been extensively characterized. Previous work from several investigators suggested that the peptides modulate cardiac functions of the catecholamines. There is considerable conflicting results among these reports. Results from the isolated rat atrial preparation indicated that enkephalins attenuated the increase in atrial rate induced by norepinephrine through restriction of the calcium fluxes. Nonetheless, others reported insensitivity of the enkephalins in similar or different test systems. We further re-examined these discrepancies using the isolated rat atrial preparation to investigate the opioid peptide modulatory effect on the cardiovascular changes induced by exogenous epinephrine. Alterations in rate and force of contraction resulting from epinephrine and the peptides were both studied in parallel. The opioid peptides used in this study were [Met 5]-enkephalin (ME), [Leu 5]-enkephalin (LE), FMRFamide, [Met 5]-enkephalyl-Arg 6-Phe 7 (MEAP), peptide E, and the non-selective opioid agonist, etorphine. We report here that none of the opioid peptides were effective in alleviating or attenuating the increase in heart rate and developed tension caused by epinephrine. The peptides did not affect the basal beating rate nor the force of contraction. Thus, the present results clearly demonstrate the insensitivity of the enkephalins in modulating the cardiac effects of epinephrine. They further indirectly support the prejunctional synaptic nerve endings as the potential peripheral site of action of the peptides.

Lack of selectivity between the uptake of [3H]adrenaline and [3H]noradrenaline into rat hypothalamic slices.

The uptake of [3H]adrenaline and [3H]noradrenaline into rat hypothalamic slices was compared for determination of whether adrenaline uptake was independent of uptake into noradrenergic neurones. Kinetic analysis revealed a similar high-affinity uptake process for both adrenaline and noradrenaline, with Km and Vmax values within similar ranges. These uptakes were inhibited by desipramine and maprotiline in a dose-dependent manner, but the selective dopamine and 5-hydroxytryptamine uptake inhibitors benztropine and fluoxetine, respectively, were without effect. Competition for uptake sites by unlabelled adrenaline with [3H]adrenaline and [3H]-noradrenaline and by unlabelled noradrenaline with [3H]-adrenaline and [3H]noradrenaline was very similar. Lesioning of the major adrenaline-containing cell group (C1 cell group) decreased the hypothalamic adrenaline concentration but had no effect on hypothalamic [3H]adrenaline or [3H]noradrenaline uptake. The results suggest that exogenous adrenaline is largely taken up by high-affinity sites on noradrenergic nerve terminals.

Effect of hormonal and substrate backgrounds on cell membrane function in normal males

Hormonal and substrate influences on in vivo cellular membrane function were evaluated in 15 healthy male volunteers. Each subject underwent serial evaluations of membrane function in the anterior tibialis muscle, as assessed by transcutaneous measurement of resting membrane potential (Em). Group A subjects (n = 9) underwent measurement of resting Em in the basal state and again during the 10th day of intravenous feeding (IVF). Group B subjects (n = 6) underwent measurement of resting Em in the basal state during epinephrine infusion and again during epinephrine infusion on the 7th day of IVF. Percutaneous needle biopsy of the vastus lateralis muscle permitted calculation of transmembrane electrolyte distribution from the Nernst equation, using the measured Em and the chloride space method. Hospitalization with intake of a defined-formula enteral diet for 3 days resulted in depolarization (P less than 0.05) of resting Em (-75.3 +/- 1.6 mV) compared with normal (-79.8 +/- 0.9 mV). Despite 10 days of subsequent IVF, further depolarization (P less than 0.05) of resting Em (-71.2 +/- 1.2 mV) was observed. In the dual presence of IVF and exogenous epinephrine infusion, there was an increase (P less than 0.05) in intracellular potassium concentration and repolarization of resting Em (-80.6 +/- 0.8 mV) to normal levels. These data indicate that hormonal background and substrate availability contribute to the in vivo modulation of cellular membrane function in human skeletal muscle, possibly through facilitation of sodium-dependent amino acid transport across the cell membrane.

Investigation of adrenomedullary function in atopic dermatitis.

Adrenomedullary function in patients with atopic dermatitis was assessed by measurement of plasma levels of catecholamines (adrenalin and noradrenalin) and cyclic AMP in response to the stimuli of standing after lying supine, and a 5-min infusion of histamine in the standing position (i.e. histamine plus standing). Plasma clearance of adrenalin was examined by measurement of plasma catecholamine and cyclic AMP levels following a 15-min intravenous infusion of adrenalin in the supine position. Resting plasma levels of adrenalin, cyclic AMP and noradrenalin were not statistically different in atopic patients and normal controls. Standing or intravenous infusion of histamine in the standing position caused a rise in plasma catecholamine levels. Plasma adrenalin, cyclic AMP and noradrenalin levels in response to these stimuli and the rate of clearance of exogenous adrenalin from the plasma were not significantly different in patients with atopic dermatitis and in normal subjects.

Epinephrine inhibits feeding nonspecifically in the rat

The hypothesis that epinephrine (EPI) and pancreatic glucagon (PG) inhibit feeding by activating a common physiological satiety mechanism was tested by comparing the two agents' behavioral effects. In several tests of specificity, EPI and PG had functionally different inhibitory actions. Intraperitoneal injection of 6.25–50 μg/kg EPI and 100–400 μg/kg PG elicited overlapping dose-related inhibitions of intake of milk diet in rats maintained ad lib on pelleted chow. Twenty-five to 50 μg/kg EPI also elicited anomalous behaviors that are not normally associated with feeding, including supine postures with limbs extended and crawling with trunk dorsoflexed and abdomen pressed against cage floor. EPI elicited similar anomalous behaviors in rats that either sham fed with open gastric cannulas, drank after water deprivation, or were presented neither food nor water. Fifty to 200 μg/kg EPI also inhibited water intake in the thirsty rats, and 25–50 μg/kg EPI inhibited sham feeding. PG, in contrast, neither elicited anomalous behaviors nor inhibited water intake nor inhibited sham feeding. These data demonstrate that the inhibitory actions of exogenous EPI and PG are functionally dissociable. We conclude that 25–200 μg/kg EPI acts nonspecifically to produce anorexia and adipsia, while PG elicits postprandial satiety.

Mechanism of action of a new prostaglandin antihypertensive, viprostol [CL 115 347; (dl)-15-deoxy-16-hydroxy-16(alpha/beta)-vinyl-prostaglandin E2 methyl ester]: (II). Effects on the adrenergic nervous system.

Viprostol [(dl)-15-deoxy-16-hydroxy-16(alpha/beta)-vinyl-prostaglandin E2 methyl ester; CL 115 347] is a new orally and transdermally active antihypertensive agent that exerts its major antihypertensive action by vasodilation. The present studies were conducted to examine its effects on the adrenergic nervous system. In cats, viprostol did not inhibit renal sympathetic nerve discharge (RSND) monitored at the postganglionic region, indicating that nerve transmission or conduction was not blocked at the ganglion or the pre- or postganglionic fibres. In cat nictitating membrane preparations in situ, viprostol partially blocked the membrane contractile response to exogenous epinephrine and norepinephrine, as well as to electrical stimulation of pre- and postganglionic fibres. In spontaneously hypertensive rats (SHR), viprostol partially blocked the vasopressor response of exogenous norepinephrine and epinephrine specifically without influencing that of angiotensin II. All these suggest that viprostol produced weak alpha-adrenoceptor blockade. Viprostol did not antagonize the tachycardia induced by stimulation of the discrete segments at C7-T1 (cardio-accelerator) of the spinal cord in pithed SHR, suggesting that viprostol did not activate the presynaptic alpha-adrenoceptors. Viprostol significantly inhibited the increase in blood pressure induced by electrical stimulation of the spinal cord at T7-T9 in pithed SHR, probably due to postsynaptic alpha-adrenoceptor blockade. In conclusion, viprostol produced weak, but statistically significant alpha-adrenoceptor blockade which may contribute partially to its antihypertensive action.

Pulmonary vascular supersensitivity to catecholamines in systemic high blood pressure

Pulmonary pressure and arteriolar resistance are elevated in uncomplicated primary systemic hypertension. This study was carried out in 16 men with this form of hypertension and in 9 healthy men to compare 1) their pulmonary vascular reactivity to endogenous catecholamines released during mental arithmetic and cold pressor tests, and 2) the dose-response relation to exogenous epinephrine and norepinephrine. Arithmetic and cold pressor tests were associated, respectively, with a predominant increase in plasma epinephrine and norepinephrine concentration; changes were significantly greater in hypertensive men. During the two tests, pulmonary arteriolar resistance in the normotensive group was reduced by 13% and augmented by 7% of baseline, respectively, whereas it was raised by 31 and 70%, respectively, in the hypertensive group. In normal subjects, the dose (microgram)-response (delta dynes) relation to epinephrine was 1 = -4, 2 = -9, 3 = -9 and 4 = -10; to norepinephrine it was 2 = +3, 4 = +6, 6 = +7 and 8 = +7. In hypertensive patients, the respective relations were 1 = +18, 2 = +44, 3 = +59 and 4 = +77; and 2 = +39, 4 = +54, 6 = +76 and 8 = +98. Group differences were highly significant. In each of these circumstances, the driving pressure across the lungs was significantly augmented in the hypertensive but not the normotensive group. Both epinephrine and norepinephrine have a vasoconstrictor influence on the lesser circulation as a consequence of vascular overreactivity. The opposite changes in resistance between normotensive and hypertensive subjects produced by epinephrine suggest that a constrictor vascular supersensitivity becomes active in the pulmonary circuit with the development of systemic high blood pressure.

Induction of delayed afterdepolarizations and triggered activity in canine Purkinje fibers by lysophosphoglycerides.

Lysophosphoglycerides accumulate in ischemic myocardium and induce electrophysiologic alterations in normoxic tissue in vitro closely resembling those seen with ischemia in vivo. Delayed afterdepolarizations and triggered activity may be particularly important in the pathogenesis of arrhythmias in the ischemic heart. The present study was performed to determine whether lysophosphatidylcholine (LPC), at concentrations comparable to those present in ischemic myocardium, can induce delayed afterdepolarizations and/or triggered activity in normoxic canine Purkinje fibers. In the present study, as little as 75 microM LPC was found to induce delayed afterdepolarizations and as little as 100 microM LPC was found to induce delayed afterdepolarizations and triggered activity even at low cycle lengths. The amplitude of the induced delayed afterdepolarizations was enhanced by augmentation of the extracellular concentration of calcium (7 mM) or by exogenous epinephrine (10(-9) to 10(-6) M). The amplitude was decreased by verapamil (1 mg/l) or Mn++ (2.5 mM). Epinephrine at a concentration of 10(-6) M also initiated triggered activity in Purkinje fibers exposed to LPC (75 microM), a response blocked by l-propranolol (2 X 10(-7) M and 10(-6) M) but not by the alpha 1-adrenergic blocking agent BE-2254 (10(-6) M). Delayed afterdepolarizations induced by LPC (75 microM) and epinephrine (10(-6) M) persisted even in the presence of acidosis (pH 6.7) and hyperkalemia ([K+]o = 7 mM). Thus, delayed afterdepolarizations and triggered activity induced by LPC may contribute to the induction and/or maintenance of arrhythmias early after the onset of myocardial ischemia. However, because of the reversal of these effects after superfusion with media devoid of LPC, they may occur with ischemia in vivo but not be seen in tissue isolated from ischemic regions and evaluated in vitro.

Factors contributing to the maintenance of arterial blood pressure in adult rats treated neonatally with guanethidine.

Adult rats treated neonatally with guanethidine had normal arterial blood pressures, but these were more dependent on the renin-angiotensin system than in control animals. Antagonism of V1-receptors (with d(CH2)5DAVP) enhanced the depressor effects of captopril in guanethidine-treated animals, but blood pressure was unaffected by this manoeuvre in control animals. Although there is evidence that sympathetic efferent vasomotor function is abolished following the schedule of neonatal guanethidine treatment used, pentolinium had a hypotensive effect in the presence of d(CH2)5DAVP and captopril, indicating that the adrenal medulla could have been contributing to the maintenance of blood pressure. This is consistent with the finding of marked supersensitivity to exogenous adrenaline in guanethidine-treated rats.

Adreneric control of insulin release from isolated islet tissue in the rainbow trout, Salmo gairdneri R

Immunoreactive insulin levels (IRI) were measured by a homologous fish insulin radioimmunoassay. An in vitro pancreatic islet superfusion technique was employed to monitor the changes in IRI in the presence and absence of specific adrenergic agonists and antagonists. Exogenous adrenaline at low concentrations (10 −10 M) inhibited IRI release but evoked an IRI stimulation at high concentrations (10 −10 M). The stimulation of IRI by adrenaline is thought to involve β-adrenoceptors located postsynaptically on the β-cell membrane as the effect of adrenaline was mimicked by the β-agonist, isoproterenol, and abolished by the β-antagonist propranolol. Phentolamine (an α-antagonist) potentiated the adrenergic stimulation of IRI, whereas yohimbine (an α 2-antagonist) was without effect. Phenylephrine ( α 1-adrenoceptor agonist) IRI release suggesting the presence of α 2-inhibitory adrenoceptors which exert a modulatory influence on adrenaline-stimulated insulin release.

Effects of Low-Dose Morphine and Fentanyl Infusions on Urinary and Plasma Catecholamine Concentrations during Scoliosis Surgery

Plasma and urinary catecholamines were measured in 20 patients during scoliosis surgery to determine whether low dose fentanyl (1.5-2.5 micrograms . kg-1 . min-1) and morphine (150-250 micrograms . kg-1 . min-1) affect the catecholamine stress response to surgery differently. In all patients epinephrine (1.6-11.4 micrograms/kg) was injected locally at the operative site for hemostasis. This was an advantage because exogenous and endogenous epinephrine undergo the same fate and the increase in epinephrine might enhance otherwise subtle effects. The data indicate that both narcotics have similar effects on catecholamine metabolism but that even low doses of fentanyl are more effective than morphine in obtunding the catecholamine response to painful stimuli. Also, postoperative differences in plasma epinephrine indicate that recovery of awareness, and thus the onset of postoperative pain, is more rapid in patients receiving fentanyl.

Postjunctional α-Adrenoceptors and the Regulation of Arteriolar Tone in Humans

The highly selective alpha-adrenoceptor agonists methoxamine (alpha 1) and azepexole (alpha 2), the less selective alpha-adrenoceptor agonists clonidine (alpha 2 greater than alpha 1) and guanfacine (alpha 2 greater than alpha 1), and the highly selective antagonists doxazosin (alpha 1) and yohimbine (alpha 2) were used to study post-junctional alpha-adrenoceptor subtypes in human blood vessels. Studies were done in healthy volunteers. All drugs were given intraarterially, and changes in forearm blood flow were measured by plethysmography. All agonists produced a significant and dose-dependent vasoconstriction. The effect of azepexole was abolished by yohimbine but hardly influenced by doxazosin. The reverse was found for methoxamine. The clonidine and guanfacine-induced vasoconstriction was partly prevented by both antagonists. Single infusions of doxazosin and yohimbine gave significant vasodilation. It is concluded that post-junctional alpha 1-and alpha 2-adrenoceptors are present in human blood vessels and contribute to basal vascular tone. In a second study, it was shown that both alpha-adrenoceptor subtypes are involved in the vasoconstrictor effect of exogenous adrenaline and noradrenaline. No preference was found of either catecholamine for the postjunctional alpha 1-and alpha 2-adrenoceptor.

PLASMA CONCENTRATIONS OF CATECHOLAMINES FOLLOWING ADRENALINE INFILTRATION DURING GYNAECOLOGICAL SURGERY

High pressure liquid chromatography with electrochemical detection has been used to measure plasma catecholamine concentrations in six gynaecological patients undergoing halothane anaesthesia for cervical cone biopsy. Mean catecholamine concentrations before infiltration were 1.01 +/- 0.23 (SEM) nmol litre-1 (185 +/- 43 pg ml-1) for adrenaline, and 2.2 +/- 0.25 nmol litre-1 (364 +/- 41 (185 +/- 43 pg ml-1) for adrenaline, and 2.2 +/- 0.25 nmol litre-1 (364 +/- 41 pg ml-1) for noradrenaline. Following infiltration with 0.5% bupivacaine 15 ml with adrenaline 1:200 000, plasma adrenaline concentrations increased to a mean peak concentration of 18.6 +/- 3.7 nmol litre-1 (3.4 +/- 0.69 ng litre-1). The lack of sympathoneuronal response was confirmed by simultaneous measurements of plasma noradrenaline concentrations, which did not change significantly. The proportion of the injected adrenaline measured in the intravascular compartment was 21.8%. The significance of intravascular absorption of exogenous adrenaline is discussed in relation to the use of halothane anaesthesia and the concurrent injection of a local anaesthetic solution.

Biphasic effect of pertussis vaccine on serum insulin in mice.

Administration of pertussis vaccine, consisting of whole, killed Bordetella pertussis organisms, causes hyperinsulinemia and enhanced secretion of insulin in response to a variety of secretagogues in rats and mice. In examining the time course and properties of this phenomenon, we discovered two distinct and separate effects of the bacteria on glucose and insulin levels in mice. First, a heat-stable (80 degrees C for 30 min) component causes a brief hyperinsulinemia which is +measureable by 1 h, maximal at 8 h, and ends in less than 48 h. This effect appears to be due to B. pertussis endotoxin, is mimicked by Escherichia coli endotoxin, and is associated with a transient, mild hypoglycemia. Second, there is a heat-labile component of the B. pertussis organism which induces a sustained (greater than 14 days), dose-dependent hyperinsulinemia which reaches a maximum at 5 to 7 days and has no associated hypoglycemia. The two effects are further distinguishable in that the early, endotoxin-induced hyperinsulinemia exhibits the normal suppressibility by exogenous epinephrine, whereas epinephrine markedly enhances the hyperinsulinemia occurring at 7 days. These two effects of B. pertussis appear to be mediated by different mechanisms and may be important in the well-recognized reactogenicity of pertussis vaccine in humans.

Effects of exogenous catecholamines on plasma catecholamines and glucose in the sea lamprey,Petromyzon marinus

The effects of exogenous dopamine (DA), norepinephrine (NE) and epinephrine (E) on endogenous catecholamine (CA) titers and glycemia were studied with a highly specific and sensitive radioenzymatic assay (REA) in cardiac-cannulated, prespawning sea lampreys. Neither DA nor NE had a specific effect on the endogenous titers of the other two CAs, or on glycemia. In contrast, E caused a strong increase of both DA and NE at three different doses, one of which must have been in the physiological ranges. This increase may be due to direct stimulation of E on the NE and DA cells. E also caused hyperglycemia 45 min after the injection; however, this effect occurred only with unphysiologically high doses. An estimation of the disappearance rate of exogenous CAs revealed a mammalian-like speed, ranging from 3–5.5 min.

Splanchnic nervous control of the stomach of the spiny dogfish, Squalus acanthias

1. The splanchnic innervation of the stomach of Squalus acanthias has been studied using an isolated perfused stomach preparation and isolated strip preparations from the smooth muscle of the stomach wall. 2. The stomach contracted in response to splanchnic nerve stimulation, and this response was mimicked by all three putative transmitter substances tested: adrenaline, acetylcholine and 5-hydroxytryptamine. The response to nerve stimulation was not affected by atropine (10 −6M) or methysergide (10 −6M), but was abolished by phentolamine (10 −6M). This concentration of phentolamine had little or no effect on the responses of the stomach to exogenous acetylcholine or 5-hydroxytryptamine, but significantly blocked the response to exogenous adrenaline. 3. The results are interpreted in favour of an adrenergic excitatory splanchnic innervation of the dogfish stomach. 4. A “rebound” contraction of the stomach normally occurred upon cessation of the splanchnic nerve stimulation. This response disappeared together with the primary contraction after administration of phentolamine, suggesting the involvement of an adrenergic mechanism in the occurrence of the “rebound” contraction in this preparation. 5. The present experiments did not provide any conclusive evidence for a splanchnic nervous control of the gastric vasculature.

Alpha- and beta-receptor control of catecholamine secretion from isolated adrenal medulla cells.

The regulation of secretion of catecholamines from bovine adrenal medulla cells was investigated by use of an improved and highly efficient method for isolating viable and responsive cells from this tissue. The method involves an in situ collagenase perfusion affecting only the connective tissue matrix of the medulla while leaving the cortex intact. The cells released both epinephrine and norepinephrine in response to stimulation by 100 microM acetylcholine. The ratio of epinephrine to norepinephrine in the medium following non-stimulated (basal) release, was similar to that found in the intact cells. On the other hand, a lower ratio of epinephrine to norepinephrine was found in the medium following stimulation by acetylcholine due mainly to preferential secretion of norepinephrine. This release ceased after 15 min of incubation and consisted of 15--20% of the catecholamines initially present in the cells. Exogenous epinephrine was found to inhibit total catecholamine secretion; however, it stimulated norepinephrine release. Addition of isoproterenol caused a stimulation of release while propranolol was inhibitory. Norepinephrine inhibited total release not favoring any specific catechol. Other alpha-agonists, such as clonidine, also had an inhibitory effect. These results suggest a receptor-mediated mechanism for the fine regulation of secretion from the adrenal medulla.

Normal responses of rats with transplanted neonatal islets to stress, epinephrine, and cortisol.

Sympathetic responses were evaluated in rats transplanted with neonatal pancreatic tissue. Transplantation was by one of two methods: mild collagenase digestion with subsequent intraportal injection, or direct placement of intact pancreatic tissue under the renal capsule. Compared with control intact animals, both groups of transplanted rats reverted to normal fasting and postglucose load plasma glucose and insulin. The hyperglycemia characteristic of stress or of injected exogenous epinephrine was present and similar in all three groups. The response to an intragastric glucose load or to tolbutamide injection during stress was similar in transplanted and in control animals. All three groups of rats responded to sustained cortisol injection with a compensatory hyperinsulinemia but maintained normal plasma glucose, water intake, and urine volume. No evidence could be detected in this study of an exaggerated "denervated" response of the transplanted islets. The normal sympathetic responses of these neonatal islet preparations in which little or no collagenase was used, in comparison with the abnormal responses previously reported for collagenase-treated individually isolated adult rat islets, suggest differences in recuperative powers of the sympathetic innervation in islets, possibly related to age and to method of preparation of transplanted tissue.

Epinephrine-induced conditioned taste aversion

The ability of the hormone epinephrine to induce a conditioned taste aversion was investigated in three experiments. Aversions induced by epinephrine appear more easily conditioned to a highly preferred sucrose solution than to a less preferred sucrose/saline solution. A dose-response curve also exists; better conditioning is evident following injection of either 0.3 or 0.5 mg/kg epinephrine than it is following a 0.1 mg/kg dose. A arcadian rhythmicity also occurs for epinephrine-induced sucrose aversions. When the endogenous levels of this hormone are at trough, conditioning is greater than when they are elevated. This circadian pattern is interpreted with respect to the degree of discriminability between endogenous and exogenous epinephrine levels.