Adecade ago, it was proposed that the primary

Transcription

1 Brief Review Epinephrine and the Genesis of Hypertension John S. Floras Several lines of evidence suggest a psychophysiological link between stress, adrenomedullary activation, and the genesis of hypertension. Experimental data support four important concepts: 1) epinephrine stimulates prejunctional ft-adrenergic receptors that facilitate norepinephrine release from sympathetic nerve endings; 2) epinephrine can be converted into a cotransmitter by neuronal uptake and on subsequent release augment the simultaneous discharge of norepinephrine; 3) exogenous epinephrine can induce sustained hypertension in rats; and 4) there is a period of critical sensitivity to endogenous epinephrine in a genetic model of rat hypertension. Plasma epinephrine concentrations are elevated in many young subjects with borderline or mild hypertension. The hypothesis that intermittent surges in epinephrine could initiate or promote the development of primary hypertension by amplifying peripheral neurotransmission, both directly (facilitative effect) and indirectly (cotransmitter action), is supported by reports that hemodynamic and noradrenergic responses to sympathetic activation can be augmented by increases in endogenous epinephrine or by its local or systemic (up to 30 ng/kg/min) infusion. Such responses have been documented in both normotensive and hypertensive subjects and can be blocked by propranolol. Although the weight of evidence (mostly indirect) indicates that epinephrine can augment norepinephrine release in humans, the epinephrine hypothesis, itself, remains unproven. Expression of hypertension by this mechanism may be restricted to a specific epinephrine-sensitive subset of individuals with a genetic predisposition to high blood pressure. (Hypertension 1992; 19:1-18) Adecade ago, it was proposed that the primary abnormality in essential hypertension may L. be increased release of epinephrine from the adrenal medulla. 1 According to this hypothesis, small repetitive increases in circulating epinephrine cause hypertension indirectly, rather than directly, by stimulating prejunctional ft-adrenergic receptors that facilitate exocytotic norepinephrine discharge from sympathetic nerve endings (Figure 1). An episodic or sustained increase in norepinephrine release could then initiate, permit, or promote the development of high blood pressure. Folkow 2 has identified "excitatory psychoemotional influences" as one of two environmental stimuli that may reinforce or sometimes precipitate primary hypertension in those with a genetic predisposition for high blood pressure. The concept From the Division of Cardiology and the Centre for Cardiovascular Research, Toronto General Hospital, University of Toronto, Toronto, Canada. Supported by grants-in-aid from the Heart and Stroke Foundation of Ontario. J.S.F. is the recipient of a Career Scientist Award from the Ministry of Health of the Province of Ontario. The opinions expressed in this review are those of the author and not of the Ministry. Address for correspondence: Dr. John S. Floras, Division of Cardiology, 12 EN-234, Toronto General Hospital, 200 Elizabeth Street, Toronto, Ontario M5G 2C4, Canada. Submitted March 5, 1991; accepted in revised form Jury 16, that episodic increases in plasma epinephrine, if of sufficient magnitude, could act as the physiological link between such "stressful" environmental influences and the development of hypertension in predisposed individuals is intuitively attractive. Thresholds for epinephrine's cardiovascular and metabolic actions are at plasma concentrations that can be replicated by mental stress. 3 " 8 Early hypertension in many subjects is associated with a pattern of enhanced autonomic cardiovascular drive at rest and increased sympathoadrenal reactivity, analogous to the hypothalamic defense reaction, in response to mental stress but not in response to physical tasks In the mind of the lay public, "stress" has long been associated with "adrenaline" and the genesis of hypertension. This concept has been reinforced by reports of hypertension after the stress of battle 10 or exposure to noise, 1112 descriptions of relations between stress, personality, and blood pressure and plasma epinephrine concentrations, and documentation in prospective studies of environmental changes that promote or protect against primary hypertension A facilitative effect on peripheral noradrenergic transmission has been observed both during and after exposure to epinephrine. Aftereffects of epinephrine are presumed to be through conversion of epinephrine from a hormone to a cotransmitter since this catecholamine can be incorporated into postgangli-

2 Hypertension Vol 19, No 1 January 1992 briiuqun/ Facftaflon ofnonpinspmn* ofnorapkwphrt fls of EpMis (Humor*) ofnorepkiaphrfiv Auj nailed Et»dor (MResporaa CelRuponsa FIGURE 1. Schematic representation (much simplified) of prejunctional and postjunctional adrenergic receptors at the sympathetic neuroeffector junction in the peripheral nervous system. Left panel: Low level nerve stimulation releases norepinephrine (NE), which has a higher affinity for inhibitory a r than facilitative firadrenergic receptors. Center panel: Direct activation of prejunctional Pradrenergic receptor by circulating epinephrine ([EPI]) facilitates neurotransmitter release. Right panel: Sustained aftereffects of [EPIJ incorporated into sympathetic varicosity (neuronal uptake) and converted from hormone to cotransmitter. Positive feedback loop by which activation of prejunctional ^-receptor by neurally released [EPI] facilitates NE release and augments effector cell response. Effector cell responses to facilitation of norepinephrine release are increased vascular contraction (a/) and increased heart rate (f),). onic sympathetic nerves and released, with norepinephrine, up to 24 hours after its uptake Through this local positive feedback mechanism, endogenous epinephrine, within the synaptic cleft, could augment norepinephrine release both during episodes of sympathoadrenaj activation and afterwards when plasma concentrations have returned to basal levels 18 * (Figure 1). The sustained, indirect aftereffects on norepinephrine release of brief, repetitive increases in plasma epinephrine concentrations may be of greater functional importance in the genesis of hypertension than its immediate direct, and often brief, cardiovascular actions as a circulating hormone. 1 ' In addition to these actions at the neuroeffector junction, stimulation of ft-adrenergic receptors by epinephrine also facilitates hypothalamic 34 and adrenal catecholamine release. As both norepinephrine and epinephrine are released from these tissues, these observations suggest a potential positive feedback loop that would allow epinephrine to augment its own release via "autoreceptor" stimulation. In this way, a relatively small increase in epinephrine could amplify neurotransmission through several closely linked central as well as peripheral 0-adrenergic mechanisms. Although these two components (facilitative and cotransmitter) (Figure 1) of the epinephrine hypothesis are supported by a number of studies in isolated tissues, intact normotensive animals, experimental models of hypertension, and in humans, not all experimental data are consistent with the epinephrine hypothesis. A review of these experimental data will be piesented briefly with principal emphasis on evidence for facilitated release of norepinephrine and induction of experimental hypertension by epinephrine. Review of investigations in humans will focus on 1) evidence for increased circulating epinephrine in primary hypertension, 2) evidence for enhanced prejunctional /^-responsiveness in early hypertension, 3) evidence for facilitated release of norepinephrine by epinephrine in normal and hypertensive humans, and 4) the functional significance of augmented sympathetic neurotransmission by epinephrine as it relates to circulatory regulation. Experimental Evidence for Facilitated Release of Norepinephrine by Epinephrine Direct Effects of Epinephrine on Norepinephrine Release Evidence that epinephrine and other /3-adrenergic receptor agonists can facilitate exocytotic norepinephrine release from sympathetic varicosities by stimulating prejunctional /^-receptors (Figure 1) has been extensively reported and reviewed ^0-32^6-46 Facilitative prejunctional ft-receptors have also been described. 47 " 49 Functionally, one could consider three potential roles for such receptors: as autoreceptors 50 activated by neuronally released norepinephrine, as "heteroreceptors" 50 activated by epinephrine, as a circulating hormone, or as autoreceptors stimulated by epinephrine transformed from a hormone to a neurotransmitter by its uptake and storage in sympathetic varicosities (Figure 1). There is little evidence that the neurotransmitter norepinephrine potentiates

3 Floras Epinephrine and Hypertension Electrical Stimulation, 2Hz Norepinaphrint, 5xlO' T M to o o. to tu a o Z O "OOX IBS- O «5j o* K>O% - a itt i SIQ Human Z UJ o <fc (u 0. 0 L > LOG CONCENTRATION OF ISOPROTERENOL,» FIGURE 2. Line plots show comparison of the effects of increasing concentrations of isoproterenol on the contractile responses of human and canine saphenous veins to electrical activation of the adrenergic nerve endings (left panel) and exogenous norepinephrine (right panel). Data are expressed as percent of the maximal contractile response to the stimulation used and are shown as mean±sem. Numbers indicate how many preparations from different donors were used at each concentration. In the canine vein, the effect of isoproterenol was comparable during contractions evoked by electrical stimulation and exogenous norepinephrine, except that at 3xlO~ 6 M, the depression in the presence of the latter was significantly larger. 'Change caused by isoproterenol is statistically significant from control (p<0.05). Reproduced with permission. 30 its own release. 28 ' Rather, epinephrine, whose affinity for ft-adrenergic receptors is more than 200- fold greater than that of norepinephrine, 51 appears to be the "natural agonist" for these receptors in humans ^3 Prejunctional ft-adrenergic receptors may respond preferentially to circulating as opposed to neurally released catecholamines. 54 Facilitated release of [ 3 H]norepinephrine from tissues preloaded with the labeled neurotransmitter has been documented in vitro at epinephrine concentrations in the range of 10" 10 to lo^m. 42 This is within the physiological range of plasma epinephrine concentrations. 55 At higher concentrations (more than 10" 8 M), epinephrine, like norepinephrine, acts on prejunctional a r adrenergic receptors to inhibit sympathetic neurotransmission ^3 (Figure 1). That this facilitative action is a nonspecific effect of epinephrine is unlikely since it is abolished or attenuated by ft-adrenergic receptor blockade ^ Although prejunctional ft-adrenergic receptor-mediated facilitation of norepinephrine release appears due to increased Ca 2+ entry into nerve terminals, resulting in increased intraneural cyclic AMP, the exact mechanism by which a 2 -receptor stimulation inhibits neurotransmitter release has not been established. 56 ' 57 ' 59 The maximal enhancement of norepinephrine release by epinephrine is relatively small (from 13 to 133%, depending on tissue and preparation) when compared with the effect of atj-adrenergic receptor antagonists on stimulated norepinephrine release, 42 ' 45 and the functional consequence of ftadrenergic receptor stimulation varies from species to species. 2^33-36 ' 39 ' 42 ' The facilitative effect of isoproterenol on electrically stimulated norepinephrine release from adrenergic nerves of saphenous veins is greater in humans than in the dog and is associated with an augmented (human) as opposed to a diminished (canine) contractile response when these two interventions are combined 30 (Figure 2). Such quantitative differences make it difficult to predict responses to epinephrine in humans from experimental observations in other species. Inhibitory prejunctional a 2 -receptors and facilitative ft-adrenergic receptors have been documented in a number of human vascular tissues, including omental arteries and veins, digital arteries, and saphenous veins, ' 53 ' and functionally important ft-receptors nave been demonstrated in the human heart. 65 Whether prejunctional inhibitory aradrenergic receptors (autoreceptors) must first be blocked to detect the facilitative effects of ^-adrenergic receptor (heteroreceptor) stimulation on norepinephrine release in humans 66 will depend on a number of factors (such as the ratio and sensitivity to agonists of prejunctional a^ft-adrenergic receptors and the width of the junctional cleft) that vary from tissue to tissue The strength of negative feedback regulation by a 2 - a dr ener gi c receptors, for example, is inversely proportional to the width of the junctional cleft It should not be surprising, therefore, to find variations in the effects of epinephrine on neu-

4 Hypertension Vol 19, No 1 January 1992 retransmitter release or neurogenic vasoconstriction in different vascular beds in humans. 41 ' 42 " Sustained Aftereffects of Epinephrine on Norepinephrine Release In 1932, Burn 68 observed augmented neurogenic vasoconstriction in vivo both during and after exposure to epinephrine, but the latter response fatigued after successive nerve stimulation. Burn suggested this effect of epinephrine was due to its storage and subsequent release. The affinity of neuronal uptake-1 for epinephrine is about half that for norepinephrine. 23 In the mouse, about 35% of injected epinephrine and about 55% of injected norepinephrine appear to be taken up and released by this neuronal mechanism. 69 Myocardial extraction and stimulated release of epinephrine have been documented in dogs and in human subjects Approximately 75% of this extraction can be blocked by desipramine. 70 In dogs, the ratio of epinephrine to norepinephrine released on stellate stimulation is comparable to the ventricular content of these catecholamines Esler and his colleagues 70 have documented cardiac epinephrine spillover, averaging 2 ng/min or 2% of the corresponding norepinephrine spillover, in six patients with untreated congestive heart failure (but not in resting healthy volunteers), and higher levels of neuronal epinephrine release from their gut, liver, lungs, and kidneys. Approximately 25% of the total plasma epinephrine appearance rate in these patients with heart failure could be attributed to regional overflows from these organs. In marked contrast to its plasma half-life, measured in minutes, 71 the half-life of neuronal epinephrine is approximately 4 hours. 20 Since the release of epinephrine may be detected up to 24 hours after its neuronal uptake, the potential for neurally released epinephrine to increase the simultaneous discharge of norepinephrine by stimulating prejunctional /32-adrenergic receptors, and thereby augment neurogenic vasoconstriction at the neuroeffector junction, should persist long after a surge in endogenous epinephrine subsides and its local concentration returns to basal levels To illustrate this concept, Majewski et al 72 demonstrated in anesthetized rabbits a 50% increase in norepinephrine spillover into plasma after an epinephrine infusion. This was at a time when tissue epinephrine concentrations were elevated, but plasma epinephrine concentrations were not. These effects on norepinephrine release could be attenuated by propranolol and abolished by prior uptake-1 blockade (desipramine). 72 Whether neurally released epinephrine exerts such positive feedback in humans remains controversial Molderings and colleagues 41 could not detect any facilitate aftereffect of newly taken up (tritiated) epinephrine (10 nmol/1) on norepinephrine release from human saphenous veins. Rat Models of Hypertension Several groups have shown that infusion or depot implantation of epinephrine in doses that do not increase its plasma concentration can cause sustained elevation of blood pressure in normotensive rats. This conclusion has been challenged by others. 75 The increase in blood pressure in two of these protocols could be blocked if metoprolol was coadministered These particular experiments suggested that a threshold uptake into atria of 46 pmol epinephrine/g tissue was sufficient (but 14 pmol/g insufficient) to generate an autofacilitative feedback loop. 29 Stress-induced hypertension (immobilization) in normal rats appears also critically dependent on the presence of epinephrine. 76 Spontaneously hypertensive rats (SHR) are similar to humans with primary hypertension in that neurohumoral alterations and excitatory psychosocial stimuli reinforce their genetic predisposition to high blood pressure. 2 SHR display several characteristics of augmented sympathoadrenal activity. Plasma and ganglionic epinephrine concentrations, epinephrine synthesis, 79 sympathetic ganglionic ft-receptor concentrations, 78 and adrenergic responses to stress 80 are all increased when compared with Wistar-Kyoto (WKY) rats, and central epinephrine concentrations and PNMT activity rise in parallel with blood pressure. 81 Surgical resection of the adrenal medullae of 4-6- week-old SHR attenuates the development of hypertension without affecting these animals' growth or heart rates. 52 This is the period of most rapid growth and development in this strain; increased vascular resistance is already established by 4 weeks of age. 82 The pressor response to neurally released norepinephrine in these experiments was reduced in the isolated perfused mesenteric bed of demedullated rats and could be augmented by adding epinephrine to the perfusate. In contrast, there was no apparent difference between sham and demedullated SHR in postjunctional responsiveness to norepinephrine or epinephrine in this vascular bed. 52 The antihypertensive effect of demedullation could be abolished by selective ft-agonists or epinephrine depot implants and restored by concurrent selective ft-adrenergic receptor blockade. 83 Because these observations contradict an earlier report that adrenal demedullation does not affect established hypertension in adult SHR 85 and are inconsistent with a large body of evidence for reduced sensitivity and responsiveness of postjunctional /J-adrenergic receptors in SHR (particularly older SHR 86 ), Borkowski 52 has proposed that /^-adrenergic receptor-mediated facilitation of norepinephrine release by epinephrine contributes to the development (rather than the maintenance) of hypertension in this model during a period of "critical sensitivity." A number of other approaches have been used to detect prejunctional facilitation of norepinephrine release in SHR. These include quantitation of neu-

5 Floras Epinephrine and Hypertension rotransmitter release or neurogenic vasoconstriction in pithed rats or in isolated vascular beds. In one such study, acute bilateral adrenal demedullation reduced pressor responses in pithed SHR, whereas the subsequent infusion of epinephrine (50 ng/min) enhanced pressor responses to nerve stimulation without affecting those induced by norepinephrine. This potentiation by epinephrine could be abolished by pretreatment with ICI 118,551 (a ^-selective antagonist) but not by atenolol. 87 Several groups have described facilitated norepinephrine release and augmented neurogenic vasoconstriction by /3-adrenergic receptor agonists in the mesenteric, renal, 36 and portal vascularure 89 of SHR and WKY rats; responses in the mesenteric vascular bed were potentiated in SHR Interpretation of responses in isolated vascular beds is complicated by tissue and species variation in prejunctional p- and a-adrenergic receptor distribution, density, or responsiveness to agonists. At low concentrations (5.5 nm/1), epinephrine potentiates electrically induced norepinephrine overflow from mesenteric sympathetic nerves of SHR and augments the pressor response to this stimulus in this vascular bed but has the opposite effect in normotensive WKY rats. 90 Both effects of epinephrine are antagonized by propranolol. At higher doses (14 nm/1), epinephrine decreases norepinephrine spillover in both strains, but this attenuation is muted in SHR and can be reversed by administration of the prejunctional a 2 - receptor antagonist yohimbine. 90 Feldman 86 has reviewed evidence for reduced postjunctional ft-adrenergic receptor sensitivity and responsiveness in adult SHR. Whether a parallel reduction in prejunctional ^-adrenergic receptor responsiveness occurs with age and stage of hypertension is unclear. Prejunctional responsiveness to the infused ft-receptor agonists salbutamol and procaterol can be demonstrated at concentrations below those at which their postjunctionally mediated vasodilator effects become apparent. 91 That prejunctional ^-responsiveness decreases with age is supported by two observations. Facilitation by isoproterenol of electrically stimulated norepinephrine release is greater in superfused spleen strips of SHR than WKY rats at 5 weeks of age but is similar in preparations obtained from 15-week-old animals. 92 Second, isoproterenol-induced vasodepressor responses are intact in 6-8-week-old SHR (i.e., in the developmental stages of hypertension) but are attenuated in 16-week-old SHR. 93 Epinephrine would therefore appear to play an important role at a critical stage of enhanced &- adrenergic receptor responsiveness in the initiation and development (but not in the maintenance) of spontaneous hypertension in rats. Plasma Catecholamine Concentrations in Primary Hypertension Investigation of the contribution of the sympathetic nervous system to the initiation and maintenance of primary hypertension has focused principally on indirect measures of sympathetic traffic such as venous plasma norepinephrine concentrations. This is despite recognized limitations of this approach Although few clear-cut differences between normotensive subjects and patients with high blood pressure have emerged from these studies, 97 virtually all group comparisons in subjects 40 years or younger have demonstrated increased venous plasma norepinephrine concentrations in those with primary hypertension. 98 " 100 Organ-specific evaluation of sympathetic activity by the radiotracer kinetic technique has also documented increased total and regional (cardiac, renal) norepinephrine spillover into plasma in young hypertensive subjects. 101 More recently, microneurographic studies have demonstrated increased sympathetic traffic to muscle in young subjects with borderline hypertension These observations suggest an important sympathoneural contribution to the initiation of hypertension in many of these young subjects. What is the evidence that epinephrine release, either intermittent or sustained, is increased in younger subjects during the developmental stages of hypertension or that increased plasma norepinephrine concentrations and the augmented norepinephrine spillover in these subjects are a consequence of facilitated neurotransmitter release by epinephrine? In normotensive subjects, plasma epinephrine concentrations remain stable or decrease with age, whereas norepinephrine concentrations tend to rise 98-io5 The positive correlation between age and norepinephrine is absent in hypertensive subjects. 106 If epinephrine release were relatively enhanced in young individuals genetically predisposed to hypertension, its central and peripheral actions could explain several of these age-related observations. A central action 34 could augment efferent sympathetic outflow, its facilitative effects at the adrenergic terminal could augment norepinephrine spillover, and its action on adrenal ft-receptors could autofacilitate its own release Elevated venous plasma epinephrine concentrations have been documented in borderline hypertension, in hypertensive subjects with rapid resting heart rates, 98 and in many subjects with wellestablished essential hypertension, at rest, in response to mental stress, and during activities such as submaximal exercise and the cold pressor test Although these increases in resting venous plasma epinephrine concentrations are modest, with average values of 13 pg/ml or about 30% above those recorded in normotensive subjects, they are reported in most studies. 98 Kjeldsen and his colleagues 112 have detected elevated arterial as well as venous plasma epinephrine and norepinephrine concentrations in middle-aged men with long-standing untreated essential hypertension. Arterial plasma epinephrine concentrations were increased by 73 pg/ml or 80% above those in age-matched normotensive subjects and were signif-

6 Hypertension Vol 19, No 1 January 1992 icantly greater than venous concentrations in both groups, indicating uptake of epinephrine by forearm tissue. The arteriovenous difference for epinephrine was greater, in absolute terms, in the hypertensive subjects. Of particular interest, in the context of the epinephrine hypothesis, was the observation that venous norepinephrine concentrations were slightly below arterial levels in the normotensive subjects, whereas they were 10% above arterial concentrations in the hypertensive group. Since forearm extraction of epinephrine and norepinephrine is similar, 113 these findings are consistent with both increased adrenal epinephrine release and augmented norepinephrine release from this specific vascular bed in these hypertensive subjects, but they do not demonstrate a causal relation between the two observations. One problem with the epinephrine hypothesis is that not all studies (only three of 15 reviewed by Goldstein and Lake to ) show distinct increases in both catecholamines in hypertension. Subjects were not always subgrouped according to age in these reports. A second difficulty is that plasma epinephrine concentrations in borderline hypertensive subjects have not been shown to predict the subsequent development of sustained hypertension. 114 Unfortunately, most investigators have studied venous not arterial epinephrine concentrations, and because venous measurements often differ substantially from arterial epinephrine concentrations, particularly during stimuli such as mental stress or orthostasis, they will underestimate adrenomedullary epinephrine release It is therefore difficult to mount a strong challenge to the epinephrine hypothesis based on apparently negative observations derived from venous sampling. 96 To obtain an integrated measure of catecholamine output over time, Brown and his colleagues 116 obtained sequential 24-hour urine collections for epinephrine and norepinephrine from 268 placebotreated patients (mean age 51 years) enrolled in the Medical Research Council Trial for Hypertension and compared these values with those of similar collections from age- and sex-matched normotensive controls. 116 Urine catecholamines were not increased, but since few of the hypertensive subjects were aged less than 40 years, these data provide little insight into epinephrine's potential contribution to the initiation or development of high blood pressure. ft-adrenergic Receptors in Primary Hypertension Whether the facilitative effect on norepinephrine release of normal or increased circulating epinephrine is potentiated early in the development of primary hypertension should depend on the responsiveness of prejunctional ft-adrenergic receptors to this agonist at this time. Many young patients with borderline or mild established hypertension have characteristics consistent with increased postjunctional 02-adrenergic receptor responsiveness, such as increased heart rate, cardiac output, or forearm blood flow. Often these patterns can be "normalized" by y3-blockade Although increased forearm vascular, natriuretic, and platelet responses to low-dose epinephrine infusion have been documented in one group of 40-year-old hypertensive men, 121 the majority of studies of older subjects and those with established hypertension have demonstrated decreased postjunctional /3-adrenergic responsiveness to infused agonists Since most investigators have documented increased lymphocyte ft-receptor density in hypertension, 124 " 127 reported reductions in postjunctional ftreceptor responsiveness in these subjects could represent a progressive defect in ft-adrenergic signal transduction as hypertension becomes established. 123 Indeed, Feldman et al 128 described similar preceptor densities in a relatively small number of normotensive and borderline hypertensive subjects, but lymphocyte ^-receptor-stimulated adenylate cyclase activity was reduced, indicating a functional uncoupling of the lymphocyte ^-adrenergic receptor from the adenylate cyclase complex in borderline hypertensive subjects. More recently, Feldman 129 has documented a reduction in isoproterenol-mediated venodilation in young subjects (less than age 35 years) with borderline-to-mild hypertension. Both defects could be reversed by sodium restriction A difficulty in interpreting such comparisons of normotensive and hypertensive subjects is that lymphocyte /3-receptor numbers and /3-adrenergic-stimulated adenylate cyclase activity are, themselves, subject to modification by exogenous or endogenous epinephrine A transient doubling of /3-receptor binding sites in mononuclear cells and a greater heart rate response to a pulse of isoproterenol can be induced by a prior minute infusion of epinephrine or isoproterenol. 130 A single session of exhaustive exercise has similar effects. 132 Thus, as well as facilitating noradrenergk transmission, a transient rise in epinephrine might also amplify acutely /3-adrenergic receptor-mediated chronotropic responses through such a postjunctional mechanism. Although some of these observations could be considered consistent with the concept of a period of increased sensitivity to epinephrine during the developmental stages of hypertension, the assumption that prejunctional and postjunctional ft-adrenergic receptor number and responsiveness are altered in parallel in primary hypertension is, as yet, unproven. Prejunctional Modulation of Norepinephrine Release by Epinephrine in Humans Direct Effects of Epinephrine on Norepinephrine Release Musgrave et al 133 detected increases in both arterial and venous norepinephrine concentrations in response to 10 ng/kg/min epinephrine. A venousarterial norepinephrine gradient was present at rest in their hypertensive but not their normotensive subjects, and this gradient increased during the infusion. Nezu et al 100 documented higher venous

7 Floras Epinephrine and Hypertension norepinephrine concentrations during an epinephrine infusion of jig/min (about 20 ng/kg/ min) and prevented this epinephrine-stimulated increase in plasma norepinephrine concentration by /3-adrenergic receptor blockade. Despite these promising observations, on the whole systemic infusions of /3-receptor agonists such as epinephrine, isoproterenol, and salbutamol have not had consistent effects on plasma norepinephrine concentrations, and whenever increases in norepinephrine concentrations have been observed, these have been similar in magnitude in normal and hypertensive subjects This lack of consensus reflects two major difficulties with this experimental approach. First, when epinephrine is infused systemically in doses even as low as 0.05 nmol/kg/min (8.5 ng/kg/min) (i.e., to replicate plasma concentrations documented during mental stress or physical exercise 138 ), plasma cyclic AMP concentrations (a marker for stimulation of ft-adrenergic receptor) rise by 50%, 6 but heart rate, stroke volume, and pulse pressure also increase. 36 These hemodynamic perturbations will alter sympathetic outflow reflexiveh/ 139 and also modify the release and/or clearance of norepinephrine independently of any prejunctional influence on noradrenergic transmission Second, high plasma or local concentrations of epinephrine could inhibit norepinephrine release by stimulating prejunctional a 2 - a( irenergic receptors. 20 ' Those studies that have demonstrated increased norepinephrine concentrations have infused epinephrine at doses of 30 ng/kg/min or less, whereas Brown et al 66 did not detect any increase in plasma norepinephrine concentrations in normal subjects during epinephrine infusions of 50 ng/kg/min unless idazoxan, an a 2 -blocker, was administered concurrently. An alternative experimental strategy has been to determine whether endogenous or exogenous epinephrine alters the noradrenergic response to stimuli that increase central sympathetic outflow. De Champlain 108 documented augmented noradrenergic responses to standing in hypertensive subjects with high plasma epinephrine concentrations. In another study by Vincent and his colleagues, 140 increases in plasma norepinephrine during a cold pressor test and isometric exercise were augmented during systemic infusion of 30 ng/kg/min (or 0.16 pmol/kg/min) epinephrine (Figure 3). This effect of epinephrine appeared to be mediated via a /3-adrenergic mechanism because it could be blocked if propranolol was coadministered. On the other hand, Lenders et al 141 have recently shown that the pressor and noradrenergic responses to mental arithmetic, head-up tilt, and the cold pressor test are not attenuated in adrenalectomized women treated with 9-a-fluorocortisol and glucocorticoids. Since these are more potent stimuli to neural norepinephrine than to adrenal epinephrine release, this apparent challenge to the epinephrine hypothesis may simply reflect the failure of such stimuli to raise plasma, myocardial, or vascular epinephrine concentrations in adrenalry in- NORAORENALINE pg/nl SCO _ txfort aftir COLD PRESSOR TEST bvfora afttr ISOMETRIC EXERCISE FIGURE 3. Line plots show plasma norepinephrine concentrations before and after cold pressor and isometric exercise test, during systemic infusion of saline (o) or epinephrine (30 ng/kg/min) ( ). Blood pressure responses to these two stimuli (not shown) were also augmented by epinephrine. *p<0.05; **p<0.01. Reproduced with permission. 140 tact normotensive women above a threshold level needed to demonstrate facilitated norepinephrine release. To quantitate its effects on central sympathetic outflow, plasma norepinephrine concentrations, and norepinephrine spillover, Persson et al 139 recorded arterial and central venous pressure, arterial and femoral venous catecholamines, peroneal muscle sympathetic nerve activity (microneurography), and norepinephrine spillover (radiotracer technique) during the stepwise infusion of epinephrine in doses selected to replicate its physiological plasma concentration in healthy volunteers ( nmol/kg/min; i.e., about ng/kg/min). Although limited by its small sample size and lack of a control study day, this technically demanding investigation is noteworthy in several respects: even at its lowest infusion rate epinephrine decreased diastolic blood pressure, and increased sympathetic traffic to muscle, norepinephrine spillover, and venous norepinephrine concentration from pre-epinephrine values was larger than the fractional increase in total muscle sympathetic nerve activity (Figure 4). The authors' interpretation of these observations was that epinephrine, at physiological concentrations, could augment norepinephrine spillover by two distinct mechanisms: increased central sympathetic nerve outflow and facilitated norepinephrine release.

8 8 Hypertension Vol 19, No 1 January Time (min) FIGURE 4. Line plot shows ratio between the mean fractional increases in venous plasma norepinephrine concentrations (pna%) and the mean fractional increase in total muscle sympathetic nerve activity (MSA%) during and after stepwise infusion of epinephrine from 0.05 to 0.6 nmol/kg/min. Reproduced with permission. 139 Sustained Aftereffects of Epinephrine on Norepinephrine Release That endogenous epinephrine, transported into sympathetic varicosities, could subsequently increase the simultaneous neural discharge of norepinephrine in humans was demonstrated by Nezu et al, 100 who increased endogenous epinephrine concentrations by injecting 1.0 mg glucagon. After an initial brisk peak, epinephrine concentrations subsided to basal levels within 20 minutes, whereas norepinephrine concentrations remained elevated. This pattern differs from the parallel time course of glucagon-stimulated norepinephrine and epinephrine release from the isolated adrenal gland. 142 Propranolol did not influence the immediate surge in both catecholamines but abolished the subsequent sustained elevation of norepinephrine. Noradrenergic responses to glucagon and epinephrine were similar in normotensive and hypertensive subjects. 100 Persson et al 139 documented greater increases in muscle sympathetic nerve activity and norepinephrine spillover after, not during, the epinephrine infusion described earlier, but the fractional increase in venous norepinephrine at this particular time was less than the fractional increase in muscle sympathetic nerve activity (Figure 4). Thus, there was no evidence for any additional prejunctional influence of neuronally released epinephrine on norepinephrine spillover. Rather, changes in norepinephrine spillover, at this time, appeared to be a reflex response to an abrupt drop in central venous pressure when epinephrine was stopped. Both Brown et al 66 and Musgrave et al 133 have commented on a dissociation between plasma norepinephrine (increased) and epinephrine concentration (basal) after systemic infusion of epinephrine. Because central hemodynamics were not measured in these studies, these observations may also be explained simply by a fall in central venous pressure and do not prove sustained facilitated norepinephrine release by epinephrine. Functional Consequences of Facilitated Norepinephrine Release by Epinephrine in Humans Because of difficulties in interpreting noradrenergic responses to epinephrine infusions, more recent efforts have focused on the functional cardiovascular consequences of increases in plasma epinephrine concentrations. The general purpose of these experiments was to determine whether hemodynamic responses (e.g., changes in blood pressure, heart rate, and forearm vascular resistance) to stimuli that increase central sympathetic outflow could be augmented during or after an infusion of epinephrine. Direct Effects of Epinephrine For example, systemic infusion of epinephrine (30 ng/kg/min) has been reported to amplify increases in both blood pressure and plasma norepinephrine during a cold pressor test and isometric exercise, suggesting that the augmented pressor responses to these stimuli are due to facilitated norepinephrine release (Figure 3). These effects of epinephrine could be blocked by propranolol. 140 On the other hand, epinephrine did not affect heart rate, possibly reflecting the contribution of parasympathetic withdrawal, as opposed to sympathetic activation, to the chronotropic responses to these particular activities. Our strategy has been to evaluate in humans the functional significance of the reported facilitated release of norepinephrine by epinephrine as it relates to neurogenic vasoconstriction. 33 In our first study, in normotensive men, we examined the effect of intraarterial epinephrine and isoproterenol on ipsilateral forearm vascular responses to vasoconstriction evoked by applying lower body negative pressure (LBNP), a reflex stimulus to norepinephrine release. Forearm blood flow was measured by strain gauge plethysmography, and blood pressure was recorded directly from the brachial artery. Forearm vasoconstrictor responses to LBNP at 40 mm Hg were compared with responses to intra-arterial infusion of the neurotransmitter norepinephrine before, during, and 30 minutes after infusion of epinephrine (50 ng/min) or isoproterenol (10 or 25 ng/min) into the brachial artery. We reasoned that if stimulation of prejunctional /3-adrenergic receptors by epinephrine or isoproterenol was functionally important, these agonists should potentiate the vasoconstrictor response, mediated via the neural release of norepinephrine, to LBNP but should not augment the forearm vascular response to norepinephrine infusion, which would be mediated via stimulation of postjunctional a-adrenergic receptors. The ratio of vasoconstrictor responses to LBNP/norepinephrine could therefore be used as an index of the neural release of the neurotransmitter norepinephrine. The forearm vasoconstrictor response to LBNP was preserved during intra-arterial infusions of epinephrine and isoproterenol, whereas the response to infused norepinephrine was significantly attenuated (Figure 5). Consequently, the ratio of vasoconstrictor responses to LBNP/norepinephrine

9 Floras Epinephrine and Hypertension EXPERIMENTAL ARM CONTROL ARM HYPMTttSn e HI UICTS I J 4 ) ' > LBNP Baseline FVR (R.U ) LBNP n NE Period 1 (Pre EPl) Period 2 (During EPD Period 3 (Posl EPl) Period 1 Period 2 Peflod * * ? FIGURE 5. Bar graphs show baseline forearm vascular resistance (FVR) measured in resistance units (R U.) and forearm vasoconstrictor responses to 40 mm Hg lower body negative pressure (LBNP) and to 18 ng/min intra-arterial norepinephrine (NE) before (control Period 1), during (Period 2), and 30 minutes after (Period 3) the intra-arterial infusion of 50 ng/min epinephrine (EPl) into the experimental arm (left panel). Responses to LBNP only are in the contralateral arm (right panel). (*p<0.05 compared with Period 1.) In the experimental arm, the ratio of repsonses to LBNP/NE was significantly increased in Periods 2 and 3, as compared with Period 1 (see Figure 8). Responses to LBNP in the contralateral arm did not differ during the three periods and indicate a reproducible effect of LBNP on the absence of intra-arterial EPl (from Reference 33). increased threefold during the infusion of epinephrine and more than 20-fold during isoproterenol33 (Figure 6). These observations provide indirect evidence that /S-adrenergic receptor agonists can facilitate the release of norepinephrine from the adrenergic terminal in humans. Two distinctive features of this study, relative to previous investigations in humans, deserve emphasis. These apply to the interpretation of vasoconstrictor responses during as well as 30 minutes after these infusions (see below). First, to localize their influence to peripheral neuroeffector sites and avoid potential effects on central and ganglionic sites or sensory afferents, we infused these catecholamines into the brachial artery at doses that restricted theneffects to the experimental arm. Measurement of central venous pressure, systemic blood pressure, heart rate, and forearm blood flow in the contralateral arm confirmed these agonists had no detectable systemic effects. Second, we measured forearm blood flow simultaneously in the experimental and the contralateral arm to confirm that LBNP caused a stable reproducible neurogenic vasoconstriction over the course of these experiments in the absence of local interventions (Figure 5). Differences between responses observed before, during, and after epinephrine or isoproterenol in the experimental arm could thus be ascribed to the local effects of these intra-arterial agonists rather than to systemic or "time" effects. In subsequent experiments, we set out first to determine if these local effects of epinephrine could be replicated by systemic infusion to achieve plasma Pit EPI During m : n PT. m Outtg EH n» rot En FIGURE 6. Bar graphs show ratio of vasoconstrictor responses to 40 mm Hg of lower body negative pressure (LBNP)/18 ng/min of intra-arterial norepinephrine (NE) before, during, and 30 minutes after the intra-arterial infusion of 50 ng/min epinephrine (EPl) into the experimental arm of normotensive (left panel) and borderline hypertensive (right panel) subjects. Ratio of forearm vasoconstrictor responses to LBNP/NE in the experimental arm was used as an index of the neural release of the neurotransmitter norepinephrine. This ratio increased during epinephrine infusion in both normotensive (+340%; p<0.01 compared with Period 1) and borderline hypertensive (+512%) subjects (NS). Thirty minutes after epinephrine infusion, this ratio was again increased in both normotensive (+353%; p<0.05 compared with Period 1) and borderline hypertensive (+262%; p<0.05 compared with Period 1) subjects compared with Period 1. These observations indicate facilitated release of norepinephrine in response to LBNP both during and after epinephrine. However, these effects of epinephrine are not exaggerated in borderline hypertension (from References 33, 145). FVR, forearm vascular resistance. levels achieved during physiological stress and second to determine whether heart rate responses to LBNP at -40 mm Hg are also augmented during epinephrine.143 Infusion of 1.5 ^ig/min epinephrine (about 20 ng/kg/min) led to forearm vasodilation and a subtle but significant increase in heart rate. Since the forearm vasoconstrictor response to intra-arterial norepinephrine is proportional to the initial forearm vascular resistance,144 we anticipated that by causing forearm vasodilation, epinephrine would also attenuate the vasoconstrictor response to LBNP unless countered by increased norepinephrine release during this stimulus. The vasoconstrictor response to LBNP was not attenuated but rather augmented by about 50% during the epinephrine infusion. The heart rate response to LBNP was not altered by epinephrine. We attributed this observation to the strong influence of vagal withdrawal on heart rate responses to this stimulus in healthy young men, but these contrasting effects of epinephrine on heart rate and vascular responses to LBNP might also be explained by differences in the ratio of responsiveness to epinephrine of facilitative ^-inhibitory/ajprejunctional adrenergic receptors in vascular as opposed to cardiac sympathetic nerves. Direct Effects of Epinephrine in Primary Hypertension There is no clear consensus on this point. Kjeldsen et al,121 who compared responses to epinephrine

10 10 Hypertension Vol 19, No 1 January 1992 u OK 12 l< 16 II TIME OF DAY t t >- - FIGURE 7. Line plot shows mean hourfy values of percent change in the intra-arterial mean arterial pressure as compared with baseline during and after 6-hour infusion of epinephrine (15 nglkglmin) or norepinephrine (30 ng/kg/min). Values corrected for percent change from baseline during and after 5% dextrose. *p<0.01; tp< Reproduced with permission, I47 infusion (from 10 to 40 ng/kg/min) in age-matched (40 years old) normotensive and hypertensive subjects, documented a fall in blood pressure, forearm vasodilation, and increased natriuresis and kaliuresis in the latter group; they therefore postulated a hyperresponsiveness to low-dose epinephrine infusion in mild essential hypertension. On the other hand, the effects, in our protocol, of intra-arterial epinephrine on neurogenic vasoconstriction and by inference neurotransmitter release appeared to be similar in age-matched normotensive and young borderline hypertensive subjects 145 (Figure 6). Sustained Aftereffects of Epinephrine Several groups have observed sustained increases, often lasting several hours, in blood pressure after epinephrine infusion into normotensive subjects *- 148 The dose and duration of epinephrine given in these studies ranged from 20 to 40 ng/kg/ min for minutes to 15 ng/kg/min sustained for 6 hours. 147 In the latter study, 24-hour intra-arterial ambulatory blood pressure was used to document, in normal subjects, the effect of 6-hour infusions of epinephrine, norepinephrine (30 ng/kg/ min), and dextrose administered in a random crossover design. During its infusion, epinephrine caused a 10-fold increase in its plasma concentration and lowered mean arterial pressure. Epinephrine concentrations returned to baseline levels less than 5 minutes after the infusion stopped. The subsequent 18-hour recording documented persistent elevation of mean arterial pressure by as much as 12% above those levels recorded on the day dextrose was infused (Figure 7). The pressor effects of prior epinephrine infusion were particularly evident during episodes of documented sympathetic activation. Between 40% and 70% of arterial epinephrine is removed in one circulation through the forearm (more than half via uptake ), and approximately 45% of arterial epinephrine is extracted in a single passage through the heart and kidney Therefore, it seems reasonable to assume that both cardiac and vascular mechanisms could contribute to any sustained pressor response subsequent to epinephrine. 147 Brown et al documented persistent tachycardia after a 2-hour infusion of epinephrine of 100 ng/kg/min was stopped, even though its plasma concentration returned to basal levels within minutes Sustained tachycardia was not observed in a small number of subjects pretreated with desmethylimipramine to reduce neuronal uptake of epinephrine 31 or when isoproterenol, which is not taken up by sympathetic nerves, was substituted for epinephrine. Brown et al proposed that the dissociation between the short plasma half-life of epinephrine and the slow decay of the tachycardia after its infusion was due to a cyclical process of reuptake and release of epinephrine from cardiac nerves. However, there are problems with this interpretation. Resting tachycardia is not seen after epinephrine infusions of 50 ng/kg/min or less Other potential causes of this sustained tachycardia, such as metabolic, 153 hemodynamic, reflex, or central effects of infused epinephrine were not excluded, and the potential effect of prior epinephrine infusion on postjunctional responses to the neurotransmitter norepinephrine was not examined in these studies. Earlier, we presented our experimental approach to the evaluation in humans of the functional significance of the reported facilitated release of norepinephrine by epinephrine. 33 In these experiments, we also explored the possibility that neuronally extracted epinephrine might have a sustained aftereffect on the forearm vasoconstrictor response to neurogenic stimulation. Our hypothesis was that if epinephrine were incorporated into the sympathetic nerve endings of the forearm during its arterial infusion and later released as a cotransmitter with norepinephrine, we might find a selective augmentation of reflex vasoconstriction after an epinephrine infusion. Further, we hypothesized that such augmentation would not be detected after an isoproterenol infusion. Thirty minutes after these infusions, resting forearm vascular resistance was at its initial, preinfusion level. From similar baselines, the forearm vasoconstrictor response to LBNP increased by about 65% 30 minutes after epinephrine, but this response was not affected by prior infusion of isoproterenol (Figure 5). The ratio of forearm vasoconstrictor responses to LBNP/norepinephrine was not altered by prior infusion of isoproterenol but was more than threefold higher 30 minutes after epinephrine 33 (Figure 6). These observations provide further support for the concept that epinephrine augments neurogenic vasoconstriction by a peripheral neuroeffector action. Since epinephrine has a half-life in plasma of only a

11 Floras Epinephrine and Hypertension 11 IX t 120 ; no in m I 70 j I " w a» so u 6 ft ISO so 10 GO SO JO 30 «SO M 70 «JO SO GO FIGURE 8. Line graphs show blood pressure (BP) (mm Hg) and pulse rate (bpm) responses to the intravenous infusion of epinephrine ( figlmin) in normotensive subjects (left panel) and in hypertensive subjects (right panel) before ( ) or after (o) ^-blockade with propranolol. Each circle and vertical line represents mean±sem. 'Statistically significant when compared with baseline values. Adapted with permission of the American Heart Association.' 00 few minutes, we attributed the augmented ratio of responses to LBNP/norepinephrine 30 minutes after epinephrine (but not isoproterenol) to facilitated release of norepinephrine by epinephrine that had been taken up into the adrenergic terminal, converted into a cotransmitter, and coreleased with norepinephrine during subsequent reflex stimulation. To test this hypothesis, eight normal men were studied on 2 separate days at least 1 week apart, 2.5 hours after taking at random either placebo or a dose of desipramine (125 mg p.o.), shown to block completely neuronal uptake of norepinephrine in the forearm On the placebo day, the forearm vasoconstrictor response to LBNP -40 mm Hg was significantly augmented by about 40% 30 minutes after an hour of systemic infusion of epinephrine at 1.5 ^tg/min (about 20 ng/kg/min) but tended to be less after epinephrine and desipramine. 148 Resting mean arterial pressure and the heart rate response to LBNP after epinephrine administration were also greater on the placebo day. 148 Thus, systemic as well as local infusion of epinephrine in a dose known to achieve plasma concentrations documented during physiological stresses 6 ' and to increase plasma norepinephrine concentrations by a /3-adrenergic mechanism can have sustained aftereffects on blood pressure and on specific chronotropic and vascular responses to reflex sympathetic activation. Fellows et aj 152 reported similar heart rate and forearm vascular responses to LBNP (-50 mmhg) applied 15 minutes after a 30-minute infusion of epinephrine (50 ng/kg/min). Since resting heart rates were not increased after epinephrine, doses administered in these two studies may not have been sufficient to maintain synaptic concentrations of epinephrine at levels necessary to facilitate norepinephrine release at rest; yet the doses caused sufficient loading of the adrenergic terminal to achieve such concentrations locally during the reflex stimulus to norepinephrine release. Sustained Aftereffects of Epinephrine in Primary Hypertension Whether the aftereffects of epinephrine infusion differ in normal and hypertensive subjects has not been firmly established. Although Nezu et al 100 documented similar increases in systolic and diastolic blood pressure in normotensive and hypertensive subjects up to 90 minutes after an epinephrine infusion (20 ng/kg/min) (Figure 8), persistent elevation of blood pressure and heart rate 20 minutes after epinephrine (39 ng/kg/min) was seen in another study 146 only in a hypertensive group even though arterial epinephrine concentrations had returned to basal levels by this time. Epinephrine clearance was identical in the two groups of subjects. 146 Because elevated plasma epinephrine concentrations and characteristics consistent with increased /3-adrenergjc receptor responsiveness are often seen in borderline hypertension, we hypothesized that the local direct and sustained aftereffects of epinephrine on neurogenic vasoconstriction would be exaggerated in such subjects. However, when tested, this was not the case: the effect of prior intra-arterial epinephrine infusion on the ratio of forearm vasoconstrictor responses to the LBNP/nor-

12 12 Hypertension Vol 19, No 1 January 1992 epinephrine was similar in normotensive and borderline hypertensive subjects (Figure 6). We considered several possible explanations for this observation. Responses to epinephrine might only be enhanced in a subgroup of borderline hypertensive patients with increased arterial epinephrine concentrations and/or "hyperadrenergic" characteristics. The latter were not prominent in our subjects. 145 Responses to epinephrine might have been similar because of substantial overlap in prejunctional -adrenergic receptor density, affinity for agonist, or /3-adrenergic-mediated adenylate cyclase activity in our two groups or because of the functional uncoupling of the lymphocyte ft-adrenergic receptor from the adenylate cyclase complex reported to occur in young subjects with borderline hypertension Such functional uncoupling may have minimized any quantitative differences in /3-adrenergic receptor number or affinity for agonist in our young normotensive and hypertensive subjects. 129 Indeed, a specific comparison of the effects of infused isoproterenol on plasma norepinephrine concentrations did not detect any difference in this response between normotensive and borderline hypertensive male subjects. 135 Last, exaggerated responses in patients with borderline hypertension may reflect increased central neural (presynaptic) 34 or peripheral ganglionic rather than peripheral neuroeffector (prejunctional) sensitivity to epinephrine. Evidence in support of this latter possibility was provided by the comparison, in young normotensive and borderline hypertensive men, of the effects of prior administration of epinephrine (1.5 /xg/min for 30 minutes) on muscle sympathetic nerve responses to LBNP at -10 mm Hg and to a cold pressor test. 107 Interventions were applied before and 30 minutes after epinephrine. The immediate direct effects of epinephrine on systolic and pulse pressure, heart rate, and postganglionic muscle sympathetic nerve activity were similar in the two groups, but the aftereffects of the infusion were only seen in the borderline hypertensive subjects: pressor and chronotropic responses to the cold pressor test were augmented, and the sympathoneural responses to both interventions more than doubled (Figure 9). Potential Contribution of Epinephrine to the Genesis of Hypertension: Summary, Implications, and Future Directions Several lines of evidence now suggest a psychophysiological link between stress, adrenomedullary activation, and the genesis of hypertension in some humans genetically predisposed to the development of high blood pressure. These include results of experiments in a number of animal preparations that support the facilitative and cotransmitter components of the epinephrine hypothesis, are consistent with the concept of a period of critical sensitivity to epinephrine in a genetic model of hypertension (the SHR), or demonstrate that exogenous epinephrine can induce sustained hypertension. s I COLD PRESSOR TEST i p < att i at via ao 0 LOWER BODY NEGATIVE PRESSURE MO WHO) FIGURE 9. Bar graph shows effects of prior infusion of epinephrine (EPI) (1.5 ug/min) on increases in muscle sympathetic nerve activity (MSNA), expressed as units per minute, in seven borderline hypertensive subjects. In contrast to normotensive controls, sympathetic outflow was augmented (p<0.01) after systemic EPI. Adapted from Reference 107. Evidence in support of the hypothesis that epinephrine participates in the initiation of primary hypertension by amplifying peripheral neurotransmission may be summarized as: 1) Resting and stimulated arterial and venous plasma epinephrine concentrations are elevated in many young subjects with borderline or mild essential hypertension. 7 ' ' ' 157 Mental stress is one such provocative stimulus. Increased plasma norepinephrine concentrations and regional norepinephrine spillover have also been documented in many young borderline or mildly hypertensive patients. 98 " 101 Because the results of some investigations differ from these general conclusions, ' 158 it should be emphasized that this hypothesis is not contingent on documentation of increased plasma epinephrine concentrations alone. The magnitude of neuronal epinephrine uptake as a reflection of the frequency and intensity of adrenomedullary epinephrine release may be more important than basal epinephrine concentrations. 2) Increases in endogenous epinephrine cause sustained increases in plasma norepinephrine and blood pressure in normotensive and hypertensive subjects. These effects can be abolished by propranolol ) Endogenous epinephrine is taken up into human peripheral sympathetic nerves. 70 ' Increased forearm uptake of epinephrine in hypertension may be associated with greater forearm norepinephrine release ) Exogenous epinephrine, infused at a rate of 30 ng/kg/min or less, has functionally significant immediate (direct) effects and subsequent (indirect) aftereffects on norepinephrine concentrations ' 140 that are similar in magnitude to its effects on norepinephrine release as documented in experimental preparations. 42 Direct effects have been observed at rest and during provocative maneuvers that increase efferent sympathetic traffic such as isometric exercise or the cold pressor test. These effects are also blocked by propranolol. 140 Increases in norepinephrine concentrations persist after plasma epinephrine concentrations return to baseline levels These effects are not i n

13 Floras Epinephrine and Hypertension 13 seen at higher doses of epinephrine unless a 2 - receptors are blocked. 66 5) Exogenous epinephrine has functionally significant immediate (direct) effects and subsequent (indirect) aftereffects on neurogenic circulatory regulation that are similar in magnitude to its effect on neurotransmitter release as documented in experimental preparations. 42 Direct effects include greater blood pressure responses to handgrip and the cold pressor test 140 and a relative augmentation of neurogenic forearm vasoconstriction Local infusion of epinephrine has sustained aftereffects on neurogenic vasoconstriction in the forearm Sustained aftereffects of its systemic infusion appear to be dosedependent: at low doses (15-40 ng/kg/min) the aftereffects include a persistent increase in blood pressure, augmentation of neurogenic vasoconstriction, and neurogenic tachycardia ; at higher doses (100 ng/kg/min), prior epinephrine infusion causes persistent tachycardia ) Aftereffects of epinephrine on sympathetic outflow are enhanced in borderline hypertensive subjects. 107 These studies have defined two temporally and functionally distinct roles for epinephrine in the regulation of vascular resistance. As a circulating hormone, it causes an immediate and transient vasodilation via its direct action on postjunctional vascular /^-receptors. Its humoral actions are brief, include additional circulatory response analogous to the defense reaction, and dissipate rapidly. As an autocrine agent, it causes indirectly a delayed and sustained vasoconstriction. These aftereffects have been interpreted as caused by the peripheral conversion of epinephrine from a circulating hormone to a cotransmitter, which when released neuronally, acts on prejunctional neuronal /^-receptors. By augmenting the simultaneous discharge of norepinephrine neurotransmitter release, epinephrine could raise blood pressure transiently through both cardiac and vascular mechanisms during periods of increased sympathetic outflow or at rest after its plasma concentrations subside to basal levels. Through this mechanism, repetitive surges in endogenous epinephrine could initiate, promote, or permit the development of primary hypertension. In addition, the trophic effects of intermittently increased neurotransmitter release, 159 when continued over months or years, could initiate early structural changes in resistance and capacitance vessels that would intensify responses to constrictor signals 4160 and allow for sustained elevations in blood pressure independent of this neurogenic trigger to hypertension. This neurogenic mechanism is attractive but as yet unproven. Evidence challenging the hypothesis that epinephrine participates in the initiation of primary hypertension by amplifying peripheral neurotransmission may be summarized as: 1) The conclusion that augmented neurogenic vasoconstriction during and after local epinephrine infusion is due to increased norepinephrine release is based on indirect evidence. There have been no published reports of forearm catecholamine uptake and release in this model, the effect of local /3-adrenergic receptor blockade on these responses has not been assessed, and it is not known whether their kinetics are altered in young hypertensive subjects. Furthermore, indirect actions of epinephrine that could augment neurogenic vasoconstriction, such as local induction of endothelin, 161 or that could facilitate norepinephrine release, such as generation of angiotensin II 162 " 165 or alteration of ionic fluxes across the adrenergic terminal, have not been specifically excluded in this preparation. 2) Pressor and noradrenergic response to mental stress, tilt, and the cold pressor test are not blunted in adrenalectomized women when compared with responses in women with intact adrenal function. 141 Given the unique opportunity presented by such patients, it would be of interest to compare the direct effects and aftereffects of both brief and sustained epinephrine infusion on responses to these interventions in these two groups. 3) Selective ft-adrenergic receptor blockade has not proved to be particularly effective in the management of hypertension Majewski and Murphy 44 have proposed that this is because the potential antihypertensive effect of such drugs on neurotransmitter release is offset by concurrent abolition of ft-adrenergic receptor-mediated peripheral vasodilation. 4) Immediate or delayed noradrenergic, pressor, and neurogenic chronotropic or vascular responses to epinephrine (and other ^-agonists) are not always augmented in subjects with borderline or mild essential hypertension ' 145 5) There are no prospective evaluations of the predictive value of increased epinephrine concentrations in childhood or adolescence as a marker for later hypertension development. These points highlight a number of present difficulties with the epinephrine hypothesis and suggest a number of future experiments. Additional questions would have to be addressed in humans before epinephrine could be considered a causative mechanism of primary hypertension. A general series of questions pertains to those phenomena attributed to the facilitative and cotransmitter effects of exogenous epinephrine themselves. Can they be replicated by increases in endogenous epinephrine? How long after surges in epinephrine does this exaggerated response to neural stimulation persist? How quickly does it decay? A second series of questions pertains to other potential functional consequences of facilitated neurotransmitter release by epinephrine (e.g., effects on venous tone, renal hemodynamics, and renal tubular sodium absorption and renin release). Given the primacy of the kidney in long-term blood pressure regulation, 171 future assessment of the potential contribution of epinephrine to the genesis of primary hypertension should include a characterization of any

14 14 Hypertension Vol 19, No 1 January 1992 effects and in particular any sustained aftereffects it may have on renal hemodynamics and function. Finally, as suggested by Folkow, 2 expression of hypertension by this mechanism may be restricted to a specific subset of individuals with a genetic predisposition to high blood pressure, who have more frequent or exaggerated surges of epinephrine in response to psychological or physiological stimuli or greater central or peripheral sensitivity to this catecholamine. Such individuals may manifest sustained increases in heart rate and cardiac output after episodic increases in epinephrine and greater neurogenic vasoconstrictor responses to reflex stimuli than those without this predisposition. A third or more of primary hypertensive individuals may fall into this category. However, if a period of "critical prejunctional /^-adrenergic receptor sensitivity" to the facilitative effects of epinephrine on norepinephrine release, similar to that described in SHR, 52 precedes the clinical detection of borderline or mild essential hypertension or if the prejunctional ft-receptor is functionally uncoupled from its adenylate cyclase complex once structural vascular changes have developed, 128 the most compelling tests of the epinephrine hypothesis, whether longitudinal or interventional (e.g., inhibition of adrenal epinephrine syntheses or peripheral ft-antagonism), will require identification of such individuals at the earliest and mildest stage of their disease, well before the development of clinically established hypertension. Acknowledgments I am grateful to Beverley Morris-Senn for her technical support and to Mary Anne Beddows for her perseverance through many drafts and revisions of this review. References 1. Brown MJ, Macquin I: Is adrenaline the cause of essential hypertension? Lancet 1981;2: Folkow B: Physiological aspects of primary hypertension. Physiol Rev 1982;62: Clutter WE, Bier DM, Shah SD, Cryer PE: Epinephrine plasma metabolic clearance rates and physiologic thresholds for metabolic and hemodynamic action in man. / Clin Invest 198O Staten MA, Mathews DE, Cryer PE, Bier DM: Physiological increments in epinephrine stimulate metabolic rate in humans. Am J Physiol 1987;253:E322-E Julius S, Johnson EH: Stress, autonomic hyperactivity and essential hypertension: An enigma. J Hypertens 1985;3(suppl 4):S11-S17 6. Freyschuss U, Hjemdahl P, Juhlin-Dannfelt A, Linde B: Cardiovascular and metabolic responses to low dose adrenaline infusion: An invasive study in humans. Can Sci 1986; 70: Eliasson K, Hjemdahl P, Kahan T: Circulatory and sympatho-adrenal responses to stress in borderline and established hypertension. / Hypertens 1983;1: Dimsdale JE, Moss J: Plasma catecholamines in stress and exercise. JAMA 198Ct243: Brod J, Fend V, Heji Z, Jirka J: Circulatory changes underlying blood pressure elevation during acute emotional stress (mental arithmetic) in nonnotensive and hypertensive subjects. Clin Sd 1959;18: Graham JDP: High blood pressure after a battle. Lancet 1945;l: Jonsson A, Hansson L: Prolonged exposure to a stressful stimulus (noise) as a cause of raised blood pressure in man. Lancet 1977;l: Andren L, Hansson L, Bjorkman M, Jonsson A: Noise as a contributing factor in the development of elevated blood pressure. Acta Med Scand 1980;2O7: Eliot RS, Buell JC: The heart and emotional stress, in Hurst JW (ed): 77K; Heart. New York, McGraw-Hill Book Co., 1982, pp Glass DC: Physiological and psychological responses of individuals displaying type A behaviour. Acta Med Scand 1982; 660(suppl): Rand MJ, Majewski H: Adrenaline mediates a positive feedback loop in noradrenergic transmission: Its possible role in development of hypertension. Clin Exp Hypertens [A] 1984;6: Poulter N, Khaw KT, Hopwood BEC, Mugambi M, Peart WS, Sever PS: Determinants of blood pressure changes due to urbanization a longitudinal study. J Hypertens 1985; 3(suppl 3):S375-S Timk) M, Verdecchia P, Venanzi S, Gentili S, Ronconi M, Francucci B, Montanari M, Bichtsao E: Age and blood pressure changes: A 20-year follow-up study in nuns of a secluded order. Hypertension 1988;12: Quinn P, Borkowski KR, Collis MG: Epinephrine enhances neurogenic vasoconstriction in the rat perfused kidney. Hypertension 1984;7: Rosell SJ, Axelrod J, Kopin LJ: Release of tritiated epinephrine following sympathetic nerve stimulation. Nature 1964; 201: Majewski H, Rand MJ, Tung L-H: Activation of prejunctional /3-adrenoceptors in rat atria by adrenaline applied exogenously or released as a co-transmitter. Br J Pharmacol 1981;73: Berecek KH, Brody MJ: Evidence for a neurotransmitter role for epinephrine derived from the adrenal medulla. Am J Physiol 1982;242(//<wrt Ore Physiol ll):h593-h Peronnet F, Nadeau R, Boudreau G, Cardinal R, LaMontagne D, Yamaguchi N, de Champlain J: Epinephrine release from the heart during left stellate ganglion stimulation in dogs. Am J Physiol 1988$54(RegulaU)ry Integrative Comp Physiol 23):R659-R Iversen LL: Catecholamine uptake processes. Br Med Bull 1973^9: Adler-Graschinsky E, Langer SZ: Possible role of a /J-adrenoceptor in the regulation of noradrenaline release by nerve stimulation through a positive feedback mechanism. Br J Pharmacol 1975^3: Bolli P, Erne P, Ji BH, Block LH, Kiowski W, Buhler FR: Adrenaline induces vasoconstriction through post-junctional alpha-2 adrenoceptors and this response is enhanced in patients with essential hypertension. J Hypertens 1984;2(suppl 3):S115-S Stjame L, Brundin J: /3-2-Adrenoceptors facilitating noradrenaline secretion from human vasoconstrictor nerves. Acta Physiol Scand 1976^7: Majewski H, McCullough MW, Rand MJ, Story DF: Adrenaline activation of prejunctional beta-adrenoceptors in guinea-pig atria. BrJ Pharmacol 1980;71: Majewski H: Modulation of norepinephrine release through activation of presynaptic /3-adrenoceptors. JAuton Pharmacol 1983^: Majewski H, Tung LH, Rand MJ: Adrenaline activation of prejuuctional 0-adrenoceptors and hypertension. / Cardiovasc Pharmacol 1982;4: Verbeuren TJ, Lorenz RR, Aarhus LL, Shepherd JT, Vanhoutte PM: Prejunctional beta-adrenoceptors in human and canine saphenous veins. JAuton Nav Syst 1983;8: Brown MJ, Dollery CT: Adrenaline and hypertension. Clin Exp Hypertens [A] 1984;6: Schmitt HHWW, Schurr C, Hedler L, Majewski H: Local modulation of norepinephrine release in vivo: Presynaptic

15 Floras Epinephrine and Hypertension IS 02-adrenoceptors and endogenous epinephrine. / Cardiovasc Pharmacol 1984;6: Floras JS, Ayfward PE, Victor RG, Mark AL, Abboud FM: Epinephrine facilitates neurogenic vasoconstriction in humans. / Clin Invest 1988;81: Ueda H, Goshima Y, Kubo T, Misu Y: Involvement of epinephrine in the presynaptic beta-adrenoceptor mechanism of norepinephrine release from rat hypothalamic slices. J Pharmacol Exp Ther 1985^232^ Foucart S, Nadeau R, de Champlain J: Local modulation of adrenal catecholamines release by beta-2 adrenoceptors in the anaesthetized dog. Naunyn Schmiedebergs Arch Pharmacol 1988^37: Ekas RD, Steenberg ML, Wood MS, Lokhandwala MF: Presynaptic a- and /3-adrenoceptor stimulation and norepinephrine release in the spontaneously hypertensive rat. Hypertension 1983^: Weicker H, Feraudi M, Hagele H, Pluto R: Electrochemical detection of catecholamines in urine and plasma after separation with HPLC. Clin Chim Acta 1984;4: Feldman RD, Limbird LE, Nadeau J, Robertson D, Wood AJJ: Leukocyte ^-receptor alterations in hypertensive subjects. / Clin Invest 1984;73: Langer SZ: Presynaptic regulation of the release of catecholamines. Pharmacol Rev 1981^32: Keeton TK; Pindolol increases plasma norepinephrine concentration by stimulation of /3 2-adrenoceptors in conscious spontaneously hypertensive rats. / Cardiovasc Pharmacol 199O;16: Molderings G, likungu J, Zerkowski HR, Gothert M: Presynaptic /32-adrenoceptors on the sympathetic nerve fibres of the human saphenous vein: No evidence for involvement in adrenaline-mediated positive feedback loop regulating noradrenergic transmission. Naunyn Schmiedebergs Arch Pharmacol 1988;337: Nedergaard OA, Abrahamsen J: Modulation of noradrenaline release by activation of presynaptic /3-adrenoceptors in the cardiovascular system. Ann N Y Acad Sci 1990;604: Yamaguchi N, de Champlain J, Nadeau RA: Regulation of norepinephrine release from cardiac sympathetic fibers in the dog by presynaptic a- and /3-adrenoceptors. Ore Res 1977;41: Majewski H, Murphy TV: Beta-adrenoceptor blockade and sympathetic neurotransmission in the pithed rat. J Hypertens 1989;7: Osswald W, Guimaraes S: Adrenergjc mechanisms in blood vessels: Morphological and pharmacological aspects. Rev PhysM Biochem Pharmacol 1983;96: Majewski H, Tung L-H, Rand MI: Adrenaline activation of presynaptic 0-adrenoceptors. / Auton Pharmacol 1983;3: Dahlof C: Studies on beta-adrenoceptor mediated facilitation of sympathetic neurotransmission. Acta PhysM Scand 1981;500<suppl):l Majewski H, Tung LH, Rand MJ: Adrenaline induced hypertension in rats. / Cardiovasc Pharmacol 1981^3: Heimburger M, Montero MJ, Fougeres V, Beslot F, Davy M, Midol-Monnet M, Cohen Y: Presynaptic beta-adrenoceptors in rat atria: Evidence for the presence of stereoselective beta,-adrenoceptors. BrJ Pharmacol 1989;98: Kalsner S: Heteroreceptors, autoreceptors, and other terminal sites. Ann N YAcad Sci 1990;604:l Lands AM, Arnold A, McAuliff JP, Luduena FP, Brown TG: Differentiation of receptor systems activated by sympathomimetic amines. Nature 1967^14: Borkowski KR: The effect of adrenal demudullation and adrenaline on hypertension development and vascular reactivity in young spontaneously hypertensive rats. J Auton Pharmacol 1991;11: Stjarse L, Brundin J: Dual adrenoceptor mediated control of noradrenaline secretion from human vasoconstrictor nerves: Facilitation by /3-receptors and inhibition by cr-receptors. Acta PhysM Scand 1975^4: Ariens EJ, Simonis AM: Physiological and pharmacological aspects of adrenergic receptor classification (commentary). Biochem Pharmacol 1983^ Cryer PE: Physiology and pathophysiology of the human sympathoadrenal neuroendocrine system. NEnglJMed 1980; 303: Dies P, Weber H-D, Neuburger J, Bucher B, Regenold JT, Norenberg W: Receptor interactions at noradrenergic neurones. Ann N YAcad Sci 1990,604: Majewski H, Costa M, Foucart S, Murphy TV, Musgrave IF: Second messengers are involved in facilitatory but not inhibitory receptor actions at sympathetic nerve endings. Ann N Y Acad Sci 1990;604: Alberts P, Ogren VR, Sellstrom AI: Role of adenosine 3',5'-cyclic monophosphate in adrenoceptor-mediated control of 3 H-noradrenaline secretion in guinea-pig ileum myenteric nerve terminals. Naunyn Schmiedebergs Arch Pharmacol 1985^330: Stjarne L, Msghina M, Stjarne E: "Upstream" regulation of the release probability in sympathetic nerve varicosities. Neuroscience 1990;36: Borkowski KR: Pre- and postjunctional beta-adrenoreceptors and hypertension. J Auton Pharmacol 1988;8: Kawasaki H, Cline WH, Su C: Enhanced presynaptic beta adrenoceptor-mediated modulation of vascular adrenergic neurotransmission in spontaneously hypertensive rats. / Pharmacol Exp Ther 1982;223: Moulds RFW, Stevens MJ: Facilitatory prejunctional /3-adrenoceptors in human arteries and veins. Gen Pharmacol 1983;14: Gothert M, Hentrich F: Identification of presynaptic ft-adrenoceptors on the sympathetic nerve fibres of the human pulmonary artery. Br J Pharmacol 1985;85: Stevens MJ, Rittinghausen RE, Medcalf RL, Moulds RFW: Prejunctional /3-adrenoceptors in human digital arteries. Eur J Pharmacol 1982;83: Hall JA, Petch MC, Brown MJ: Intracoronary injections of salbutamol demonstrate the presence of functional ft-adrenoceptors in the human heart. Ore Res 1989;65: Brown MJ, Struthers AD, Burrin JM, Di Silvio L, Brown DC: The physiological and pharmacological role of presynaptic a- and ^-adrenoceptors in man. Br J Clin Pharmacol 1985;2O: Bevan JA, Bevan RD, Duckies SP: Adrenergic regulation of vascular smooth muscle, in Shepherd JT, Abboud FM (eds): Handbook of Physiology, Section 2: The Cardiovascular System, Volume III. Bethesda, Md, American Physiological Society, 1983, pp Burn JH: On vasodilatorfibresin the sympathetic, and on the effect of circulating adrenaline in augmenting the vascular response to sympathetic stimulation. / PhysM (Lond) 1932; 75: Iversen LL: Role of transmitter uptake mechanisms in synaptic neurotransmissions. Br J Pharmacol 1971;41: Esler M, Eisenhofer G, Chin J, Jennings G, Meredith I, Cox H, Lambert G, Thompson J, Dart A: Release of adrenaline from the human sympathetic nervous system. Clin Auton Res 1991;l: Ferreira SH, Vane JR: Half-lives of peptides and amines in the circulation. Nature 1967^15: Majewski H, Hedler L, Starke K: The noradrenaline release rate in the anaesthetized rabbit: Facilitation by adrenaline. Naunyn Schmiedebergs Arch Pharmacol 1982^21: Johnson MD, Grignolo A, Kuhn CM, Schandberg SM: Hypertension and cardiovascular hypertrophy during chronic catecholamine infusion in rats. Life Sci 1983^33: Tung L-H, Rand MJ, Majewski H: Adrenaline-induced hypertension in rats. Clin Sci 1981;61:191s-193s 75. Zabludowski J, Clark S, Ball SG, Brown AJ, Inglis GC, Lever AF, Murray G: Pressor effects of brief and prolonged infusions of epinephrine in the conscious rat. Am J PhysM 1984;246{/f(Wrt Ore PhysM 15):H683-H689

16 16 Hypertension Vol 19, No 1 January Majewski H, Alade PI, Rand MJ: Adrenaline and stressinduced increases in blood pressure in rats. Clin Exp Pharmacol PhysUA 1986;13: Vlachakis ND, Alexander N, Maronde RF: Increased plasma normetanephrine in spontaneously hypertensive rats. Clin Exp Hypenens [A] 1980;2: Saavedra JM, Pinto JEB, Flugge G, Torda T: Increased ft-adrenoceptor number in peripheral sympathetic ganglia of spontaneously hypertensive rats. Am J Hypertens 1990;3: Morisawa T: On the noradrenaline reaction of the adrenal medulla in experimental hypertensive rats, especially in spontaneously hypertensive rats: II. Effects of various experimental conditions on the noradrenaline storing islets of the adrenal medulla in spontaneously hypertensive rats. Jpn Ore /1968;32: McCarty R, Kopin IJ: Alterations in plasma catecholamines and behaviour during acute stress in spontaneously hypertensive and Wistar-Kyoto normotensive rats. Life Sci 1960,22: Versteeg DHG, Petty MA, Bohus B, dejong W: The central nervous system and hypertension: The role of catecholamines and neuropeptides, in dejong W (ed): Handbook of Hypertension Volume 4: Experimental and Genetic Models of Hypertension. New York/Amsterdam, Elsevier Science Publishing Co, Inc, 1984, pp Adams MA, Bobik A, Komer PI: Differential development of vascular and cardiac hypertrophy in genetic hypertension. Hypertension 1989;14: Borkowski KR, Quinn P: Adrenaline and the development of spontaneous hypertension in rats. JAuton Pharmacol 1985;5: Borkowski KR, Quinn P: Tne effect of bilateral adrenal demedullation on vascular reactivity and blood pressure in spontaneously hypertensive rats. Br J Pharmacol 1983;80: Finch L, Leach GDH: Does the adrenal medulla contribute to the maintenance of experimental hypertension? Eur J Pharmacol 1970;l 1: Feldman RD: /3-Adrenergic receptor alterations in hypertension Physiological and molecular correlates. Can J PhysM Pharmacol 1987;65: Borkowski KR, Quinn P: ft-adrenoceptors mediate adrenaline's facilitation of neurogenic vasoconstriction. EurJ Pharmacol 1984;103: dine WH: Enhanced epinephrine-induced facilitation in vivo of mesenteric vascular adrenergjc neurotransmission in spontaneously hypertensive rats (abstract). Fed Proc 1985;44: Westfall TC, Meldrum MJ: Alterations in the release of norcpinephrine at the vascular neuroeffector junction in hypertension. Annu Rev Pharmacol Tcadcol 1985;25: Tsuda K, Kuchii M, Nishio I, Masuyama Y: Effects of epinephrine and dopamine on norepinephrine release from the sympathetic nerve endings in hypertension. J Hypertens 1986;4(suppl 5):S45-S Borkowski KR, Quinn P: Facilitation by beta-adrenoreceptors of stimulation-induced vasoconstriction in pithed spontaneously hypertensive rats. / Auton Pharmacol 1984;4: Kubo T, Kuwahara M, Misu Y: Effect of isoproterenol on vascular adrenergic neurotransmission in prehypertensive and hypertensive spontaneously hypertensive rats. Jpn J Pharmacol 1984^6: Toal CB, Leenen FHH: Blood pressure responsiveness to isoproterenol during the development of hypertension in conscious spontaneously hypertensive rats. Blood Vessels 1984;21: Folkow B, DiBona G, Hjemdahl P, Thoren PN, Wallin BG: Measurements of plasma norepinephrine concentrations in human primary hypertension: A word of caution on their applicability for assessing neurogenic contributions. Hypertension 1983^: Floras J, Jones JV, Hassan MO, Osikowska BA, Sever PS, Sleight P: Failure of plasma norepinephrine to consistently reflect sympathetic activity in humans. Hypertension 1986;8: Kjeldsen SE, Westheim A, Aakesson I, Eide I, Leren P: Plasma adrenaline and noradrenaline during orthostasis in man: The importance of arterial sampling. Scand J Clin Lab Invest 1986;46: Goldstein DS: Plasma catecholamines and essential hypertension. An analytical review. Hypertension 1983^: Goldstein DS, Lake CR: Plasma norepinephrine and epinephrine levels in essential hypertension. Fed Proc 1984;43: Masuo K, Ogihara T, Kumahara Y, Yamatodani A, Wada H: Increased plasma norepinephrine in young patients with essential hypertension under three sodium intakes. Hypertension 1984;6: Nezu M, Miura Y, Adachi M, Adachi M, Kimura S, Toriyabe S, Ishizuka Y, Ohashi H, Sugawara T, Takahashi M, Noshiro T, Yoshinaga K: The effects of epinephrine on norepinephrine release in essential hypertension. Hypertension 1985;7: Esler M, Jennings G, Korner P, Willett I, Dudley F, Hasking G, Anderson W, Lambert G: Assessment of human sympathetic nervous system activity from measurements of norepinephrine turnover. Hypertension 1988;ll: Anderson EA, Sinkey CA, Lawton WJ, Mark AL: Elevated sympathetic nerve activity in borderline hypertensive humans: Evidence from direct intraneural recordings. Hypertension 1989;14: Yamada Y, Miyajima E, Tochikubo O, Matsukawa T, Shionoiri H, Ishii M, Kaneko Y: Impaired baroreflex changes in muscle sympathetic nerve activity in adolescents who have a family history of essential hypertension. J Hypertens 1988; 6(suppl 4):S525-S Floras JS, Senn BL: Absence of postexercise hypotension and sympathoinhibition in normal subjects: Additional evidence for increased sympathetic outflow in borderline hypertension. Can J Cardiol 1991;7: Kjeldsen SE, Eide I, Christensen C, Westheim A, Muller O: Renal contribution to plasma catecholamines Effect of age. Scand J Clin Lab Invest 1982;42: Lake CR, Ziegler MG, Coleman MD, Kopin IJ: Age adjusted plasma norepinephrine levels are similar in normotensive and hypertensive subjects. N Engl J Med 1977;296: Grassi GM, Rea RF, Somers VK, Abboud FM, Mark AL: Epinephrine acts centrally to facilitate reflex sympathetic nerve responses in borderline hypertension (abstract). Hypertension 1988;12: de Champlain J: The contribution of the sympathetic nervous system to arterial hypertension. Can J Physiol Pharmacol 1978^6: BQhler FR, Amann FW, Bolli P, Hulthen L, Kiowski W, Landmann R, Burgisser E: Elevated adrenaline and increased a-adrenoceptor-mediated vasoconstriction in essential hypertension. / Cardiovasc Pharmacol 1982;4: S134-S Franco-MorseUi R, Elghozi JL, Joly E, di Giuilli S, Meyer P: Increased plasma adrenaline concentrations in benign essential hypertension. Br Med J 1977;2: Bertel O, Buhler FR, Kiowski W, Lutold BE: Decreased beta-adrenoreceptor responsiveness as related to age, blood pressure, and plasma catecholamines with essential hypertension. Hypertension 1980;2: Kjeldsen SE, Flatten B, Eide I, Helgeland A, Leren P: Increased peripheral catecholamine release in patients with long-standing, untreated essential hypertension. Ada Med Scand 1983;677(suppl): Brown MJ, Jenner DA, Allison DJ, Dollery CT: Variations in individual organ release of noradrenaline measured by an improved radioenzymatic technique: Limitations of peripheral venous measurements in the assessment of sympathetic nervous activity. Clin Sci 1981;61:

17 Floras Epinephrine and Hypertension Bohm RO, van Baak MA, van Hooff ME, Mooy J, Rahn KH: A long-term study of plasma catecholamine levels and plasma renin activity in borderline hypertension. / Hypertens 1987;5: Jorgensen LS, Bonlokke L, Christensen NJ: Plasma adrenaline and noradrenaline during mental stress and isometric exercise in man: The role of arterial sampling. Scand J Clin Lab Invest 1985;45: Brown MJ, Causon RC, Barnes VF, Brennan P, Bames G, Greenberg G: Urinary catecholamines in essential hypertension: Results of 24-hour urine catecholamine analyses from patients in the Medical Research Council trial for mild hypertension and from matched controls. Q J Med 1985;57: Amery A, Bossaert H, Verstraete M: Muscle blood flow in normal and hypertensive subjects: Influence of age, exercise and body position. Am Heart J 1969;78: Frohlich EH, Tarazi RC, Dustan HP: Hyperdynamic B-adrenergic circulatory state: Increased beta-receptor responsiveness. Arch Intern Med 1969;123:l Julius S, Esler M: Autonomic nervous system in borderline hypertension. Am J Cardiot 1975^35: Levenson J, Simon AC, Safar ME, Bouthier JD, London GM: Elevation of brachial arterial blood velocity and volumic flow mediated by peripheral beta-adrenoceptors in patients with borderline hypertension. Circulation 1985;71: Kjeldsen SE, Os I, Westheim A, Lande K, Gjesdal K, Hjermann I, Eide I: Hyperresponsiveness to low-dose epinephrine infusion in mild essential hypertension. / Hypertens 1988;6{suppl 4):S581-S Leenen FHH, Boer P, Dorhout Mees EJ: Peripheral betaadrenoceptor responsiveness in young normotensive and hypertensive subjects. Clin Exp Hypertens [A] 1981^^ McAllister RG, Love DW, Guthrie GP, Dominic JA, Kotchen TA: Peripheral ^-receptor responsiveness in patients with essential hypertension. Arch Intern Med 1979; 139: Middeke M, Remien J, Block LH, Kirzinger S, Landrock A, Holzgreve H: Beta-2-adrenoceptor density on membranes and on intact mononuclear cells in essential hypertension. Res Exp Med (Berl) 1983;183: Brodde O-E, Daul A, O'Hara N, Bock KD: Increased density and responsiveness of a-2 and 8-2 adrenoceptors in circulating blood cells of essential hypertensive patients. / Hypertens 1984;2(suppl 3): Brodde O-E, Prywarra A, Daul A, Anlauf M, Bock KD: Correlation between lymphocyte ft-adrenoceptor density and mean arterial blood pressure: Elevated /3-adrenoceptors in essential hypertension. / Cardiovasc Pharmacol 1984;6: Michel MC, Brodde O-E, Insel PA: Peripheral adrenergic receptors in hypertension. Hypertension 1990;16:2: Feldman RD, Lawton WJ, McArdle WL: A low sodium diet corrects the defect in lymphocyte beta-adrenergic receptors in hypertensive subjects. / Clin Invest 1987;79: Feldman RD: Defective venous /3-adrenergic response in borderline hypertensive subjects is corrected by a low sodium diet. / Clin Invest 1990;85: Tohmeh JF, Cryer PE: Biphask adrenergic modulation of ^-adrenergic receptors in man: Agonist-induced early increment and late decrement in ^-adrenergic receptor number. / Clin Invest 1980,65: Maisel AS, Harris T, Rearden CA, Michel MC: 0-Adrenergjc receptors in lymphocyte subsets after exercise: Alterations in normal individuals and patients with congestive heart failure. Circulation 1990;82: Burman KD, Ferguson EW, Djuh Y-Y, Wartofsky L, Latham K: Beta receptors in peripheral mononuclear cells increase acutely during exercise. Ada Endocrinol (Copenh) 1985;109: Musgrave IV, Bachmann AW, Gordon RD: Increased plasma noradrenaline during adrenaline infusion in man. J Hypertens 1984;2(suppl 3): Brown MJ, Brown DC, Murphy MB: Hypokalemia and beta 2-receptor stimulation by circulating epinephrine. N EngI J Med 1983;309: Vincent HH, Man In't Veld AJ, Boomsma F, Wenting GJ, Schalekamp MADH: Elevated plasma noradrenaline in response to /3-adrenoceptor stimulation in man. Br J Clin Pharmacol 1982;13: Fellows IW, Bennett T, Macdonald IA: The effect of adrenaline upon cardiovascular and metabolic functions in man. Clin Sci 1985;69: Bergel DH, Davies CL, Spougos HP: Changes in plasma noradrenaline concentration following intravenous isoprenaline in man and dogs. BrJ Pharmacol 1981;74:823P 138. Vendsalu A: Studies on adrenaline and noradrenaline in human plasma. Ada Physiol Scand 1960,173: Persson B, Andersson OK, Hjemdahl P, Wysocki M, Agerwall S, WaUin G: Adrenaline infusion in man increases muscle sympathetic activity and noradrenaline overflow to plasma. / Hypertens 1989;7: Vincent HH, Boomsma F, Man in't Veld AJ, Schalekamp MADH: Stress levels of adrenaline amplify the blood pressure response to sympathetic stimulation. J Hypertens 1986;4: Lenders JWM, Peters JHM, Pieters GFF, Willemsen JJ, Thien T: Hemodynamic reactivity to sympathoadrenal stimulation in adrenalectomized women. / Clin Endocrinol Metab 1988;67: Scian LF, Westermass CD, Verdesca AS, Hilton JG: Adrenocortical and medullary effects of glucagon. Am J Physiol 1960,199: Floras JS: Epinephrine facilitates reflex neurogenic vasoconstriction and tachycardia in humans. Trans Assoc Am Physicians 1987;C: Abboud FM, Schmid PG, Eckstein JW: Vascular responses after alpha adrenergic receptor blockade: I. Responses of capacitance and resistance vessels to norepinephrine in man. / Clin Invest 1968;47:l Floras JS, Aylward PE, Mark AL, Abboud FM: Adrenaline facilitates neurogenic vasoconstriction in borderline hypertensive subjects. / Hypertens 1990,8: Bachmann AW, Gordon RD, Bathgate RAD, Thompson RE: Effect of graded adrenaline infusion on arterial adrenaline clearance in normotensive and hypertensive man. Clin Exp Pharmacol Physiol 1990;17: Blankestijn PJ, Man In't Veld AJ, Tulen J, Van Den Meiracker AH, Boomsma F, Moleman P, Van Eck HJR, Derkx FHM, Mulder P, Lamberts SJ, Schalekamp MADH: Support for adrenaline-hypertension hypothesis: 18 hour pressor effect after 6 hours adrenaline infusion. Lancet 1988;2: Floras JS, Sole MJ, Morris BL: Desipramine blocks augmented neurogenic vasoconstrictor responses to epinephrine. Hypertension 1990,15: Chang PC, van der Krogt JA, Bmmmelen P: Demonstration of neuronal and extraneuronal uptake of circulating norepinephrine in the forearm. Hypertension 1987;9: Halter JB, Pflug AE, Tolas AG: Arterial-venous differences of plasma catecholamines in man. Metab Clin Exp 1980,29: Hjemdahl P, Eklund B, Kaijser L: Catecholamine handling by the human forearm at rest and during isometric exercise and lower body negative pressure (abstract). Br J Pharmacol 1982;77324P 152. Fellows IW, Macdonald IA, Bennett T, O'Donoghue DP: Effect of intravenous infusion of adrenaline on the cardiovascular responses to distal body subatmospheric pressure in man. Clin Sci 1988;75: Macdonald IA, Bennett T, Fellows IW: Catecholamines and the control of metabolism in man. Clin Sci 1985;68: Esler M, Jackman G, Leonard P, Skews H, Bobik A, Komer P: Effect of norepinephrine uptake blockers on norepinephrine kinetics. Clin Pharmacol Ther 1981^9: Goldstein DS, Zimlichman R, Stull R, Folio J, Levinson PD, Keiser HR, Kopin LI: Measurement of regional neuronal

18 18 Hypertension Vol 19, No 1 January 1992 removal of norepinephrine in man. / Clin Invest 1985;76: Stratton JR, Pfcifer MA, Ritchie JL, Halter JB: Hemodynamic effects of epinephrine: Concentration-effect study in humans. JAppl Physiol 1985^5: Eliasson K: Borderline hypertension: Circulatory sympathoadrenal and psychological reactions to stress. Ada MedScand 1985;692(suppl):l Warren JB, O'Brien M, Dalton N, Turner CT: Sympathetic activity in benign familial tremor. Lancet 1984; 1: Bevan RD: Trophic effects of peripheral adrenergic nerves on vascular structure. Hypertension 1984;6(suppl III): m-19-m Mark AL: Structural changes in resistance and capacitance vessels in borderline hypertension. Hypertension 1984;6(suppl ni):in-69-iii Kohno M, Murakawa K, Yokokawa K, Yasunari K, Horio T, Kurihara N, Takeda T: Production of endothelin by cultured porcine cndothelial cells: Modulation by adrenaline. / Hypertens 1989;7(suppl 6):S13O-S Kawasaki H, Cline WH, Su C: Involvement of the vascular renin-angiotensin system in /3-adrenoceptor-mediated facilitation of vascular neurotransmission in spontaneously hypertensive rats. / Pharmacol Exp Ther 1984;231: Zimmerman BG: Actions of angiotensin on adrenergic nerve endings. Fed Proc 1978^37: Gothert M, Kollecker P: Subendothelial /Jj-adrenoceptors in the rat vena cava: Facilitation of noradrenaline release via local stimulation of angiotensin II synthesis. Naunyn Schmiedebergs Arch Pharmacol 1986^34: Webb DJ, Seidelin PH, Benjamin N, Collier JG, Struthers AD: Sympathetically mediated vasoconstriction is augmented by angiotensin II in man. / Hypertens 1988;6(suppl 4):S542-S Vincent HH, Boomsma F, Man in't Veld AJ, Derkx FHM, Wenting GH, Schalekamp MADH: Effects of selective and nonselective 0-agonists on plasma potassium and norepinephrine. J Cardiovasc Pharmacol 1984;6: Sweadner KJ: Ouabain-evoked norepinephrine release from intact rat sympathetic neurons: Evidence for carriermediated release. / Neurosd 1985^: Baron DN, Green RJ, Khan FA: Adrenaline and ion flux in isolated human leucocytes. Clin Sci 1985;68: Dahlof B, Andren L, Svensson A, Hansson L: Antihypertensive mechanism of beta-adrenoceptor antagonism: The role of beta r blockade. / Hypertens 1983;l(suppl 2): Vincent HH, Man in't Veld AJ, Boomsma F, Derkx FHM, Schalekamp MADH: Is ft-antagonism essential for the antihypertensive action of 0-blockers? Hypertension 1987;9: Guyton AC The surprising kidney-fluid mechanism for pressure control Its infinite gain! Hypertension 1990;16: KEYWORDS hypertension system ^-adrenergic receptor epinephrine essential norepinephrine stress sympathetic nervous