- Email: [email protected]
The role of beta1 and beta2 adrenoceptors in isoproterenol-induced drinking
BRAIN RESEARCH ELSEVIER
Brain Research 656 (1994) 79-84
Research report
The role of beta I and beta 2 adrenoceptors in isoproterenol-induced drinking Robert F. Kirby **, Colleen M. Novak, Robert L. Thunhorst, Alan Kim Johnson * Departments of Psychology and Pharmacology and the Cardiovascular Center, The University of lowa, Iowa City, IA 52242-1407, USA Accepted 24 May 1994
Abstract
The present study examined the contribution of beta t and beta 2 adrenoceptor activation to drinking behavior and the stimulation of plasma renin activity produced by the mixed beta adrenoceptor agonist, isoproterenol. The stimulation of drinking by beta adrenoceptor activation could occur via two independent pathways; by either directly stimulating renal beta 1 adrenoceptors on the juxtaglomerular cells to release renin or by stimulating vascular beta e adrenoceptors that would decrease blood pressure and activate afferent neural and humoral mechanisms. Selective pharmacological antagonism of each adrenoceptor type was achieved by administering atenolol (2.5 mg/kg), a beta I adrenoceptor antagonist, or ICI 118,551 (1 mg/kg), a beta 2 adrenoceptor antagonist, before treatment with isoproterenol (25 tzg/kg). Neither adrenoceptor mechanism alone could account for all of the water intake or stimulation of plasma renin activity due to isoproterenol treatment. Cardiovascular recordings confirmed the selectivity of the antagonists to their respective receptor subtypes, with atenolol blocking the beta 1 adrenoceptormediated heart rate increases and ICI 118,551 blocking the beta 2 adrenoceptor-mediated depressor response to isoproterenol. The results provide evidence that the stimulation of both beta~ and beta 2 adrenoceptors by isoproterenol acts in a synergistic manner to induce drinking and renin-angiotensin system activation. Key words: Isoproterenol; Drinking; Hypotension; Adrenergic /3 receptor; Renin-angiotensin system
1. Introduction
Beta adrenoceptor activation is known to induce drinking behavior in rats [2,7]. These findings have been based primarily on drinking induced by the mixed beta adrenoceptor agonist, isoproterenol [7,10,12,13, 17]. Different mechanisms for isoproterenol-induced thirst have been proposed. One mediator of the drinking effect may be the release of renin leading to the formation of angiotensin II ( A N G II), which, in turn, acts on the brain to produce drinking [5,8,10,12,14]. It has also been proposed that the depressor effect of isoproterenol, mediated through non-renal mechanisms [6,15], is an important stimulus for the drinking response. Both hypotheses focus on the peripheral effects of isoproterenol; either the hormonal effect of
* Corresponding author. Fax: (1) (319) 335-2507. ** Reprint requests.
0006-8993/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0 0 0 6 - 8 9 9 3 ( 9 4 ) 0 0 6 5 3 - T
renin or the depressor effect without a significant contribution of the renin-angiotensin system (RAS). Postsynaptic beta I and beta 2 adrenoceptors likely to influence drinking behavior are located in various sites throughout the cardiovascular system. Beta I adrenoceptors, located in the kidney, mediate renin release from the juxtaglomerular cells into the blood. Beta 2 adrenoceptors, located in the peripheral vasculature, stimulate vasodilation of the skeletal muscle to induce hypotension. Isoproterenol-induced drinking behavior may therefore result from the activation of either beta adrenoceptor subtype and the attendent stimulation of neural and humoral afferent signals to the central nervous system. The following study investigated the role of beta1 and beta 2 adrenoceptors in drinking, RAS activation, and cardiovascular changes produced by isoproterenol treatment. Selective pharmacological blockade of either receptor subtype was used to isolate the contribution of each beta adrenergic pathway to RAS activation and drinking produced by isoproterenol. Cardio-
R.F. Kirby et al. /Brain Research 656 (1994) 79 84
80 vascular pressor
responses effect
were
specificity of each respective
2. Materials
subtype.
that
in t h e d r i n k i n g
isoproterenol
to address
and
beta-adrenoceptor
receptor
ies demonstrate pate
examined
of isoproterenol
both and
The
antagonist pathways
activation
to its partici-
produced
by
treatment.
and
2.5. Experiment .7
the
results of these stud-
receptor RAS
the de-
to determine
methods
p
2.1. Subjects Adult male Sprague-Dawley rats (Harlan; Indianapolis, IN) weighing 270-360 g were used. Animals were maintained in wire mesh cages with Purina chow (5021) and tap water available ad libitum. A 12 h - 1 2 h light-dark cycle was employed and testing was conducted during the light phase. Temperature and humidity were held constant in the colony.
Expt. 3 investigated the cardiovascular responses to isoproterenol and determined antagonist selectivity. Animals were anesthetized with equithesin (0.33 m l / 1 0 0 g b.wt.) and a catheter, made of PE-50 with a PE-10 tip, was placed in the carotid artery. The catheter exited through an incision (/).5 cm) made on the dorsal surface of the neck. Catheters were filled with heparinized (100 I U / m l ) saline and the animals were returned to their home cage until tested 4 to 5 days later. For testing, each animal was placed in a test cage, their arterial catheter was attached to a Beckman polygraph, and left to acclimate for 30 min to their new environment. Baseline measures of blood pressure and heart rate were recorded, followed by a pretreatment injection of atenolol, ICI 118,551, or vehicle to each animal (n = 10 for each pretreatment). Each animal received only one pretreatment condition in the course of this study. After 15 min, blood pressure and heart rate were recorded and each animal was administered isoproterenol. A control group received a vehicle injection instead of isoproterenol injection (n = 4). Cardiovascular responses were measured at 1, 5, 10, 15, 20, 30, 60, and 90 min post isoproterenol or vehicle treatment.
2. 6. Statistical analysis 2.2. Drug administrations Two beta adrenoceptor antagonists were used as pretreatment drugs. Atenolol, which blocks beta I adrenoceptors, was given at a dose of 2.5 m g / k g . ICI 118,551, which blocks beta 2 adrenoceptors, was given at a dose of 1 m g / k g . Pretreatment drugs and isoproterenol were dissolved in distilled water, which was used for control injections. All injections were given subcutaneously at 1 m l / k g . lsoproterenol (lsuprel HC1; Sanofi Winthrop Pharmaceuticals) was purchased and ICI 118,551 and atenolol were donated by ICI and Stuart Pharmaceuticals, respectively.
2.3. Experiment 1 Expt. 1 examined the drinking behavior induced by isoproterenol. Three pretests were used to secure an optimal isoproterenol dose for stimulating water intake. In the first 2 trials, 2 5 / z g / k g isoproterenol was given; in the third trial, 50 p,g/kg was given. Water intake was recorded at 30, 60, and 90 min post treatment. Greater water intakes were found to the lower isoproterenol dose and therefore 25 #zg/kg was used for all further studies. The contribution of beta 1 or beta 2 adrenoceptors to drinking behavior induced by isoproterenol was then examined in the same animals. Vehicle, atenolol, or 1CI 118,551 was administered followed 15 rain later by isoproterenol. Water volumes were recorded at 30, 60, and 90 min after isoproterenol treatment. Testing consisted of three trials, with each animal receiving each pretreatment. At least 2 days passed between testing periods.
Water intakes were analyzed using a repeated measures analysis of variance ( A N O V A ) with the cumulative intakes as one factor and treatment as another factor. The P R A data were subjected to an A N O V A with three treatment levels. Heart rate and M A P data were subjected to 2-way repeated measures A N O V A s over four groups. To examine differences between treatment groups, pairs of group means were subjected to l-way repeated measures A N O V A s and followed up with post-hoc analyses at each time point. Fisher's test of least significant difference (LSD) was used for post-hoc analyses.
3. Results
Water and
intakes
isoproterenol
[IlL
are
•
vL
•
/'xt(; r',oiol L I(I i 5 L:~}1
•
following illustrated
vehiclein F i g .
1.
u~:
~J
0
////~--
e-
_JJ
L CG}
/
c}
j j,
/
2.4. Experiment 2
J
'
Expt. 2 examined the activation of the RAS using the animals from Expt. 1 and the same procedure was followed: all animals received pretreatment injections of either atenolol, ICI 118,551, or vehicle, followed by isoproterenol treatment. Trunk blood was collected 30 min post isoproterenol treatment into chilled tubes containing a m m o n i u m heparin (20 U / m l ) . Samples were centrifuged for 10 min at 3000 rpm to separate plasma for the assay of renin activity. Plasma renin activity (PRA) was measured at the University of Iowa Cardiovascular Center Core Radioimmuno-Assay laboratory.
to
antagonist-pretreatment
(~
/ •' ,/•
, ! ,
•
•
}')
,k i
[
T _ira_
••
60
90
,
•
Mir~tltes Fig. t. Cumulative water intake following isoproterenol in animals pretreated with vehicle, or the beta~ and beta 2 adrenoceptor antagonists, atenolol and ICI 118,551, respectively. Selective blockade of either beta adrenoceptor subtype significantly reduced isoproterenol-induced drinking ( P < 0.01 ).
R.F. Kirby et al. / Brain Research 656 (1994) 79-84 130
35
o~ I 120
I_
3C
81
F
~
¢
.
~
?
~
~
~
vE 110
T
25
1
03 c
2C
T
~d 15 121 c
&
5
80-
2 Y3 c o
kl 1 ~ 1 70-
o Veh-Veh
H/I
• Veh - I s o p • A t e n - Isop • ICI - Isop
60-
tt
10
L
E ffl cl
90
c
>
E
k
T
Boseline
5
15
30
5 o
60 Time
Veh
Aten
ICl
Pret reatment Fig. 2. Plasma renin activity at 30 min post injection of isoproterenol. Beta 1 and beta 2 adrenoceptor blockade decreased the P R A response to isoproterenol ( P < 0.01). In addition, beta I blockade produced a greater inhibition of the P R A response to isoproterenol than did beta 2 blockade ( P < 0.05).
Pretreatment with the specific beta-adrenoceptor antagonists produced significant decreases in water intake to isoproterenol, (F2,46 = 14.96, P < 0.001) and changed the pattern of drinking during the test session (F4,92 = 2.56, P < 0.05). Both the atenolol- and ICI-pretreatments significantly decreased water intake compared to the vehicle-pretreatment. Water intake of the ICI l18,551-pretreated group was slightly greater than the atenolol-pretreated group but this did not reach significance ( P < 0.10). The stimulation of P R A following isoproterenol for vehicle- and antagonist-pretreated animals is illustrated in Fig. 2. The vehicle-pretreated group had greater PRA than both the atenolol-pretreated group (t22 = 10.47, P < 0.01) and the ICI l18,551-pretreated group (t22 = 7.51, P < 0.01). The ICI 118,551-pretreated group also had greater P R A than the atenololpretreated group (t22 = 2.96, P < 0.01). Mean arterial pressure (MAP) for each group is shown in Fig. 3. Isoproterenol treatment produced a fall in blood pressure that was influenced by the pretreatment conditions (F3,30 = 7.53, P < 0.01). Blood pressure of isoproterenol-treated animals was significantly lower than vehicle-treated animals from 1 to 30 min post treatment. This fall in MAP was unaffected by pretreatment with the beta I adrenoceptor antagonist, atenolol. In contrast, pretreatment with the beta 2 adrenoceptor antagonist, ICI 118,551, blocked the depressor effect of isoproterenol. The MAPs of ICI 118,551-pretreated animals administered isoproterenol were significantly higher than the vehicle-pretreated animals that received isoproterenol (Fo,18 = 17.91, P <
90
(rain)
Fig. 3. Mean arterial pressure following isoproterenol or vehicle treatment. Blood pressure fell comparably in vehicle and beta I adrenoceptor antagonist pretreated animals to isoproterenol treatment. However, the depressor response was blocked by beta z adrenoceptor antagonism.
0.01). Comparison of the MAPs between ICI 118,551pretreated/isoproterenol and vehicle-pretreated/ vehicle-treated group demonstrated no significant main effects of treatment or time, or interaction of treatment by time. Heart rate responses to isoproterenol are illustrated in Fig. 4. There was a significant effect of drug treatment on heart rate that varied with time for the four treatment groups ( P ' s < 0.01). Isoproterenol treatment increased heart rate compared to vehicle-treated animals (F1j 2 = 22.18, P < 0.01). However, isoproterenol did not increase heart rate in atenolol-pretreated animals compared to vehicle/vehicle-treated animals (Fl2,1 = 1.76, n.s.). Treatment with ICI 118,551 did not alter the heart rate response to isoproterenol when compared to the vehicle-pretreated/isoproterenoltreated group (F1j 8 = 4.66, n.s.).
o veh-ven
550
525
~\l
n~ 5(?o m
• Veh - Isop • Aten-lsop
~
~
*
Icl 1sop
475
450 425 L
400
T
375" 350"
x
325"
Basefine
5
15
30
60
Time
90
(rain)
Fig. 4. Heart rate following isoproterenol or vehicle treatment. Heart rate increased comparably in vehicle and beta 2 adrenoceptor-pretreated animals to isoproterenol treatment. This heart rate increase was blocked by beta 1 adrenoceptor antagonism.
g2
R.F. Kirby et al. / Brain Research 656 (1994) 79-84
4. Discussion
Drinking elicited by isoproterenol was found to be significantly decreased by blockade of either beta l or beta 2 adrenoceptors. Blockade of these same beta adrenoceptor subtypes also diminished peripheral RAS activation following isoproterenol treatment. Cardiovascular responses to the treatments confirmed that the antagonists were selective for their respective adrenoceptor subtypes. Specifically, the beta 2 adrenoceptor antagonist, ICI 118,551, selectively blocked the depressor response of isoproterenol, but did not affect the heart rate response. In contrast, the betal adrenoceptor antagonist, atenolol, selectively attenuated the heart rate effect of isoproterenol but did not affect the depressor response to isoproterenol. Taken together, the present results demonstrate that both beta 1 and beta 2 adrenoceptor-associated mechanisms mediate the drinking response to isoproterenol. Falk and Tang [2] previously attempted to identify a role for beta 2 adrenoceptors in drinking behavior in rats. Two beta 2 adrenoceptor agonists, salbutamol and quinterenol, successfully induced the animals to drink. However, the cardiovascular responses to these challenges were not measured and the selectivity of the drugs to their respective beta adrenoceptor subtype was not monitored. Falk and Tang used large doses of salbutamol (from 0.25 m g / k g to 8 mg/kg) and quinterenol (from 2 m g / k g to 20 mg/kg), which would not have been selective to beta 2 adrenoceptors at such large doses [9]. Therefore, drinking to the agonists most likely involved activation of both beta~ and beta 2 adrenoceptor subtypes. In the present study, the cardiovascular measures provide strong evidence that the beta~ and beta 2 adrenoceptor blockers were selective to their respective receptor subtypes. The results of the present study are the first to demonstrate a role for both beta~ and beta 2 adrenoceptors in drinking behavior. Several studies give evidence that renin release and the subsequent formation of A N G II is the primary mechanism in isoproterenol-induced drinking. The onset of drinking to isoproterenol is correlated with renin release [10], the amount of A N G II formed is in the physiological range to induce drinking [11,13], and isoproterenol-induced drinking may be diminished by RAS blockade [1] or nephrectomy [7,14]. These studies provide evidence for an important role of the renal RAS in isoproterenol-induced drinking. The results of the present study are consistent with the hypothesis that the peripheral RAS mediates drinking to isoproterenol, and demonstrate that both beta adrenoceptor subtypes participate in the RAS response and the drinking behavior. Isoproterenol may also stimulate drinking independent of its effects on the renal RAS. While multiple
studies have demonstrated that bilateral nephrectomy and RAS blockade inhibit isoproterenol drinking (see above), many studies have found residual drinking after these same manipulations [4,16,17]. This has led to the suggestion that depressor responses to isoproterenol may stimulate central thirst mechanisms through afferent baroreceptor input, with no role or only a permissive role for the renal RAS [6,15]. The studies of Rettig, Ganten, and Johnson [13] support the hypothesis that both renal RAS and nonrenal mechanisms are involved in isoproterenol-induced thirst. These authors found that a low dose of isoproterenol (25 ~ g / k g ) did not induce drinking in nephrectomized animals. However, a high dose of isoproterenol (330 p~g/kg) was effective at stimulating drinking after bilateral nephrectomy. Mean arterial pressure (MAP) in bilaterally nephrectomized animals given the low dose rapidly dropped to about 90 mmHg and then slowly rose across the 1-h test period. In contrast, pressure dropped to below 80 mmHg in response to the high dose and did not rise across the test period. They interpreted these findings to indicate that the RAS is the primary mediator of isoproterenol-induced drinking in intact animals but that the extreme depressor effect of high isoproterenol doses may involve extrarenal factors, such as arterial baroreceptors, in mediating drinking. The inability of RAS inhibition to abolish drinking to isoproterenol in some studies has led a number of investigators to question the role of A N G II in isoproterenol-induced drinking [4,16,17]. However, studies that use nephrectomy or peripheral RAS blockade are complicated by the fact that extrarenal factors may be brought into play due to the elimination of circulating ANG II vasoconstrictor actions, leading to greater depressor responses to isoproterenol [13]. Given that hypotension stimulates drinking in nephrectomized rats [13], greater falls in blood pressure to isoproterenol in the absence of the vasoconstrictor action of ANG II can influence the mechanisms that mediate drinking to isoproterenol. A recent study by Fitts [3] has circumvented the common difficulty associated with peripheral blockade studies to provide convincing evidence that the peripheral RAS is the primary mediator of drinking produced by low isoproterenol doses. Circulating A N G II stimulates drinking through the activation of A N G II receptors in the subfornical organ (SFO). Therefore, selective blockade of A N G II receptors in the SFO would eliminate the receptor population mediating the drinking response without blocking the receptors mediating the vasoconstriction. Fitts [3] demonstrated that either lesions of the SFO, or direct application of the A N G II receptor antagonist Sar ~, ThrS-ANG II, abolished or severely diminished drinking to isoproterenol (10-50 ~g/kg). However, drinking to isoproterenol was not
R.F. Kirby et al. / Brain Research 656 (1994) 79-84
diminished by blockade of the brain RAS by intraventricular Sar 1, Thr8-ANG II infusion at a dose 25 times greater than that used in the SFO. These results suggest that isoproterenol produces drinking primarily by activation of the peripheral RAS which stimulates ANG II receptors located outside the blood-brain barrier in the SFO. Isoproterenol administration mimics in many ways the hypovolemic situation that is proposed to elicit RAS-dependent thirst. Hypovolemia increases sympathetic drive to the kidney, increasing renin release, and as volume loss increases, hypotension contributes a second stimulus to increase RAS activity. In the present study, isoproterenol treatment increased activity of the RAS via both beta 1 adrenoceptor stimulation (increased sympathetic drive) and beta 2 adrenoceptormediated hypotension. Blockade of either beta adrenoceptor pathway significantly decreased renin release and drinking behavior, indicating that neither receptor pathway alone could account for all of the RAS stimulation or drinking after isoproterenol. Most importantly, the RAS activation and water intake produced by isoproterenol with both receptors active is greater than the individual contribution of beta~ and beta 2 adrenoceptors added together. These results indicate that sympathetic drive to the kidney and hypotension act synergistically to increase RAS activity and thereby, enhance drinking behavior. Two explanations support a synergistic effect of beta~ and beta 2 adrenoceptor activation by isoproterenol. First, Thames and DiBona [18] found that decreased blood pressure increased the release of renin to renal nerve stimulation. This indicates that the depressor response to isoproterenol may exaggerate the capacity of beta~ adrenoceptor to stimulate renin release. Second, unloading of low pressure/volume receptors increases the dipsogenic responsiveness to ANG II. Reductions in arterial pressure below resting levels increased the drinking response to ANG II administration [19]. The drinking response to isoproterenol could be facilitated, therefore, by reduced pressure increasing the amount of ANG II produced in the circulation or by increasing the sensitivity of central pathways related to the mediation of angiotensin-induced drinking behavior. The present results demonstrate that beta I and beta 2 adrenoceptor subtypes contribute to isoproterenol-induced drinking. The localization of these receptor subtypes and the control mechanisms for renin release support the idea that isoproterenol increases PRA via the direct activation of beta1 adrenoceptors on the juxtaglomerular apparatus and by stimulating vasodilatory beta 2 adrenoceptors located in the skeletal muscle vasculature that results in decreased renal perfusion pressure. The consistent reductions in PRA and drinking behavior following selective beta adrenoceptor an-
83
tagonism support the hypothesis that isoproterenol-induced drinking is mediated by the renal RAS. These results clarify the role of both neural and non-neural mechanisms, and their interactive nature, in the drinking elicited by isoproterenol.
Acknowledgements This research was supported by National Heart, Lung and Blood Institute, NIH Grants HL 14338 and HL 44546.
References [1] Evered, M.D. and Robinson, M.M., The renin-angiotensin system in drinking and cardiovascular responses to isoprenaline in the rat, J. Physiol., 316 (1981) 357-367. [2] Falk, J.L. and Tang, M., Salbutamol and quinterenol: dispogenic action produced by beta-adrenergic stimulants, Pharmacol. Biochem. Behav., 2 (1974) 413-415. [3] Fitts, D.A., Angiotensin II receptors in SFO but not in OVLT mediate isoproterenol-induced thirst, Am. J. Physiol., 267 (1994) RXX-RXX. [4] Fregly, M.J. and Rowland, N.E., Effect of DuP 753, a nonpeptide angiotensin II receptor antagonist, on the drinking responses to acutely administered dipsogenic agents in rats, P.S.E.B.M, 199 (1992) 158-164. [5] Fregly, M.J., Threatte, R.M., Barney, C.C. and Katovich, M.J., Effect of acute administration of isoproterenol and angiotensin II, seperately and in combination, on water intake and blood pressure in rats, Brain Res. Bull., 10 (1983) 327-332. [6] Hosutt, J.A., Rowland, N. and Stricker, E.M., Hypotension and thirst in rats after isoproterenol treatment, Physiol. Behav., 21 (1978) 593-598. [7] Houpt, K.A. and Epstein, A.N., The complete dependence of beta-adrenergic drinking on the renal dipsogen, Physiol. Behav., 7 (1971) 897-902. [8] Hubbard, J.I., Lin, N. and Sibbald, J.R., Subfornical organ lesions in rats abolish hyperdipsic effects of isoproterenol and serotonin, Brain Res. Bull., 23 (1989) 41-45. [9] Kirby, R.F., Woodworth, C.W., Woodworth, G.G. and Johnson, A.K., Beta-2 adrenoceptor mediated vasodilation: role in cardiovascular responses to acute stressors in spontaneously hypertensive rats, Clin. Exp. Hyper.-Theory Practice, A13 (1991) 10591068. [10] Leenen, F.H.H. and McDonald, R.H.J., Effect of isoproterenol on blood pressure, plasma renin activity, and water intake in rats, Eur. Z Pharmacol., 26 (1974) 129-135. [ll] Mann, J.F., Johnson, A.K. and Ganten, D., Plasma angiotensin II: dipsogenic levels and angiotensin-generating capacity of renin, Am. J. Physiol., 238 (1980) R372-R377. [12] Meyer, D.K. and Hertting, G., Drinking induced by direct or indirect stimulation of beta-receptors: evidence for involvement of the renin-angiotensin system. In G. Peters, J.T. Fitzsimons and L. Peters-Haefeli (Eds.), Control Mechanisms of Drinking, Springer-Verlag, New York, 1975, pp. 89-92. [13] Rettig, R., Ganten, D. and Johnson, A.K., Isoproterenol-induced thirst: renal and extrarenal mechanisms, Am. Z Physiol., 241 (1981) R152-R157. [14] Rolls, B.J. and Ramsey, D.J., The elevation of endogenous angiotensin and thirst in the dog. In G. Peters, J.T. Fitzsimons
84
R.F. Kirby et al. /Brain Research 656 (1994) 79 84
and L. Peters-Haefeli (Eds.), Control Mechanisms of Drinking, Springer-Verlag, New York, 1975, pp. 74-78. [15] Stricker, E.M., The renin/angiotensin system and thirst: a reevaluation. II. Drinking elicited in rats by caval ligation or isoproterenol, J. Comp. Physiol. Psych., 91 (1977) 1220-1231. [16] Szczepanska-Sadowska, E. and Fitszimons, J.T., The effects of angiotensin I1, renin and isoprenaline on drinking in the dog. In G. Peters, J.T. Fitzsimons and L. Peters-Haefeli (Eds.), Control Mechanisms of Drinking, Springer-Verlag, New York, 1975, pp. 69-73.
[17] Tang, M. and Falk, J.L., SarI-Ala s angiotensin II blocks reninangiotensin but not beta-adrenergic dispogens, Pharmacol. Biochem. Behat,., 2 (1974) 401-418. [18] Thames, M.D. and DiBona, G.F., Renal nerves modulate the secretion of renin mediated by nonneural mechanisms, Circ. Res., 44 (1979) 645-652. [19] Thunhorst, R.L. and Johnson, A.K., Effects of arterial pressure on drinking and urinary responses to intracerebroventricular angiotensin 11, Am. J. PhysioL, 264 (1993) R211-R217.
Brain Research 656 (1994) 79-84
Research report
The role of beta I and beta 2 adrenoceptors in isoproterenol-induced drinking Robert F. Kirby **, Colleen M. Novak, Robert L. Thunhorst, Alan Kim Johnson * Departments of Psychology and Pharmacology and the Cardiovascular Center, The University of lowa, Iowa City, IA 52242-1407, USA Accepted 24 May 1994
Abstract
The present study examined the contribution of beta t and beta 2 adrenoceptor activation to drinking behavior and the stimulation of plasma renin activity produced by the mixed beta adrenoceptor agonist, isoproterenol. The stimulation of drinking by beta adrenoceptor activation could occur via two independent pathways; by either directly stimulating renal beta 1 adrenoceptors on the juxtaglomerular cells to release renin or by stimulating vascular beta e adrenoceptors that would decrease blood pressure and activate afferent neural and humoral mechanisms. Selective pharmacological antagonism of each adrenoceptor type was achieved by administering atenolol (2.5 mg/kg), a beta I adrenoceptor antagonist, or ICI 118,551 (1 mg/kg), a beta 2 adrenoceptor antagonist, before treatment with isoproterenol (25 tzg/kg). Neither adrenoceptor mechanism alone could account for all of the water intake or stimulation of plasma renin activity due to isoproterenol treatment. Cardiovascular recordings confirmed the selectivity of the antagonists to their respective receptor subtypes, with atenolol blocking the beta 1 adrenoceptormediated heart rate increases and ICI 118,551 blocking the beta 2 adrenoceptor-mediated depressor response to isoproterenol. The results provide evidence that the stimulation of both beta~ and beta 2 adrenoceptors by isoproterenol acts in a synergistic manner to induce drinking and renin-angiotensin system activation. Key words: Isoproterenol; Drinking; Hypotension; Adrenergic /3 receptor; Renin-angiotensin system
1. Introduction
Beta adrenoceptor activation is known to induce drinking behavior in rats [2,7]. These findings have been based primarily on drinking induced by the mixed beta adrenoceptor agonist, isoproterenol [7,10,12,13, 17]. Different mechanisms for isoproterenol-induced thirst have been proposed. One mediator of the drinking effect may be the release of renin leading to the formation of angiotensin II ( A N G II), which, in turn, acts on the brain to produce drinking [5,8,10,12,14]. It has also been proposed that the depressor effect of isoproterenol, mediated through non-renal mechanisms [6,15], is an important stimulus for the drinking response. Both hypotheses focus on the peripheral effects of isoproterenol; either the hormonal effect of
* Corresponding author. Fax: (1) (319) 335-2507. ** Reprint requests.
0006-8993/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0 0 0 6 - 8 9 9 3 ( 9 4 ) 0 0 6 5 3 - T
renin or the depressor effect without a significant contribution of the renin-angiotensin system (RAS). Postsynaptic beta I and beta 2 adrenoceptors likely to influence drinking behavior are located in various sites throughout the cardiovascular system. Beta I adrenoceptors, located in the kidney, mediate renin release from the juxtaglomerular cells into the blood. Beta 2 adrenoceptors, located in the peripheral vasculature, stimulate vasodilation of the skeletal muscle to induce hypotension. Isoproterenol-induced drinking behavior may therefore result from the activation of either beta adrenoceptor subtype and the attendent stimulation of neural and humoral afferent signals to the central nervous system. The following study investigated the role of beta1 and beta 2 adrenoceptors in drinking, RAS activation, and cardiovascular changes produced by isoproterenol treatment. Selective pharmacological blockade of either receptor subtype was used to isolate the contribution of each beta adrenergic pathway to RAS activation and drinking produced by isoproterenol. Cardio-
R.F. Kirby et al. /Brain Research 656 (1994) 79 84
80 vascular pressor
responses effect
were
specificity of each respective
2. Materials
subtype.
that
in t h e d r i n k i n g
isoproterenol
to address
and
beta-adrenoceptor
receptor
ies demonstrate pate
examined
of isoproterenol
both and
The
antagonist pathways
activation
to its partici-
produced
by
treatment.
and
2.5. Experiment .7
the
results of these stud-
receptor RAS
the de-
to determine
methods
p
2.1. Subjects Adult male Sprague-Dawley rats (Harlan; Indianapolis, IN) weighing 270-360 g were used. Animals were maintained in wire mesh cages with Purina chow (5021) and tap water available ad libitum. A 12 h - 1 2 h light-dark cycle was employed and testing was conducted during the light phase. Temperature and humidity were held constant in the colony.
Expt. 3 investigated the cardiovascular responses to isoproterenol and determined antagonist selectivity. Animals were anesthetized with equithesin (0.33 m l / 1 0 0 g b.wt.) and a catheter, made of PE-50 with a PE-10 tip, was placed in the carotid artery. The catheter exited through an incision (/).5 cm) made on the dorsal surface of the neck. Catheters were filled with heparinized (100 I U / m l ) saline and the animals were returned to their home cage until tested 4 to 5 days later. For testing, each animal was placed in a test cage, their arterial catheter was attached to a Beckman polygraph, and left to acclimate for 30 min to their new environment. Baseline measures of blood pressure and heart rate were recorded, followed by a pretreatment injection of atenolol, ICI 118,551, or vehicle to each animal (n = 10 for each pretreatment). Each animal received only one pretreatment condition in the course of this study. After 15 min, blood pressure and heart rate were recorded and each animal was administered isoproterenol. A control group received a vehicle injection instead of isoproterenol injection (n = 4). Cardiovascular responses were measured at 1, 5, 10, 15, 20, 30, 60, and 90 min post isoproterenol or vehicle treatment.
2. 6. Statistical analysis 2.2. Drug administrations Two beta adrenoceptor antagonists were used as pretreatment drugs. Atenolol, which blocks beta I adrenoceptors, was given at a dose of 2.5 m g / k g . ICI 118,551, which blocks beta 2 adrenoceptors, was given at a dose of 1 m g / k g . Pretreatment drugs and isoproterenol were dissolved in distilled water, which was used for control injections. All injections were given subcutaneously at 1 m l / k g . lsoproterenol (lsuprel HC1; Sanofi Winthrop Pharmaceuticals) was purchased and ICI 118,551 and atenolol were donated by ICI and Stuart Pharmaceuticals, respectively.
2.3. Experiment 1 Expt. 1 examined the drinking behavior induced by isoproterenol. Three pretests were used to secure an optimal isoproterenol dose for stimulating water intake. In the first 2 trials, 2 5 / z g / k g isoproterenol was given; in the third trial, 50 p,g/kg was given. Water intake was recorded at 30, 60, and 90 min post treatment. Greater water intakes were found to the lower isoproterenol dose and therefore 25 #zg/kg was used for all further studies. The contribution of beta 1 or beta 2 adrenoceptors to drinking behavior induced by isoproterenol was then examined in the same animals. Vehicle, atenolol, or 1CI 118,551 was administered followed 15 rain later by isoproterenol. Water volumes were recorded at 30, 60, and 90 min after isoproterenol treatment. Testing consisted of three trials, with each animal receiving each pretreatment. At least 2 days passed between testing periods.
Water intakes were analyzed using a repeated measures analysis of variance ( A N O V A ) with the cumulative intakes as one factor and treatment as another factor. The P R A data were subjected to an A N O V A with three treatment levels. Heart rate and M A P data were subjected to 2-way repeated measures A N O V A s over four groups. To examine differences between treatment groups, pairs of group means were subjected to l-way repeated measures A N O V A s and followed up with post-hoc analyses at each time point. Fisher's test of least significant difference (LSD) was used for post-hoc analyses.
3. Results
Water and
intakes
isoproterenol
[IlL
are
•
vL
•
/'xt(; r',oiol L I(I i 5 L:~}1
•
following illustrated
vehiclein F i g .
1.
u~:
~J
0
////~--
e-
_JJ
L CG}
/
c}
j j,
/
2.4. Experiment 2
J
'
Expt. 2 examined the activation of the RAS using the animals from Expt. 1 and the same procedure was followed: all animals received pretreatment injections of either atenolol, ICI 118,551, or vehicle, followed by isoproterenol treatment. Trunk blood was collected 30 min post isoproterenol treatment into chilled tubes containing a m m o n i u m heparin (20 U / m l ) . Samples were centrifuged for 10 min at 3000 rpm to separate plasma for the assay of renin activity. Plasma renin activity (PRA) was measured at the University of Iowa Cardiovascular Center Core Radioimmuno-Assay laboratory.
to
antagonist-pretreatment
(~
/ •' ,/•
, ! ,
•
•
}')
,k i
[
T _ira_
••
60
90
,
•
Mir~tltes Fig. t. Cumulative water intake following isoproterenol in animals pretreated with vehicle, or the beta~ and beta 2 adrenoceptor antagonists, atenolol and ICI 118,551, respectively. Selective blockade of either beta adrenoceptor subtype significantly reduced isoproterenol-induced drinking ( P < 0.01 ).
R.F. Kirby et al. / Brain Research 656 (1994) 79-84 130
35
o~ I 120
I_
3C
81
F
~
¢
.
~
?
~
~
~
vE 110
T
25
1
03 c
2C
T
~d 15 121 c
&
5
80-
2 Y3 c o
kl 1 ~ 1 70-
o Veh-Veh
H/I
• Veh - I s o p • A t e n - Isop • ICI - Isop
60-
tt
10
L
E ffl cl
90
c
>
E
k
T
Boseline
5
15
30
5 o
60 Time
Veh
Aten
ICl
Pret reatment Fig. 2. Plasma renin activity at 30 min post injection of isoproterenol. Beta 1 and beta 2 adrenoceptor blockade decreased the P R A response to isoproterenol ( P < 0.01). In addition, beta I blockade produced a greater inhibition of the P R A response to isoproterenol than did beta 2 blockade ( P < 0.05).
Pretreatment with the specific beta-adrenoceptor antagonists produced significant decreases in water intake to isoproterenol, (F2,46 = 14.96, P < 0.001) and changed the pattern of drinking during the test session (F4,92 = 2.56, P < 0.05). Both the atenolol- and ICI-pretreatments significantly decreased water intake compared to the vehicle-pretreatment. Water intake of the ICI l18,551-pretreated group was slightly greater than the atenolol-pretreated group but this did not reach significance ( P < 0.10). The stimulation of P R A following isoproterenol for vehicle- and antagonist-pretreated animals is illustrated in Fig. 2. The vehicle-pretreated group had greater PRA than both the atenolol-pretreated group (t22 = 10.47, P < 0.01) and the ICI l18,551-pretreated group (t22 = 7.51, P < 0.01). The ICI 118,551-pretreated group also had greater P R A than the atenololpretreated group (t22 = 2.96, P < 0.01). Mean arterial pressure (MAP) for each group is shown in Fig. 3. Isoproterenol treatment produced a fall in blood pressure that was influenced by the pretreatment conditions (F3,30 = 7.53, P < 0.01). Blood pressure of isoproterenol-treated animals was significantly lower than vehicle-treated animals from 1 to 30 min post treatment. This fall in MAP was unaffected by pretreatment with the beta I adrenoceptor antagonist, atenolol. In contrast, pretreatment with the beta 2 adrenoceptor antagonist, ICI 118,551, blocked the depressor effect of isoproterenol. The MAPs of ICI 118,551-pretreated animals administered isoproterenol were significantly higher than the vehicle-pretreated animals that received isoproterenol (Fo,18 = 17.91, P <
90
(rain)
Fig. 3. Mean arterial pressure following isoproterenol or vehicle treatment. Blood pressure fell comparably in vehicle and beta I adrenoceptor antagonist pretreated animals to isoproterenol treatment. However, the depressor response was blocked by beta z adrenoceptor antagonism.
0.01). Comparison of the MAPs between ICI 118,551pretreated/isoproterenol and vehicle-pretreated/ vehicle-treated group demonstrated no significant main effects of treatment or time, or interaction of treatment by time. Heart rate responses to isoproterenol are illustrated in Fig. 4. There was a significant effect of drug treatment on heart rate that varied with time for the four treatment groups ( P ' s < 0.01). Isoproterenol treatment increased heart rate compared to vehicle-treated animals (F1j 2 = 22.18, P < 0.01). However, isoproterenol did not increase heart rate in atenolol-pretreated animals compared to vehicle/vehicle-treated animals (Fl2,1 = 1.76, n.s.). Treatment with ICI 118,551 did not alter the heart rate response to isoproterenol when compared to the vehicle-pretreated/isoproterenoltreated group (F1j 8 = 4.66, n.s.).
o veh-ven
550
525
~\l
n~ 5(?o m
• Veh - Isop • Aten-lsop
~
~
*
Icl 1sop
475
450 425 L
400
T
375" 350"
x
325"
Basefine
5
15
30
60
Time
90
(rain)
Fig. 4. Heart rate following isoproterenol or vehicle treatment. Heart rate increased comparably in vehicle and beta 2 adrenoceptor-pretreated animals to isoproterenol treatment. This heart rate increase was blocked by beta 1 adrenoceptor antagonism.
g2
R.F. Kirby et al. / Brain Research 656 (1994) 79-84
4. Discussion
Drinking elicited by isoproterenol was found to be significantly decreased by blockade of either beta l or beta 2 adrenoceptors. Blockade of these same beta adrenoceptor subtypes also diminished peripheral RAS activation following isoproterenol treatment. Cardiovascular responses to the treatments confirmed that the antagonists were selective for their respective adrenoceptor subtypes. Specifically, the beta 2 adrenoceptor antagonist, ICI 118,551, selectively blocked the depressor response of isoproterenol, but did not affect the heart rate response. In contrast, the betal adrenoceptor antagonist, atenolol, selectively attenuated the heart rate effect of isoproterenol but did not affect the depressor response to isoproterenol. Taken together, the present results demonstrate that both beta 1 and beta 2 adrenoceptor-associated mechanisms mediate the drinking response to isoproterenol. Falk and Tang [2] previously attempted to identify a role for beta 2 adrenoceptors in drinking behavior in rats. Two beta 2 adrenoceptor agonists, salbutamol and quinterenol, successfully induced the animals to drink. However, the cardiovascular responses to these challenges were not measured and the selectivity of the drugs to their respective beta adrenoceptor subtype was not monitored. Falk and Tang used large doses of salbutamol (from 0.25 m g / k g to 8 mg/kg) and quinterenol (from 2 m g / k g to 20 mg/kg), which would not have been selective to beta 2 adrenoceptors at such large doses [9]. Therefore, drinking to the agonists most likely involved activation of both beta~ and beta 2 adrenoceptor subtypes. In the present study, the cardiovascular measures provide strong evidence that the beta~ and beta 2 adrenoceptor blockers were selective to their respective receptor subtypes. The results of the present study are the first to demonstrate a role for both beta~ and beta 2 adrenoceptors in drinking behavior. Several studies give evidence that renin release and the subsequent formation of A N G II is the primary mechanism in isoproterenol-induced drinking. The onset of drinking to isoproterenol is correlated with renin release [10], the amount of A N G II formed is in the physiological range to induce drinking [11,13], and isoproterenol-induced drinking may be diminished by RAS blockade [1] or nephrectomy [7,14]. These studies provide evidence for an important role of the renal RAS in isoproterenol-induced drinking. The results of the present study are consistent with the hypothesis that the peripheral RAS mediates drinking to isoproterenol, and demonstrate that both beta adrenoceptor subtypes participate in the RAS response and the drinking behavior. Isoproterenol may also stimulate drinking independent of its effects on the renal RAS. While multiple
studies have demonstrated that bilateral nephrectomy and RAS blockade inhibit isoproterenol drinking (see above), many studies have found residual drinking after these same manipulations [4,16,17]. This has led to the suggestion that depressor responses to isoproterenol may stimulate central thirst mechanisms through afferent baroreceptor input, with no role or only a permissive role for the renal RAS [6,15]. The studies of Rettig, Ganten, and Johnson [13] support the hypothesis that both renal RAS and nonrenal mechanisms are involved in isoproterenol-induced thirst. These authors found that a low dose of isoproterenol (25 ~ g / k g ) did not induce drinking in nephrectomized animals. However, a high dose of isoproterenol (330 p~g/kg) was effective at stimulating drinking after bilateral nephrectomy. Mean arterial pressure (MAP) in bilaterally nephrectomized animals given the low dose rapidly dropped to about 90 mmHg and then slowly rose across the 1-h test period. In contrast, pressure dropped to below 80 mmHg in response to the high dose and did not rise across the test period. They interpreted these findings to indicate that the RAS is the primary mediator of isoproterenol-induced drinking in intact animals but that the extreme depressor effect of high isoproterenol doses may involve extrarenal factors, such as arterial baroreceptors, in mediating drinking. The inability of RAS inhibition to abolish drinking to isoproterenol in some studies has led a number of investigators to question the role of A N G II in isoproterenol-induced drinking [4,16,17]. However, studies that use nephrectomy or peripheral RAS blockade are complicated by the fact that extrarenal factors may be brought into play due to the elimination of circulating ANG II vasoconstrictor actions, leading to greater depressor responses to isoproterenol [13]. Given that hypotension stimulates drinking in nephrectomized rats [13], greater falls in blood pressure to isoproterenol in the absence of the vasoconstrictor action of ANG II can influence the mechanisms that mediate drinking to isoproterenol. A recent study by Fitts [3] has circumvented the common difficulty associated with peripheral blockade studies to provide convincing evidence that the peripheral RAS is the primary mediator of drinking produced by low isoproterenol doses. Circulating A N G II stimulates drinking through the activation of A N G II receptors in the subfornical organ (SFO). Therefore, selective blockade of A N G II receptors in the SFO would eliminate the receptor population mediating the drinking response without blocking the receptors mediating the vasoconstriction. Fitts [3] demonstrated that either lesions of the SFO, or direct application of the A N G II receptor antagonist Sar ~, ThrS-ANG II, abolished or severely diminished drinking to isoproterenol (10-50 ~g/kg). However, drinking to isoproterenol was not
R.F. Kirby et al. / Brain Research 656 (1994) 79-84
diminished by blockade of the brain RAS by intraventricular Sar 1, Thr8-ANG II infusion at a dose 25 times greater than that used in the SFO. These results suggest that isoproterenol produces drinking primarily by activation of the peripheral RAS which stimulates ANG II receptors located outside the blood-brain barrier in the SFO. Isoproterenol administration mimics in many ways the hypovolemic situation that is proposed to elicit RAS-dependent thirst. Hypovolemia increases sympathetic drive to the kidney, increasing renin release, and as volume loss increases, hypotension contributes a second stimulus to increase RAS activity. In the present study, isoproterenol treatment increased activity of the RAS via both beta 1 adrenoceptor stimulation (increased sympathetic drive) and beta 2 adrenoceptormediated hypotension. Blockade of either beta adrenoceptor pathway significantly decreased renin release and drinking behavior, indicating that neither receptor pathway alone could account for all of the RAS stimulation or drinking after isoproterenol. Most importantly, the RAS activation and water intake produced by isoproterenol with both receptors active is greater than the individual contribution of beta~ and beta 2 adrenoceptors added together. These results indicate that sympathetic drive to the kidney and hypotension act synergistically to increase RAS activity and thereby, enhance drinking behavior. Two explanations support a synergistic effect of beta~ and beta 2 adrenoceptor activation by isoproterenol. First, Thames and DiBona [18] found that decreased blood pressure increased the release of renin to renal nerve stimulation. This indicates that the depressor response to isoproterenol may exaggerate the capacity of beta~ adrenoceptor to stimulate renin release. Second, unloading of low pressure/volume receptors increases the dipsogenic responsiveness to ANG II. Reductions in arterial pressure below resting levels increased the drinking response to ANG II administration [19]. The drinking response to isoproterenol could be facilitated, therefore, by reduced pressure increasing the amount of ANG II produced in the circulation or by increasing the sensitivity of central pathways related to the mediation of angiotensin-induced drinking behavior. The present results demonstrate that beta I and beta 2 adrenoceptor subtypes contribute to isoproterenol-induced drinking. The localization of these receptor subtypes and the control mechanisms for renin release support the idea that isoproterenol increases PRA via the direct activation of beta1 adrenoceptors on the juxtaglomerular apparatus and by stimulating vasodilatory beta 2 adrenoceptors located in the skeletal muscle vasculature that results in decreased renal perfusion pressure. The consistent reductions in PRA and drinking behavior following selective beta adrenoceptor an-
83
tagonism support the hypothesis that isoproterenol-induced drinking is mediated by the renal RAS. These results clarify the role of both neural and non-neural mechanisms, and their interactive nature, in the drinking elicited by isoproterenol.
Acknowledgements This research was supported by National Heart, Lung and Blood Institute, NIH Grants HL 14338 and HL 44546.
References [1] Evered, M.D. and Robinson, M.M., The renin-angiotensin system in drinking and cardiovascular responses to isoprenaline in the rat, J. Physiol., 316 (1981) 357-367. [2] Falk, J.L. and Tang, M., Salbutamol and quinterenol: dispogenic action produced by beta-adrenergic stimulants, Pharmacol. Biochem. Behav., 2 (1974) 413-415. [3] Fitts, D.A., Angiotensin II receptors in SFO but not in OVLT mediate isoproterenol-induced thirst, Am. J. Physiol., 267 (1994) RXX-RXX. [4] Fregly, M.J. and Rowland, N.E., Effect of DuP 753, a nonpeptide angiotensin II receptor antagonist, on the drinking responses to acutely administered dipsogenic agents in rats, P.S.E.B.M, 199 (1992) 158-164. [5] Fregly, M.J., Threatte, R.M., Barney, C.C. and Katovich, M.J., Effect of acute administration of isoproterenol and angiotensin II, seperately and in combination, on water intake and blood pressure in rats, Brain Res. Bull., 10 (1983) 327-332. [6] Hosutt, J.A., Rowland, N. and Stricker, E.M., Hypotension and thirst in rats after isoproterenol treatment, Physiol. Behav., 21 (1978) 593-598. [7] Houpt, K.A. and Epstein, A.N., The complete dependence of beta-adrenergic drinking on the renal dipsogen, Physiol. Behav., 7 (1971) 897-902. [8] Hubbard, J.I., Lin, N. and Sibbald, J.R., Subfornical organ lesions in rats abolish hyperdipsic effects of isoproterenol and serotonin, Brain Res. Bull., 23 (1989) 41-45. [9] Kirby, R.F., Woodworth, C.W., Woodworth, G.G. and Johnson, A.K., Beta-2 adrenoceptor mediated vasodilation: role in cardiovascular responses to acute stressors in spontaneously hypertensive rats, Clin. Exp. Hyper.-Theory Practice, A13 (1991) 10591068. [10] Leenen, F.H.H. and McDonald, R.H.J., Effect of isoproterenol on blood pressure, plasma renin activity, and water intake in rats, Eur. Z Pharmacol., 26 (1974) 129-135. [ll] Mann, J.F., Johnson, A.K. and Ganten, D., Plasma angiotensin II: dipsogenic levels and angiotensin-generating capacity of renin, Am. J. Physiol., 238 (1980) R372-R377. [12] Meyer, D.K. and Hertting, G., Drinking induced by direct or indirect stimulation of beta-receptors: evidence for involvement of the renin-angiotensin system. In G. Peters, J.T. Fitzsimons and L. Peters-Haefeli (Eds.), Control Mechanisms of Drinking, Springer-Verlag, New York, 1975, pp. 89-92. [13] Rettig, R., Ganten, D. and Johnson, A.K., Isoproterenol-induced thirst: renal and extrarenal mechanisms, Am. Z Physiol., 241 (1981) R152-R157. [14] Rolls, B.J. and Ramsey, D.J., The elevation of endogenous angiotensin and thirst in the dog. In G. Peters, J.T. Fitzsimons
84
R.F. Kirby et al. /Brain Research 656 (1994) 79 84
and L. Peters-Haefeli (Eds.), Control Mechanisms of Drinking, Springer-Verlag, New York, 1975, pp. 74-78. [15] Stricker, E.M., The renin/angiotensin system and thirst: a reevaluation. II. Drinking elicited in rats by caval ligation or isoproterenol, J. Comp. Physiol. Psych., 91 (1977) 1220-1231. [16] Szczepanska-Sadowska, E. and Fitszimons, J.T., The effects of angiotensin I1, renin and isoprenaline on drinking in the dog. In G. Peters, J.T. Fitzsimons and L. Peters-Haefeli (Eds.), Control Mechanisms of Drinking, Springer-Verlag, New York, 1975, pp. 69-73.
[17] Tang, M. and Falk, J.L., SarI-Ala s angiotensin II blocks reninangiotensin but not beta-adrenergic dispogens, Pharmacol. Biochem. Behat,., 2 (1974) 401-418. [18] Thames, M.D. and DiBona, G.F., Renal nerves modulate the secretion of renin mediated by nonneural mechanisms, Circ. Res., 44 (1979) 645-652. [19] Thunhorst, R.L. and Johnson, A.K., Effects of arterial pressure on drinking and urinary responses to intracerebroventricular angiotensin 11, Am. J. PhysioL, 264 (1993) R211-R217.