Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The initial subclassification of -adrenoceptors (-ARs) by Lands et al. (1967) into 1- and 2-subtypes permitted us to understand the effects of catecholamines on effectors during the last two decades. The existence of an atypical -adrenoceptor distinct from 1- and 2-AR, named 3-AR, has been demonstrated in some tissues by both pharmacological (Bojanic et al., 1985; Hollenga and Zaagsma, 1989; Mc Laughlin and MacDonald, 1990; Langin et al., 1991; Holloway et al., 1992; Tavernier et al., 1992; Berlan et al., 1993) and molecular approaches (Emorine et al., 1989; Granneman et al., 1991; Muzzin et al., 1991; Tate et al., 1991). Using various synthetic 3-AR agonists (Arch et al., 1984; Croci et al., 1991; Holloway et al., 1991), several groups have shown that these compounds are potent stimulators of metabolic and functional processes in various tissues. These effects cover the control of metabolic events in white and brown adipose tissues, i.e., stimulation of lipolysis and thermogenesis (Arch et al., 1984; Hollenga and Zaagsma, 1989; Langin et al., 1991; Arch and Kaumann, 1993; Berlan et al., 1993), inhibition of intestinal motility in rats, guinea pigs, and rabbits (Bianchetti and Manara, 1990; Mc Laughlin and MacDonald, 1990; Norman and Leathard, 1990; Arch and Kaumann, 1993), or stimulation of insulin secretion in pancreas -cells (Yoshida et al., 1991). In addition, 3-AR agonist administration can induce modifications of cardiovascular parameters in conscious dogs (Donckier et al., 2001). Namely, a positive chronotropic effect unrelated to a direct effect on the heart but due to a baroreflex mechanism (Tavernier et al., 1992) or to a direct central 3-AR effect was reported (Montastruc et al., 1999). Moreover, it was found that 3-AR agonists could exert a potent vasodilating action on some arterial and/or venous smooth muscles in specific vascular beds, which cause a reactional increase in heart rate (Berlan et al., 1993; Shen et al., 1994). Nevertheless, from a physiological point of view, the role of 3-ARs is currently poorly understood. They appear to exert a redundant action in target tissue, which also possess 1- and 2-ARs. In vitro studies have demonstrated that in rat (Granneman, 1992) and dog (Galitzky et al., 1993a) fat cells, catecholamines stimulate 3-AR at higher concentrations than those required to activate 1- or 2-ARs. The following schedule was performed in normal dogs to determine the "in vivo" conditions for 3-AR stimulation of various endocrino-metabolic and cardiovascular effects. For this purpose, we studied the pharmacological effects induced by a low and high dose of the nonselective -agonist isoproterenol with and without nadolol blockade, a potent 1- and 2-AR antagonist, which displays a low affinity for 3-AR (Galitzky et al., 1993b). In these conditions, the effects that were still observed under nadolol blockade were considered to involve only 3-AR activation, whereas others, masked by the -blockade, were regarded as 1- and/or 2-AR-dependent. To confirm this pharmacological hypothesis, we also evaluated the effects of nadolol plus SR59230A (a selective 3-AR antagonist) combination both on cardiovascular and metabolic effects induced by the high dose of isoproterenol. Finally, to confirm the real role played by the 3-adrenoceptor in these responses, we investigated the cardiovascular and metabolic effects of SR58611A, a drug known to be a selective 3-agonist (Manara and Bianchetti, 1990).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. A total of 23 normal male beagle dogs (weighing 11 to 14 kg) were used. Animals were anesthetized with an i.v. bolus of -chloralose (100 mg · kg¹ i.v at 9:00 AM), after 10 to 12 h fasting. General anesthesia is necessary to obtain stable and reproducible blood flow measurements. All animal procedures were performed in accordance with the official regulations of the French Ministry of Agriculture for Animal Experimentation and conforms with the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (publication no. 85-23, revised 1996).

General Procedure. After anesthesia, dogs were placed on a Pavlov table, and catheters were introduced into the jugular vein and antebrachial vein for blood sampling and drug infusion, respectively. Arterial blood pressure and heart rate were recorded by a catheter introduced in the femoral artery according to Sedlinger's method and connected to a Statham P23ID transducer (Gould, Inc., Cleveland, OH) on a Honeywell recorder (Honeywell Phillips, Corvallis, OR). Heart rate was obtained using a heart period meter triggered by blood pressure. Arterial blood pressure and heart rate were computed for each cycle and extracted at regular intervals of 0.5 s. Mean values of systolic and heart rate were calculated on 2-min periods. A Laser-Doppler Flowmeter (Periflux PF2B instrument; Perimed, Stockholm, Sweden) was used for blood flow measurement. This instrument measures vascular blood cell perfusion through cutaneous tissue. The probe holder was placed on the internal part of the abdominal skin before hand was shaved (Berlan et al., 1993). The results are presented in arbitrary units, which correspond to the displacement (centimeters) of the marker pen on the recorder (maximal amplitude, 20 cm). Jugular blood samples were collected in heparinized tubes and centrifuged immediately, and plasma was collected and frozen at 20°C until analysis. Plasma glucose and nonesterified fatty acids (NEFA) were determined by a glucose oxidase method (commercial kit; Biotrol, Paris, France) and an enzymatic method (commercial kit; Unipath, Dardilly, France), respectively. Glycerol plasma levels were determined by an ultrasensitive radiometric method in 10-µl fractions of plasma (Bradley and Kaslow, 1989).

Experimental Design. All experimental protocols were performed in the same randomized fashion. After a 10-min resting period and cardiovascular stabilization (i.e., 20-30 min after anesthesia induction), at T₀ an intravenous injection (saline or pretreatment) was performed. Ten minutes later, isoproterenol or SR58611A was infused during 20 min. Venous blood samples for biochemical determinations were performed at T₅ (i.e., basal value), T₊₁₀ (i.e., just before drug infusion), T₊₂₀, T₊₃₀, and T₊₄₀ (i.e., 10 min after the end of drug infusion). Cardiovascular parameters (mean arterial blood pressure, heart rate, and cutaneous blood flow) were continuously recorded and analyzed at T₅ (i.e., basal value), T₊₁₀ (i.e., pretreatment value), T₊₂₀, T₊₃₀, and T₊₄₀. In the first experimental protocol (n = 6), isoproterenol was infused at a low (0.4 nmol/kg/min) or high dose (4 nmol/kg/min) 10 min after a 0.9% NaCl (3 ml, i.v.) or nadolol (1 mg/kg, i.v.) injection. In the second protocol (n = 6), nadolol (1 mg/kg, i.v.) and SR59230A (1 mg/kg, i.v.) were injected as a bolus before isoproterenol infusion (4 nmol/kg/min). In the third protocol (n = 5), after a saline bolus (3 ml), SR58611A was infused at a dose of 10 nmol/kg/min. All doses of drugs were chosen in accordance with the conclusions of previous report in which dose responses toward lipid mobilization (Galitzky et al., 1993a) and peripheral cardiovascular effects (Montastruc et al., 1999) were performed in dogs. In each experimental protocol, changes in cardiovascular and endocrino-metabolic parameters were calculated using their own basal values.

Chemicals. ()-Isoproterenol bitartrate and nadolol came from Sigma-Aldrich (St. Louis, MO) and Squibb (Paris, France), respectively. SR58611A and SR59230A were kindly provided by Dr. Manara (Sanofi-Midy Group, Milano, Italy). Adenosine 5'-triphosphate, tetra-(triethylammoniun) salt, and -³²P (3000 Ci/mmol) came from DuPont de Nemours products (les Ulis, France). Other chemicals were reagent grade.

Statistical Analysis. All statistical comparisons were performed after examination of homoscedasticity. Pretreatment (saline or -AR antagonist) influences on basal values were analyzed with a paired Student's t test. In each protocol, cardiovascular and endocrino-metabolic changes induced by isoproterenol or SR58611A infusion were statistically evaluated using ANOVA for repeated measures and followed, when required, by a Dunnett's post hoc test using T₁₀ as the control group. Interprotocol differences were evaluated using a two-way ANOVA. All results are depicted as the mean ± S.E.M. Differences were considered significant when P was less than 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Pretreatment Effects on Cardiovascular and Endocrino-Metabolic Basal Values. Basal values of cardiovascular and endocrino-metabolic parameters did not significantly differ between the different experimental protocols. As depicted in Table 1, all -AR antagonist pretreatment significantly decreased heart rate. Mean arterial blood pressure was decreased by nadolol (1/2 antagonist) pretreatment, but remained unchanged by nadolol plus SR59230A (a selective 3 antagonist) pretreatment. Cutaneous blood flow, NEFA, and glycerol plasma levels were significantly increased by pretreatment with nadolol + SR59230A. Cutaneous blood flow, NEFA, and glycerol plasma levels were not significantly changed by nadolol pretreatment. Moreover, -AR antagonist pretreatments induced no changes in glucose plasma levels (Table 1).

View this table: [in this window] [in a new window]   TABLE 1 Basal values of cardiovascular and endocrino-metabolic parameters and changes induced by saline bolus (NaCl 0.9%; 3 ml, i.v.), nadolol (a 1- and 2-antagonist; 1 mg/kg, i.v.), and nadolol (1 mg/kg, i.v.) plus SR59230A (1 mg/kg, i.v.) All results are depicted as the mean ± S.E.M. Pretreatment (saline or -AR antagonist) influences on basal values were analyzed with a paired Student's t test.

Cardiovascular and Endocrino-Metabolic Effects of a Low Dose of Isoproterenol. Saline bolus failed to induce any significant change in cardiovascular and endocrino-metabolic parameters in dogs (Table 1). At the low dose, isoproterenol increased heart rate and decreased mean arterial blood pressure and cutaneous blood flow (Fig. 1). Isoproterenol at 0.4 nmol/kg/min also induced a significant increase in NEFA, glycerol, and glucose plasma levels (Fig. 2). As shown in Figs. 1 and 2, all cardiovascular and endocrino-metabolic significant effects induced by the low dose of nonselective -AR agonist were blocked by nadolol.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Changes in heart rate (beats per minute), mean arterial blood pressure (mm Hg), and cutaneous blood flow (arbitrary units) induced by isoproterenol infusion (0.4 nmol/kg/min) 10 min after saline bolus (3 ml, i.v.; closed columns) or nadolol (1 mg/kg i.v.) pretreatment (open columns) in dogs. Values are the mean ± S.E.M. *, indicates a significant difference (P < 0.05) using ANOVA for repeated measures followed, when required, by a Dunnett's post hoc test using T10 as control group. , indicates a significant difference (P < 0.05) between treatments evaluated by a two-way ANOVA.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Changes in nonesterified fatty acid (micromolar), glycerol (micromolar), and glucose (millimolar) plasma levels induced by isoproterenol infusion (0.4 nmol/kg/min) 10 min after saline bolus (3 ml, i.v.; closed columns) or nadolol (1 mg/kg i.v.) pretreatment (open columns) in dogs. Values are the mean ± S.E.M. *, indicates a significant difference (P < 0.05) using ANOVA for repeated measures followed, when required, by a Dunnett's post hoc test using T10 as control group. , indicates a significant difference (P < 0.05) between treatments evaluated by a two-way ANOVA.

Cardiovascular and Endocrino-Metabolic Effects of a High Dose of Isoproterenol. Under nadolol blockade, a high dose (4 nmol/kg/min) of isoproterenol induced a significant tachycardia associated with a decrease in mean arterial blood pressure and an increase in cutaneous blood flow (Fig. 3). Moreover, in these conditions NEFA and glycerol plasma levels were increased, but glucose plasma levels remained unchanged (Fig. 4). All modified parameters progressively returned to resting values shortly after discontinuation of the infusion. As shown in Figs. 3 and 4, all cardiovascular and endocrino-metabolic significant effects induced by the high dose of isoproterenol under nadolol pretreatment were blunted by the association nadolol plus SR59230A (i.e., 1-, 2-, and 3-AR antagonist).

View larger version (20K): [in this window] [in a new window]   Fig. 3.   Changes in heart rate (beats per minute), mean arterial blood pressure (mm Hg), and cutaneous blood flow (arbitrary units) induced by isoproterenol (4 nmol/kg/min) infusion 10 min after nadolol (1 mg/kg i.v.; open columns) or nadolol (1 mg/kg, i.v.) plus SR59230A (1 mg/kg, i.v.) pretreatment (hatched columns) and by SR58611A (10 nmol/kg/min) infusion 10 min after saline bolus (3 ml, i.v.; black columns) in dogs. Values are the mean ± S.E.M. *, indicates a significant difference (P < 0.05) using ANOVA for repeated measures followed, when required, by a Dunnett's post hoc test using T10 as control group.

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Changes in nonesterified fatty acid (micromolar), glycerol (micromolar), and glucose (millimolar) plasma levels induced by isoproterenol (4 nmol/kg/min) infusion 10 min after nadolol (1 mg/kg i.v.; open columns) or nadolol (1 mg/kg, i.v.) plus SR59230A (1 mg/kg, i.v.) pretreatment (hatched columns) and by SR58611A (10 nmol/kg/min) infusion 10 min after saline bolus (3 ml, i.v.; black columns) in dogs. Values are the mean ± S.E.M. *, indicates a significant difference (P < 0.05) using ANOVA for repeated measures followed, when required, by a Dunnett's post hoc test using T10 as control group.

Cardiovascular and Metabolic Effects of SR58611A Infusion. Before infusion, heart rate, mean arterial blood pressure, and cutaneous blood flow were 121 ± 9 bpm, 149 ± 5 mm Hg, and 2.8 ± 0.7 arbitrary units, respectively. Furthermore, basal plasma levels of NEFA, glycerol, and glucose were 285 ± 75 µM, 55.7 ± 8.5 µM, and 5.58 ± 0.31 mM, respectively. Except for glycemia, these basal values were not significantly different from those of the other experimental groups.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Using isoproterenol as a nonselective -adrenergic agonist, the present study evaluated the in vivo conditions for 3-AR stimulation. Under distinct pharmacological blockade, we have compared the doses of isoproterenol able to exert cardiovascular and various endocrino-metabolic effects known to be under 1-, 2-, and/or 3-AR control. Thus, we show that a high dose of isoproterenol is required to obtain in vivo vasodilation and lipolysis 3-AR dependence.

At the low dose of 0.4 nmol/kg/min, isoproterenol infusion induced both an increase in heart rate and a decrease in blood pressure in dog, although we observed a significantly slight decrease in cutaneous blood flow. As previously demonstrated, the decrease in arterial blood pressure induced by isoproterenol was mainly due to reduction of resistances in the skeletal, renal, or mesenteric vascular bed (Lundvall et al., 1981). In consequence, this vasodilation produces a decrease in arterial blood pressure and, through baroreflex activation, a positive chronotropic effect and a vasoconstriction in cutaneous blood vessels as shown in anesthetized rhesus monkey (Hom et al., 2001) and in dogs (Berlan et al., 1993). Nevertheless, the tachycardia directly and indirectly induced by isoproterenol remained insufficient to compensate for the decrease in systemic vascular resistance, thus explaining the fall in arterial blood pressure. After nadolol pretreatment, cardiovascular effects of the low isoproterenol dose were dramatically diminished or abolished. The choice of nadolol for 1- and 2-AR blockade was deduced from previous studies, which demonstrated that nadolol is a nonselective -antagonist displaying low antagonistic effects toward selective 3-AR in dog fat cells (Galitzky et al., 1993b). Thus, we show that the cardiovascular effects observed with the low isoproterenol dose are only 1- and/or 2-AR-dependent. Moreover, the lack of isoproterenol effect on heart rate after nadolol pretreatment indicated that 1- and 2-AR blockade was effective.

Under 1- and/or 2-AR blockade, a 10-fold higher dose (4 nmol/kg/min) of isoproterenol induced a cutaneous vasodilation and a decrease in blood pressure with a slight tachycardia. The remaining positive chronotropic effect of isoproterenol observed could be due to a baroreflex mechanism inducing a partial inhibition of the vagal tone (Tavernier et al., 1992). Inasmuch as an hypotensive effect persisted after 1- and 2-AR blockade, it was deduced that the high isoproterenol dose induces a vasodilation through a 1- and/or 2-vascular mechanism. In fact, it has been demonstrated that vasodilation induced by 3-AR stimulation occurs primarily in skin (Berlan et al., 1993; Shen et al., 1994). Using a radioactive microsphere technique, however, Shen et al. (1994) demonstrated that vasodilation induced by 3-AR stimulation was mainly achieved in cutaneous and adipose tissues. Different other tissues seem to be devoid of vascular 3-AR (Berlan et al., 1993). The recruitment of vascular 3-AR by the high isoproterenol dose was confirmed by its blockade by SR59230A, a selective 3-AR antagonist (De Ponti et al., 1996; Nisoli et al., 1996; Horinouchi and Koike, 2001). Previous study by our group has shown that SR59230A alone (at the used dose) induced a slight but not significant increase in heart rate, with no effect in mean arterial blood pressure. Moreover, SR59230A has no effect on plasma glucose levels. On the contrary, glycerol and NEFA plasma levels were significantly increased after SR59230A bolus; the action is not completely understood, but a recent report described a partial -agonist activity of SR59230A in the digestive tract (Horinouchi and Koike, 2001). Moreover, the pattern of cardiovascular effects produced by the high isoproterenol dose under nadolol pretreatment was similar to that observed with an infusion of SR58611A, a selective 3-AR agonist (Manara and Bianchetti, 1990). This compound is known to induce a peripheral 3-AR vasodilation (blunted by SR59230A; data not shown) and subsequently a tachycardia through a baroreflex activation (Tavernier et al., 1992; Berlan et al., 1993; Montastruc et al., 1999). Nevertheless, the long-acting effect of SR58611A, a chemical compound, could be explained by its pharmacokinetic properties, which remain unknown.

Reflected by NEFA and glycerol plasma levels, lipolytic effects observed with the low isoproterenol dose were blunted under 1- and 2-AR blockade. Even under nadolol pretreatment at a dose that prevents increase in heart rate (mainly 1-AR stimulation), the high dose of isoproterenol is still able to induce lipomobilization. Lipolytic effects of a high isoproterenol dose is blunted by a selective 3-AR antagonist and mimicked by SR58611A infusion. The coexistence of 3-AR with 1- and 2-AR in white fat cells from rodent and dog adipose tissue is now well documented (Arch and Kaumann, 1993; Galitzky et al., 1993a; Lafontan and Berlan, 1993). As well as for vascular 3-AR, our results show that the 3-AR-dependent lipolysis appears to be activated only through a high dose of catecholamines. "In vitro" studies have shown that the 3-AR subtype present on animal tissues (Granneman, 1992; Galitzky et al., 1993b) or in transfected cells (Marullo et al., 1989; Nantel et al., 1993) was activated by higher concentrations of catecholamines (isoproterenol, adrenaline, or noradrenaline) than those necessary for 1- or 2-AR stimulation. Thus, the vascular and/or adipocyte 3-AR stimulation, which was observed only with the high isoproterenol dose, could be explained by a lower affinity of this receptor for catecholamine than 1- or 2-AR. Another original observation from our study concerns glycogenolysis observed with the low isoproterenol doses, which were blunted under 1- and 2-AR blockade. After nadolol pretreatment, even the high dose of isoproterenol is not able to modify glycemia. Moreover, SR58611A failed to change basal glycemia. Previous studies from our group have established that a selective 2-AR agonist increases glycemia through a hepatic glycogenolysis (Taouis et al., 1989). This observation appears more consistent with the involvement of 2-AR rather than 3-AR in the hepatic glycogenolysis in dog. Thus, our results suggest that 3-AR is not involved in in vivo glycogenolysis.

The role of 3-AR in tissues also expressing 1- and/or 2-AR are not clearly understood. 3-ARs are mainly located in peripheral vascular beds (Shen et al., 1994; Trochu et al., 1999), in adipose tissue, or in the heart (Gauthier et al., 1998; Donckier et al., 2001). Our results are consistent with that; only a high isoproterenol dose is able to induce an in vivo vasodilation and lipolysis 3-AR dependence. Thus, from a physiological point of view, it could be hypothesized that 3-AR is a "back-up" receptor activated during extreme or stressful conditions. This hypothesis is consistent with recent studies that reported up-regulation of cardiac 3-ARs in the failing canine (Cheng et al., 2001) or human (Moniotte et al., 2001) myocardium.