Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Neuropharmacological and behavioral studies have previously demonstrated that besipirdine (HP 749) can stimulate adrenergic and serotonergic neurotransmission (Cornfeldt et al., 1990; Huger et al., 1990; Zaczek et al., 1993; Smith et al., 1994; Woods et al., 1995). These reports have shown besipirdine to antagonize alpha-2 adrenoceptors, inhibit both NE and 5HT uptake and stimulate [³H]NE release in vitro (Huger et al., 1990; Zaczek et al., 1993; Smith et al., 1994). The ability of besipirdine to activate central adrenergic and serotonergic function in vivo may underlie its efficacy in animal models of cognition, antidepression and obsessive-compulsive disorder (Santucci et al., 1991; Woods-Kettelberger et al., 1996; Smith et al., 1996).

Pharmacokinetic evaluation of besipirdine in rat, dog and man indicates that the parent compound is rapidly absorbed after oral administration and extensively metabolized via N-dealkylation to P7480, the primary metabolite (fig. 1) (Dileo et al., 1991; Hsu et al., 1991; Hubbard et al., 1991a, 1995). Pharmacologically, P7480 displays potent alpha-2 adrenoceptor antagonist and alpha-1 adrenoceptor agonist properties (Huger et al., 1990; Hubbard et al., 1991a, b). Therefore, it is possible that the behavioral efficacy of besipirdine in preclinical animal models could be partially mediated by the active metabolite P7480.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   The structure of besipirdine and metabolite P7480, which is formed by N-dealkylation (E).

Because both central and peripheral adrenergic neurons are involved in the control of cardiac function, vasomotor tone and arterial pressure regulation (McCall et al., 1982; McCall, 1988), it was important to determine whether besipirdine and its N-dealkylated metabolite P7480 possess cardiovascular effects in vitro and in vivo. The pharmacological studies described herein were designed to characterize the cardiovascular effects of besipirdine in rat and dog models and to ascertain the contribution of the parent and metabolite to the observed cardiovascular effects.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

In Vitro Assays

Radioligand binding assays. Male Wistar rats (175-300 g; Charles River Laboratories, Wilmington, MA) were used in all experiments. Radioligand binding of [³H]prazosin (70-75 Ci/mmol, Du Pont-NEN Corp., Boston, MA) to cortical alpha-1 adrenoceptors was performed according to Morrow and Creese (1986). Briefly, cerebral cortices were homogenized in 50 volumes of ice-cold 50 mM Tris buffer (pH 7.7) using a Tekmar homogenizer (setting 8) for 10 to 15 sec, then centrifuged at 48,000 × g for 10 min. The pellet was resuspended in fresh 50 mM Tris buffer and centrifuged a second time. After washing, the pellet was suspended in ice-cold 50 mM Tris buffer (pH 7.7) to a final tissue concentration of 3 to 5 mg/ml. Binding was determined by incubating 750 µl tissue homogenate, 150 µl [³H]prazosin (0.2 nM final concentration) and 100 µl of either buffer (total binding) or 10 µM phentolamine (nonspecific binding) or various concentrations of the test agents. After equilibration for 30 min at 30°C, samples were rapidly filtered through Whatman GF/B filters and washed three-times with 5 ml ice-cold 0.05 M Tris buffer. The filters were counted in 5 ml of Liquiscint scintillation fluid. Specific binding was defined as the difference between total binding and that displaced by 10 µM phentolamine. K_I values were calculated by nonlinear regression analysis to a one site model (GraphPad Software, San Diego, CA) using a K_d value of 0.12 nM for [³H]prazosin (derived from saturation analysis).

Preparation of rat aortic rings. Aortic rings from male, Long-Evans rats (300-350 g) were prepared as previously described (Vargas et al., 1993). After mounting, rings equilibrated at a basal tension of 2 g for 90 min in oxygenated Krebs buffer (37°C) before experiments commenced. Isometric tension (g) was measured with Grass (Charles River Laboratories, Wilmington, MA) FTO3C transducers coupled to a Beckmann R611 dynograph. All experiments were carried out on sets of four rings from the same aorta. One ring served as a time control to assess changes in vessel sensitivity associated with the duration of the experiment. Dose-response curves for NE, PE, besipirdine and P7480 were constructed in cumulative fashion. Graded concentrations of agonist were added sequentially at 5-min intervals or when the plateau response occurred. Two compounds were tested on each ring (e.g., NE or PE and besipirdine or P7480); the NE or PE curve was prepared first, followed by 60 min of fresh buffer washout. For normalization purposes, the maximum contraction (E_max) attained with NE was considered 100%. EC₅₀ values were calculated by fitting the data to a standard logistic equation using iterative nonlinear regression analysis (GraphPad Software).

Stimulated guinea pig ileum. Male Hartley guinea pigs (200-300 g, Charles River, Wilmington, MA) were fasted overnight. On the experimental day, were stunned with a blow to the head and killed by cervical dislocation. The ileum was located and cut approximately 15 cm proximal to the cecum. The tissue was placed in oxygenated Krebs buffer (37°C) and cut into four pieces 2-cm long. The luminal contents were gently flushed out of the tissue using a Pasteur pipette. Sutures were tied to each end of the tissue and it was transferred to a 20-ml bath containing oxygenated Krebs buffer at 37°C. One suture was secured to a stationary hook on platinum wire electrodes and the other suture was connected to a Grass FT03 force displacement transducer. The electrodes were connected to a Grass S88 stimulator via a constant current unit and stimulated at 0.1 Hz, 0.5 msec at a supramaximal current (40 mA). After the twitch height (developed tension) had plateaued, cumulative concentrations of clonidine (0.0043-13 µM) were added to the bath. After stable responses had been obtained, the clonidine was rinsed out of the bath and replaced with fresh buffer. The tissues were allowed to return to baseline developed tension and then either besipirdine or P7480 was added to the bath in a concentration of 0.1 µM. The tissues were incubated with the test compound for 30 min and then the clonidine dose-response was repeated. The responses are expressed as a percent of the maximal inhibition to clonidine. A linear regression analysis was used to determine the concentration of each drug that produced a 50% inhibition (IC₅₀) in developed tension. The effects of besipirdine and P7480 on clonidine-induced inhibition of developed tension were quantified by parallel line assessment (Tallarida and Murray, 1987).

Spontaneously contracting guinea pig atria. Male Hartley guinea pigs (200-300 g, Charles River) were stunned with a blow to the head and the hearts were rapidly excised and dissected in a petri dish containing oxygenated ice-cold Krebs solution. The ventricle was carefully trimmed away from the atria and a 4-0 silk ligature was attached to the apex of each atrium. The right atria were suspended in a 20 ml tissue bath containing oxygenated Krebs solution at 37°C. One ligature was attached to a force transducer (model FTO3; Grass Instruments, Quincy, MA) and the other was anchored to a glass support rod. The resting tension was adjusted to 1.0 to 1.5 gm and the tissues were allowed to equilibrate for 60 min until a stable atrial rate was obtained. Besipirdine and P7480 were added to the tissue bath every 15 min in cumulative dose ranges from 0.04 to 120 µM and 0.01 to 100 µM, respectively. For reference, isoproterenol and acetylcholine were also evaluated. Time control atria were incubated with vehicle throughout the course of the experiment. The analog output of the contractile force was summed by a cardiotachometer (type 9857B; Beckman Instruments) which continuously displayed the atrial rate.

[³H]NE release from rat cortical slices. The method for determining [³H]NE release from rat cortical slices has been previously described (Smith et al., 1994). Briefly, the cerebral cortices of male Wistar rats were cooled on ice, then coronally sliced (0.4 mm) with a McIlwain tissue chopper. After a 30-min preincubation period in oxygenated Krebs buffer, the slices were incubated for 30 min with buffer (35°C; pH 7.4) containing 25 nM [³H]NE (37 Ci/mmol). The slices were placed in glass superfusion chambers containing platinum electrodes and perfused (0.7 ml/min) in buffer for 60 min. Stimulation periods (5 Hz, 60 sec, 2-msec duration) occurred at 60-min intervals. Idazoxan (0.1 µM) and P7480 (0.1-3 µM) were each added to the chambers between stimulation periods. Compounds were introduced at fraction 14 (28 min after the first stimulation). The overflow of [³H]NE induced by electrical stimulation was calculated as the total increase in radioactivity above the resting outflow obtained in the sample immediately preceding the onset of the first stimulation (S1, performed in the absence of drug) and the second stimulation (S2, performed in the presence of drug). Control values were determined by calculating the S2/S1 ratio when both stimulations were performed in the absence of drug. Drug effects were determined by calculating the S2/S1 ratio when only the S2 was performed in the presence of drug.

In Vivo Assays

All surgical techniques and postoperative care of the animals used in these studies were in accordance with the National Institute of Health Guidelines for the Care and Use of Laboratory Animals (NIH, Department of Health and Human Services publication no. 85-23, revised 1985).

Pithed rats. Male Long Evans rats (350-425 g; Charles River) were pithed and instrumented for arterial pressure recording and i.v. injection as previously described (Vargas et al., 1994). Thirty min after pithing, the ability of i.v. PE (0.003-0.1 mg/kg), besipirdine (0.1-10 mg/kg) and P7480 (0.01-3 mg/kg) to increase diastolic pressure (mm Hg) was evaluated. In these studies, all pithed rats were treated with PE, then after a 30-min recovery period, either besipirdine or P7480. Because of its short duration of action, PE was administered in noncumulative fashion at 10- to 15-min intervals. Besipirdine and P7480 were administered in cumulative fashion at 3-fold increments and doses were administered at 1- to 2-min intervals or when the peak effect occurred. In some studies, prazosin (0.1-0.3 mg/kg, i.v.) was administered after the peak pressor response to P7480. In separate experiments, the pressor effects of PE, besipirdine and P7480 were evaluated in pithed rats pretreated with reserpine (5 mg/kg, i.p.; 24 hr). This dose of reserpine was previously shown to significantly inhibit (>80%) the pressor response to tyramine (0.1 and 0.3 mg/kg, i.v.). In the besipirdine groups, all rats were treated with SKF 525A (30 mg/kg, i.p.; 15 min), an inhibitor of hepatic cytochrome P450 enzyme function. This treatment was used to reduce the cardiovascular activity resulting from the enzymatic formation of P7480, an active metabolite of besipirdine found in the rat, dog, monkey and man (DiLeo et al., 1991; Hsu et al., 1991; Hubbard et al., 1995).

Conscious rats. Male Long Evans Rats (300-500 g; Charles River) were premedicated with atropine sulfate (0.1 mg/kg, s.c.) and anesthetized with sodium pentobarbital (50 mg/kg, i.p., Abbott, Chicago, IL). Polyvinyl-Teflon tipped catheters were implanted in the abdominal aorta and inferior vena cava via punctures in the femoral artery and vein, respectively. The catheters were secured and externalized on the top of the head. All animals were given 0.2 ml/sc of antibiotic (DiTrim, Syntex, Palo Alto, CA) and allowed to recover for 2 days before obtaining MAP and HR measurements. The arterial and venous catheters were flushed daily with 0.25 ml of heparinized saline (500 U/ml). On the day of the experiment, the catheter was attached to a miniature pressure transducer (model P10EZ, Spectramed, Oxnard, CA) and MAP and HR recordings were continuously displayed on a stripchart recorder (Beckmann R611 Dynograph). Data were collected for 1 hr before and 4 hr after administration of besipirdine (2, 4 and 10 mg/kg, p.o.) and P7480 (0.3, 3 and 10 mg/kg, p.o.). In some studies, animals were pretreated with a single dose of SKF 525A (30 mg/kg, i.p.; 15 min) as above.

Conscious dogs. Beagle dogs of either sex (10-15 kg) were sedated with xylazine (Rompun, 2.5 mg/kg, i.m., Miles Laboratories, Shawnee, KS) and acepromazine (PromAce, 1 mg/kg, i.m., Fort Dodge Laboratories, Fort Dodge, IA) and surgical anesthesia was maintained with sodium pentobarbital (20-30 mg/kg, i.v.). Using sterile surgical techniques, a small incision was made on the right lateral side of the neck and the omocervical artery was exposed and cannulated (Micro-renathane tubing, 1.02-1.27 mm). The catheter was attached to a specially fabricated stainless steel valve and externalized in the midscapular region. All incisions were closed and the dogs were placed in a temperature controlled recovery chamber for 24 hr. During the recovery period, each dog received a 1-ml injection (s.c.) of DiTrim antibiotic for 5 to 7 days. Body temperature and hematology were monitored weekly throughout the life of each dog. The arterial catheter was flushed three times per week with 2 ml of heparinized saline (500 U/ml).

Anesthetized dogs. Beagles of either sex (8-15 kg) were anesthetized with sodium pentothal (15 mg/kg, i.v.) sodium barbital (200 mg/kg, i.v.) and sodium pentobarbital (60 mg, i.v.). The trachea was isolated through a midline incision, intubated with a cuffed endotracheal tube and the dogs were artificially ventilated at 16 cycles/sec with a tidal volume of 15 to 20 ml/kg and arterial PO₂ and PCO₂ values of 90 to 100 mm Hg and 30 to 40 mm Hg, respectively (Harvard Respirator Pump, South Natick, MA). In specified treatment conditions, the right and left vagii were isolated and cut immediately below the carotid bifurcation. The femoral artery and vein were exposed and cannulated with polyethylene catheters for direct measurement of arterial pressure and administration of drugs, respectively. The arterial catheter was connected to a pressure transducer (model P23Gb, Statham) and the phasic arterial pressure was continuously monitored. The HR was derived from the pulsatile arterial pressure signal via a cardiotachometer (model 9875B, Beckman Instruments). A midline incision was made and a polyethylene catheter was inserted into the isolated duodenum above the pyloric sphincter for administration of besipirdine.

Drugs and Solutions

Idazoxan hydrochloride (Research Biochemicals, Inc., Natick, MA); acetylcholine chloride, prazosin hydrochloride, phentolamine hydrochloride, hexamethonium bromide, reserpine benzoate, desmethylimipramine hydrochloride, deoxycorticosterone acetate; isoproterenol bitartrate, norepinephrine tartrate, phenylephrine hydrochloride, (±)propranolol; Tween 80 (Sigma Chemical Co., St. Louis, MO); normetanephrine (Aldrich); SKF 525A (SmithKline & French, King of Prussia, PA). In all studies, fresh solutions of besipirdine hydrochloride (molecular weight = 287.7) and P7480 maleate (molecular weight = 325.3) were prepared daily and used within 60 min of preparation. These latter agents were synthesized at Hoechst Marion Roussel, Inc. (Bridgewater, NJ) according to the method described by Effland et al. (1990). For all in vitro studies, besipirdine and P7480 were dissolved in triple-distilled water containing 1% acetic acid. For oral and intraduodenal administration, all compounds were dissolved in triple-distilled water containing 1 or 2 drops of Tween 80 and administered in a volume of 5 to 10 ml/kg. All doses are expressed as mg of drug free base per kg of body weight. Krebs buffer contained (mM): NaCl, 118; KCl, 4.7; CaCl₂, 2.5; MgSO4, 1.2; KH₂PO₄, 1.2; NaHCO₃, 25; glucose, 11; disodium EDTA, 0.03; (±) propranolol, 0.001 (beta adrenoceptor antagonist); normetanephrine, 0.001 (catechol-O-methyltransferase inhibitor); desmethylimipramine, 0.0001; deoxycorticosterone acetate, 0.01 (neuronal and extraneuronal reuptake inhibitors, respectively) and continuously gassed with 95% O₂/5% CO₂ (37°C). In [³H]NE release studies, the antagonists (propanolol, normetanephrine) and uptake inhibitors (desmethylimipramine and deoxycorticosterone acetate) were not added to the Krebs buffer.

Statistical Analysis

The data are presented as mean ± S.E.M. Within and between treatment statistical comparisons were carried out using paired and unpaired Student's t test, respectively (Tallarida and Murray, 1987). The significance of treatment-related effects across time was assessed using a two-way analysis of variance with repeated measures (Winer, 1971). Significant main effects within a treatment group were analyzed by Dunnett's post hoc test for multiple comparisons with the pretreatment baseline (Tallarida and Murray, 1987). Results with P  0.05 were considered to be statistically significant.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

In Vitro Assays

Radioligand binding assays. Besipirdine and P7480 exhibited affinity for both alpha-1 and alpha-2 adrenergic receptors (table 1). Besipirdine and P7480 had approximately 76- and 3-fold lower affinity for alpha-2 adrenoceptors than idazoxan, respectively, e.g., K_I values of 380, 18 and 5 nM. Besipirdine and P7480 displayed 15- and 80-fold higher affinity for alpha-2 vs. alpha-1 adrenoceptors, respectively. The affinities of besipirdine and P7480 for cortical alpha-1 adrenoceptors were in the 1 to 5 µM range (table 1). Besipirdine and P7480 exhibited negligible affinity (IC₅₀ >10 µM) for beta-1/2 adrenergic, D₂-DA, 5-HT₂, 5-HT_1B receptors and monoamine oxidase A and B (data not shown). Besipirdine displayed low affinity for muscarinic (IC₅₀: 2 µM) and 5-HT_1A (IC₅₀: 17 µM) receptors. P7480 also had low affinity for the latter two receptors (muscarinic IC₅₀: 15 µM; 5-HT_1A IC₅₀: 3 µM).

Effect of NE, besipirdine and P7480 on rat aortic rings. In control aortic rings, NE and PE had EC₅₀ values of 0.044 ± 0.019 (N = 14) and 0.26 ± 0.07 µM (N = 4; fig. 2), respectively and their peak contractions (NE: 3.3 ± 0.4 g; PE: 3.0 ± 0.1 g) were equivalent. Besipirdine (0.01-100 µM) did not effect basal smooth muscle tension, but P7480 did produce aortic contraction (fig. 2). P7480 behaved as a partial agonist in this tissue since its peak contractile effect was 30% (1.1 ± 0.2 g; N = 14) of the NE E_max. P7480 (EC₅₀: 0.071 ± 0.050 µM; N = 14) had similar potency as NE and was three times more potent than PE.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Concentration-response effect of norepinephrine (NE), phenylephrine (PE), P7480 and besipirdine (HP 749) in rat aortic rings. The number of observations in each group was: NE, 18; PE, 4; P7480, 14; besipirdine, 4. Each point represents the mean and each bar the S.E.M.

Effect of besipirdine and P7480 on NE-induced aortic contractions in vitro. The affinity of besipirdine, P7480 and prazosin for aortic alpha-1 adrenoceptors was determined by Schild analysis (table 2). Besipirdine (5, 30 and 75 µM), P7480 (0.1, 1, and 10 µM) and prazosin (3, 10 and 100 nM) antagonized NE-induced contractions in a concentration-dependent manner. The parameters of the analysis indicated that all three compounds competitively antagonized aortic alpha-1 adrenoceptors in vitro because the slopes of the plots were near unity (table 2). Comparison of the K_b values for besipirdine and P7480 indicated that these agents had 2800- and 80-fold lower affinity for alpha-1 adrenoceptors than prazosin, respectively (table 3).

Stimulated guinea pig ileum. Clonidine caused a concentration-related decrease in the developed tension of the electrically stimulated guinea pig ileum (IC₅₀: 25 ± 0.8 nM, N = 8; fig. 3). Besipirdine (0.1 µM) did not significantly inhibit clonidine's effect on the ileum. In contrast, P7480 (0.1 µM) caused a significant 89-fold parallel shift to the right in the clonidine dose-response curve. The IC₅₀ for clonidine in the presence of P7480 was 2.23 µM.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Percent inhibition of developed tension in isolated guinea pig ileum (n = 4 per treatment condition) after ascending concentrations of clonidine alone or plus either besipirdine (HP 749; 0.1 µM; left) or P7480 (0.1 µM; right). All points are mean ± S.E.M. Clonidine produced significant (P < .05) inhibition of developed tension between 0.01 and 30 µM. The IC50 for clonidine was 25 nM. Pretreatment with P7480 (0.1 µM) shifted the IC50 for clonidine to 2.23 µM.

Spontaneously contracting guinea pig atria. The basal beating rate of the isolated right atria was 125 ± 9 bpm (N = 14). Figure 4 shows the direct effect of ascending concentrations of besipirdine and P7480 on the intrinsic rate of spontaneously contracting guinea pig atria. Neither compound significantly altered spontaneous rate at concentrations less than 100 µM. However, a significant 30 ± 11% reduction in atrial rate (P < .05) was seen at the 120 µM concentration of besipirdine. Vehicle treatment did not significantly alter intrinsic atrial rate throughout the course of the experiment. For reference purposes, the beta-1 adrenoceptor agonist isoproterenol (0.01-100 µM) increased atrial rate by 162 ± 5 bpm (at 10 µM) and had an EC₅₀ of 0.22 ± 0.06 µM (N = 6). In contrast, the cholinergic agonist acetylcholine (0.05-500 µM) reduced atrial rate by 83 ± 9 bpm (at 50 µM) and had an EC₅₀ of 9.1 ± 2.7 µM (N = 4).

View larger version (18K): [in this window] [in a new window]   Fig. 4.   Mean ± S.E.M. percent change in the rate of contraction of isolated guinea atria following ascending concentrations of either besipirdine, P7480, or vehicle. N = four to six per treatment condition. * P < .05 as compared to the vehicle control group.

[³H]NE release from rat cortical slices. Table 3 shows the effects of idazoxan and P7480 on the release of [³H]NE. Idazoxan and P7480 each significantly (P < .05) increased the S2/S1 ratio. P7480 showed concentration related increases from 0.1 to 3 µM, with a peak increase of approximately 122% as compared to control. Significant effects of P7480 were also seen at 0.3 and 1 µM concentrations.

In Vivo Assays

Pithed rat. Intravenous administration of PE, P7480 and besipirdine caused dose-dependent elevations of diastolic pressure in the pithed rat (fig. 5). In control rats, the ED₅₀ and E_max values for i.v. PE were 0.037 ± 0.006 µmol/kg and 119 ± 6 mm Hg (N = 8), respectively. P7480 was three times less potent than PE (0.090 ± 0.007 µmol/kg, i.v.; N = 4), but produced similar maximal pressor responses as PE (118 ± 7 mm Hg, N = 4; P = .46). The pressor effects of PE and P7480 were unaffected by pretreatment with reserpine (fig. 5). The P7480 pressor response was mediated by postsynaptic alpha-1 adrenoceptor stimulation since prazosin (0.1-0.3 mg/kg, i.v.), administered immediately after the highest dose of P7480, rapidly reversed the elevated arterial pressure to basal values (control diastolic pressure before P7480: 32 ± 4 mm Hg; diastolic pressure after P7480/prazosin: 53 ± 6 mm Hg; N = 6). In contrast with PE and P7480, the pressor response to i.v. besipirdine was less marked (E_max: 21 ± 4 mm Hg; N = 4) and occurred at a higher dose range (1-34 µmol/kg, i.v.; fig. 5). The pressor effect of besipirdine was significantly antagonized by reserpine pretreatment.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Pressor effect of i.v. phenylephrine (PE), P7480 and besipirdine (HP 749) in control or reserpinized pithed rats. The mg/kg doses were as follows: PE: 0.003, 0.01, 0.03, 0.1; P7480: 0.01, 0.03, 0.1, 0.3, 1, 3; besipirdine: 0.1, 0.3, 1, 3, 10. There were four animals in each group, with the exception of the PE control and reserpine groups, which had eight. Each point represents the mean and each bar the S.E.M. * P < .05, unpaired t test.

Conscious rat. Both besipirdine and P7480 caused vasopressor responses and bradycardia upon oral administration to conscious rats. Besipirdine caused a significant decrease in HR at both 2 and 10 mg/kg, p.o. doses, with decreases of 42 ± 10 and 74 ± 19 bpm, respectively (table 4). A significant 44 ± 5 mm Hg (P < .05) increase in MAP was seen after the 10 mg/kg dose of besipirdine. In contrast, P7480 caused MAP to increase significantly at both 3- and 10-mg/kg doses. At 3 mg/kg, MAP was increased by 46 ± 3 mm Hg and HR was decreased by 108 ± 16 bpm. At 10 mg/kg, MAP was increased by 67 ± 7 mm Hg and HR to decreased by 100 ± 40 bpm at 15 min postdosing (table 5). Based on these results, P7480 was approximately 3-fold more potent as a vasopressor when compared to besipirdine.

View larger version (22K): [in this window] [in a new window]   Fig. 6.   Change in mean arterial pressure (MAP) and heart rate (HR) in conscious rats after oral administration of either besipirdine (HP 749; 10 mg/kg, p.o.), SKF 525A (30 mg/kg, i.p.) or pretreatment with SKF 525A 15 min before the same dose of besipirdine. * P < .05 as compared to the pretreatment baseline or across treatment conditions. N = 4 in each group.

View larger version (23K): [in this window] [in a new window]   Fig. 7.   Change in mean arterial pressure (MAP) and heart rate (HR) in conscious rats after oral administration of either P7480 (10 mg/kg, p.o.), SKF 525A (30 mg/kg, i.p.) or pretreatment with SKF 525A 15 min before the same dose of P7480. * P < .05 as compared to the pretreatment baseline or across treatment conditions. N = 4 in each group.

Conscious dog. Besipirdine (0.3, 1 and 2 mg/kg, p.o.) caused a significant dose-related pressor response from 15 to 180 min (P < .05) with a peak pressor response of 35 ± 7 mm Hg at 30 to 60 min postdosing (table 6). A significant bradycardia (43 bpm; P < .05) was also observed from 30 to 240 min postdosing (table 6). Pretreatment of the dogs with prazosin (3 mg/kg, p.o.) completely antagonized the pressor effect of besipirdine, but did not alter the bradycardia caused by the compound (fig. 8).

View larger version (25K): [in this window] [in a new window]   Fig. 8.   Change in mean arterial pressure (MAP) and heart rate (HR) in conscious dogs (N = 3) after administration of either besipirdine (HP 749; 2 mg/kg, p.o.) or pretreatment with prazosin (3 mg/kg, p.o.) followed 30 min later by the same dose of besipirdine. * P < .05 as compared to the pretreatment baseline. All points represent the mean ± S.E.M.

View larger version (21K): [in this window] [in a new window]   Fig. 9.   Change in mean arterial pressure (MAP) and heart rate (HR) in conscious dogs (N = 3) after administration of either besipirdine (HP 749; 2 mg/kg, p.o.), hexamethonium (10 mg/kg, i.v.) or pretreatment with hexamethonium followed 15 min later by the same dose of besipirdine. * P < .05 as compared to the pretreatment baseline or across treatment conditions. All points represent the mean ± S.E.M.

Anesthetized dog. Besipirdine (2 mg/kg, i.d.) caused a significant 43 ± 8 bpm peak decrease in HR and 31 ± 17 mm Hg increase in MAP from 30 to 60 min postdose (fig. 10). Bilateral vagotomy and pretreatment with atropine did not significantly alter the response to besipirdine. In contrast, pretreatment with atenolol (beta-1 adrenergic blockade) completely blocked the bradycardia (9 ± 3 bpm) to a level that was not significantly different from the nontreated control group (10 ± 2), but did not alter the pressor response to besipirdine. Combined cholinergic and beta-1 adrenergic blockade was not significantly different from beta-1 adrenergic blockade alone. Thus, the reduction in HR caused by besipirdine appears to be due to a withdrawal of sympathetic drive to heart.

View larger version (27K): [in this window] [in a new window]   Fig. 10.   The effects of besipirdine (2 mg/kg, i.d.) on the heart rate (HR; upper) and mean arterial pressure (MAP; lower) of anesthetized dogs with no pretreatment or after sequential pharmacological and surgical autonomic blockade. Data are shown as mean ± S.E.M. with N = 3 per treatment condition. * P < .05 as compared to the saline control treatment.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

We have previously demonstrated that besipirdine and its N-despropyl metabolite P7480 produce marked cardiovascular effects in the conscious monkey (Hubbard et al., 1991b, Hubbard et al., 1995). In that study, i.v. (10 mg/kg) and p.o. (10, 20 and 40 mg/kg) administration of besipirdine caused a significant pressor effect minutes after dosing. A significant tachycardia resulted from i.v. administration, but not after p.o. administration of the drug. An i.v. bolus of P7480 (0.1 mg/kg) also caused a rapid increase in arterial pressure and a reflex decrease in heart rate. Although this study clearly showed that besipirdine and P7480 cause a sympatho-excitatory (pressor) response, the mechanism of action of these drugs and their central or peripheral locus of action was not determined. Our studies were designed to pharmacologically characterize the adrenergic effects of besipirdine and P7480 and determine the site(s) of drug action underlying the pressor and bradycardic effects of these agents in reflex-intact conscious rats and dogs. Mechanistic studies were also performed in anesthetized dogs, reflex-compromised pithed rats and isolated tissues.

As in the monkey, oral or intraduodenal besipirdine consistently elevated arterial pressure in conscious rats and dogs and anesthetized dogs. In awake rats, a pressor effect was clearly evident at the 10 mg/kg dose, although hypertensive effects were observed at 1 and 2 mg/kg in conscious and anesthetized dogs. The pressor effect of besipirdine were mediated pharmacologically by the activation of peripheral vascular alpha-1 adrenoceptors since pretreatment with prazosin, but not hexamethonium (ganglion blocker), abolished besipirdine's ability to increase arterial pressure in conscious dogs. The hypertensive effect of besipirdine appears to be causally linked to the formation of the vasopressor metabolite P7480, because pretreatment with the hepatic metabolic enzyme inhibitor SKF 525A prevented besipirdine's hypertensive effect in the rat. We focussed on the pressor metabolite P7480 for two reasons. First, this agent is the primary metabolite found in rat, dog, monkey and human plasma after oral besipirdine administration (DiLeo et al., 1991; Hsu et al., 1991; Hubbard et al., 1995). Second, P7480 exhibited potent activity as an alpha-1 adrenergic agonist and alpha-2 adrenergic antagonist (see below).

Previous functional studies on besipirdine indicated that this compound directly facilitated the release of [³H]NE through the inhibition of neuronal NE reuptake and the blockade of presynaptic alpha-2 adrenoceptors in rat cortical slices in vitro (Smith et al., 1994). As with the parent compound, the metabolite P7480 significantly increased [³H]NE release from preloaded rat cortical slices. However, the maximal stimulatory effect of P7480 on NE release was smaller (122%) than that induced by besipirdine (534%; Smith et al., 1994). This quantitative difference suggests that P7480 facilitates NE release primarily through presynaptic alpha-2 adrenoceptor antagonism, especially because this agent is a weak inhibitor of [³H]NE uptake (IC₅₀ = 2500 nM; Klein et al., 1996) compared to besipirdine (IC₅₀ = 560 nM; Smith et al., 1994). However, besipirdine increases electrically-stimulated [³H]NE release via a combination of NE uptake inhibition and alpha-2 adrenoceptor antagonism (Smith et al., 1994). In whole animals, this combination of properties results in higher extracellular NE levels than alpha-2 adrenoceptor antagonism alone (Dennis et al., 1987).

In our study, P7480 displaced [³H]clonidine from rat cortical alpha-2 adrenoceptors with higher affinity (K_I, 18 nM) than besipirdine (K_I, 380 nM). P7480 also showed greater potency than besipirdine as an antagonist of prejunctional alpha-2 adrenoceptors in the guinea pig ileum. In that assay, P7480 (0.1 µM) competitively blocked the neuroinhibitory effect of clonidine on electrically stimulated acetylcholine release and caused a significant shift to the right in the clonidine concentration-response curve. In contrast, besipirdine displayed comparatively weak presynaptic antagonist activity in this test. Together, the finding that besipirdine and metabolite P7480 antagonize presynaptic alpha-2 adrenoceptors and enhance presynaptic NE release suggests that this pharmacological action may underlie or contribute to the cardiovascular effects observed in animal models.

As with idazoxan, besipirdine and P7480 showed lower affinity (K_I: 1-5 µM) for cortical alpha-1 adrenoceptors and had a high degree of selectivity for alpha-2 adrenoceptors (154-, 15- and 83-fold, respectively). Contraction studies in the rat aorta indicated that besipirdine was devoid of intrinsic activity in this tissue and that P7480 was a partial alpha-1 adrenoceptor agonist. Contractions induced by P7480 were mediated by alpha-1 adrenoceptors because they were sensitive to blockade by a low concentration (3 nM) of prazosin. Similarly, the affinity constant (K_b) of prazosin for aortic alpha-1 adrenoceptors was similar when determined with either NE (0.4 nM) or P7480 (0.23 nM) as the agonist, confirming that NE and P7480 interacted with alpha-1 adrenoceptors. These values are equivalent with prior affinity estimates of prazosin functionally derived in this tissue (Aboud et al., 1993; Mir and Fozard, 1990). The partial agonist properties of P7480 were demonstrated by its ability to antagonize NE-induced aortic contractions.

It is interesting to point out that the affinity estimate of besipirdine for alpha-1 adrenoceptors as determined by competition binding with [³H]prazosin (K_I, 5.6 µM) was close to its affinity determined functionally in the rat aorta (K_b, 1.7 µM). However, P7480 demonstrated a 30-fold higher affinity for aortic (K_b, 0.048 µM) than cortical alpha-1 adrenoceptors (K_I, 1.5 µM). The higher affinity of P7480 for aortic alpha-1 adrenoceptors suggests that this compound may preferentially bind alpha-1D adrenoceptors, the subtype primarily found in rat aorta (see Vargas and Gorman, 1995), whereas the rat cortex contains approximately equal proportions of alpha-1A and alpha-1B adrenoceptor subtypes based on radioligand binding (Morrow and Creese, 1986). Further binding studies on the cloned alpha-1 adrenoceptor subtypes will be necessary to examine this subtype preference in detail. Because P7480 is a novel alpha-1 adrenoceptor agonist unrelated chemically to the phenethylamines and imidazolines (DeMarinis et al., 1987), its intrinsic efficacy at each alpha-1 adrenoceptor subtype also needs to be fully assessed.

In the pithed rat model, P7480 behaved as an agonist at vascular alpha-1 adrenoceptors and was slightly less potent than PE. The peak pressor effect (E_max) of P7480 was equivalent with the peak PE response, a finding that indicates that this new agent behaved as a full alpha-1 adrenoceptor agonist in vivo. The expression of full agonist activity in the pithed rat suggests that vascular alpha-1 adrenoceptors were able to maximally transduce the stimulus triggered by the partial agonist. Attainment of the full pressor response could be related to the occupation of a greater fraction of the vascular alpha-1 adrenoceptor pool (vs. the isolated aorta) or possibly, the receptor-mediated activation of both intracellular and extracellular calcium pathways (Timmermans and Thoolen, 1987). Nonetheless, the dose-dependent pressor response caused by i.v. P7480 in the pithed rat or oral administration in conscious rats corroborates the observation that i.v. P7480 elevated arterial pressure in conscious monkeys (Hubbard et al., 1995). As in the aorta, the vasoconstrictor effects of P7480 in the pithed rat were mediated by vascular alpha-1 adrenoceptor activation because it was sensitive to prazosin blockade. The dose-response curves for P7480 and PE in reserpinized pithed rats were identical to responses in untreated rats, further proof that these agents selectively stimulated postsynaptic alpha-1 adrenoceptors and that neuronal NE release does not contribute to their pressor effects in vivo. Intravenous besipirdine did produce a weak pressor effect in pithed rats that was abolished by reserpinization, indicating that the indirect sympathomimetic activity of the parent compound was mediated by NE release from sympathetic nerve terminals.

Although various pharmacological studies have shown that besipirdine has activity on multiple neurotransmitter systems (Cornfeldt et al., 1990; Huger et al., 1990; Smith et al., 1994; Woods-Kettelberger et al., 1996), the results of our study demonstrate that the cardiovascular effects in rats and dogs are predominantly due to the adrenergic properties of besipirdine and the major metabolite P7480 (DiLeo et al., 1991; Hsu et al., 1991). We were unable to demonstrate a direct cholinergic effect of the compound on the heart, e.g., did not reduce atrial beating, or vasculature when assessed both in vitro and in vivo. A high concentration of besipirdine (120 µM) was found to slightly lower atrial rate, but the mechanism for that direct effect was not characterized.

It should be mentioned that besipirdine lowered heart rate in conscious rats (2 and 4 mg/kg) and conscious and anesthetized dogs (2 mg/kg). In the rat, the bradycardia observed at low doses was not a reflex response since arterial pressure was not elevated. Furthermore, besipirdine-induced bradycardia was still evident in rats after pretreatment with SKF 525A (fig. 6) and dogs pretreated with prazosin (fig. 7). These observations suggest that besipirdine-induced bradycardia may be a distinct central action of the compound, possibly the withdrawal of cardiac sympathetic tone. This conclusion is based on autonomic blockade experiments in the anesthetized dog that showed that besipirdine-induced bradycardia was inhibited by beta adrenoceptor blockade and was not altered by vagotomy (plus muscarinic receptor antagonism). However, in conscious monkeys bradycardia was not observed due to the strong sympatho-excitatory effect at a high i.v. dose of 10 mg/kg (Hubbard et al., 1995). Therefore, we speculate that besipirdine may reduce cardiac sympathetic nerve tone through the activation of central noradrenergic function resulting from the inhibition of NE reuptake and alpha-2 adrenoceptor blockade. For example, systemic administration of the NE uptake inhibitor desipramine reduced renal and splanchnic sympathetic nerve activity through the depression of locus ceruleus nerve cell firing (Svensson and Usdin, 1978; Lavian et al., 1991). It seems plausible that besipirdine may have a similar action in the central nervous system, but direct brain microinjection studies are needed to verify this possibility. A central action of besipirdine is supported by recent microdialysis evidence that showed that this agent (2.5 mg/kg) produced a 3-fold increase in synaptic NE levels in the hippocampus of the rat (Smith et al., 1996). In contrast to besipirdine, the metabolite P7480 (i.v. or oral administration) appears to act directly on vascular smooth muscle as an alpha-1 adrenoceptor agonist and cause a concomitant reflex decrease in HR.

In conclusion, it appears that the primary cardiovascular effects of besipirdine in whole animals are bradycardia and hypertension. These effects are likely due to a central sympatho-inhibitory effects of the parent compound on the cardiac sympathetic nerve tone and alpha-1 adrenergic agonist activity of P7480 on vascular smooth muscle, respectively. The effects of besipirdine on other transmitter systems do not appear to contribute to these cardiovascular responses at the pharmacological doses examined in our study.