Introduction

Top Abstract Introduction Methods Results Discussion References

The human prostatic smooth muscle contains high densities of alpha-1-adrenoceptors (Muramatsu et al., 1994; Hieble et al., 1985; James et al., 1989) and several alpha-1-adrenoceptor subtypes have been identified and their heterogeneity revealed both pharmacologically and by molecular cloning (Michel et al., 1993; Price et al., 1993). The three alpha-1-adrenoceptor subtypes with high affinity for prazosin, so far identified, i.e., alpha-1A, alpha-1B and alpha-1D-adrenoceptors, have been cloned (alpha-1a, alpha-1b and alpha-1d). The pharmacology of the cloned receptors has been characterized against that of their native forms (Michel and Insel, 1994; Blue et al., 1995). The alpha-1a-adrenoceptor subtype has been described to be predominant in the human prostate, either using molecular biology techniques (Price et al., 1993; Faure et al., 1994; Forray et al., 1994) or autoradiography (Kobayashi et al., 1993). However, whether this adrenoceptor subtype plays a prevalent role in mediating contractile responses of the prostatic tissue remains a controversial issue. Initial in vitro studies on the efficacy of various alpha-1-adrenoceptor antagonists at inhibiting noradrenaline or phenylephrine-mediated contraction suggested that contractile potency is well correlated with binding affinities for the alpha-1a subtype (Forray et al., 1994). Nevertheless, recent data using more selective alpha-1a-adrenoceptor antagonists suggest that the alpha-1-adrenoceptor subtype that mediates prostatic contraction does not have all the pharmacological characteristics of the alpha-1A-adrenoceptor (Ford et al., 1996; Kenny et al., 1996, Van Der Graaf et al., 1996).

In recent years, attempts have been made to evaluate the selective effects of alpha-1-adrenoceptor antagonists on prostatic tissue compared to side effects, most particularly the effects on blood pressure. Only a limited number of in vivo methodologies are available to specifically evaluate the mechanisms that explain the preferential action of some alpha-1-adrenoceptor antagonists at the prostatic level (Kenny et al., 1994; Lefèvre-Borg et al., 1992, 1993) and all reported results appeared to depend on the species and experimental conditions. Use of exogenous stimulation or interspecies comparisons lead to difficulties in analyzing the respective action of the tested drugs on the vasculature and on the urinary tract. We have recently developed a new model that allows the simultaneous measurement of urethral and arterial pressures in conscious male rats (Martin et al., 1995) therefore enabling a direct estimation of functional uroselectivity.

The therapeutical value of alpha-1-adrenoceptor antagonists in the treatment of BPH is well established (Jonler et al., 1994; Eri et al., 1995) and recent development in the identification and distribution of alpha-1-adrenoceptors have revitalized the question of adrenoceptor subtype selectivity and functional uroselectivity of available drugs. Our study was undertaken to assess the functional uroselectivity of several alpha-1-adrenoceptor antagonists in the conscious male rat under normal adrenergic tone and to relate this uroselectivity with their affinities for the cloned alpha-1-adrenoceptor subtypes and with their efficacy to antagonize phenylephrine-induced contractions of prostatic tissue in vitro.

	Methods

Top Abstract Introduction Methods Results Discussion References

Receptor Binding Assays

Cell culture and membrane preparations. HeLa cell-lines stably expressing the bovine brain alpha-1-adrenoceptor, the hamster smooth muscle alpha-1b-adrenoceptor and a rat fibroblast cell-line stably expressing the rat cerebral cortex alpha-1d-adrenoceptor were purchased from Tulco (Durham, NC). Transfected cells were cultured in monolayers in Dubelco modified medium with 4500 mg/liter glucose supplemented with 10% fetal calf serum, penicillin (50 U/ml), streptomycin (50 U/ml) and geniticin (400 µg/ml). Cells were harvested in phosphate-buffered saline, collected and centrifuged at 140 × g for 5 min. Subsequently, the cells were homogenized in 5 mM Tris-HCl buffer, pH 7.4, containing 5 mM EDTA and centrifuged at 46,000 × g for 20 min. The pellets were resuspended in 50 mM Tris-HCl buffer, pH 7.4, recentrifuged, pooled in buffer and frozen at 80°C until binding studies.

Radioligands binding assays. Ligand binding studies were performed in 50 mM Tris-HCl buffer, pH 7.4. Cell membrane preparations (4 µg protein) were incubated with [³H] prazosin at 25°C for 1 hr in a total volume of 2 ml. Parameters of [³H] prazosin binding have been presented elsewhere (Faure et al., 1994). Phentolamine (10 µM) was used to define nonspecific binding. Incubations were terminated by rapid filtration using a Brandel Cell Harvester (Neurolab, Paris, France) through Whatman GF/B glass fiber filters. The filters were washed with 2 × 5 ml of buffer, dried and the radioactivity was measured in a liquid scintillation counter.

Rabbit Isolated Prostate Preparation

Male New Zealand rabbits (~18 wk; ESD, Chatillon Sur Chalaroyne, France) were killed by cervical dislocation and exsanguination and the prostate was removed and placed in modified Krebs-Henseleit solution (KHS) of the following composition (mM): NaCl:114.0, NaHCO₃:25.0, KCl:4.7,KH₂PO₄:1.2, MgSO₄:1.2, glucose:11.7 and ascorbic acid:1:1. Tissues were cleared of surrounding adipose tissue, cut in strips of ~2 cm and mounted between two stainless-steel hooks in 20 ml organ baths containing modified KHS (gassed with 95% O₂ and 5% CO₂ and thermostatically controlled at 37°C). Tissue responses were measured continuously using Grass FT03 isometric transducers and displayed on a Grass 7D polygraph (Quincy, MA).

After application of 1 g resting tension, tissues were allowed to equilibrate for 90 min during which time the organ bath fluid was replaced regularly. Subsequently, tissues were exposed to two consecutive concentrations of noradrenaline (30 µM), separated by 60-min washout periods. After an additional 60-min washout period, a first E/[A] curve was constructed by cumulative additions as half-log unit concentration increments. Tissues were then washed for 60 min and incubated for 30 min with antagonist or vehicle before a second phenylephrine E/[A] curve was done. Only one concentration of antagonist was studied in each tissue. Propranolol (1 µM) was present in all experiments to block -adrenoceptors.

Individual phenylephrine E/[A] curves were fitted to the Hill equation using the Allfit program to provide estimates of midpoint location (EC₅₀). Each pair of EC₅₀ values obtained in the absence and presence of antagonist was then used to calculate the concentration ratio (EC₅₀ in the presence of antagonist divided by the EC₅₀ of the control curve) and antagonist dissociation equilibrium constants (K_B, represented as the negative logarithm, that is pK_B) were calculated according to the method of Schild (Jenkinson, 1991).

In Vivo Methods

Simultaneous assessment of blood pressure and urethral pressure was undertaken as recently described (Martin et al., 1995). Briefly, male Wistar rats (350-400 g, Charles River, Saint Aubin, France) were anesthetised with pentobarbital (40 mg/kg, i.p.). A polyethylene catheter was inserted into the abdominal aorta via a femoral artery and a venous catheter placed into a jugular vein for drug injection. Through a laparotomy, the bladder was exposed and, via an incision in the bladder wall, another polyethylene catheter was positioned into the urethra below the bladder neck and held in place by a suture on the bladder wall. Such a positioning of the urethral catheter does not interfere with normal micturition (Martin et al., 1995). The animals were allowed to recover for 48 hr before drug testing. At the day of the experiments, the animals were partially restrained in a restraining device and mean arterial BP, HR and UP were continuously monitored.

A control period of at least 20 min was established before i.v. drug injection. Seven experimental groups were randomly assigned for vehicle or drug treatment (n = 4-7 per group). Compounds were given as a single dose infused i.v. over 5 min and effects were analyzed at 5, 15, 30 and 60 min. Pressure changes were expressed as percentage changes from base-line values and statistical significance assessed by a one-way analysis of variance.

Dose range of compounds was selected according to effects on urethral pressure and the uroselectivity profile of alfuzosin, doxazosin, tamsulosin (3, 10, 30 and 100 µg/kg), prazosin (3, 10 and 100 µg/kg), 5-Me-urapidil (30, 100 and 300 µg/kg) and terazosin (10, 30 and 100 µg/kg) were investigated.

Estimation of Uroselectivity

The simultaneous measurement of BP and UP in the same animal allows us to propose a definition of uroselectivity based on a reduction in UP without modification in BP (full uroselectivity), a reduction in UP associated with a decrease in BP less than 10% (partial uroselectivity) or a decrease in UP (or no modification of UP) associated with a decrease in BP equal or greater than 10% (no uroselectivity).

Compounds Used

All tested adrenoceptor antagonists were synthesized at Synthelabo Recherche (Rueil-Malmaison, France), except for 5 Me-urapidil which was obtained from Research Biochemicals International (Illkirch, France). All compounds were dissolved in normal saline or in a 5% glucose solution. Phenylephrine hydrochloride was obtained from Sigma Chemical Co. (St Louis, MO) and [³H]-prazosin from Amersham (Les Ulis, France).

	Results

Top Abstract Introduction Methods Results Discussion References

In vitro. Phenylephrine (0.1-300 µM) produced a concentration-dependent contraction of rabbit isolated prostate strips with a pEC₅₀ of 5.02 ± 0.08. All the antagonists investigated produced parallel, rightward shift of the phenylephrine E/[A] curve, indicative of competitive antagonism (fig. 1). The antagonist pK_B estimates are given in table 1.

View larger version (K): [in this window] [in a new window]   Fig. 1.   Effects of alpha-1-adrenoceptor antagonists on phenylephrine-induced contractions of rabbit prostatic strips. Results are expressed as percentage of the maximum response to phenylephrine in the absence of adrenoceptor antagonists (%Emax phenylephrine).

In vivo. Evaluation of uroselectivity was carried out in conscious male rats by the simultaneous measurement of blood and uretral pressures. Throughout the observation period (60 min) no significant changes occurred in the vehicle-treated group (fig. 2; baseline: BP = 118.6 ± 3.6 mmHg, UP = 10.9 ± 1.3 cm H₂O, HR = 448 ± 28 beats/min; at 60 min postdose BP = 118.9 ± 2.9 mmHg, UP = 10.4 ± 1.0 cm H₂O, HR = 408 ± 19 beats/min, n = 11 for all parameters except for HR, n = 4). Base-line values of all parameters in the various experimental groups were not different (data not shown). HR was not significantly modified by any of the drugs studied (table 2).

View larger version (K): [in this window] [in a new window]   Fig. 2.   Time-course effects of selective alpha-1a-adrenoceptor antagonists on urethral (filled symbols) and arterial blood pressure (open symbols) in conscious male rats. All drugs were administered by the i.v. route over a 5-min period. Results are expressed as percentage variation from control readings obtained prior to drug infusion (% variation).

View larger version (K): [in this window] [in a new window]   Fig. 3.   Time-course effects of nonselective alpha-1-adrenoceptor antagonists on urethral (filled symbols) and arterial blood pressure (open symbols) in conscious male rats. All drugs were administered by the i.v. route over a 5-min period. Results are expressed as percentage variation from control readings obtained before drug infusion (% variation).

	Discussion

Top Abstract Introduction Methods Results Discussion References

Alpha-1 adrenoceptor classification has evolved over the past years and although today the three cloned subtypes, i.e., alpha-1a, alpha-1b and alpha-1d are well characterized, their tissue distribution and function remain to be established. The specific contribution of alpha-1A-adrenoceptors to prostatic contractility has been intensively investigated due to the availability of new adrenoceptor antagonists with variable degree of selectivity for this subtype (Michel and Insel, 1994; Kawabe, 1995; Testa et al., 1994, 1996). Tamsulosin was one of the first antagonists for which selectivity for the alpha-1A-adrenoceptor subtype was reported (Michel et al., 1993) with selectivity ratios for the alpha-1a over the alpha-1b and alpha-1d-adrenoceptor ranging between 12 to 38 and 1 to 10, respectively (Michel et al., 1993, Ford et al., 1996). 5-Me-urapidil has been described as having high selectivity for the alpha-1a-adrenoceptor with selectivity ratios for the alpha-1a over the alpha-1b subtype ranging from 45 to 83 (Faure et al., 1994; Forray et al., 1994). Attempts to evaluate the relationship between selectivity for a given alpha-1-adrenoceptor subtype and functional in vitro and in vivo selectivity for the urinary tract led to different conclusions. They depend on the compounds evaluated, the type of receptors used to estimate affinities (cloned vs. native, human vs. animal) and the way selectivity was assessed (in vitro or in vivo). The binding data described in our study are in agreement with the values previously published for human cloned alpha-1-adrenoceptor subtypes and confirm the selectivity of tamsulosin and 5-Me-urapidil for the alpha-1a-adrenoceptors (Kenny et al., 1996).

In our study, the alpha-1-adrenoceptor antagonists tested dose-dependently shifted the dose-response contraction curves to phenylephrine of rabbit prostatic tissue. The antagonist affinities (pK_B) were well correlated with affinity for the cloned alpha-1a-adrenoceptor, and agree with previous data reported for human (Forray et al., 1994, Marshall et al., 1995), rat (Chess-Williams et al., 1994), rabbit (Auguet et al., 1995) or dog (Kenny et al., 1996; Testa et al., 1993) lower urinary tract tissues. These results suggest that alpha-1A-adrenoceptors may mediate phenylephrine-induced contractions of lower urinary tract tissues.

However, more recent studies have suggested that the characteristics of the functional alpha-1-adrenoceptor subtype in human prostate are not consistent with those of the alpha-1A-adrenoceptor subtype. This functional alpha-1-adrenoceptor subtype seems to better correspond with a subtype having low affinity for prazosin (Muramatsu et al., 1994) which is discriminated by RS-17053 (Ford et al., 1996). The data obtained with RS-17053 by Ford et al. (1996), and further confirmed by Kenny et al., (1996) with additional selective antagonists and by us (Van Der Graaf et al., 1996) clearly indicated that cloned alpha-1a-adrenoceptors and alpha-1-adrenoceptors mediating contractile responses to norepinephrine in vitro were different. In our study, the pK_B values, expressing efficacy of the tested compounds to relax phenylephrine-induced contractions, are consistently lower (from 0.6 to 1.9 log unit) than the pK_i values expressing the affinity of this drugs for the alpha-1a-adrenoceptor and this difference was indeed greater for the quinazoline compounds. The actual potency displayed by the antagonists tested is therefore 4- to 80-fold lower than their affinity for the alpha-1a-adrenoceptor and this shift could suggest that the receptor mediating smooth muscle contraction of the urethra does not have all the characteristics of the alpha-1a-adrenoceptor subtype. These results are also in agreement with those reported by Ford et al. (1996a, 1996b), Van Der Graaf et al. (1996) and Muramatsu et al. (1994) suggesting the involvement of the alpha-1L-adrenoceptor in the contraction of the lower urinary tract tissue.

To further evaluate the functional role of alpha-1-adrenoceptor subtypes, in the lower urinary tract, comparative in vivo studies with selective antagonists were carried out. Indeed, published data on respective effects on vasculature and lower urinary tract do not allow us to draw conclusions concerning functional uroselectivity. Few direct comparisons of in vivo efficacy and uroselectivity of the existing alpha-1-adrenoceptor antagonists are available. In the anesthetized dog, Kenny et al. (1994) showed that the potencies of several alpha-1-adrenoceptor antagonists to inhibit the phenylephrine-induced increases in prostatic pressure were similar to their antagonist effect on increased blood pressure. In this study, 5-Me-urapidil was the only adrenoceptor antagonist to show uroselectivity, providing further contrast between the dog and the rat studies. However, Testa et al. (1994) showed that tamsulosin and rec15/2739 inhibited noradrenaline-induced increases in prostatic pressure with a respective selectivity ratio of 10 and 100 over the reduction in basal blood pressure. Accordingly, Shibashaky et al. (1992) showed in female dogs that R()YM 12617 (tamsulosin) was more potent to reduce phenylephrine-induced increases in urethral than arterial blood pressure. The use of agonists with a different specificity for alpha-adrenoceptor subtypes (phenylephrine vs. noradrenaline) or end organ responses (prostatic urethra vs. urethra) may account for such discrepancies. Recently, Brune et al. (1996), investigated in conscious dogs, the effects of tamsulosin, doxazosin and terazosin on phenylephrine-induced increases in UP and BP and reported a weak uroselectivity for tamsulosin at one of the doses tested. Alfuzosin, but not prazosin, was shown to be uroselective when its effects on blood pressure in spontaneously hypertensive rats were compared to its effects on electrically stimulated urethral pressure in cats (Lefèvre-Borg et al., 1993). In anesthetized cats, doxazosin and terazosin were shown to be active on the sympathetic drive to the bladder and on blood pressure within the same dose range, even though doxazosin induced less tachycardia than terazosin (Ramage et al., 1995). In all these studies, with the exception of the latter, adrenoceptor antagonists were tested against increases in urethral and arterial blood pressure induced by agonist injection, electrical stimulation or against a genetically induced hypertension. Our study, as well as a previous one (Martin et al., 1995), are to our knowledge, first attempts to evaluate the direct effects of alpha-1-adrenoceptor antagonists in conscious animals under normal adrenergic tone. The obtained results, together with antagonist affinities (pK_i) therefore provide new information on the functional role of alpha-1-adrenoceptor subtypes.

In our study we found that the two compounds with the highest affinity for the alpha-1a-adrenoceptor subtype, tamsulosin (pK_i = 10.4) and prazosin (pK_i = 9.73) were the most potent in affecting blood pressure as from the 3 µg/kg dose and the least uroselective. Compounds with higher functional uroselectivity, i.e., alfuzosin and doxazosin have a similar binding profile which is devoid of alpha-1-adrenoceptor subtype selectivity. 5-Me-urapidil was the compound with the highest selectivity for the alpha-1a-adrenoceptor subtype and, in our work, was devoid of significant activity on urethral and blood pressures. Its affinity for the alpha-1a-adrenoceptor subtype (pK_i = 8.29) was similar to that of doxazosin and, if this subtype was functionally relevant, an effect should have been seen at the doses tested (up to 300 µg/kg). Terazosin only induced minimal changes on urethral pressure at the intermediate dose we tested whereas all other compounds studied had a significant effect on urethral pressure at this dose. Other studies have clearly demonstrated the efficacy of terazosin to reduce urethral pressure. Nevertheless, all previous studies were carried out under conditions where urethral pressure was increased either by electrical stimulation (Lefèvre-Borg et al., 1993) or agonist infusion (Brune et al., 1996). It is not clear, whether the weakness of effect seen in our study is related to the absence of urethral hypertonia or due to differences in plasma clearance or plasma protein binding. Finally, no clear dose-response relationship was observed on urethral pressure, mainly between the two higher doses tested. At the higher dose tested, a marked decrease in blood pressure was obtained and it can be hypothetized that the vascular responses have blunted the urethral responses by decreasing the drug availability for the urethral tissue.

In our study, the selectivity for the alpha-1a-adrenoceptor subtype has been assessed using cloned bovine receptors and functional uroselectivity in conscious rats. The question of species differences could therefore be raised as a putative explanation for this lack of relationship between alpha-1a-adrenoceptor selectivity and functional uroselectivity. Affinities for the alpha-1a-adrenoceptor subtype of a variety of alpha-1-adrenoceptor antagonists are similar whether evaluated in native tissue (Testa et al., 1996), in cloned bovine receptors (Testa et al., 1996) or in human cloned receptors (Testa et al., 1995; Kenny et al., 1996). It seems therefore unlikely that species differences in binding data were of importance.

In conclusion, our results demonstrate that functional uroselectivity of adrenoceptor antagonists cannot be extrapolated from either receptor affinity or receptor subtype selectivity data. In the conscious male rat under normal adrenergic tone, the most uroselective drugs were alfuzosin and, over a more limited dose-range, doxazosin whereas tamsulosin, 5-Me-Urapidil, prazosin and terazosin where either partially uroselective or not uroselective. This uroselectivity is therefore not correlated with a selective affinity for the alpha-1a-adrenoceptor subtype.