Introduction

Top Abstract Introduction Methods Results Discussion References

Considerable effort has been made toward developing selective water diuretics for hyponatremic disorders and congestive heart failure. Because the primary control of water homeostasis involves vasopressin, development for water diuretic activity has been directed toward inhibition of vasopressin activity. This has involved two primary strategies: inhibition of vasopressin secretion with kappa opioid agonists and inhibition of vasopressin activity with antagonists to the vasopressin V2 receptor. Development of both classes of agent has been complicated, with vasopressin receptor antagonists being associated with species-dependent activity (Ruffolo et al., 1991) and kappa opioid agonists possessing undesirable central nervous system activity (Peters and Gaylor, 1989; Dionne et al., 1991). In recent years, some of the newer kappa opioid agonists (Reece et al., 1994; Ohnishi et al., 1994) and vasopressin receptor antagonists (Ohnishi et al., 1993) have been advanced into clinical studies; however, to date, no direct comparison of these has been made. In the present study, we have therefore evaluated the effect of the vasopressin receptor antagonist, OPC 31260 (Ohnishi et al., 1993) (fig. 1), and the kappa receptor agonists, niravoline (Sinnassamy et al., 1994) (fig.1), SB 215519 and SB 215520 (Vecchietti et al., 1994) (fig.1) in conscious dogs.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Chemical structures.

	Methods

Top Abstract Introduction Methods Results Discussion References

All experiments were approved by the Institutional Animal Care and Use Committee and were in accordance with National Institutes of Health guidelines for the care and use of animals.

Opioid binding assays. Opioid receptor binding experiments were carried out with use of guinea pig (mu and kappa-1 assays) and rat (kappa-2 assay) brain homogenates. Tissues were prepared according to methods described previously (Kosterlitz et al., 1981). Male guinea pigs (300-350 g, Dunkin-Hartley, Charles River Italia SpA, Calco, Italy) and male rats (200-250 g, Sprague-Dawley, Charles River Italia SpA, Calco, Italy) were sacrificed by decapitation and the brains, without cerebella, were removed rapidly. The tissues were homogenized in 10 volumes of Tris HCl (50 mM, pH 7.4) by use of a PBI politron, and were centrifuged at 49,000 × g for 10 min at 4°C. The resultant pellets were resuspended in 10 volumes of Tris buffer, incubated at 37°C for 45 min and centrifuged at 49,000 × g for 10 min. The resultant pellets were finally resuspended in 100 volumes of Tris buffer (final concentration, 0.5-0.6 mg of protein/ml), and 1.9-ml aliquots were used for the assay.

Rat diuretic activity. Diuretic activity was assessed in normally hydrated rats by use of the methods described by Leander (1983). Rats were removed from their home cage, weighed, administered drug and placed in metabolism cages for 5 hr. Urine was collected into graduated cylinders and activity expressed as milliliters of urine output in 5 hr. ED₃₀₀ values, based on the dose of drug which resulted in a 3-fold increase in urine output in comparison with the urine output of control animals, was determined by regression analysis.

Rat incoordination/ataxia activity. The degree of motor incoordination/ataxia induced by test compounds was evaluated by use of an accelerating rotarod (U. Basile, Varese, Italy) as described by Hayes and Tyers (1983). Groups of six male Sprague-Dawley rats (Charles River) weighing between 180 and 350 g were used for all experimental procedures. Rats were trained on the rotarod during six 5-min sessions separated by 1 hr, on 2 consecutive days before testing on day 3 (Iwamoto, 1981; Sutters et al., 1990). On day 1 during the first three sessions of training, the rotarod accelerated from 2 to 20 rpm in a period of 5 min; whereas for the remaining sessions (day 2) the belt-gear ratio was set so that the rotor would accelerate from 3 to 30 rpm in the 5-min period. These latter conditions were adopted for the experiment itself. Thirty minutes after drug or vehicle administration, rats were placed on the rotarod and the animal's performance (i.e., time on the rotarod) was recorded. The degree of the motor incoordination/ataxia afforded by the compound was determined by the formula: % activity = (1  T/C) × 100, where T = mean performance time (sec) on the rotarod of the treated group and C = mean performance time (sec) on the rotarod of the control group.

Dog diuretic activity. Four female mongrel dogs (11.7-14.85 kg) were instrumented with arterial Vascular-Access-Ports. Animals were anesthetized with sodium biotol (for induction) and isofluorane. By use of standard surgical aseptic techniques, a subcutaneous pocket on the left flank was created, and the access portion of the Vascular-Access-Port was sutured to the underlying muscle in the area of the paralumbar fossa. The catheter end of the Vascular-Access-Port was tunneled to the area of the left femoral triangle. A second incision was made, and the catheter was inserted in the femoral artery to the length of approximately 12 cm and secured with silk sutures. The animals were trained to lie lightly restrained on their right side on a cradle-like board and were deprived of food and water for 18 hr before each test. Before the dogs were placed in their cradles, a catheter was inserted into the left cephalic vein (Cathlon IV) for drug or vehicle infusion. Once the dogs had been lightly restrained in their cradles, the left saphenous vein was catheterized (Intracath) for obtaining blood samples and the urinary bladder (14 Fr Foley) for urine collection. After a 20- to 40-min equilibration period, urine flow was measured for three 20-min periods, with the final 20-min period being designated the control period. Subsequently, drug or vehicle was infused. For each experiment, three doses of drug were infused intravenously for a minimum of 60 min. If a diuretic response to an individual dose of drug was observed and the maximum effect had not been reached within 60 min, infusion of that dose of drug was continued until a maximum response had been observed. Usually the maximum response would be observed within the 60-min infusion period. SB 215519 was given at 0.01 to 1 µg/kg·min at an infusion rate of 0.01 ml/kg·min; SB 215520 was given at 1 to 10 µg/kg·min at an infusion rate of 0.03 ml/kg·min; niravoline was given at 0.01 to 1 µg/kg·min at an infusion rate of 0.01 ml/kg·min; and OPC 31260 was given at 3 to 30 µg/kg.min at an infusion rate of 0.02 ml/kg·min. In control tests, vehicle (0.9% NaCl) was administered at 0.01 ml/kg·min for 3 hr. The order of administration of vehicle and test compound was randomized with an interval of at least 2 weeks between each study. If during the study significant side effects were observed, subsequent doses of compound were not administered (this only occurred with SB 215519). Urinary fluid loss was not replaced during the study. Blood samples were taken at the end of each 20-min period. Urine and plasma electrolytes and creatinine were determined with the Synchron Clinical Systems AS-8.

Data analysis. Data are reported as means ± S.E. Osmolar clearance (Cosm, ml/min) was calculated as Uosm·V/Posm, where V is the urine flow rate (ml/min) and Uosm and Posm are osmolality of the urine and plasma (as measured by freezing point depression), respectively. Free water clearance (C H₂O; ml/min) was calculated as V  Cosm. Data were analyzed statistically by an analysis of variance for repeated measures. P < .05 was considered statistically significant.

Materials. The kappa agonists, SB 215519, SB 215520 and niravoline, and the vasopressin receptor antagonist, OPC 31260, were synthesized at SmithKline Beecham, S.p.A., Milan, Italy.

	Results

Top Abstract Introduction Methods Results Discussion References

The three kappa agonists evaluated, SB 215519, SB 215520 and niravoline, are all potent kappa receptor agonists, with binding affinities to the kappa-1 receptors in the low or subnanomolar range (table 1). SB 215520 and SB 215519 are diastereoisomers, with SB 215520 having approximately 15 times less affinity for the kappa-1 receptors than SB 215519. None of the kappa agonists used in the present study had significant activity at the kappa-2 or mu receptors (table 1). In a model to evaluate ataxia and locomotor incoordination (the rat rotarod), SB 215519 was the most active and SB 215520 was the least active (table 1). A similar rank order potency was observed with rat diuretic activity (table 1); however, the ratio of activities in the rat rotarod and rat diuresis was significantly greater for SB 215520, which perhaps suggested a greater separation in the water diuretic and sedation/ataxic activities for this compound compared with both SB 215519 and niravoline (table 1).

When SB 215519 was administered to conscious dogs, no effects were seen on urine flow and free water clearance at doses less than 0.3 µg/kg·min; however, an antinatriuresis was observed at doses as low as 0.03 µg/kg·min (fig. 2). Only one dog received the highest dose (1 µg/kg·min) of SB 215519 because of significant side effects; however, this dog responded with a brisk water diuresis with urine flow increasing and urine osmolality decreasing, which led to a positive free water clearance (fig. 2). In this dog, an increase in heart rate was observed before any change in urine flow (fig. 2). Higher doses of SB 215520 were required to cause an increase in urine flow and a decrease in urine osmolality; however, at no time was SB 215520 able to cause a positive free water clearance, i.e., a free water clearance >0 (fig. 3). In addition, SB 215520 resulted in both a reduction in sodium excretion and a modest increase in heart rate (fig. 3). At the doses used, niravoline had no significant effect on urine flow, urine osmolality or free water clearance (fig. 4); however, a decrease in sodium excretion was observed and, at the highest dose evaluated (1 µg/kg·min), an increase in heart rate (fig. 4).

View larger version (29K): [in this window] [in a new window]   Fig. 2.   Effect of the kappa opioid agonist, SB 215519, on (A) urine flow (V) and urine osmolality (Uosm); (B) free water clearance (C H2O) and osmolal clearance (Cosm); (C) urinary sodium excretion (UNaV) and glomerular filtration rate (GFR); and (D) mean arterial pressure (MAP) and heart rate (HR) in four conscious hydropenic dogs. Clearance measurements were performed over 20-min time periods. Blood pressure and heart rate were measured at the end of each period. *P < .05 versus C. (n = 1-4 observations/dose).

View larger version (26K): [in this window] [in a new window]   Fig. 3.   Effect of the kappa opioid agonist, SB 215520, on (A) urine flow (V) and urine osmolality (Uosm); (B) free water clearance (C H2O) and osmolal clearance (Cosm); (C) urinary sodium excretion (UNaV) and glomerular filtration rate (GFR); and (D) mean arterial pressure (MAP) and heart rate (HR) in four conscious hydropenic dogs. Clearance measurements were performed over 20-min time periods. Blood pressure and heart rate were measured at the end of each period. *P < .05 versus C. (n = 4 observations/dose).

View larger version (27K): [in this window] [in a new window]   Fig. 4.   Effect of the kappa opioid agonist, niravoline, on (A) urine flow (V) and urine osmolality (Uosm); (B) free water clearance (C H2O) and osmolal clearance (Cosm); (C) urinary sodium excretion (UNaV) and glomerular filtration rate (GFR); and (D) mean arterial pressure (MAP) and heart rate (HR) in four conscious hydropenic dogs. Clearance measurements were performed over 20-min time periods. Blood pressure and heart rate were measured at the end of each period. *P < .05 versus C. (n = 4 observations/dose).

All three kappa opioid agonists resulted in overt side effects, specifically increased heart rate as described above and trembling. Trembling was first observed with SB 215519 at a dose of 0.1 µg/kg·min, with SB 215520 at 10 µg/kg·min and with niravoline at 1 µg/kg·min. At no time were overt side effects observed with administration of OPC 31260.

Administration of the vasopressin receptor antagonist, OPC 31260, to conscious hydropenic dogs resulted in a significant water diuresis, with an increase in urine flow, a decrease in urine osmolality and an increase in free water clearance to >0 (fig. 5). In addition, no decrease in sodium excretion was observed nor were there any changes in heart rate (fig. 5) or any overt side effects.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Effect of the vasopressin V2 receptor antagonist, OPC 31260, on (A) urine flow (V) and urine osmolality (Uosm); (B) free water clearance (C H2O) and osmolal clearance (Cosm); (C) urinary sodium excretion (UNaV) and glomerular filtration rate (GFR); and (D) mean arterial pressure (MAP) and heart rate (HR) in four conscious hydropenic dogs. Clearance measurements were performed over 20-min time periods. Blood pressure and heart rate were measured at the end of each period. *P < .05 versus C. (n = 4 observations/dose).

In time control studies in which no drug was administered, no significant changes were observed in any of the parameters measured (table 2).

	Discussion

Top Abstract Introduction Methods Results Discussion References

In the present study, we have demonstrated that in the conscious hydropenic dog, kappa opioid agonists and the nonpeptide vasopressin V2 receptor antagonist can increase urine flow and decrease urine osmolality. The data indicate, however, that under the conditions studied, the vasopressin receptor antagonist, OPC 31260, caused a more consistent positive free water clearance than the kappa agonists and that this water diuresis was not associated with any change in sodium excretion or systemic hemodynamics. Conversely, the changes in urine flow and urine osmolality induced by the kappa agonists, SB 215519, SB 215520 and niravoline, were accompanied by sodium retention, tachycardia and apparent central nervous system side effects, e.g., trembling. The less consistent changes in free water clearance induced by the kappa agonists in dogs are consistent with studies in rats where niravoline was unable to cause a positive free water clearance in normal rats, but resulted in a dramatic water diuresis in cirrhotic rats (Bosch-Marce et al., 1995).

Kappa opioid agonists have long been recognized to cause sedation, ataxia and dysphoria; indeed, these activities are the primary reason that development of this class of compounds as novel analgesics has, to date, been unsuccessful. To develop kappa agonists as water diuretics, it was necessary to address the issue of these unwanted activities. Because there was evidence that kappa agonists might exert their water diuretic activity by a peripheral mechanism of action, initially there was a hope that an agonist that would not cross the blood-brain barrier would represent a selective water diuretic agent.

Evidence to suggest a peripheral site of action of kappa agonists to inhibit the antidiuretic activity of vasopressin includes the observations that there are kappa opioid receptors in the kidney (Slizgi and Ludens, 1985); that the kappa opioid prototype, ethylketocyclazocine, inhibits the antidiuretic activity of exogenous vasopressin in Brattleboro diabetes insipidus rats (Slizgi and Ludens, 1986); and the observation that U62066 can cause a water diuresis in human volunteers without a significant change in plasma vasopressin concentration (Rimoy et al., 1991). Previously, however, we were unable to observe any significant inhibition of vasopressin-induced adenylate cyclase activity in rat collecting tubules in vitro (Brooks et al., 1993), consistent with the observation that the kappa agonist, U62066, did not inhibit vasopressin-stimulated water permeability in isolated perfused inner medullary collecting ducts (Yamada et al., 1989).

In an attempt to provide definitive evidence of whether kappa opioid agonists caused a water diuresis by a central site of action and thus presumably by inhibition of vasopressin secretion, we previously evaluated the water diuretic activity of three opioid agonists with variable ability to cross the blood-brain barrier (Brooks et al., 1993). BRL 52974, a kappa agonist that, based on computational (cLogP), physicochemical (DLogP) and functional (rat rotarod) determination, does not readily cross the blood-brain barrier, indicated that sites within the blood-brain barrier are required for the major component of kappa agonist-induced water diuresis. These observations are consistent with previous reports that both peripheral and intracerebroventricular administration of the kappa opioid agonist, U50488, can inhibit both hemorrhage- and osmotic-induced vasopressin secretion. Subsequently, Rossi and Brooks (1996), with compartmentalized rat hypothalamo-neurohypophyseal explants in culture, were able to confirm that kappa opioid agonists do indeed inhibit vasopressin preferentially at hypothalamic sites. What little peripheral activity kappa opioid agonists may have to cause a water diuresis in rats appears to involve adrenal release of epinephrine and subsequent stimulation of renal alpha-2 receptors (Wang et al., 1994); this response may be limited to the rat (Brooks, 1996).

It appears, therefore, that a peripherally acting kappa opioid agonist would not be a potent diuretic and that water diuretic activity depends on the ability to penetrate the blood-brain barrier and inhibit vasopressin secretion. Nonetheless, studies in rats suggested that despite the requirement for a central site of action, the water diuretic activity could be separated from some of the side effects, e.g., ataxia/sedation. Indeed, in the present study, the ratio in a rat diuretic model and a rat sedation/ataxia model of the three kappa agonists differed. Thus, SB 215520 had a ratio seven to nine times greater than niravoline and the diastereoisomer, SB 215519, respectively. In the dog, SB 215520 resulted in a water diuresis at a higher dose than either SB 215519 or niravoline, consistent with its lower affinity for the kappa-1 receptor. Nonetheless, the water diuretic activity was associated with sodium retention, an increase in heart rate and, at the highest dose studied, trembling. The sodium retention observed with all three kappa agonists has been reported previously and appears to involve stimulation of central kappa opioid receptors and increased renal sympathetic nervous system activity (Kapusta and Obih, 1993). It is well known that inhibition of vasopressin activity with vasopressin receptor antagonists do not alter osmolar clearance (Brooks et al., 1991), and thus the slightly lower initial sodium excretions observed in dogs receiving the vasopressin antagonist is unlikely to explain any lack of response to OPC 31260.

In contrast to the water diuretic activity of the kappa agonists, the vasopressin receptor antagonist, OPC 31260, was not associated with any sodium retention, consistent with a selective action to inhibit vasopressin activity at the renal vasopressin V2 receptors (Yamamura et al., 1992). The effect of OPC 31260 in dogs is similar to that observed previously with the peptide vasopressin receptor antagonist, SK&F 105494 (Caldwell et al., 1988). Development of vasopressin receptor antagonists, however, has been complicated by tremendous species differences. Peptides that could antagonize vasopressin-induced antidiuresis in rats were first identified in 1981 (Manning et al., 1981; Sawyer et al., 1981); however, subsequent studies in dogs (Stassen et al., 1983) and squirrel monkeys (Kinter et al., 1987) failed to demonstrate significant vasopressin receptor activity in vivo. Further substitutions resulted in the identification of SK&F 101926 that exhibited potent vasopressin receptor antagonist activity in nonrodent species and no agonist activity. When tested in normal male volunteers, however, SK&F 101926 did not reduce urine osmolality or increase urine flow but possessed apparent agonist activity (Allison et al., 1988). This agonist activity was later observed in the indomethacin-treated conscious dog (Albrightson-Winslow et al., 1989) and the conscious rhesus monkey (Brooks et al., 1988), models which helped in the subsequent identification of SK&F 105494 which was a potent water diuretic in both models. When tested in human volunteers, however, SK&F 105494, like its predecessor, resulted in a full antidiuretic response (Ilson et al., 1990). The mechanism by which these apparent vasopressin receptor antagonists result in antidiuretic activity in man is unclear but may involve the peptide nature of these agents because subsequent development of nonpeptide vasopressin receptor antagonists has not been complicated by such dramatic species variability. Thus, OPC 31260, that demonstrated potent water diuretic activity in the present study, is a potent nonpeptide vasopressin receptor antagonist which showed selectivity for the V2 receptor (Yamamura et al., 1992). Administration of OPC 31260 to humans has demonstrated that this nonpeptide receptor antagonist can result in a significant increase in urine volume and decrease in urine osmolality, thus reflecting water diuretic activity (Ohnishi et al., 1993).

Despite the difficulties in developing a water diuretic agent, both kappa opioid receptor agonists and a vasopressin receptor antagonist have been shown to cause a water diuresis in man; however, based on the present study, it appears that the water diuresis induced by the vasopressin receptor antagonist may be more consistent and associated with less unwanted activities such as possible central nervous system side effects. The sodium-retaining activity observed with administration of the kappa agonists could be a disadvantage if observed in patients with congestive heart failure. Patients with the syndrome of inappropriate secretion of antidiuretic hormone, however, are often sodium depleted; and thus an agent which resulted in a water diuresis and sodium retention might be of interest.