Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

During the past decade, several clinical and experimental investigations indicated that antidiuretic hormone (ADH) was a major causative factor of the water retention occurring in cirrhosis with ascites (Arroyo et al., 1994). This finding laid the theoretic substrate to examine new therapeutical strategies addressed to correct this severe renal excretory dysfunction. Two families of pharmacological tools inhibiting ADH activity have emerged. The first type includes those that reduce the circulating levels of the hormone (Hamon and Jouquey, 1990), and the second includes a group of compounds that specifically block ADH V₂ receptors (Sawyer et al., 1981; Yamamura et al., 1992).

Plasma ADH levels can be reduced by administering -opioid receptor agonists. These substances increase urine volume and decrease urinary osmolality (U_Osm), in humans and experimental animals, by virtue of their inhibitory effect on ADH release (Slizgi and Ludens, 1982). On the other hand, several peptide and nonpeptide receptor antagonists have proved to be highly selective and efficacious at antagonizing ADH V₂ receptors in experimental animals (Sawyer et al., 1981; Yamamura et al., 1992). However, only nonpeptide ADH V₂ receptor antagonists are useful in humans because peptide ADH V₂ receptor antagonists may behave as ADH agonists when given to healthy subjects (Stassen et al., 1983).

Studies investigating the renal effects of a single dose of these compounds recently demonstrated that both the -opioid receptor agonist niravoline (Bosch-Marcé et al., 1995; Moreau et al., 1996) and the nonpeptide ADH V₂ receptor antagonist {(±)-5-dimethylamino-1-(4-[2-methybenzoylamine]benzoyl)-2,3,4,5-tetrahydro-1H-benzazepin hydrochloride} (OPC-31260; Tsuboi et al., 1994) induce a marked aquaretic effect in experimental cirrhosis. However, there is no investigation comparing the therapeutical efficacy of chronically giving a -opioid receptor agonist or a nonpeptide ADH V₂ receptor antagonist in cirrhosis. Therefore, the current study was designed to compare the renal, hormonal, and hemodynamic effects induced by the chronic oral administration of niravoline, OPC-31260, or vehicle in rats with CCl₄-induced cirrhosis, ascites, and severe water retention. This is an intriguing question, particularly considering the profound differences existing in the mechanisms by which these drugs may induce aquaresis in advanced liver disease.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

The study was performed with 40 conscious adult male Wistar rats with cirrhosis, ascites, and impaired free water excretion and with 44 control Wistar rats. Both groups were fed ad libitum with standard chow and distilled water containing phenobarbital (0.3 g/liter). Cirrhosis was induced by CCl₄, which was administered twice weekly (Monday and Friday) following a method described elsewhere (Jiménez et al., 1995). Cirrhotic rats were obtained from a group of 163 animals submitted to the cirrhosis induction protocol. One hundred twenty-three of these animals could not be included in the study for several reasons: 66 rats died before the development of impairment of water excretion; 43 died before completing the experimental protocol; and 14 rats showed a normal ability to excrete free water 30 weeks after starting the cirrhosis induction program. Five weeks after starting the study, rats receiving CCl₄ and control rats were placed in individual metabolic cages. One week after developing ascites, the renal ability to excrete free water was determined once weekly (Tuesday) in each rat submitted to the cirrhosis induction program as follows: 2 h after the removal of food and water, a water load (50 ml/kg/b.wt.) was administered via a gastric tube inserted with the animals under light ether anesthesia. Immediately afterward, they were reintroduced into their metabolic cages, where each volume of spontaneously voided urine was collected separately. After 3 h and after an abdominal massage, a final urine sample was obtained. The osmolality of each urine sample was measured. Total volume was measured gravimetrically. The renal ability to excrete free water was estimated through the minimum urinary osmolality (mU_Osm) of spontaneously voided samples obtained after the water load and by calculating the percentage of the water load excreted during the 3-h urine collection period. When a significant impairment in the renal ability to excrete free water was detected (percentage of water load excreted <60% and mU_Osm >160 mOsm/kg) animals were included in the protocol. All studies in cirrhotic rats were performed 24 h after a 2-min CCl₄ reexposure to avoid spontaneous improvement in free water excretion. Because rats treated with CCl₄ and phenobarbital developed the impairment in free water excretion within 12 to 26 weeks after starting the cirrhosis induction program, controls rats were investigated 17 to 25 weeks after being included in the study.

Protocol I: Effect of Chronic Administration of Niravoline, OPC-31260 or Vehicle on Renal Sodium and Water Metabolism, Urinary Excretion of ADH and Aldosterone, and Hypothalamic ADH mRNA Expression in Cirrhotic Rats with Ascites and Water Retention

Animals included in the protocol were randomly assigned to one of the following groups: (A) intragastric administration of niravoline (3 mg/kg, dissolved in 3 ml/kg water) administered daily for 10 days (seven cirrhotic and nine control rats), (B) intragastric administration of OPC-31260 (5 mg/kg, dissolved in 3 ml/kg water) administered daily for 10 days (eight cirrhotic and eight control rats), and (C) intragastric administration of water (3 ml/kg) administered daily for 10 days (seven cirrhotic and nine control rats). The dose of niravoline was selected based on previous investigations showing that oral administration of this compound to normal conscious rats triplicates urine volume and markedly reduces the mU_Osm, without significantly affecting sodium excretion (Ginés et al., 1998). Similarly, the amount of OPC-31260 chosen was identical to that previously used by Tsuboi et al. (1994) in cirrhotic rats. These authors demonstrated that after the administration of 5 mg/kg OPC-31260, the percentage of water load excreted was >200%.

Measurements of 24-h urine volume and U_Osm and urinary sodium excretion (U_NaV) were made 1 day before the water overload and for 9 consecutive days after animals were included in the protocol. An aliquot of each 24-h urine collection was frozen at 30°C until analyzed to determine U_Osm, urinary excretion of ADH (U_ADHV), and urinary excretion of aldosterone (U_ALDV). Urine creatinine concentration was determined in urine from the ninth day.

On the 10th day, animals were submitted to a second water overload, as previously described, and the mU_Osm and percentage of water excreted were determined in the 3-h urine collection. Thereafter, animals were sacrificed by decapitation, and the hypothalamus was dissected, quickly frozen in dry ice, and stored in liquid nitrogen until further extraction of total RNA. Trunk blood was collected to measure serum osmolality and creatinine.

Protocol II: Effect of Chronic Administration of Niravoline, OPC-31260, or Vehicle on Arterial Pressure, Cardiac Output, and Peripheral Resistance in Cirrhotic Rats with Ascites and Water Retention

Control (n = 18) and cirrhotic (n = 18) rats were randomly distributed as described in protocol I, with six cirrhotic and six control rats included in each treatment group. At the ninth day of the protocol, animals were anesthetized with ketamine (50 mg/kg) and prepared with polyvinyl tubes (PE-50) as previously described (Castro et al., 1993). Briefly, a catheter was inserted in the left femoral artery for mean arterial pressure (MAP) and heart rate (HR) measurement. A second catheter was inserted in the right jugular vein for saline infusion, and a thermodilution catheter was inserted in the right carotid artery and advanced up to the aortic arch for cardiac output (CO) measurement. Catheters were tunneled s.c., exteriorized in the nape of the neck, and run through a flexible stainless steel sheath that was attached to a harness made of polystyrene that was worn by the animal. MAP, HR, and CO were determined in a microcomputer system (Cardiomax IIR; Columbus Instruments, Columbus, OH) and recorded in a multichannel system (MX4P and MT4; Lectromed Ltd., Jersey Channel Islands, UK). CO was determined by thermodilution after the administration of a bolus of 200 µl of Ringer's solution (20-23°C) into the right atrium. A spring-loaded syringe was used (Hamilton syringe, model CR-700-200) to ensure a constant injection rate and volume. Total peripheral resistance (TPR) was estimated with the formula TPR = MAP/CO. Each value represents the average of three measurements. The coefficient of variability of the method was calculated to be 2.9% and 5.5% in control and cirrhotic rats, respectively.

Animals were placed in rectangular cages with no restriction of movement and allowed to recover from surgery and anesthesia with free access to standard chow and water. At the 10th day, after complete recovery of anesthesia and 24 h after catheter insertion, rats received the corresponding dose of niravoline, OPC-31260, or vehicle, and MAP and HR were monitored until stable for at least 30 min; then, CO was assessed, and a final value was obtained by averaging three separate measurements.

The protocols were performed according to the criteria of the Committee for the Care and Use of Laboratory Animals in the Hospital Clínic Universitari.

Measurements and Statistical Analysis

RNA Isolation. Total RNA was extracted from the hypothalamus of cirrhotic and control rats using a commercially available kit (Trizol Reagent; Life Technologies, GIBCO BRL, Gaithersburg, MD). The final RNA pellets were resuspended in diethylpyrocarbonate-treated water and stored at 80°C until use. The amount of RNA was evaluated by absorption at 260 nm (Uvikon Spectrophotometer; Kontron Instruments, Milano, Italy). The ratio of absorption (260:280 nm) of all preparations was between 1.8 and 2.1. Moreover, all samples were subjected to gel electrophoresis and stained with ethidium bromide to prove the integrity of the 18S and 28S rRNA.

ADH mRNA Expression. ADH mRNA was determined by Northern blot hybridization. Total RNA (3 µg) isolated from hypothalamus was denatured by heating (65°C), separation in 1% agarose gel containing 50% (v/v) formamide and 2.2 M formaldehyde, and vacuum transfer to a nylon membrane. The blots were baked at 80°C for 2 h and then prehybridized for 15 min and overnight hybridization at 65°C in Rapid-Hyb buffer (Amersham Iberica, Madrid, Spain) containing the ³²P-labeled cDNA probe. The probe is the 480-bp cDNA probe generated by polymerase chain reaction (PCR) of the cDNA base sequence of rat ADH. Filters were also probed with a 309-bp reverse transcription-PCR product of the cDNA base sequence of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), a constitutively expressed gene, as a control. The probes were labeled using the Rediprime DNA labeling system (Amersham International, Buckinghamshire, UK). After hybridization, blots were washed three times as follows: 2× standard saline citrate (SSC)/0.1% SDS at room temperature for 20 min, 1× SSC/0.1% SDS at room temperature for 15 min, and 0.5× SSC/0.1% SDS at room temperature for 10 min and then washed twice in 0.1× SSC/0.1% SDS at 65°C for 5 min. Blots were dried and exposed to X-ray film in the presence of an intensifying screen for 12 to 16 h.

Probes. A 480-bp cDNA probe generated by PCR was used for analysis of ADH mRNA expression. PCR was performed using a DNA amplification reagent kit (Life Technologies, GIBCO BRL, Gaithersburg, MD) with rat ADH primers prepared by a DNA synthesizer (392 DNA/RNA Synthesizer; Applied Biosystems, Foster City, CA). The sense and antisense sequences are 5'-TGATGCTCAACACTACGCTC-3' and 5'-TGGCAGAATCCACGGACTCT-3', respectively. The cDNA amplification product was predicted to be 480 bp corresponding to bases 11 to 490 in the cDNA base sequence of rat ADH (Foo et al., 1991).

Other Measurements. Serum osmolality and U_Osm were determined from osmometric depression of the freezing point (Osmometer 3 MO; Advanced Instruments, Needham Heights, MA) and sodium concentration by flame photometry (IL 943; Instrumentation Laboratory, Lexington, MA). Urinary ADH was determined by radioimmunoassay (Bühlman Laboratories AG, Basel, Switzerland) of unextracted samples as previously described (Camps et al., 1987). The urinary concentration of aldosterone was measured with the use of a commercial kit (Coat-A-Count Aldosterone; Diagnostic and Products Corporation, Los Angeles, CA) in urine samples (0.5 ml) adjusted to pH 1.0 with 1 ml of 0.2 N HCl and kept for 20 h at 30°C. With this procedure, most aldosterone-18-glucuronide is transformed into aldosterone (Jiménez et al., 1985). Creatinine was measured by the Ektachem Clinical Chemistry Slide method (Johnson & Johnson Clinical Diagnostic Inc., Rochester, NY).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Histological examination of the liver specimens obtained from CCl₄-treated rats showed cirrhosis in all animals, with no significant difference among rats receiving niravoline, OPC-31260, or vehicle. Control rats showed no appreciable alterations in liver histology.

Table 1 shows that cirrhotic rats included in protocol I were investigated after they had developed marked sodium retention, severely impaired renal ability to excrete free water, hyperaldosteronism, and nonosmotic hypersecretion of ADH. The impairment of water excretion occurred within the range of 12 to 26 weeks after starting the cirrhosis induction program. Ascites preceded the impairment in free water excretion by at least 2 weeks. No differences were found in the baseline renal response to water overload, U_NaV, U_ALDV, and U_ADHV in cirrhotic rats receiving the -opioid receptor agonist, the ADH V₂ receptor antagonist, or the vehicle. In addition, no significant differences were found in any of these parameters among the three groups of control rats.

The effect of niravoline, OPC-31260, or vehicle on urine flow and osmolality in control and cirrhotic rats is shown in Figs. 1 and 2, respectively. Neither -opioid receptor agonist nor ADH V₂ receptor antagonist administration was associated with any significant change in these parameters in control rats (Fig. 1). In contrast, both agents significantly increased urine flow and decreased U_Osm in cirrhotic rats. Marked differences were observed in the response pattern to each drug in these animals. The effect of OPC-31260 was intense during the first 2 days of treatment. Subsequently, urine volume and U_Osm returned to pretreatment values, being similar to those of cirrhotic animals receiving vehicle. In contrast, the aquaretic effect of niravoline, which was similar to that of OPC-31260 during the first day, persisted during the entire period of treatment. Cirrhotic rats receiving vehicle did not experience any noticeable change in urine volume and U_Osm during the days of treatment.

View larger version (42K): [in this window] [in a new window]   Fig. 1.   Urine volume (UV) and UOsm under basal conditions and during entire period of treatment in control rats receiving niravoline (3 mg/kg/b.wt.) and OPC-31260 (5 mg/kg/b.wt.). Shaded area corresponds to mean ± S.E.M. of control rats receiving vehicle (3 ml H2O/kg/b.wt.)

View larger version (38K): [in this window] [in a new window]   Fig. 2.   Urine volume (UV) and urinary osmolality UOsm under basal conditions and during entire period of treatment in cirrhotic rats with ascites and water retention receiving niravoline (3 mg/kg/b.wt.) and OPC-31260 (5 mg/kg/b.wt.). Shaded area corresponds to mean ± S.E.M. of cirrhotic rats with ascites and water retention receiving vehicle (3 ml H2O/kg/b.wt.). #p < .01 and ##p < .05 versus basal values (one-way ANOVA and Newman-Keuls tests). ap < .005 and bp < .05 versus vehicle on same day of treatment (unpaired Student's t test).

After completion of the study, no significant differences were observed in the renal response to the water load and serum osmolality among the three groups of control rats (Table 2). In contrast, cirrhotic rats with ascites receiving the aquaretic agents showed a significantly higher percentage of water load excreted and lower mU_Osm than cirrhotic animals receiving vehicle. This improvement in renal water metabolism, however, was associated with a normalization in serum osmolality only in cirrhotic rats treated with niravoline. Figure 3 shows the individual values of the percentage of water load excreted obtained under baseline conditions and after completion of the treatment in the three groups of cirrhotic rats included in protocol I. The mean values are given in Tables 1 and 2. The administration of vehicle did not produce any significant change in this parameter. Treatment with OPC-31260 or niravoline was associated with a significant increase in water excretion (p < .05 and p < .01, respectively). However, although niravoline normalized the percentage of water load excreted in four of seven animals, this occurred in only one of the eight rats treated with OPC-31260.

View larger version (75K): [in this window] [in a new window]   Fig. 3.   Individual values of percentage of water load excreted observed in three groups of rats with cirrhosis, ascites, and impaired water excretion before and after completion of daily chronic oral administration of vehicle (3 ml of H2O/kg/b.wt.), OPC-31260 (5 mg/kg/b.wt.), and niravoline (3 mg/kg/b.wt.). Shaded area denotes the values below which renal ability to excrete water was considered markedly impaired.

The average values of U_NaV and U_ALDV during the entire period of treatment and creatinine clearance at the end of the study in control and cirrhotic rats receiving vehicle, OPC-31260, or niravoline are shown in Table 3. Control rats receiving the -opioid receptor agonist or the V₂ receptor antagonist showed similar values of U_NaV and creatinine clearance and slightly, albeit significantly, decreased U_ALDV values than those receiving vehicle. In contrast, cirrhotic rats treated with niravoline or OPC-31260 had significantly higher values of U_NaV and lower values of U_ALDV than cirrhotic animals receiving vehicle. Cirrhotic rats receiving niravoline had significantly higher values of creatinine clearance than cirrhotic rats receiving OPC-31260 or vehicle. In these latter two groups of animals, creatinine clearance also was significantly reduced compared with control animals.

View this table: [in this window] [in a new window]   TABLE 3 Average values of UNaV and UALDV during entire period of treatment and creatinine clearance (CCreat) at end of study in control and cirrhotic rats of protocol I receiving -opioid receptor agonist niravoline, the V2 receptor antagonist OPC-31260, or vehicle

A representative Northern blot analysis of total RNA from hypothalamic tissue of control and cirrhotic rats with ascites after completion of the different treatments is shown in Fig. 4. A single band of ADH mRNA corresponding to the size of ~700 bp was detected in the hypothalamus of all the animals. The hybridization products corresponding to GAPDH mRNA that were used as an internal control are also shown (top). The densitometric analysis of Northern blot autoradiographies from animals receiving vehicle demonstrated that cirrhotic rats with ascites have higher ADH transcript abundance than control rats. This difference between cirrhotic and control ADH mRNA expression was also observed in rats receiving the V₂ receptor antagonist. The most interesting finding was that the administration of the -opioid receptor agonist reduced ADH mRNA abundance in cirrhotic rats with ascites, thus resulting in no differences in ADH transcript expression between cirrhotic and control rats chronically treated with niravoline (Fig. 4). The U_ALDV also was diminished only in cirrhotic rats receiving niravoline (Fig. 5).

View larger version (43K): [in this window] [in a new window]   Fig. 4.   Representative Northern blot analysis of ADH mRNA in three groups of control and cirrhotic rats with ascites and water retention after completion of daily oral administration of vehicle, OPC-31260, or niravoline for 10 days. Northern blot was performed with total RNA (3 µg/lane) isolated from the hypothalamus. Membrane was probed with a -32P-labeled 480-bp fragment of rat ADH generated by PCR. Top, expression of rat GAPDH in the same membrane. Bottom, densitometric analysis of ADH Northern blot corrected for GAPDH expression. Northern blot analysis was made with samples from 12 control and 12 cirrhotic rats with ascites.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   Average values of UADHV during entire period of treatment in cirrhotic and control rats receiving vehicle, the V2 receptor antagonist (OPC), or -opioid receptor agonist (niravoline). #p < .001 versus control rats receiving the same treatment. ap < .05 versus cirrhotic rats receiving vehicle (unpaired Student's t test).

As shown in Table 4, no differences were found in the renal response to the basal water overload among the three groups of cirrhotic rats included in protocol II. In fact, similar values of percent water excreted and mU_Osm were recorded during the 3-h collection period after the water load in cirrhotic rats receiving the -opioid receptor agonist, the V₂ ADH receptor antagonist, or the vehicle. On starting the study, no significant differences were found in any of these parameters among the three groups of control rats (Table 4).

The effect on systemic hemodynamics induced by the chronic administration of niravoline, OPC-31260, or vehicle to control and cirrhotic rats is shown in Table 5. Neither -opioid receptor agonist nor V₂ ADH receptor antagonist administration was associated with any significant change in these parameters in control rats. As anticipated, cirrhotic rats receiving vehicle have a marked hyperkinetic circulatory syndrome characterized by arterial hypotension, high CO, and low peripheral resistance. This characteristic circulatory dysfunction in cirrhotic animals did not undergo any significant change after the long-term administration of OPC-31260. In contrast, the chronic administration of the -opioid receptor agonist to cirrhotic rats with ascites and water retention was associated with a remarkable amelioration of the hyperkinetic circulation because these animals had significantly higher TPR and lower CO values than cirrhotic rats receiving either vehicle or OPC-31260.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Three investigations have been performed assessing the effects of a single dose of the nonpeptide ADH V₂ receptor antagonist OPC-31260 or the -opioid receptor agonist niravoline in cirrhotic rats. Tsuboi et al. (1994) examined the renal response to a water overload after the administration of OPC-31260 (5 mg/kg/b.wt.) to rats with CCl₄-induced cirrhosis. These animals experienced a 4-fold increase in water excretion and a 30% reduction in mU_Osm, which was associated with an increase in serum sodium concentration and plasma osmolality over control values. On the other hand, we have shown that niravoline, at a dose of 1 mg/kg/b.wt., increases urine volume and free water clearance; does not change sodium and potassium excretion; and significantly reduces the circulating levels of ADH in cirrhotic rats with ascites and water retention (Bosch-Marcé et al., 1995). Finally, the aquaretic effect of niravoline in experimental cirrhosis has recently been confirmed by Moreau et al. (1996), who demonstrated that this drug induces water diuresis in bile duct-ligated rats. Therefore, an important issue arising from these investigations is a comparison of the long-term effects of the ADH V₂ receptor antagonist and the -opioid receptor agonist on renal water handling in cirrhotic rats.

The chronic administration of both OPC-31260 and niravoline did not produce any significant effect on renal excretory function in control animals. This finding differs from previous investigations in which acute administration of these compounds induced aquaresis in euvolemic rats. Yamamura et al. (1992) showed that oral administration of 5 mg/kg OPC-31260 to normal hydrated conscious rats significantly increased urine volume and decreased U_Osm, although these effects lasted for only 4 h. Similarly, the aquaretic effect of a single oral dose of 3 mg/kg niravoline to conscious rats occurs within 2 h after the administration of this agent (Ginés et al., 1998). Moreover, in previous investigations, we also observed that the diuretic effect of i.v. niravoline in control rats lasts for 1 h and is followed by a decrease in urine volume below baseline values (Bosch-Marcé et al., 1995). This early effect of the aquaretic agents followed by a compensatory reduction in urine volume, probably secondary to a decrease in extracellular fluid volume, explains why the aquaretic effect is not detected in control rats when the urine collection periods are 24 h. In fact, Fujisawa et al. (1993) did not observe changes in the 24-h urine volume after the oral administration of 5 mg/kg OPC-31260 to normal rats.

ADH V₂ receptor antagonist administration to cirrhotic rats resulted in an acute increase in the urinary flow rate and a reduction in U_Osm that lasted for only 2 days. The absence of diuretic activity during the remaining days of treatment could be the consequence of different circumstances. First, it is possible that short-term administration of OPC-31260 was sufficient to correct the impairment in water excretion in cirrhotic rats. Therefore, further administration of this compound would not result in any additional diuretic effect. However, this explanation is unlikely because at the end of the study, cirrhotic rats receiving the ADH V₂ receptor antagonist still showed hypoosmolality and ascites. The second possibility is that as previously reported in studies of acute administration (Ohnishi et al., 1993), the chronic administration of OPC-31260 induced a marked increase in ADH levels, thus counteracting the pharmacological effects of the drug. However, this also seems unlikely because we were unable to detect any further increase in U_ADHV during the entire period of treatment in cirrhotic rats receiving the ADH V₂ receptor antagonist. The most reliable possibility, therefore, is that chronic administration of OPC-31260 induces some tachyphylactic response, thus neutralizing the beneficial effects of this drug on renal water metabolism in cirrhotic rats. The long-term administration of the ADH V₂ receptor antagonist also produced a moderate improvement in U_NaV. This natriuretic effect of OPC-31260 was associated with an important reduction in the U_ALDV. This effect was not observed in cirrhotic rats receiving vehicle.

Cirrhotic rats treated with niravoline experienced a rapid improvement in renal water metabolism. However, the -opioid receptor agonist displayed aquaretic efficacy during the entire period of treatment, and at the end of the investigation, cirrhotic rats receiving niravoline did not show hypo-osmolality: in four of seven, the percentage of the water load excreted was normalized, and all except one displayed values of mU_Osm within the normal range.

The aquaretic activity of -opioid receptor agonists has mainly been attributed to the interaction of these compounds with -opioid receptors at hypothalamic sites, thus resulting in the inhibition of ADH secretion (Leander et al., 1985; Yamada et al., 1989; Rossi and Brooks, 1996). This concept is consistent with the results of the current investigation demonstrating that long-term administration of niravoline to cirrhotic rats with impaired water excretion and nonosmotic hypersecretion of ADH significantly reduces U_ALDV. These findings, however, do not exclude the possibility that niravoline also may have an extra hypothalamic site of action inhibiting ADH activity.

Confirming a previous investigation by Kim et al. (1993), cirrhotic rats receiving vehicle or OPC-31260 showed increased ADH mRNA expression compared with control animals under the same treatment. In contrast, no differences were observed in ADH transcript abundance between cirrhotic and control rats after a 10-day treatment with niravoline. However, whether this reduction in ADH mRNA abundance in cirrhotic rats results from a direct effect of niravoline on ADH transcript expression cannot be elucidated from the current experiment. In vitro studies have shown that -opioid receptor agonist may inhibit ADH release in hypothalamoneurohypophyseal preparations (Rossi and Brooks, 1996), but so far there are no data demonstrating that these compounds may influence ADH transcription.

An unexpected result of the current investigation was the remarkable improvement in sodium excretion and renal perfusion induced by niravoline in cirrhotic rats. In fact, cirrhotic animals chronically treated with the -opioid receptor agonist had normalized U_NaV and creatinine clearance. These results differ from previous investigations in which acute niravoline administration to healthy subjects (Bellisant et al., 1996) or cirrhotic rats with ascites (Bosch-Marcé et al., 1995) resulted in a diminution of or no changes in sodium excretion, respectively. The sodium-retaining properties of -opioid receptor agonists appear to be mediated by activation of the sympathetic nervous system and a concomitant increase in systemic vascular resistance (Kapusta and Obih, 1993; Bellisant et al., 1996). Under normal conditions, these kinds of effects are associated with antinatriuresis. Cirrhosis with ascites, however, is characterized by arterial hypotension, high CO, and low peripheral resistance (Guevara et al., 1998), and it is well known that any maneuver toward correcting arteriolar vasodilation is usually associated with an amelioration in renal excretory function (Nichols et al., 1986; Lenz et al., 1989; Guevara et al., 1998). Therefore, the normalization of sodium excretion and creatinine clearance in cirrhotic rats with ascites chronically treated with niravoline could be explained by an improvement in the deranged systemic hemodynamics, which may be secondary to the activation of the sympathetic nervous system induced by this agent. The marked inhibition of aldosterone excretion observed in cirrhotic rats receiving the -opioid receptor agonist further supports this hypothesis.

The results of protocol II also would be consistent with this interpretation because the administration of niravoline to cirrhotic rats with ascites was associated with an improvement in the systemic hemodynamic parameters. However, it should be pointed out that this ameliorated cardiovascular function is mainly due to a profound reduction in CO, which decreased to values close to those found in control animals. Therefore, an alternative explanation is that the enhancement in renal excretory function produced by niravoline in cirrhotic animals tends to normalize extracellular fluid volume and, as a consequence, ventricular load. However, regardless of the mechanism, the most important finding of this protocol is that the chronic administration of niravoline to cirrhotic rats with ascites is associated with a significant improvement in cardiovascular function rather than with a further deterioration in systemic hemodynamics.

In summary, the present study indicates that in rats with cirrhosis, ascites, and water retention, the long-term oral administration of either -opioid receptor agonists or nonpeptide ADH V₂ receptor antagonists ameliorates water metabolism, increases urine flow and sodium excretion, and reduces U_ALDV. The data indicate, however, that under the conditions studied, the beneficial effects of the -opioid receptor agonist niravoline are more consistent than those of the V₂ receptor antagonist OPC-31260. In fact, at the end of the 10 days of treatment, cirrhotic rats receiving niravoline had normalized serum osmolality, sodium excretion, and creatinine clearance; improved systemic hemodynamics; and showed a significant reduction in hypothalamic mRNA expression and U_ALDV. Conversely, the diuretic effect of OPC-31260 in cirrhotic rats was evident only during the first 2 days of treatment, and no significant effects on serum osmolality, renal perfusion, and systemic hemodynamics were recorded on completion of the 10-day administration period. However, because -opioid agonists may have some adverse effects that are probably related to their central mechanism of action (Bellisant et al., 1996), it is important to emphasize that caution should be taken on extrapolation of these results to the clinical arena.