Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

A primary defect in the ability of the kidney to excrete sodium associated with a secondary increase in the plasma level of the so-called OLF has been proposed as a possible cause of primary hypertension (Blaustein, 1977). This hypothesis predicts that the pressor mechanisms derive from the increased neurovascular reactivity caused by OLF inhibition of the Na-K pump at the sympathetic nervous terminals or the vasculature. Consequently, inhibition of the OLF effect on the Na-K pump should prevent that portion of the rise in blood pressure triggered by such mechanisms.

This hypothesis is supported by much experimental and clinical evidence (Haddy and Overbeck, 1976; Ferrandi et al., 1992; Doris, 1994; Leenen et al., 1994; Kojima et al., 1982; Pamnani et al., 1989; Moreth et al., 1987; Hamlyn et al., 1982; Rossi et al., 1995). Some disagreement still exists over the methodology for measuring OLF in body fluids and tissues, and its chemical structure is still a subject of debate (Lewis et al., 1994; Kelly and Smith, 1992; Gomez-Sanchez et al., 1994). We have recently provided consistent data on the appropriate methodology for measuring OLF both in rats and in humans (Ferrandi et al., 1997). Other research groups have demonstrated that the structure of OLF must be very close, if not identical, to that of ouabain (Mathews et al., 1991; Zhao et al., 1995). In addition, Manunta et al. (1994) have demonstrated that chronic infusion of low doses of ouabain in normal Sprague-Dawley rats increases plasma levels of ouabain by 4- to 5-fold and induces a sustained but reversible form of hypertension. The pressor effect of ouabain has not been consistently demonstrated in other species (Pidgeon et al., 1996), which is not surprising, because according to Blaustein's hypothesis (Blaustein, 1977), both a peculiar renal defect in Na⁺ handling and a high level of OLF, or ouabain, must be present to induce the pressor effect.

A direct consequence of these findings is that OLF could represent a new pharmacological target in the treatment of those forms of volume-dependent hypertension where it plays a pathogenic role. In the past it has been hypothesized that the antihypertensive activity of K-Canrenoate (Pamnani et al., 1990; de Mendoca et al., 1988) might be due not only to its antimineralcorticoid activity but also to a partially antagonistic action on ouabain (or OLF) at the Na-K pump site (Finotti and Palatini, 1981; Garay et al., 1985). However, K-Canrenoate cannot be considered an ideal anti-ouabain-type antihypertensive drug because of its lack of selectivity and because of undesired side effects due to interactions with progesterone and androgen receptors (Corvol et al., 1975; Pita et al., 1975).

The aim of our research was therefore to synthesize a new antihypertensive compound capable of selectively antagonizing the pressor effect of ouabain without causing the cardiac or hormonal side effects typical of digitalis and antimineralcorticoid drugs. In the present report we describe, for the first time, the pharmacological characteristics of a new, orally active, antihypertensive compound, PST 2238, which selectively displaces ouabain from the purified Na-K ATPase receptor in vitro at micromolar concentrations without interacting with other receptors involved in blood pressure or hormonal regulation. Moreover, PST 2238 antagonizes the effect of ouabain on the Na-K pump in cultured cells, at nanomolar concentrations, and prevents the effects of ouabain on blood pressure and Na-K pump activity in vivo at doses of micrograms per kilogram.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Synthesis

Details of the synthesis of PST 2238 have been reported elsewhere (Quadri et al., 1997).

Materials

For in vitro receptor studies, PST 2238 was solubilized up to 3 × 10⁴ M in the appropriate buffers (see below) containing 10% DMSO. For in vivo treatments, PST 2238 was administered p.o., by gavage, in suspension with 0.5% w/v Methocel (methylcellulose) at a volume of 2 ml/kg b.wt. Ouabain was obtained from Sigma Corp. (St. Louis, MO) and solubilized up to 10² M in appropriate buffers or 0.9% NaCl solution for in vitro studies and osmotic minipump loading (see "Hypertensive models"). ³H-PST 2238 (specific activity 21.5 Ci/mmol) was synthesized by the Institute of Nuclear Chemistry of the CNR (Rome, Italy).

In Vitro Studies on Purified Receptors

Na-K ATPase purification. Renal Na-K ATPase was purified from outer medulla of dog kidney according to Jørgensen's method (Jørgensen, 1974). The tissue was homogenized in (mM) 250 sucrose, 30 histidine, pH 7.2 (1 g of tissue/10 ml) using a PCU-Kinematica Polytron. The homogenate was centrifuged at 6000 × g for 15 min at 4°C, and the pellet was discarded. The supernatant was centrifuged at 48,000 × g for 30 min at 4°C, and the pellet (microsomal fraction) was incubated for 30 min at 20°C with SDS, 0.58 mg/ml, and then layered on a discontinuous density gradient of sucrose (10, 15 and 29.4%) and centrifuged at 230,000 × g for 115 min at 4°C. The pellet obtained was resuspended in 25 mM imidazole, 1 mM EDTA solution, pH 7.5. In all experiments the protein content was measured by the Lowry method, using bovine serum albumin as reference standard (Lowry et al., 1951). The presence of the 1 Na-K ATPase isoform was verified by Western blot analysis, using a monoclonal antibody against the 1 isoform (Upstate Biotechnology, Lake Placid, NY).

Na-K ATPase receptor study. Binding displacement of ³H-ouabain from purified dog kidney Na-K ATPase was carried out by the rapid filtration technique (Noel and Godfraind, 1984). Incremental concentrations of either ouabain or PST 2238 were incubated for 45 min at 37°C with 1.2 µg of purified Na-K ATPase in 120 µl of solution A, (mM: 100 NaCl, 3 MgCl₂, 3 ATP, 50 Tris-HCl, pH 7.4) containing 25 nM ³H-ouabain (20-50 Ci/mmol specific activity, Amersham, Milan, Italy). Nonspecific binding, calculated in the presence of 10³ M unlabeled ouabain, accounted for less than 5% of total radioactivity and was subtracted from total ³H-ouabain binding. Free ³H-ouabain from membrane-bound ³H-ouabain was separated by rapid filtration on Whatman glass GF/ C fiber filter sheets using a Brandel 48R Cell Harvester apparatus (Biomedical Research and Development Laboratories, Gaithersburg, MD). Filters were washed twice with 10 ml of cold incubation solution and counted for radioactivity in a liquid scintillation  counter (Beckman LS 5000 CE).

8

6

3

5

3

Na-K ATPase inhibition. Na-K ATPase activity was assayed after the release of ³²P from ³²P-ATP (Doucet et al., 1979). Increasing concentrations of ouabain or PST 2238 were preincubated with 0.3 µg of purified dog kidney enzyme for 10 min at 37°C in a 120-µl final volume of preincubation medium (solution B; mM: 140 NaCl, 3 MgCl₂, 3 ATP, 50 HEPES-Tris, 3 ATP, pH 7.5). After the preincubation period, 10 µl of incubation solution, containing 10 mM KCl and 20 nCi of ³²P-ATP (0.5-3 Ci/ mmol, Amersham, Milan, Italy) was added, and the reaction was continued for 15 min at 37°C. The reaction was stopped by acidification with ice-cold perchloric acid solution at 30% v/v. ³²P was separated by centrifugation with activated charcoal, and the radioactivity was measured by liquid scintillation counting. The inhibitory activity was expressed as percent of a control sample that contained neither ouabain nor product.

General and hormonal receptor binding. We evaluated the in vitro interaction of PST 2238 with the following receptor sites:  and  adrenergic, D₁, D₂, D₃, 5-HT₁, 5-HT₂, H₁, H₂, M₁, M₂, A₁, A₂, associated with the Ca⁺⁺ and K⁺ channels, AT₁, AT₂, ET_A, ET_B, GABA, thromboxane and vasopressin. Crude membrane preparations were obtained from rat, guinea pig and rabbit tissues according to previously described procedures (Greenglass and Bremner, 1979; U'Prichard et al., 1977; Bylund and Snyder, 1976; Kilpatrick et al., 1986; Fields et al., 1977; Blurton and Wood, 1986; Leysen et al., 1982; Hill et al., 1978; Beaumont et al., 1978; Hill and Bowery, 1981; Speth et al., 1979; Honoré et al., 1986; Wang et al., 1987; Javitch et al., 1983; Schoemaker and Langer, 1985; Reynolds et al., 1983; Mihara et al., 1989; Patel et al., 1982; Bruns et al., 1986; Chiu et al., 1990; Whitebread et al., 1991; Williams et al., 1991; Mihara et al., 1994; Shleikh et al., 1989; Gopalakrishnan et al., 1986; Gaines et al., 1988; Vazquez et al., 1989; Catterall et al., 1979). Displacement of the specific bindings by 10⁵ to 10⁴ M PST 2238 was measured after separation of the free ligand from the receptor-ligand complex by the rapid filtration technique.

Cell Culture Studies

Cell culture. NRK-52E cells (epithelial-like cells) (De Larco and Todano, 1978) were purchased from the European Collection of Animal Cell Cultures (ECACC, CRL 1571). Cells were maintained in monolayer on plastic substrate in Dulbecco's modified Eagle's medium (Gibco BRL, Milan, Italy) supplemented with 5% fetal calf serum (Myoclone, Life Technologies), 1% nonessential aminoacids (Sigma), penicillin/streptomycin 100 IU/ml (Gibco BRL, Milan, Italy) and were kept in a 37°C humidified incubator with 5% CO₂. Medium was changed on alternate days. The cells were seeded at 5 × 10⁵ cells/cm² on Transwell filter inserts (Costar Clear, 0.4-µm pore size) in the absence or presence of ouabain or PST 2238 at different concentrations. In some studies, ouabain and PST 2238 were simultaneously present in the medium from seeding. Cells were used for measurement of Na-K pump and intracellular ion content and for cell protein determination 5 days after seeding (typically, seeding on Monday and measurements on Friday). The medium was changed on the third day.

Cell protein content. We assessed cell growth rate at the end of the incubation period, measuring the content of protein extracted by lysing the cells with NaOH 0.1 N and SDS 0.1%.

Cell Na-K pump at V_max. Cells grown on all filters were washed in K⁺-free saline and incubated without external K⁺ for 50 min to load the cells with Na⁺ and reach the maximal activation of the Na-K pump. The ouabain-sensitive component of the K⁺ transport (Na-K pump activity) was determined as the amount of ⁸⁶Rb⁺ uptake, measured in the absence or presence of 10 mM ouabain, which was added to the basolateral side 5 min before the assay to block the Na-K pump quickly and completely. PST 2238, ouabain alone or the combination of the two compounds (according to the protocol) was always present in all filters at the concentrations under study, during both the loading and the uptake procedure. ⁸⁶Rb⁺ uptake was measured during the initial 10 min from restoration of normal external K⁺ concentration (5.4 mM KCl + ⁸⁶RbCl 8 µCi/ml) as described (Bowen, 1992). At the end of the uptake period, the Transwells were washed three times in a large volume (250 ml) of saline. Radioactivity was extracted by lysing the cells with 0.1 N NaOH and SDS 0.1% and counted in a -counter (Beckman 5500). An aliquot of the lysate was used to measure the protein content of each Transwell. Ouabain-sensitive Rb⁺ uptake was expressed as the equivalent K⁺ transport in nanomoles per hour per milligram of protein. We also expressed it as rate constant for Na⁺ (hr¹), correcting the expected value of Na⁺ extrusion for the intracellular Na⁺ content (see below).

Intracellular Na⁺ content at V_max. A different set of filters were used to measured the intracellular Na⁺ content after Na⁺ loading. The filters were washed four times in a Na⁺-free medium (mM): choline-Cl 95, MgCl₂ 1, sucrose 85, glucose 10, MOPS-Tris buffer 10, pH 7.4, at room temperature. Sodium was extracted in double-distilled water and measured by atomic absorption spectrophotometry (Perkin Elmer 1100B). The intracellular Na⁺ content was expressed as nanomoles of Na⁺ per milligram of protein.

In Vivo Studies

Effect of PST 2238 on the pressor response to vasoactive compounds. Male Wistar rats, (Charles River Italia SpA, Calco, Italy), aged 4 to 5 weeks and weighing 80 to 100 g were treated p.o., daily for 6 weeks, with either vehicle (Methocel 0.5% w/v) (controls) or PST 2238 10 mg/kg/day. At the end of this treatment the rats were catheterized, under light halothane anesthesia, with a polyethylene tubing (PE50, Dow Chemical, Lepetit, Italy) via the carotid for blood pressure recording and via the jugular vein for infusions. Rats were allowed to recover from surgery for 4 hr, and then SBP, MBP, DBP and HR were recorded by connecting the carotid catheter to a pressure transducer (XCDR-P23XL, Gould, Cernusco-Milano, Italy) attached to a multichannel recorder (Gould Instrument Systems, Mod. 3400). After 1 hr of stabilization (basal recording) and 5 hr from the last PST 2238 or vehicle treatment, each rat was sequentially injected i.v. with noradrenaline (1 µg/kg), ACh (7.5 µg/kg), angiotensin II (0.1 µg/kg) and rat renin extract (3 µl/rat) purified from plasma of nephrectomized rats, according to a previously described procedure (Haas et al., 1966). Basal blood pressure and the acute pressor responses to the vasoactive substances were monitored continuously.

Hypertensive rat model. A ouabain-dependent hypertension was induced in normotensive Sprague-Dawley (Hsd:S.D.) rats (Harlan Sprague-Dawley Inc., Indianapolis, IN) according to a previously described procedure (Manunta et al., 1994). Briefly, male rats 6 to 7 weeks old and weighing 150 to 180 g were implanted s.c., under light ether anesthesia, with osmotic minipumps (Alzet, Charles River, Calco, Italy) containing a ouabain-saline solution that slow-released 50 µg/kg/day of ouabain at a mean pumping rate of 10 µl/day (OS rats). CS rats received sterile saline through osmotic minipumps. The pumps were changed every 15 (Mod 2002) or every 28 days (Mod 2004). SBP and HR were measured weekly at the tail by plethysmography ("W+W' BP Recorder, U. Basile, Varese, Italy). From 3 to 4 weeks after implantation of the minipumps, SBP rose significantly in 70% to 80% of the ouabain-treated rats, by 20 to 25 mmHg, from initial average values of 145 mmHg. In saline-treated control rats, SBP was unchanged. HR was not affected in either group. At this point, OS rats were divided into groups, which received either PST 2238 or vehicle (Methocel 0.5% w/v). SBP and HR were recorded weekly, 6 hr after treatment.

Biochemical assays: Preparation of renal outer medulla microsomes. Rats were anesthetised with ether and then sacrificed by decapitation. Kidneys were removed, weighed and sliced, and the outer medulla was dissected under a stereo microscope at 4°C, pooled, weighed and then frozen in liquid nitrogen and stored at 70°C up to the moment of microsome preparation. Kidney outer medulla slices were suspended (1 g/10 ml) in an ice-cold solution containing (mM): 250 sucrose, 30 histidine and 5 disodium EDTA, pH 7.2, and homogenized in a PCU-Kinematica Polytron. Each sample was homogenized twice for 15 sec at setting 5. The homogenate was centrifuged at 6000 × g for 15 min at 4°C (J2-21 M/E, Beckman Instruments). The supernatant fluid was decanted and saved, and the pellet was resuspended in the same solution, homogenized and centrifuged at 6000 × g for 15 min at 4°C. The second supernatant was decanted, pooled with the first and centrifuged at 48,000 × g for 30 min at 4°C. Pellets were resuspended 1:1 (w/v) in the sucrose-histidine solution. The protein content of the microsomes was assayed. Na-K ATPase enzymatic activity was determined in microsome preparations previously permeabilized with deoxycholic acid (0.65 mg DOC/mg protein) for 30 min at room temperature to obtain the maximal activation of the Na-K ATPase (Ferrandi et al., 1996). Na-K ATPase activity was assayed as already described.

Ouabain content in plasma and tissue. Ouabain was extracted from plasma and freshly thawed renal tissue and measured by RIA assay according to a previously described procedure (Ferrandi et al., 1997). Ouabain concentrations in plasma and tissues extracts were calculated as percentage displacement of the control sample, carried out in the absence of ouabain, and expressed in nM (plasma) or nanograms per gram of tissue according to a ouabain standard curve.

Statistics

Data are reported as mean ± S.E.M. Regression analysis was used to calculate IC₅₀ values in the in vitro tests (Motulsky and Ransnas, 1987). Factorial one-way analysis of variance (ANOVA) followed by Fisher's LSD was performed to test the differences among different compound concentrations in cell culture studies. Factorial two-way ANOVA for repeated measures was performed to test the interaction of time and in vivo treatments. Factorial one-way ANOVA was then performed to test the different groups vs. the control group at different times and different times vs. base-line time in each group. Dunnett's t test was used to determine the significance of the F ratio; P < 0.05 was considered significant for all comparisons.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

PST 2238 In Vitro Receptor Interaction

Interaction with the Na-K ATPase. Ouabain inhibited purified dog kidney Na-K ATPase and displaced ³H-ouabain from the receptor with a similar IC₅₀ of 2.5 × 10⁸ M (fig. 1). Under the same experimental conditions used to assay ouabain, PST 2238 inhibited the enzyme with an IC₅₀ of 2.5 × 10⁵ M and displaced ³H-ouabain with an IC₅₀ of 1.7 × 10⁶ M (fig. 1). Therefore, a 15-fold difference between the inhibitory and the displacing IC₅₀ values was observed for PST 2238. To determine whether this difference could be ascribed to the presence of K⁺ during the Na-K ATPase inhibition test, we also conducted experiments at low K⁺ concentration (0.5 mM). At this low K⁺ concentration ouabain, IC₅₀ was unchanged (2.5 × 10⁸ M) compared with that obtained in the presence of 10 mM K⁺. On the contrary, the IC₅₀ of PST 2238 at low K⁺ was 1.5 × 10⁶ M, a value 16.6 times lower than that obtained in the presence of 10 mM K⁺ and similar to that of the displacement experiment. Thus the difference between the inhibitory and displacing activities of PST 2238 should be ascribed to a decrease in affinity for the Na-K ATPase when K⁺ was added.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Inhibition (closed symbols) of Na-K ATPase activity and displacement (open symbols) of 3H-ouabain from dog kidney Na-K ATPase by ouabain (,) or PST 2238 (,). Mean ± S.E.M. of three experiments run in triplicate.

8

6

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Curves of association to (panel a) and dissociation from (panel b) dog kidney Na-K ATPase of 5 × 108 M ouabain () and 1 × 106 M PST 2238 (). The rates of association and dissociation were measured by preincubating the enzyme with 3H-ouabain or 3H-PST 2238 in the absence of K+ and then measuring the amount of traced compound bound to the enzyme at increasing times. The insets show the association and dissociation curves only for PST 2238, referred to a time scale in seconds. One of three similar experiments run in triplicate.

Interaction with general and hormonal receptors. To assess their specificity of interaction, we assayed PST 2238, ouabain and Canrenone in vitro on a panel of general and hormonal receptors. PST 2238, at fixed concentrations of either 10⁵ M or 10⁴ M, did not show any significant interactions with the tested receptors (table 1). Only with the thromboxane A₂ receptor did PST 2238 10⁴ M show any interaction (61%). Ouabain was ineffective at 10⁵ M in all the receptor binding assays (table 1). As expected, Canrenone showed a specific interaction with the mineralocorticoid receptors and also some interactions with the M₁ and M₂ receptors (table 1). It should be noted that Canrenone did not show any affinity for the Na-K ATPase receptor up to the maximum possible concentration of 5 × 10⁴ M.

PST 2238 Effect on the Na-K Pump of NRK Cells

The activity of the Na-K pump in NRK cells in culture was assessed as a function of the ouabain-sensitive ⁸⁶Rb uptake measured in Na⁺-loaded cells at V_max. In these experiments, intracellular Na⁺ concentration was raised from 140 nmole/mg protein (basal conditions) to 450 nmole/mg protein (maximal intracellular Na concentration). This resulted in a maximal activation (V_max) of the Na-K pump (+500% compared with basal conditions) and gives an indirect evaluation of the total number of activable pump sites (Bowen, 1992).

Acute effect (5-hr incubation). As shown in figure 3, ouabain inhibited the Na-K pump with IC₅₀ of 2.08 × 10⁴ M, whereas PST 2238 was ineffective up to 10⁴ M (maximal concentration tested). These results are similar to those obtained measuring the in vitro inhibitory activity of both ouabain and PST 2238 on purified rat renal Na-K ATPase: ouabain showed an IC₅₀ of 7 × 10⁵ M, as already reported (Anner et al., 1995), whereas PST 2238 did not inhibit the enzyme up to 10⁴ M. In the cell culture experiments, PST 2238 affected neither the Na-K cotransport nor the passive permeability, measured as bumetanide-sensitive, ouabain-resistant uptake and as bumetanide and ouabain-resistant ⁸⁶Rb uptake, respectively (data not shown).

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Inhibition of the Na-K pump rate by ouabain (panel a, ) or PST 2238 (panel b, ), measured at Vmax as ouabain-sensitive Rb uptake, in cultured rat renal epithelial cells. The Na-K pump rate was measured after 5 hr of incubation with different concentrations of ouabain or PST 2238. Data are reported as percentage of the Na-K pump rate at Vmax of control samples run in parallel in the absence of both compounds. Mean ± S.E.M. of three experiments, each run in triplicate. In panel b), ouabain 102 M was inserted for comparison.

Prolonged (5 days) incubation. NRK cells were incubated with both compounds separately or in combination. Under these conditions, neither ouabain (from 10¹⁴ to 10⁵ M) nor PST 2238 (from 10¹⁶ to 10⁵ M) affected the cell growth rate, as indicated by the fact that the amount of proteins found on Transwell filters was not different from controls (data not shown). Figure 4A shows that ouabain increased the rate constant of the Na-K pump at V_max by 37% at 10¹⁰ M, by 33% at 10⁹ M (P < .05) and by 24% at 10⁸ M. PST 2238 did not affect the Na-K pump rate in the range of 10¹⁶ to 10⁹ M and slightly stimulated it at 10⁶ M (+23%) and 10⁵ M (+28%, P < .05; fig. 4B). Both Na-K cotransport and the passive permeability were unaffected by PST 2238 at all the test concentrations, even after 5 days of incubation (data not shown).

View larger version (45K): [in this window] [in a new window]   Fig. 4.   Effect of 5-day incubation with different concentrations of ouabain (panel a), PST 2238 (panel b) or a combination of ouabain 109 M with PST 2238 1010 and 106 M (panel c) on the Na-K pump rate at Vmax of NRK cells. Data are reported as the percentage of the Na-K pump rate at Vmax of control samples run in parallel in the absence of both compounds. Mean ± S.E.M., number of replicates reported in the bars. * P < .05 significantly different from the control. § P < .05 significantly different from the ouabain 109 M sample.

9

9

9

10

6

14

9

16

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Regression analysis of the dose-dependent inhibition by PST 2238 () of ouabain's effect () on the Na-K pump rate of NRK cells, after 5 days of incubation. Data are mean ± S.E.M. of the percentage of 109 M ouabain samples, n = 4. r = 0.938, P < .02.

PST 2238 In Vivo Experiments

Modulation of the pressor responses to vasoactive compounds. As shown in figure 6 the i.v. injections of fixed doses of noradrenaline, ACh, angiotensin II and renin to control rats elicited a general 30% to 35% change in MBP over the basal values. These pressor responses were completely superimposable on those recorded in rats previously treated p.o. with 10 mg/kg/day of PST 2238 for 6 weeks, a dose 100 to 10,000 times higher than those active on ouabain-dependent hypertension (see the next paragraph). These data exclude the possibility that the compound interacts with pressor mechanisms dependent on the vasoactive substances used in this study.

View larger version (42K): [in this window] [in a new window]   Fig. 6.   Pressor responses (percent of basal mean blood pressure, MBP) to i.v. bolus injections of noradrenaline, ACh, angiotensin II and renin in Wistar rats treated p.o. for 6 weeks with vehicle (Methocel 0.5% w/v, open bars) or PST 2238 10 mg/kg/day (hatched bars). Two cycles of pressor responses, recorded 5 and 6 hr after the last p.o. treatment, are reported. Mean ± S.E.M., n = 8 for each group.

Antihypertensive activity in OS rats. The blood pressure-lowering effect of orally administered PST 2238 and its reversibility were measured in hypertensive OS rats treated for 4 weeks. Figure 7 shows the hypertensive effect of 8 weeks of chronic infusion of 50 µg/kg/day of ouabain in normotensive Sprague-Dawley rats. SBP started to rise after 2 to 3 weeks of ouabain infusion and reached statistically significant higher levels in OS rats (P < .05), compared with CS rats, after 4 weeks (fig. 7). The SBP of OS rats decreased to the level of CS controls after 2 weeks of treatment with PST 2238 at 0.1 mg/kg/day (fig. 7). After suspension of treatment (washout period) the SBP of PST 2238-treated OS rats returned to the high levels of OS controls within 2 weeks (fig. 7). This demonstrates that the increase in SBP caused by ouabain infusion (+22 mmHg) was abolished by PST 2238. 

View larger version (25K): [in this window] [in a new window]   Fig. 7.   SBP in adult Sprague-Dawley rats receiving, through osmotic minipumps, ouabain (50 µg/kg/day) (,; OS rats) or saline (; CS rats) for 10 weeks. Starting from the fourth week of ouabain infusion, one group of OS rats was treated p.o. with PST 2238 at 0.1 mg/kg/day (circles, ), while the second group of OS rats () and the CS rats () received only vehicle (Methocel 0.5% w/v; OS and CS controls). A washout period of 2 weeks from the PST 2238 treatment started at the end of the eighth week. Data are mean ± S.E.M. of seven rats for each group. * P < .05; ** P < .01 significantly different from OS control rats.

View larger version (47K): [in this window] [in a new window]   Fig. 8.   a) SBP in OS and CS rats after 4 weeks of p.o. treatment with vehicle (Methocel 0.5% w/v, black bars), PST 2238 0.1-1-10-100 µg/kg/day (OS) and PST 2238 100 µg/kg/day (CS). b) The Na-K ATPase activity was measured in kidney outer medulla microsomes from the same rats. The weight of the outer medulla was corrected for the total kidney weight. Data are mean ± S.E.M. of eight OS and seven CS rats for each group. * P < .05; ** P < .01 significantly different from OS control rats.

View larger version (19K): [in this window] [in a new window]   Fig. 9.   SBP during 4 weeks of p.o. treatment of OS and CS rats with PST 2238 at 0.01, 1 and 100 µg/kg/day. Data are mean ± S.E.M. of eight rats for each group. * P < .05; ** P < .01 significantly different from OS control rats.

Effect on renal Na-K ATPase and on tissue and plasma ouabain concentration. Because of the specific interaction of ouabain and PST 2238 with the Na-K ATPase receptor and their activities on the Na-K pump rate of NRK cells, we verified whether in vivo long-term treatment with these two compounds could influence the activity of the Na-K ATPase at the kidney level. The activity of the renal outer medulla Na-K ATPase was quantified in OS rats at the end of the p.o. treatment with PST 2238 at 0.1, 1, 10 and 100 µg/kg and in CS rats treated with 100 µg/kg. As shown in figure 8B, renal Na-K ATPase activity was significantly increased in OS control rats, as compared with CS controls, and was normalized to the levels of CS controls by the PST 2238 treatment at all doses tested. In contrast, PST 2238 at 100 µg/kg/day did not affect renal Na-K ATPase activity in CS rats (fig. 8B). In the third experiment, the renal Na-K ATPase activity, measured in OS rats treated with PST 2238 at the dose of 0.01 µg/kg/day (which did not affect SBP, fig. 9), was similar (1.69 ± 0.071 µmol P_i/min/mg) to that of OS controls (1.67 ± 0.065 µmol P_i/min/mg), and both values were significantly increased as compared with CS controls (1.125 ± 0.12 µmol P_i/min/mg, P < .01).

View larger version (36K): [in this window] [in a new window]   Fig. 10.   Plasma concentration (panel a) and renal content (panel b) of ouabain after 4-week p.o. treatment of OS and CS rats with different doses of PST 2238 (0.1, 1, 10 and 100 µg/kg/day). Data are mean ± S.E.M. of eight OS and seven CS rats for each group. ** P < .01; *** P < .001 significantly different from OS control rats.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

In this report we describe the in vitro and in vivo pharmacological characteristics of a new antihypertensive molecule that selectively interacts with the Na-K ATPase and antagonizes the pressor effect of ouabain. A series of new derivatives of digitoxigenin have been synthesized and screened for their selectivity for the Na-K ATPase receptor and then for their antihypertensive activity in a ouabain-hypertensive rat model (Manunta et al., 1994). Selectivity for the Na-K ATPase receptor was tested firstly by measuring the affinities of the compounds for the dog kidney enzyme. This tissue was chosen because it is known to express the 1 isoform (Sweadner et al., 1994), which binds ouabain with high affinity (K_d = 2.5 × 10⁸ M). This preparation permits screening molecules with a wide range of affinities for the Na-K ATPase 1 receptor, including those less potent than ouabain, that could be lost using tissue preparations from the so-called ouabain-resistant species, such as rat (Sweadner, 1989). Second, all the molecules that demonstrated an IC₅₀ for the Na-K ATPase receptor lower than 10⁵ M were assayed for their ability to interact with a wide panel of general and hormonal receptors known to be involved in blood pressure regulation. All the molecules that showed some affinity for receptors other than Na-K ATPase were excluded from the in vivo tests. The ability of the selected compounds to antagonize in vivo the pressor effect of ouabain was tested in the OS. Even though the rat is considered a ouabain-resistant species, it responds to very low doses of ouabain with a sustained and reversible increase in blood pressure (Manunta et al., 1994) and, as shown in this study, with an up-regulation of the renal Na-K ATPase activity. Furthermore, the ability of the selected compounds to antagonize at the cell level the long-term effects of low concentrations of ouabain on the Na-K pump rate was assayed on rat renal cells in culture, where only the 1 Na-K ATPase isoform is present (Whorwood et al., 1994).

Following this protocol, the most interesting compound selected up to now is PST 2238, which displays the following pharmacological characteristics:

	1)  Selectivity for the Na-K ATPase receptor with affinity in the micromolar range, demonstrated by the absence of a relevant interaction with other receptors known to be involved in blood pressure regulation. Despite its steroidal structure, PST 2238 does not affect the receptors for steroidal hormones such as estrogens, progestinics, androgens and mineralocorticoids. Therefore, PST 2238 might be devoid of those side effects typical of the antimineralocorticoids such as Canrenone and Spironolactone (Corvol et al., 1975).
	2)  Ability to reverse the ouabain-dependent increase in the Na-K pump rate in cultured renal cells at less than nanomolar concentrations, without itself affecting the Na-K pump.
	3)  Ability to reverse completely the two main effects produced by chronic infusion of ouabain in vivo: hypertension and up-regulation of the renal Na-K ATPase activity, at p.o. doses (0.1-1 µg/kg) compatible with the nanomolar concentrations active in intact cells. Both the results obtained in intact cultured cells and those obtained in vivo in ouabain-hypertensive rats indicate that the chronic exposure of a cellular system to low concentrations of ouabain induces an up-regulation of the Na-K ATPase activity that is selectively antagonized by low concentrations of PST 2238.

The mechanism through which ouabain produces a pressor effect in the rat has not been fully elucidated, but according to Blaustein's hypothesis (Blaustein, 1977), a chronic inhibition of the Na-K ATPase in target organs such as the vasculature and the nervous system may induce an increase in the intracellular Na⁺ and, in turn, Ca⁺⁺ concentrations. These changes are responsible for enhanced vasoconstriction and nervous reactivity, which causes an increase in the peripheral vascular resistance and thus hypertension. Moreover, it has also been shown that the prolonged and continuous infusion of ouabain in normal rats increases the concentration of this steroid in plasma, kidney, hypothalamus and pituitary (Manunta et al., 1994). The accumulation of ouabain by these organs may be relevant because these tissues are involved in the long-term regulation of blood pressure. In particular, the hypothalamus, which contains high-affinity ouabain binding sites (Sweadner, 1989), can play a role in the central regulation of blood pressure after ouabain infusion (Huang et al., 1994).

The results presented here show that chronic infusion of ouabain in rat stimulates renal Na-K ATPase activity at V_max. This means that, in vivo, low doses of ouabain can, in the long-run, affect the rat Na-K ATPase 1 isoform, which is highly resistant to ouabain binding when studied as an isolated enzyme. It is unknown whether this effect, observed in renal microsomes, mirrors an in vivo increase in the Na⁺ transport rate in intact tubular cells, but if such an increase is present, it may cause Na⁺ retention and thus hypertension. The observation that administration of digitalis-like compounds in vivo increases tissue Na-K ATPase activity is not new. In fact, an up-regulation of Na-K ATPase activity in guinea pig heart (Bluschke et al., 1976), rat liver (Lindsay and Parker, 1976) and rat heart, skeletal muscle and renal medulla (Wai Ching Li et al., 1993) has already been described after chronic digitalization. In recent years, a number of important studies have focused on the long-term mechanisms by which the Na-K pump responds to ionic stimuli or chronic inhibition to maintain transmembrane ion gradients, transepithelial transports and cell excitability (Tang and McDonough, 1992; Rayson, 1989; Pollack et al., 1981). These long-term mechanisms involve the processes of synthesis, membrane insertion, internalization and degradation of the Na-K pumps that determine the level of expression of this enzyme at the cell surface. Many experimental data support the notion that the pool of the two Na-K pump subunits,  and , changes in a coordinated way after an inhibitory stimulus (Tang and McDonough, 1992; Rayson, 1989; Pollack et al., 1981). In some cell types, such as renal tubular cells, chronic inhibition of the Na-K pump induced by low extracellular K⁺ or ouabain treatment increases the expression of the  subunit mRNA present proportionally to the final Na-K ATPase activity measurable on the cell membrane (Tang and McDonough, 1992; Rayson, 1989; Lescale-Matys et al., 1993; Pollack et al., 1981). In this way the cell can reestablish the equilibrium in Na⁺ and K⁺ ion gradients altered by pump inhibition (Rayson and Gupta, 1985). It has been suggested that coordinated increase of intracellular Na⁺ and Ca⁺⁺ is the triggering mechanism for the Na-K pump synthesis (Rayson, 1993).

The data available so far do not permit precise elucidation of the molecular mechanism through which PST 2238 antagonizes the effect of ouabain on Na-K pump expression and activity, partly because the ouabain mechanism itself has not yet been clearly explained. However, some findings related to the effects of chronic ouabain exposure on the Na-K pump may be relevant to this discussion. Long-term but not short-term incubation of cells with submaximal inhibitory concentrations of ouabain increases the affinity for digitalis (Griffiths et al., 1991). This long-term exposure is accompanied by three parallel feedback mechanisms: 1) replacement of the blocked pumps with new functioning sites and parallel internalization of the drug, 2) increased transport activity of unoccupied pump sites, transiently stimulated by an increase of intracellular Na⁺ and 3) final enhancement of the total number of pump sites (Pollack et al., 1981). This latter process is driven both by a transient increase in the rate of synthesis and by a subsequent decrease in the pump degradation rate, causing a prolongation of the time that Na-K pumps are resident on the cell membrane surface (Rayson, 1989). It is possible that PST 2238, which is more lipophylic than ouabain, reversibly binds to the Na-K pump and interferes with the effects of ouabain on Na-K pump turnover.

The possibility that PST 2238 is internalized into the cell and interacts with some unknown intracellular receptors or second messengers that affect Na-K pump synthesis cannot be excluded. In fact, PST 2238 is internalized by the cell to a concentration 2- to 3-fold higher than the extracellular concentration (L. Torielli, unpublished results).

Finally, it is possible that PST 2238, besides exerting its evident effect on the renal Na-K ATPase 1 isoform, may also act in vivo by interfering with the ouabain-sensitive Na-K ATPase isoforms (2 and 3) in target organs such as the vasculature or the nervous system. This and other hypotheses deserve further study to clarify the molecular mechanism(s) of PST 2238. However, the ability of this new compound to reverse the ouabain-dependent increase in renal Na-K ATPase suggests that it may act by normalizing the alterations in tubular ion transport caused by ouabain or OLF.

Could an antagonist of the long-term pressor effect of ouabain be useful for the therapy of human primary hypertension? It is known that increased levels of "endogenous ouabain" or OLF are present in a subgroup of essential hypertensive patients (Hamlyn et al., 1982; Rossi et al., 1995), in some cardiovascular disorders (Bagrov et al., 1994; Delva et al., 1991) and in animal models of genetic hypertension (Ferrandi et al., 1992; Doris, 1994; Leenen et al., 1994). In rats of the Milan strain, where plasma and tissue OLF levels are increased (Ferrandi et al., 1992), the development of hypertension is paralleled by an up-regulation of the expression of Na-K ATPase at the renal level (Ferrandi et al., 1996). Also in this strain, preliminary data show that PST 2238 reduces the development of hypertension and normalizes the increased renal Na-K ATPase activity (Ferrari et al., 1995). It is likely that new antihypertensive compounds, such as PST 2238, that are able to modulate the effect of ouabain or OLF on the expression of Na-K ATPase can be useful for the therapy of those patients in whom these pathogenic mechanisms are the cause of hypertension.