Alterations in phenylephrine-induced contractions and the vascular expression of Na⁺,K⁺-ATPase in ouabain-induced hypertension

Abstract

1. Hypertension development, phenylephrine-induced contraction and Na⁺,K⁺-ATPase functional activity and protein expression in aorta (AO), tail (TA) and superior mesenteric (SMA) arteries from ouabain- (25 μg day⁻¹, s.c., 5 weeks) and vehicle-treated rats were evaluated.

2. Ouabain treatment increased systolic blood pressure (127±1 vs 160±2 mmHg, n=24, 35; P<0.001) while the maximum response to phenylephrine was reduced (P<0.01) in AO (102.8±3.9 vs 67.1±10.1% of KCl response, n=12, 9) and SMA (82.5±7.5 vs 52.2±5.8%, n=12, 9).

3. Endothelium removal potentiated the phenylephrine response to a greater extent in segments from ouabain-treated rats. Thus, differences of area under the concentration-response curves (dAUC) in endothelium-denuded and intact segments for control and ouabain-treated rats were, respectively: AO, 56.6±9.6 vs 198.3±18.3 (n=9, 7); SMA, 85.5±15.4 vs 165.4±24.8 (n=6, 6); TA, 13.0±6.1 vs 39.5±10.4% of the corresponding control AUC (n=6, 6); P<0.05.

4. The relaxation to KCl (1 – 10 mM) was similar in segments from both groups. Compared to controls, the inhibition of 0.1 mM ouabain on KCl relaxation was greater in AO (dAUC: 64.8±4.6 vs 84.0±5.1%, n=11, 14; P<0.05), similar in SMA (dAUC: 39.1±3.9 vs 43.3±7.8%, n=6, 7; P>0.05) and smaller in TA (dAUC: 62.1±5.5 vs 41.4±8.2%, n=12, 13; P<0.05) in ouabain-treated rats.

5. Protein expression of both α₁ and α₂ isoforms of Na⁺,K⁺-ATPase was augmented in AO, unmodified in SMA and reduced in TA from ouabain-treated rats.

6. These results suggest that chronic administration of ouabain induces hypertension and regional vascular alterations, the latter possibly as a consequence of the hypertension.

Introduction

Na⁺,K⁺-ATPase, the biochemical expression of the electrogenic Na⁺ pump, is responsible for the maintenance of cellular membrane potential and can contribute to the regulation of tone and contractility in the vasculature (Blaustein, 1993). The existence of an endogenous circulating Na⁺,K⁺-ATPase inhibitor in humans and other mammals has been shown (Buckalew & Haddy, 1989). This inhibitor was characterized as ouabain or an isomer of ouabain (Mathews et al., 1991), and is found in plasma, brain and adrenal cortex (Ludens et al., 1992; Huang & Leenen, 1996). Elevated circulating levels of endogenous ouabain have been demonstrated in humans with essential hypertension (Hamlyn et al., 1982) and congestive heart failure (Gottlieb et al., 1992) as well as in several animal models of hypertension (Overbeck et al., 1976; Hamlyn et al., 1996). Ouabain seems to contribute to the regulation of blood pressure and is associated with the development and maintenance of hypertension by central and peripheral mechanisms (Blaustein, 1993; Huang & Leenen, 1996).

Regarding peripheral mechanisms, ouabain in the nanomolar range potentiates agonist-induced contractions in isolated vessel preparations (Rossoni et al., 1999). At micromolar concentrations, ouabain induces contraction (Marín et al., 1988) by a direct action on the vascular smooth muscle (myogenic component), or by the release of norepinephrine from the perivascular adrenergic nerve endings (neurogenic component) (Marín et al., 1988; Rodríguez-Mañas et al., 1992). In addition, this glycoside can trigger the release of endothelium-dependent relaxing factors such as endothelium-derived hyperpolarizing factor (EDHF) or nitric oxide (NO), counteracting the vasoconstriction and the sensitization of the vascular smooth muscle induced by ouabain (Xie et al., 1993; Ponte et al., 1996; Rossoni et al., 1999). Moreover, unknown endothelial factors have been suggested to possibly stimulate the sodium pump and partially protect against the inhibition of Na⁺,K⁺-ATPase produced by ouabain (see Marín & Redondo, 1999).

Several reports have shown that the chronic administration of ouabain to Wistar rats induces hypertension (Yuan et al., 1993; Manunta et al., 1994; 2000; Huang et al., 1994; Veerasingham & Leenen, 1999; Huang & Leenen, 1999), although others have not found a hypertensive effect after ouabain treatment (Li et al., 1995; Pidgeon et al., 1996; Cargnelli et al., 2000). Ouabain-induced hypertension has also been associated with an increase in sympathetic activity and impairment of the arterial baroreceptor reflex (Huang & Leenen, 1999). However, whether this altered vascular reactivity participates in ouabain-induced hypertension is still unclear.

Thus, the goal of the present study was to determine if chronic treatment with ouabain induced hypertension in Wistar rats and whether this hypertension was associated with changes in vascular reactivity and in the functional activity and protein expression of Na⁺,K⁺-ATPase. We have evaluated the effect of chronic treatment with ouabain on: (1) the vascular responses to phenylephrine and acetylcholine; (2) the endothelial modulation of vascular reactivity; (3) the relaxation to K⁺ in the absence and the presence of ouabain, as an indicator of the functional activity of the Na⁺,K⁺-ATPase; and (4) the protein expression of isoforms of Na⁺,K⁺-ATPase α subunit. Because regional differences in vascular reactivity and endothelial modulation of α-adrenoceptor agonists (Oriowo et al., 1987; Tabernero et al., 1996) as well as in the Na⁺,K⁺-ATPase activity (Webb & Bohr, 1980; Myers et al., 1987) have been reported, this study was performed in three different vessel preparations: thoracic aorta, superior mesenteric and tail artery. This approach was employed to avoid bias resulting from basing a model characterization on a single vascular bed.

Methods

Animals

Six-week-old male Wistar rats were obtained from colonies maintained at the Animal Quarters of the Facultad de Medicina of the Universidad Autónoma of Madrid. During treatment, rats were housed at a constant room temperature, humidity and light cycle (12-h light/dark). Rats had free access to tap water and were fed with standard rat chow ad libitum. All experiments comply with the current Spanish and European laws (RD 223/88 MAPA and 609/86).

Pellet implantation

Under anaesthesia with diethyl ether (Panreac, Barcelona, Spain), a small incision was made in the back of the neck and one controlled time-release pellet (Innovative Research of America, Florida, U.S.A.) containing either ouabain (0.5 mg pellet⁻¹, n=35) or vehicle (n=24) was implanted subcutaneously, as described by Huang et al. (1994). These pellets are designed to release a constant amount of ouabain (approximately 25 μg day⁻¹) or vehicle for a 60-day period.

Blood pressure measurements

Indirect systolic blood pressure was measured once a week for 5 weeks by the tail-cuff plethysmographic method (Letica, Digital Pressure Meter, LE 50000 and Pressure Cylinder LE 5100). For this, conscious rats were restrained for 5 – 10 min in a warm and quiet room, and were conditioned to numerous cuff inflation – deflation cycles by a trained operator. The average values for systolic blood pressure were subsequently obtained from 10 to 15 sequential cuff inflation – deflation cycles.

After 5 weeks, the rats were anaesthetized with diethyl ether and sacrificed by exanguination and the thoracic aorta, superior mesenteric and tail arteries were carefully dissected out and cleaned of connective tissue. For reactivity experiments the arteries were divided into segments of 4 mm in length. For analysis of Na⁺,K⁺-ATPase expression, arteries were rapidly frozen in liquid nitrogen and kept at −70°C until the day of analysis.

Reactivity experiments

For isometric tension recording, each segment was set up in an organ bath containing 5 ml of Krebs Henseleit solution (KHS) at 37°C continuously bubbled with a 95% O₂ – 5% CO₂ mixture, pH=7.4. Two horizontally arranged stainless steel pins, 150 μm in diameter, were passed through the lumen of the vascular cylinder. One pin was fixed to the organ bath wall, while the other was vertically connected to a strain gauge for isometric tension recording. The isometric contraction was recorded through a force-displacement transducer (Letica TRI 011, Barcelona, Spain) connected to a recorder (MacLab/ 41 ADInstruments Pty Ltd, Castle Hill, Australia). Thoracic aorta segments were subjected to a tension of 9.8 mN (optimal resting tension), and superior mesenteric and tail segments were subjected to a tension of 4.9 mN. This tension was readjusted every 15 min during a 45 min equilibration period before drug administration.

Vessels were initially exposed to 75 mM KCl to check their functional integrity. Afterwards, concentration-response curves to acetylcholine (0.01 nM – 10 μM) were made in arterial segments previously contracted with phenylephrine at a concentration (approximately 0.1 μM for thoracic aorta and approximately 1 μM for mesenteric and tail arteries) that produced close to 50% of the contraction induced by K⁺ (75 mM). After a washout period of 60 min, concentration-response curves to phenylephrine were constructed by its cumulative addition (0.1 nM – 0.3 mM).

The functional activity of Na⁺,K⁺-ATPase was indirectly measured in another set of experiments using the method described by Webb & Bohr (1978); after stabilization in KHS with 4.7 mM K⁺, the arteries were incubated in a K⁺-free medium for 30 min. Afterwards, the vessels were contracted with phenylephrine to obtain approximately 50% of the contraction induced by K⁺ (75 mM); once a plateau was reached, K⁺ (1 – 10 mM) was cumulatively added. To determine the effect of 0.1 mM ouabain on these responses, the curve to K⁺ was performed in another group of vessels preincubated for 30 min in a K⁺-free medium with ouabain.

In some experiments, the influence of the endothelium on the concentration-response to phenylephrine and to K⁺ was determined. In these studies, the endothelium was mechanically removed by rubbing the lumen with a needle. The absence of endothelium was confirmed by the inability of 10 μM acetylcholine to induce relaxation.

Western blot analysis of isoforms of α-subunit of the Na⁺,K⁺-ATPase

Thoracic aorta, superior mesenteric and tail arteries were homogenized in ice-cold Tris-EDTA buffer (in mM: Tris-50, EDTA-1.0, pH=7.4). The homogenates were centrifuged at 500×g for 10 min at 4°C and the protein concentration in the supernatant fraction was measured.

Rat kidney microsomal fractions were used as controls for the α₁ and brain microsomal fractions for the α₂ and α₃ isoforms (Sahin-Erdemli et al., 1994). Tissues were homogenized in ice-cold Sucrose-Tris-EDTA buffer (in mM: Tris-50, Sucrose-250, EDTA-1.0, pH=7.4). Initial centrifugation was at 10,000×g for 10 min at 4°C. The supernatant was centrifuged at 100,000×g for 60 min. The pellet, representing the microsomal fraction, was resuspended in Tris-EDTA buffer and the protein concentration was measured.

The samples (30 μg protein for thoracic aorta, 50 μg protein for superior mesenteric and tail artery per lane), as well as the corresponding controls (7 – 9 μg protein each for kidney and brain homogenates per lane) and prestained molecular SDS – PAGE standards (Bio-Rad, Laboratories, Hercules, CA, U.S.A.) were electrophoretically separated on a 7.5% SDS – PAGE and then transferred to polyvinyl difluoride membranes overnight at 4°C, using a Mini Trans-Blot Transfer Cell system (Bio-Rad) containing (in mM): Tris-25, glycine-250, methanol-20% and SDS-0.05%. Then, the membrane was blocked for 60 min at room temperature in Tris-buffered solution (in mM: Tris-25, NaCl-137, Tween 20-0.2%, pH 7.5) with 5% powdered non-fat milk. Next, the membrane was incubated for 90 min at room temperature with anti-α₁ rabbit polyclonal IgG (0.1 μg ml⁻¹ dilution), anti-α₂ rabbit polyclonal antiserum (1 : 5000 dilution) or anti-α₃ rabbit polyclonal antiserum (0.5 μg ml⁻¹ dilution), all purchased from Upstate Biotechnology (Lake Placid, NY, U.S.A.). After washing, the membrane was incubated for 90 min with an anti-rabbit IgG antibody conjugated to horseradish peroxidase (1 : 3000 dilution) (Transduction Laboratories, Lexington, U.K.). The membrane was thoroughly washed and the immunocomplexes were detected using an enhanced horseradish peroxidase/luminol chemiluminiscence system (ECL Plus, Amersham International plc, Little Chalfont, U.K.) and then subjected to autoradiography (Hyperfilm ECL, Amersham International plc) for either 1 – 2 min (α₁) or 5 – 10 min (α₂ and α₃). Signals on the immunoblot were quantified with a NIH Image V1.56 computer program.

Data analysis and statistics

Vasoconstrictor responses induced by phenylephrine were expressed as a percentage of the tone generated by 75 mM K⁺. Relaxation induced by acetylcholine or K⁺ was expressed as a percentage of the tone previously obtained with phenylephrine. The maximum response (Emax values) and the negative logarithm of concentrations of phenylephrine or acetylcholine producing 50% of maximum response (pD₂ values), were calculated by a non-linear regression analysis of each individual concentration-response curve using GraphPad Prism Software (San Diego, CA, U.S.A.).

In order to compare the effect of endothelium removal or in vitro ouabain on the responses to phenylephrine or KCl in segments from control or ouabain-treated rats, some results were expressed as ‘differences of area under the concentration-response curves' (dAUC) in control and experimental situations. AUC were calculated from the individual concentration-response curve plot using a computer program (WinNonlin v2.0, Pharsight Corp, Cary, NC, U.S.A.); the differences were expressed as a percentage of AUC of the corresponding control situation.

To compare the results for isoforms of the α-subunit of Na⁺,K⁺-ATPase protein expression within the same experiment and with others, we assigned a value of 1 to either α₁ or α₂ expression in arteries from control rats and used that value to calculate the relative density of other bands from the same gel. When loading, special care was taken to be sure of the amount of sample loaded. Preliminary experiments showed that increasing the loaded protein concentrations gave proportional actin and α₁ signals.

Results are expressed as mean±s.e.mean of the number of rats indicated in each case. Data were analysed using the Student's t-test for unpaired experiments or by one- or two-way analysis of variance (ANOVA) to compare groups. When ANOVA showed a significant treatment effect, Tukey's post hoc test was used to compare the different treatment groups. A probability value of less than 5% was considered significant.

Drugs and solutions

KHS contained (mM): NaCl 115, NaHCO₃ 25, KCl 4.7, MgSO₄. 7H₂O 1.2, CaCl₂ 2.5, KH₂PO₄ 1.2, glucose 11.1 and Na₂EDTA 0.01. The K⁺ free KHS was prepared by omitting KCl and replacing KH₂PO₄ with NaH₂PO₄. Drugs used were: acetylcholine hydrochloride, ouabain octahydrate and phenylephrine hydrochloride (Sigma Chemical Co., St. Louis, MO, U.S.A.); Tween 20, Tris, SDS and acrylamide (BioRad, Laboratories, Hercules, CA, U.S.A.) and methanol and sucrose (Merck KGaA, Darmstadt, Germany). Drug solutions were made in bidistilled water. Stock solutions were kept at −20°C and appropriate dilutions were made on the day of the experiment.

Results

Systolic blood pressure

Figure 1 shows the evolution of rat systolic blood pressures (SBP) during ouabain (25 μg day⁻¹) or vehicle treatment. A progressive increase of the SBP was observed in the ouabain-treated group from the first week. In the control group (rats treated with vehicle), there was a small increase in SBP after the third week; this rise in SBP was the same that is normally observed in growing rats. At the end of the study, the ouabain-treated rats showed significant hypertension as compared to the control group (control, 127±1, n=24; ouabain 160±2 mmHg, n=35, P<0.001).

Figure 1.

Systolic blood pressure for Wistar rats which received ouabain (n=35) or vehicle (n=24) subcutaneously for 5 weeks. Results are expressed as mean±s.e.mean for the number of animals used. ^*P<0.01 vs control rats; +P<0.05 vs 0 week.

No differences in body weight gain were observed (161±8 and 167±4 g, P>0.05, in control and ouabain-treated rats, respectively).

Vascular responses to phenylephrine and acetylcholine

To provide maximal activation for each preparation rings were contracted with 75 mM KCl. This treatment induced similar contractions in vessels from control and ouabain-treated rats (thoracic aorta: control, 22.7±1.0, n=12 vs ouabain, 26.5±1.3 mN, n=9, P>0.05; superior mesenteric artery: control, 12.5±1.3, n=12, vs ouabain, 13.4±0.8 mN, n=9, P>0.05; tail artery: control, 18.6±0.7, n=13, vs ouabain, 19.6±0.7 mN, n=9, P>0.05).

Treatment with ouabain for 5 weeks reduced the vasoconstrictor responses induced by phenylephrine (0.1 nM – 0.3 mM) in endothelium-intact aorta and mesenteric arteries but did not change this response in tail arteries (Figure 2). In aorta, ouabain treatment reduced both Emax and pD₂ whereas only Emax was reduced in mesenteric arteries (Table 1). In addition, AUC to phenylephrine was reduced in aorta (210.9±17.1, n=12, vs 109.4±17.6 arbitrary units, n=9, for control and ouabain-treated rats, respectively, P<0.001) and mesenteric arteries (158.4±23.7, n=12, vs 88.2±11.3 arbitrary units, n=9, for control and ouabain-treated rats, respectively, P<0.05) but not in tail arteries (253.1±9.0, n=12, vs 237.5±4.5 arbitrary units, n=9, for control and ouabain-treated rats, respectively, P>0.05).

Figure 2.

Concentration-response curve to phenylephrine in intact segments from thoracic aorta, superior mesenteric and tail arteries of Wistar rats which received ouabain (n=9) or vehicle (n=12) subcutaneously for 5 weeks. Results (mean±s.e.mean) are expressed as a percentage of the response to 75 mM KCl in each case. ^*P<0.01 vs control rats.

Table 1.

Maximum response (Emax) and sensitivity (pD₂) to phenylephrine (Phenyl) and acetylcholine (ACh) of intact (E+) or denuded (E−) segments from rats subcutaneously receiving ouabain (25 μg day⁻¹) or vehicle for 5 weeks

The contractions induced by 75 mM KCl were unaltered by endothelium removal (results not shown) in both control (n=12, 8, 11 for aorta, mesenteric and tail arteries, respectively) and ouabain-treated (n=15, 9, 14 for aorta, mesenteric and tail arteries, respectively) rats; however, the vasoconstrictor response to phenylephrine was potentiated in all studied arteries, in both control and ouabain-treated rats (Figure 3). In aorta and mesenteric arteries from both groups endothelium removal increased both Emax and pD₂ to phenylephrine, while only pD₂ was increased in the tail artery from control and ouabain-treated rats (Table 1). The comparison of dAUC values indicates that after endothelium removal the potentiation of the phenylephrine-induced contraction in arteries from ouabain-treated hypertensive rats was greater than that observed in arteries from control normotensive rats (Figure 3).

Figure 3.

Concentration-response curve to phenylephrine in intact (E+) and endothelium denuded (E−) segments from thoracic aorta, superior mesenteric and tail arteries of Wistar rats which received ouabain (n=6 – 9) or vehicle (n=6 – 12) subcutaneously for 5 weeks. Results (mean±s.e.mean) are expressed as a percentage of the response to 75 mM KCl in each case. ^*P<0.01 vs E+. Inserts show differences in area under the concentration-response curve (dAUC) to phenylephrine in endothelium-denuded and intact segments from control (C) and ouabain-treated (Oua) rats; dAUC are expressed as a percentage of the corresponding AUC for intact segments.+P<0.01 vs control.

The endothelium-dependent relaxation induced by acetylcholine (0.01 nM – 10 μM) was unaltered by ouabain treatment in the three arteries studied (Table 1).

Na⁺,K⁺-ATPase activity

In intact vessels previously contracted by phenylephrine after exposure to K⁺-free medium, the increase of extracellular K⁺ concentration (1 – 10 mM) induced a vasodilation that was similar in arteries from control and ouabain-treated rats (Figure 4 and AUC values: aorta, 781.3±19.4, n=6, vs 790.6±15.4, n=8, P>0.05; mesenteric, 875.0±6.2, n=6, vs 862.0±4.2, n=7, P>0.05; tail artery, 713.5±47.4, n=6, vs 713.3±56.3 arbitrary units, n=8, P>0.05, for control and ouabain-treated rats, respectively). After endothelium removal, the relaxation induced by increasing concentrations of K⁺ after K⁺-free medium exposure was reduced in the three vessels studied from both control and ouabain-treated rats (Figure 4). In these conditions, the K-induced relaxation was smaller in segments from ouabain-treated rats (Figure 4).

Figure 4.

Concentration-response curve to potassium in intact (E+) and endothelium denuded (E−) segments from thoracic aorta, superior mesenteric and tail arteries of Wistar rats which received ouabain (n=7 – 15) or vehicle (n=6 – 12) subcutaneously for 5 weeks, previously incubated in a K⁺-free medium and contracted with phenylephrine. Results (mean±s.e.mean) are expressed as a percentage of the response to phenylephrine in each case. Inserts show differences in area under the concentration response curve (dAUC) to potassium in endothelium-denuded and intact segments from control (C) and ouabain-treated (Oua) rats; dAUC are expressed as a percentage of the corresponding AUC for intact segments ^*P<0.05 vs E+; +P<0.01 vs control rats.

Preincubation of intact segments with ouabain (0.1 mM) for 30 min in K⁺-free medium induced a slight increase of vascular tone that was similar (P>0.05) in arteries from control and ouabain-treated rats (aorta: 6.2±2.8, n=11, vs 9.9±3.6, n=14; mesenteric: 23.3±3.6, n=6, vs 22.4±7.5, n=7; caudal: 7.3±3.4, n=12, vs 7.3±3.8% of 75 mM KCl response, n=13) and reduced the K⁺-induced vasodilation (Figure 5). However, long-term ouabain treatment had different actions on the inhibitory effect of in vitro ouabain on the K⁺-induced relaxation in the three isolated vascular preparations. In mesenteric arteries, this reduction was similar in control and ouabain-treated rats; in contrast, in aorta and tail arteries from ouabain-hypertensive rats this reduction was, respectively, greater and smaller than in those from control rats (Figure 5).

Figure 5.

Effect of 0.1 mM ouabain on potassium-induced relaxations in intact segments from thoracic aorta, superior mesenteric and tail arteries of Wistar rats which received ouabain (n=7 – 14) or vehicle (n=6 – 12) subcutaneously for 5 weeks, previously incubated in a K⁺-free medium and contracted with phenylephrine. Results (mean±s.e.mean) are expressed as a percentage of the response to phenylephrine in each case. Inserts show differences in area under the concentration-response curve (dAUC) to potassium in segments in the absence and in the presence of in vitro ouabain, from control (C) and ouabain-treated (Oua) rats; dAUC are expressed as a percentage of the corresponding AUC in the absence of ouabain. ^*P<0.05 vs control rats; +P<0.01 vs response in the absence of ouabain.

Western blot analysis of Na⁺,K⁺-ATPase α₁ and α₂ isoform expression

Figure 6 shows the α₁ and α₂ isoforms detected in thoracic aorta, mesenteric and tail arteries. Ouabain treatment increased α₁ and α₂ expression in aorta, reduced it in tail arteries and did not change it in mesenteric arteries (Figure 6). The α₃ isoform was undetected in the three vessels studied, although it was detected in brain microsomal fractions (Figure 6).

Figure 6.

Upper panel. Representative Western blot for α₁, α₂ and α₃ isoforms protein expression in intact segments from thoracic aorta, superior mesenteric and tail artery of Wistar rats which received ouabain (+) or vehicle (−) subcutaneously for 5 weeks. The first lane shows the corresponding positive control for each protein (kidney for α₁ and brain for α₂ and α₃). Lower panel. Densitometric analysis of the Western blot of α₁ and α₂ isoform protein expression in intact segments from thoracic aorta, superior mesenteric and tail arteries of Wistar rats subcutaneously receiving ouabain (n=7 – 8) or vehicle (n=5 – 6) for 5 weeks. Results (mean±s.e.mean) are expressed as the ratio of the signal of the α₁ or α₂ protein of the corresponding segments from ouabain-treated versus that of control rats. ^*P<0.05 vs Control.

Discussion

The results obtained in the present study show that chronic treatment with ouabain induces the development of time-dependent hypertension. This hypertension is associated with a decrease in the phenylephrine-induced contractions, probably due to an increase of the endothelial modulation of this contraction. Moreover, this hypertension is associated with alterations in the ouabain-sensitive Na⁺,K⁺-ATPase activity and protein expression of α-isoforms. Regional differences on the vasopressor responses to phenylephrine and in the ouabain-sensitive Na⁺,K⁺-ATPase activity and protein expression resulting from ouabain treatment are found.

Ouabain is an endogenous digitalis compound found in the plasma of several mammals (Overbeck et al., 1976; Hamlyn et al., 1996); endogenous levels have been reported to be increased in some pathological conditions such as hypertension (Hamlyn et al., 1982; 1996; Gottlieb et al., 1992). Central mechanisms (Huang et al., 1994; Veerasingham & Leenen, 1999; Huang & Leenen, 1999) and peripheral mechanisms, such as the increase of myocardial contraction and vascular smooth muscle tone (Marín et al., 1988; Blaustein, 1993) as well as increased vascular reactivity to pressor agents (Vassallo et al., 1997; Rossoni et al., 1999; Davel et al., 2000), have all been proposed as mechanisms for ouabain induction of hypertension.

Contributing to the hypothesis that ouabain could be associated with the genesis of hypertension, our results show that peripheral administration of ouabain for 5 weeks induced hypertension in Wistar rats, which was evident from the first week of treatment. There are several reports showing that chronic treatment with ouabain induces hypertension in rats (Yuan et al., 1993; Manunta et al., 1994; 2000; Huang et al., 1994; Veerasingham & Leenen, 1999; Huang & Leenen, 1999; Kimura et al., 2000), although other studies found a lack of a hypertensive effect after ouabain administration in rats (Li et al., 1995; Cargnelli et al., 2000) and sheep (Pidgeon et al., 1996). These contradictory results could be due to differences in the duration, dosage, route of ouabain administration or differences in rat strains, or age at the beginning of the treatment.

To determine the possible participation of the vascular component in the ouabain-induced hypertension, we investigated the contraction induced by phenylephrine in rings from thoracic aorta, superior mesenteric and tail artery. There was a gradient in the effect of ouabain on the phenylephrine response in rings from the three vessels studied. Thus, in aorta, both Emax and sensitivity were reduced, in mesenteric artery only Emax was reduced, while in the tail artery no differences in either parameter were found. Nevertheless, the contractile response to high potassium remained unmodified after ouabain treatment in the three studied vessels; this suggests that the treatment does not impair the ability of smooth muscle cells to contract. The results from aortic rings agree with those of Cargnelli et al. (2000), despite the absence of any increase in systolic blood pressure. However, another recent study has proposed that chronic treatment with ouabain induces hypertension and increases the contraction evoked by potassium and phenylephrine in renal arteries (Kimura et al., 2000).

It is usually assumed in different animal models of hypertension as well as in essential human hypertension, there is some impairment of endothelium-dependent vasodilator responses, although enhancement or no modification have also been described (Marín & Rodríguez-Martínez, 1997). After ouabain treatment, we observed no change of the endothelium-dependent relaxation to acetylcholine in any of the vessels studied, as was also reported for renal arteries (Kimura et al., 2000). Thus, our results cannot support an association between these effects and the observed increase in systolic blood pressure, although they do suggest a compensatory mechanism.

One hypothesis to explain the reduced response to phenylephrine in arteries from ouabain-treated rats is that this treatment increases the production of endothelium-derived relaxing factor(s). It is known that negative endothelial modulation of vasoconstrictor responses, including α-adrenergic responses (Carrier & White, 1985), is increased in spontaneously hypertensive rats (SHR) (Arribas et al., 1994), although a decrease of negative endothelial modulation in SHR has also been described (Dohi et al., 1996). On the other hand, Rossoni et al. (1999) have recently suggested that nanomolar concentrations of ouabain induce the release of an endothelium-derived relaxing factor that seems to open potassium channels. Other reports have also suggested that ouabain induces the release of nitric oxide in cultured endothelial cells (Xie et al., 1993) or that this compound stimulates inducible NO synthase expression in vascular smooth muscle cells (Pacheco et al., 2000). In endothelium-denuded segments from aorta and superior mesenteric artery there was an increase in the vasoconstrictor response to phenylephrine, and this increase was stronger in segments from ouabain-treated rats. These results contrast with those of Cargnelli et al. (2000) who found that endothelium removal reduced the contractile response to phenylephrine in aorta from ouabain-treated rats. Our results suggest that after ouabain treatment there is an increase in negative endothelial modulation of phenylephrine-induced responses. This phenomenon could explain the hyporesponsiveness to phenylephrine-induced contraction that we observed. Nevertheless, additional experiments are necessary to identify the endothelial factor(s) that might be increased after ouabain treatment.

The increase of the phenylephrine response after endothelial denudation was greater in tail artery segments from ouabain-treated rats, but still much lower than that found in aorta and mesenteric artery. If ouabain induces the release of an endothelial vasodilator factor and the endothelium modulates vasoconstrictor responses to a lesser extent in the tail artery than in other vascular beds, it is likely that the response to α-adrenergic stimulation is less affected by ouabain treatment. Interestingly, the magnitude of the increase of the phenylephrine-induced contraction after endothelium removal seems to be related to the ability of acetylcholine to induce relaxation, and this may reflect the ability of endothelium to release vasoactive factors. However, as mentioned above, the endothelium-dependent relaxation to acetylcholine was not modified after ouabain treatment. Thus, although ouabain increases endothelial vasodilator factors release in response to phenylephrine, this increase does not occur in the response to acetylcholine, suggesting differences in the mechanisms of endothelial factors release.

As the effects of ouabain are usually explained by its ability to inhibit Na⁺,K⁺-ATPase, the activity of this pump was measured by a usual procedure described by Webb & Bohr (1978). The addition of K⁺ elicited concentration-dependent relaxations that almost fully reversed the tone induced by phenylephrine in a K⁺-free medium in the three studied vessels. It has been proposed that endothelial factors could stimulate the Na pump and partially protect it against inhibition by ouabain (see Marín & Redondo, 1999). Our results confirm the existence of a positive endothelial modulation of the Na pump in the three vessels studied. This endothelial modulation was slightly stronger in segments from ouabain-treated rats, supporting the observed higher endothelial modulation found in these arteries. The inhibitory effect of a high concentration of ouabain on the relaxation induced by K⁺ was different in intact segments from ouabain-treated and control rats. Ouabain reduced the K⁺-elicited relaxation more in thoracic aorta from ouabain-treated rats than in those from control rats. In superior mesenteric arteries, the effect of ouabain was similar, but in the tail artery, it was smaller in ouabain-treated rats than in controls. These results suggest that ouabain-sensitive Na⁺,K⁺-ATPase activity was increased in aorta, unaltered in mesenteric rings and reduced in tail artery from the ouabain-induced hypertensive rats. Na⁺ pump activity has been reported to be reduced in vessels from SHR (Ponte et al., 1996) and renal hypertensive dogs (Overbeck et al., 1976), although increased activity has also been reported in SHR (Cox, 1986). Redondo et al. (1995) found that the Na⁺ pump activity in aortic cultures of smooth muscle from SHR and Wistar Kyoto rats was similar, with no differences in either the number of Na⁺ pump sites and/or activity, although the regulation of its activity seems to be significantly different.

The Na⁺ pump is composed of two major subunits, α and β, the α subunit being the catalytic component (Blanco & Mercer, 1998; Marín & Redondo, 1999). A third subunit, named γ, has been also reported, but its role, if any, is not well defined (see Blanco & Mercer, 1998; Marín & Redondo, 1999). Four different α isoforms, α₁, α₂, α₃, α₄, of Na⁺,K⁺-ATPase, with different affinities for digitalics, and expressed in species- and tissue-specific manners, have been identified (Sweadner, 1989; Blanco & Mercer, 1998). Rat vascular tissues have been reported to only express the α₁, α₂ and α₃ isoforms (Sahin-Erdemli et al., 1994). Our results showed that, in rat aorta, superior mesenteric and tail arteries only the expression of the α₁ and α₂ isoforms could be detected. Small amounts of the α₃ isoform in vascular smooth muscle and a very low sensitivity of the antibody used could result in cross-reaction with the α₂ isoform, which would explain the lack of α₃ expression detection. Chronic ouabain administration changed the protein expression of the α-subunit isoforms of Na⁺,K⁺-ATPase, depending on the vessel studied in a way similar to the changes observed in ouabain-sensitive Na⁺,K⁺-ATPase activity. Thus, in the thoracic aorta from ouabain-treated rats both α₁ and α₂ expressions were increased; in superior mesenteric arteries, no modification of either isoform was found, while in the tail artery from ouabain-treated rats, the α₁ and α₂ protein expression was reduced. In agreement, a recent report suggests that chronic treatment with either ouabain or digoxin induces tissue specific differential regulation of the Na⁺ pump α subunit mRNA (Wang et al., 2000). In addition, earlier reports had already suggested alterations of the expression of α isoforms in various pathologies, including hypertension (Herrera et al., 1988; Sahin-Erdemli et al., 1995). An upregulation of the α₁ and α₂-subunit isoforms of Na⁺,K⁺-ATPase has been reported after 1 to 4 days of ouabain treatment in a culture of aortic smooth muscle cells (Liu & Songu-Mize, 1998). An upregulation was also obtained in cultured aortic smooth muscle cells by mechanical strain (Songu-Mize et al., 1996), which is another procedure that increases intracellular Na⁺, and these increases stimulate the transcription of α₁ and α₂ subunits of the sodium pump in smooth muscle cell cultures (Yamamoto et al., 1994).

We do not know where the changes in Na⁺ pump are localized in the vascular wall. However, given the higher amount of smooth muscle cells when compared with the endothelium, it seems likely that the altered protein expression could occur in the muscular layer. On the other hand, it is difficult to explain the different effects of ouabain treatment in the three vessels studied. Some investigators have suggested that the activity of Na⁺ pump in thoracic aorta, mesenteric and tail artery is inversely correlated with the catecholamine content (Myers et al., 1987). It is possible that differences in adrenergic innervation as well as structural differences between blood vessels could be related to the different changes observed in Na pump activity and expression in this hypertensive model.

The regional alterations of the ouabain-sensitive Na⁺,K⁺-ATPase activity and protein expression might help explain the differences in the phenylephrine-induced contraction observed in the vessels from ouabain-induced hypertensive rats. The increase in the ouabain sensitive Na⁺,K⁺-ATPase activity that would promote hyperpolarization, together with the increase of endothelial modulation, would also result in the reduction of both maximal response and sensitivity to phenylephrine observed in aorta from ouabain-treated rats. In the tail artery, the activity of ouabain-sensitive Na⁺,K⁺-ATPase was reduced, leading to higher intracellular Na⁺ and Ca²⁺ concentrations, thus increasing contractile activity. However, the enhanced endothelial function observed in this vessel counteracted the contractile activity resulting in a similar phenylephrine-induced contraction in control and ouabain-treated rats. In mesenteric arteries, where endothelial modulation is increased but changes in Na⁺ pump activity are not observed, only the maximal response was reduced by ouabain treatment.

In conclusion, the results described here indicate that chronic administration of ouabain induces hypertension. This finding supports the hypothesis that the endogenous Na⁺ pump inhibitor, characterized as ouabain or an isomer (Mathews et al., 1991), could be related to the genesis and/or maintenance of hypertension. Although there are proposals suggesting the involvement of central mechanisms in this hypertension model (Huang et al., 1994; Veerasingham & Leenen, 1999; Huang & Leenen, 1999), vascular mechanisms with regional alterations of vasopressor responses to phenylephrine as well as of the ouabain sensitive Na⁺ pump activity and protein expression are also present.

Abbreviations

dAUC

difference of area under the concentration-response curves

EDHF

endothelium-derived hyperpolarizing factor

EDTA

ethylenediamine-tetraacetic acid

Emax

maximum response

KHS

Krebs Henseleit solution

NO

nitric oxide

PAGE

polyacrylamide gel

pD₂

negative logarithm of concentrations producing 50% of maximum response

SBP

systolic blood pressure

SDS

sodium lauryl sulphate

SHR

spontaneously hypertensive rats